case study on greenhouse effect

• school Campus Bookshelves
• menu_book Bookshelves
• perm_media Learning Objects
• login Login
• how_to_reg Request Instructor Account
• hub Instructor Commons

Margin Size

• Download Page (PDF)
• Download Full Book (PDF)
• Periodic Table
• Physics Constants
• Scientific Calculator
• Reference & Cite
• Tools expand_more
• Readability

selected template will load here

This action is not available.

8.3: Case Study- Greenhouse Gases and Climate Change

• Last updated
• Save as PDF
• Page ID 12061

• Heriberto Cabezas
• Georgia College and State University via GALILEO Open Learning Materials

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)

\( \newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\)

( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\)

\( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

\( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\)

\( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)

\( \newcommand{\Span}{\mathrm{span}}\)

\( \newcommand{\id}{\mathrm{id}}\)

\( \newcommand{\kernel}{\mathrm{null}\,}\)

\( \newcommand{\range}{\mathrm{range}\,}\)

\( \newcommand{\RealPart}{\mathrm{Re}}\)

\( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

\( \newcommand{\Argument}{\mathrm{Arg}}\)

\( \newcommand{\norm}[1]{\| #1 \|}\)

\( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\AA}{\unicode[.8,0]{x212B}}\)

\( \newcommand{\vectorA}[1]{\vec{#1}}      % arrow\)

\( \newcommand{\vectorAt}[1]{\vec{\text{#1}}}      % arrow\)

\( \newcommand{\vectorB}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vectorC}[1]{\textbf{#1}} \)

\( \newcommand{\vectorD}[1]{\overrightarrow{#1}} \)

\( \newcommand{\vectorDt}[1]{\overrightarrow{\text{#1}}} \)

\( \newcommand{\vectE}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash{\mathbf {#1}}}} \)

Introduction

If increased greenhouse gas emissions from human activity are causing climate change, then how do we reduce those emissions? Whether dictated by an international, national, or local regulation or a voluntary agreement, plans are needed to move to a low-carbon economy. In the absence of federal regulation, cities, states, government institutions, and colleges and universities, have all taken climate action initiatives. This case study provides two examples of climate action plans – one for a city (Chicago) and one for an institution (the University of Illinois at Chicago).

Chicago’s Climate Action Plan

Urban areas produce a lot of waste. In fact, 75 percent of all greenhouse gas emissions are generated in urban areas. Therefore, it is important for cities to develop plans to address environmental issues. The Chicago Climate Action Plan (Chicago CAP) is one such example. The mid-term goal of this plan is a 25 percent reduction in greenhouse gas emissions by 2020 and final goal is 80 percent reduction below 1990 GHG levels by the year 2050.

The Chicago CAP outlines several benefits of a climate action plan. The first would obviously be the reduction of the effects of climate change. Under a higher emissions scenario as per the Intergovernmental Panel on Climate Change (IPCC), it is predicted that the number of 100 degree Fahrenheit days per year would increase to 31, under the lower emissions scenario it would only be eight. Established by the United Nations Environment Programme (UNEP), the IPCC is the leading international body that assesses climate change through the contributions of thousands of scientists.

Second, there is an economic benefit derived from increased efficiencies that reduce energy and water consumption. Third, local governments and agencies have great influence over their city’s greenhouse gas emissions and can enhance energy efficiency of buildings through codes and ordinances so they play a key role in climate action at all governmental levels. Finally, reducing our dependence on fossil fuels helps the United States achieve energy independence.

Designing a Climate Action Plan

A good climate action plan includes reporting of greenhouse gas emissions, as far back as there is data, preferably to 1990. Figure \(\PageIndex{1}\) depicts the emissions calculated for Chicago through 2005. From that point there is an estimate (the dotted line) of a further increase before the reductions become evident and the goals portrayed can be obtained. The plan was released in September 2008 and provides a roadmap of five strategies with 35 actions to reduce greenhouse gas emissions (GHG) and adapt to climate change. The strategies are shown in Table \(\PageIndex{1}\). Figure \(\PageIndex{2}\) identifies the proportion of emissions reductions from the various strategies.

In 2010 CCAP put out a progress report wherein progress is measured by the many small steps that are being taken to implement the plan. It is not translated exactly to emissions reductions but reports on progress for each step such as the number of residential units that have been retrofitted for energy efficiency, the number of appliances traded in, the increase in the number of rides on public transit, and the amount of water conserved daily.

University Climate Action Plan

Several factors caused a major Chicago university to develop a climate action plan. As part of the American College and University Presidents’ Climate Commitment (ACUPCC), nearly 670 presidents have signed a commitment to inventory their greenhouse gases, publicly report it, and to develop a climate action plan. Part of the Chicago CAP is to engage businesses and organizations within the city in climate action planning. In order to be a better steward of the environment, the University of Illinois at Chicago (UIC) developed a climate action plan . The goals are similar to Chicago’s: a 40 percent GHG emissions reduction by 2030 and at least 80 percent by 2050, using a 2004 baseline. The strategies align with those of the city in which the campus resides (see Table \(\PageIndex{1}\)). UIC’s greenhouse gas reports are also made publically available on the ACUPCC reporting site . Figure \(\PageIndex{3}\) displays UIC’s calculated emissions inventory (in red) and then the predicted increases for growth if activities continue in a “business as usual (BAU)” approach. The triangular wedges below represent emissions reductions through a variety of strategies, similar to those of the wedge approach that Professors Sokolow and Pacala proposed. Those strategies are displayed in Table \(\PageIndex{1}\), alongside Chicago’s for comparative purposes.

Table \(\PageIndex{1}\) Alignment of the Chicago and UIC Climate Action Plans Source: C. Klein-Banai using data from Chicago Climate Action Plant and UIC Climate Action Plan

There is no one approach that will effectively reduce greenhouse gas emissions. Climate action plans are helpful tools to represent strategies to reduce emissions. Governmental entities such as nations, states, and cities can develop plans, as can institutions and businesses. It is important that there be an alignment of plans when they intersect, such as a city and a university that resides within it.

Study Shows Economic Impacts of Greenhouse Gas Emissions

Climate-warming activity from five countries caused $6 trillion in global losses.

Finding Predictability in the Uncertainty of Snow Drought

A sound scientific basis exists for climate liability claims between individual countries, according to a Dartmouth study released today.

The study is the first to assess the economic impacts that individual countries have caused to other countries through their contributions to global warming. The research draws direct connections between cumulative emissions per nation of heat-trapping gases to losses and gains in gross domestic product in 143 countries for which data are available.

The study, published in the journal  Climatic Change , provides an essential basis for nations to make legal claims for economic losses tied to emissions and warming, according to the researchers.

“Greenhouse gases emitted in one country cause warming in another, and that warming can depress economic growth,” says  Justin Mankin , an assistant professor of geography and senior researcher of the study. “This research provides legally valuable estimates of the financial damages individual nations have suffered due to other countries’ climate-changing activities.”

Among the data, the research found that five national emitters of greenhouse gases caused $6 trillion in global economic losses through warming from 1990 to 2014.

According to the study, emissions from the U.S. and China, the world’s two leading emitters, are responsible for global income losses of over $1.8 trillion each in the 25-year period from 1990. Economic losses caused by Russia, India, and Brazil individually exceed $500 billion each for the same years. The $6 trillion in cumulative losses attributable to the five countries equals about 11% of annual global GDP within the study period.

“This research provides an answer to the question of whether there is a scientific basis for climate liability claims—the answer is yes,” says  Christopher Callahan , Guarini ’23, first author of the study. “We have quantified each nation’s culpability for historical temperature-driven income changes in every other country.”

This research provides an answer to the question of whether there is a scientific basis for climate liability claims—the answer is yes.

Warmer temperatures can cause economic losses for a country through many pathways, such as lowering agricultural yields, reducing labor productivity, and decreasing industrial output.

In addition to losses, the research also values the economic benefits derived from warming caused by country-level emissions but highlights that the large gains disproportionately benefitting some countries do not negate the losses suffered in others.

The study focuses on the economic impacts of temperature change as a consequence of emissions, not other effects of emissions such as those on air quality. Data presented in the study quantifies economic impacts based on distinct greenhouse gas emissions accounting schemes, considering those emissions that happened within a country’s territory versus the emissions embodied in international trade.

The research shows that the distribution of warming impacts from emitters is highly unequal, with the top 10 global emitters causing more than two-thirds of losses worldwide. Countries that lose income are warmer and poorer than the global average and are generally located in the tropics and the global South. Countries that gain income are cooler and wealthier than the global average and are generally located in the middle latitudes and the North.

“Irrespective of the accounting, warm counties have warmed and lost income because of it, while colder countries have warmed but enjoyed economic gains,” says Mankin. “The responsibility for the warming rests primarily with a handful of major emitters, and this warming has resulted in the enrichment of a few wealthy countries at the expense of the poorest people in the world.”

For years, researchers have worked to establish direct legal links between economic loss and emissions of greenhouse gases such as carbon dioxide, methane, and nitrous oxide. Previous studies have provided estimates on the total, global level of economic loss but could not determine the warming attributable to individual nations, undermining efforts to hold emitting countries accountable for legal damages because of the uncertainties involved.

By creating an analytical framework that links emissions from individual countries to the losses and gains in every other country, the Dartmouth research team hopes to help resolve questions of climate liability and national accountability to inform climate policy.

“For the first time, we have been able to show clear and statistically significant linkages between the emissions of specific countries and historical economic losses experienced by other countries,” says Callahan. “This is about the culpability of one country to another country, not the effect of overall global warming on a country.”

The team from the  Climate Modeling & Impacts Group  says that the study discredits the idea that climate mitigation is simply a “collective action problem,” where no one country acting alone can have an effect on the impacts of global warming.

“Until now, the complexity of the carbon cycle, natural variations in climate, and uncertainties in models have provided emitters with plausible deniability for individual damage claims. That veil of deniability has now been lifted,” says Mankin.

According to the team, identifying national culpability demonstrates that individual countries can have large, attributable impacts from warming due to their emissions; the actions of individual nations do matter; and country-level mitigation, even if pursued alone, would limit measurable harms to others.

“Nations need to work together to stop warming, but that doesn’t mean that individual countries can’t take actions that drive change,” says Callahan. “This research upends the notion that the causes and impacts of warming only occur at the global level.”

A major challenge for the research was to account for large uncertainties at each step in the causal chain from emissions to global warming, from warming to country-level temperature changes, and from country-level temperature changes to impact.

To overcome this difficulty, the research team combined historical data with climate models in an integrated framework to quantify each nation’s culpability for historical temperature-driven income changes in every other country.

The study sampled 2 million possible values for each country-to-country interaction. In total, 11 trillion values were calculated on a supercomputer operated by  Dartmouth’s Research Information, Technology and Consulting .

“This is the first research to integrate and quantify all of the uncertainties in each step of the chain between emissions and economic impact,” says Callahan. “We are not addressing the question of whether fossil fuels have been good or bad for economic growth, but how to compensate for the damage caused by the warming from those emissions.”

According to the research team, future work can use the same analytical approach to determine the contribution of specific emitters, including individual corporations, to economic loss and gain.

The research was funded by the  Wright Center for the Study of Computation and Just Communities , a research center in the  Neukom Institute for Computational Science , and the National Science Foundation.

David Hirsch can be reached at  [email protected] .

• Society & Culture
• Department of Geography
• Press Release
• Neukom Institute for Computational Science
• Wright Center for the Study of Computation and Just Communities

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 29 January 2021

Greenhouse gas emissions from the water–air interface of a grassland river: a case study of the Xilin River

• Xue Hao 1 ,
• Yu Ruihong 1 , 2 ,
• Zhang Zhuangzhuang 1 ,
• Qi Zhen 1 ,
• Lu Xixi 1 , 3 ,
• Liu Tingxi 4 &
• Gao Ruizhong 4  

Scientific Reports volume  11 , Article number:  2659 ( 2021 ) Cite this article

10k Accesses

29 Citations

16 Altmetric

Metrics details

• Carbon cycle
• Climate change

Greenhouse gas (GHG) emissions from rivers and lakes have been shown to significantly contribute to global carbon and nitrogen cycling. In spatiotemporal-variable and human-impacted rivers in the grassland region, simultaneous carbon dioxide, methane and nitrous oxide emissions and their relationships under the different land use types are poorly documented. This research estimated greenhouse gas (CO 2 , CH 4 , N 2 O) emissions in the Xilin River of Inner Mongolia of China using direct measurements from 18 field campaigns under seven land use type (such as swamp, sand land, grassland, pond, reservoir, lake, waste water) conducted in 2018. The results showed that CO 2 emissions were higher in June and August, mainly affected by pH and DO. Emissions of CH 4 and N 2 O were higher in October, which were influenced by TN and TP. According to global warming potential, CO 2 emissions accounted for 63.35% of the three GHG emissions, and CH 4 and N 2 O emissions accounted for 35.98% and 0.66% in the Xilin river, respectively. Under the influence of different degrees of human-impact, the amount of CO 2 emissions in the sand land type was very high, however, CH 4 emissions and N 2 O emissions were very high in the artificial pond and the wastewater, respectively. For natural river, the greenhouse gas emissions from the reservoir and sand land were both low. The Xilin river was observed to be a source of carbon dioxide and methane, and the lake was a sink for nitrous oxide.

Similar content being viewed by others

Andean headwater and piedmont streams are hot spots of carbon dioxide and methane emissions in the Amazon basin

Diurnal and seasonal variability of CO2 and CH4 concentration in a semi-urban environment of western India

Globally elevated greenhouse gas emissions from polluted urban rivers

Introduction.

As the concentration of greenhouse gas increases, the global warming effect becomes more pronounced 1 . Carbon dioxide (CO 2 ), methane (CH 4 ) and nitrous oxide (N 2 O) have been shown to dominate the well-mixed greenhouse gas (GHG), contributing 80% of the positive radiative forcing that drives climate change 2 , 3 . CO 2 has long been known as an important greenhouse gas, and CH 4 is also an important greenhouse gas. The global warming potential of CH 4 in 100 years is 25 times that of CO 2 , and the contribution rate to the greenhouse effect is approximately 22% 4 , 5 . The N 2 O molecule is a powerful greenhouse gas that has a global warming potential 296 times greater than that of CO 2 6 . Global warming could have a significant impact on local and regional climatic regimes, which would in turn impact hydrological and water resources systems 1 , 7 . The terrestrial ecosystem carbon cycle and its driving mechanisms are important components of current global change research. They are the key to predicting future atmospheric CO 2 changes and global warming. The carbon cycle of the terrestrial ecosystem is mainly reflected in the exchange of CO 2 on land and in lakes, rivers and the atmosphere, as well as in the direct transport of carbon to the ocean by river action 8 , 9 .

The river system connects the two carbon banks of land and sea. It is a key link in the global carbon cycle and the main channel for land-based carbonaceous materials to enter the sea 10 . The river carbon cycle refers to the entire process of carbon sources from different sources in the terrestrial system entering the river network system in a variety of forms under the influence of machinery, biochemistry and human activities. Rivers are significant source of greenhouse gas emissions. It is estimated that aquatic systems contribute more than 50% to atmospheric CH 4 , and global river N 2 O emissions have gradually exceeded 10% of human emissions 11 , 12 . The greenhouse gas emissions of urban rivers are more significant compared with natural rivers. The N 2 O, CO 2 and CH 4 escaping from rivers are mainly derived from microbial processes such as microbial degradation, acetic acid fermentation, ammonia oxidation and the denitrification of sediments 13 , 14 . As a reaction matrix, the increase in soluble inorganic nitrogen and soluble organic carbon stimulates microbial activity in the aquatic environment and promotes CO 2 , CH 4 and N 2 O production 15 , 16 .

Inland waters (streams, rivers, lakes and reservoirs) have been gradually recognized as important sources of greenhouse gas release into the atmosphere 17 . Many regional studies on inland waters have proposed a specific focus on the emissions of CO 2 , CH 4 , and N 2 O 18 . However, only a few studies have assessed the three GHG concentrations together in a river system. Most of the research on greenhouse gas emissions in grassland areas has focused on soil systems but rarely on inland river systems 19 . The transverse carbon and nitrogen cycle of an inland river is generated along with the direction of the river but disappears into the terrestrial cycle with the river. The longitudinal cycle of carbon and nitrogen is exchanged by the water–air interface. Differently from a fresh water river connecting the land and the ocean, the carbon and nitrogen cycle of an inland river does not enter the ocean system but directly enters the land system in a short time.

The Xilin River basin is located in the inland river basin of arid and semiarid steppe areas. The Xilin River is a seasonal river with a low network density, highly meandering and no obvious riverbed in the downstream, and ends with a terminal lake. The grassland regions are affected by different degrees of human activities. The greenhouse gas emissions of grassland rivers under the influence of different human activities have rarely been studied and our study intended to understand the carbon emission mechanism of rivers in grassland region and provide a reference for the greenhouse gas emissions of global grassland rivers. The specific purposes of this study were to (1) explore the spatial and temporal variations of greenhouse gas emissions at the water–air interface; (2) explore effects of land use on emissions of greenhouse gas and (3) analyze the effects of human activities on emissions of greenhouse gas.

Materials and methods

Study sites.

The Xilin River Basin is in the southeastern part of the Inner Mongolia Autonomous Region in China (E115° 00′ ~ 117° 30′ and N43° 26′ ~ 44° 39′) (Fig.  1 ). It is located at the western extension of the lower hills and hills of the greater Xingan mountains in the middle and eastern part of the Inner Mongolia plateau. In the north, it is characterized by an alternating distribution of low mountains and hills and high plains, and in the south, it is a multistage basalt platform. The middle area of these two terrains is mainly sandy dunes, and the terrain gradually declines from the east to the west 20 . The Xilin River Basin covers an area of 10,542 km 2 , and the average altitude is 988.5 m. The total length of the Xilin River is 268.1 km, with an average channel drop of 1.25%; however, it is cut off nearly 124.7 km below the Xilin Reservoir 21 . The Xilin River Basin is dominated by grasslands, followed by swamps, sand land and urban land 22 . The grassland area of the Xilin River Basin accounts for 88.35% of the total drainage area, and the water area accounts for 0.37%. The climate type of the Xilin River Basin is a temperate semiarid continental monsoon climate with climatic characteristics, for example, of less precipitation, more evaporation and greater daily temperature difference. According to the meteorological data of the Xilinhot Meteorological Station from 1968 to 2015, the annual average precipitation was 278.9 mm, the annual average evaporation was 1862.9 mm, the annual average temperature was 2.8 °C, and the annual mean wind speed was 3.4 m s −1 .

Xilin River Basin and sampling sites (generated by Arcgis10).

Sampling procedures and analysis

This study conducted four rounds of field work in April, June, August, and October in 2018. The design of the eighteen sampling points takes into account the changes of land use types. Fourteen sites on the main stem and four sites on the tributary were selected for sampling and measurement (Fig.  1 , Table 1 ). The types of land use on the tributary mainly included grassland and sand land. The upstream of the Xilin River is swamp. The Xilin River flows to the grassland section in the upper stream of the Xilin River Reservoir. The downstream of the Xilin River flows through the artificial lake in Xilinhot City.

The collected water samples were subjected to low-pressure suction filtration through Whatman GF/F filters (nominal pore diameter of 0.7 μm). The fiber filter was prefired in a muffle furnace at 450 °C. pH, water temperature (T w ), salinity (Sal), dissolved oxygen (DO), and total dissolved solids (TDS) were measured by a portable water quality analyzer. pH and T w were measured using a portable pH meter (WTW). Alkalinity (Alk) was titrated with 0.1 mol L −1 hydrochloric acid (HCl) within 10 h after sampling. HCO 3 – represents 96% of the alkalinity when the pH ranges from 7 to 10 10 . Alk was determined by on-site titration. Total nitrogen (TN) was determined by the alkaline potassium persulfate digestion-UV spectrophotometric method 23 , and total phosphorus (TP) was determined by the ammonium molybdate spectrophotometric method 24 . Flow velocity of water (V w ) was measured using a doppler portable flow meter (DPL-LS10), and the flow discharge was calculated by V w , river width and depth.

GHGs measurement

Pco 2 , pch 4 , and pn 2 o measurements.

In this study, surface water p CO 2 was calculated using the headspace equilibrium method. By using an 1100 mL conical flask, 800 mL of water was collected to the depth of 10 cm below the water surface and the remaining volume of 300 mL was filled with ambient air. The flask was immediately closed with a lid and vigorously shaken for 3 min to equilibrate the gas in the water and air. The equilibrated gas was automatically injected into the calibrated Li-7000 gas analyzer. The Li-7000 CO 2 /H 2 O analyzer was connected to a computer interface that allowed p CO 2 recording for two seconds. The measurements at each site were repeated three times and the average was calculated (analytical error below 3%). The original surface water p CO 2 was finally calculated by using solubility constants for CO 2 and the headspace ratio 10 . After shaking the conical flask, the gas extracted from flask was injected into the vacuum cylinder (Labco Exetainer). p CO 2 , p CH 4 , and p N 2 O in the water column were measured using a gas chromatograph.

p CO 2 from the water was also calculated using the CO 2 SYS program 25 , which has been widely employed for aquatic p CO 2 calculations 26 , 27 . T w , Alk and pH were essential data for such calculation 28 .

The p CO 2 calculated was slightly higher than the p CO 2 measured directly by gas chromatography (R 2  = 0.90) (Fig.  2 ). The reason for the higher calculated p CO 2 value was due to the error generated from pH and T w measurements or the artificial error that occurred during the titration. The directly measured data could be used for analysis and discussion.

Comparison of the results of river p CO 2 at all sampling sites on the Xilin River by the measured and calculated methods.

Greenhouse gas emissions calculation

In this study, F CO 2 was measured by the floating chamber method and an Li-7000 CO 2 /H 2 O analyzer (Li-Cor, USA). The Li-7000 instrument was calibrated with standard CO 2 gases of 500 ppm and 1000 ppm before each measurement. F CH 4 and F N 2 O were measured by the floating chamber method. 60 mL of gas was taken from the floating chamber every three minutes; five samplings were taken and injected into a vacuum cylinder.

The static chamber volume was 17.8 L, and the covered water area was 0.09 m 2 . The chamber was covered with tinfoil to reduce the influence of sunlight. The temperature inside the chamber was measured with a thermometer. At the beginning of each experiment, the chamber was placed in the air near the monitoring point. The instrument automatically recorded the air CO 2 concentration and ambient atmospheric pressure. When the chamber was placed on the water surface, the analyzer recorded the CO 2 concentration every two seconds, and each measurement lasted for 6–10 min.

The greenhouse gas emissions from water were calculated using the following equation 29 :

where d p GHG/dt is the slope of greenhouse gas change within the chamber (Pa d –1 ; converted from μatm min –1 ), V is the chamber volume (17.8 L), R is the gas constant, T is chamber temperature (K), and S is the area of the chamber covering the water surface (0.09 m 2 ).

Physical and chemical parameters variation of the Xilin River

During the sampling campaigns, pH ranged from 6.90 to 9.10, and the seasonal variation of pH was not obvious (Fig.  3 a). The average annual pH was 8.20 but the spatial variation was significant. In the sand land area, the pH value was the lowest (7.12 ± 0.13). The pH value from upstream to downstream showed an overall increase trend. The concentration of DO ranged from 2.23 to 16.69 mg/L, and the average concentration of DO was 8.97 mg/L (Fig.  3 b). The DO concentrations of the seasonal and spatial variables showed significant differences. In the waste water, the DO value was the highest in October and the lowest in June. The DO value in swamp and pond land use types were lower than in the other areas. T w varied from 0.30 to 31.90 °C at all sampling sites, the annual mean value was 15.40 °C, and the seasonal variation of T w was significant ( P  < 0.05); however, there was no significant difference in spatial distribution ( P  > 0.05) (Fig.  3 c).

Box plot showing seasonal and spatial variation under different land use variations in six parameters of water quality(pH ( a ), DO ( b ), T w ( c ), Alk ( d ), TDS ( e ), Sal ( f ), TN ( g ), TP ( h ), V w ( i ), Q ( j )).

The Alk concentration ranged from 1.15 to 14.01 mg/L, and the average Alk concentration was 4.93 mg/L (Fig.  3 d). There was no significant seasonal difference in Alk. The average value of four months was 4.31 mg/L in April, 6.04 mg/L in June, 4.91 mg/L in August, and 5.83 mg/L in October. Alk had a significant spatial change, and the Alk gradually increased from the upper reaches to the downstream. The variable range of TDS concentration was 147.00–2580.00 mg/L, and the average value was 782.00 mg/L (Fig.  3 e). The seasonal variation was not significant ( P  > 0.05), but the spatial change was significant, with a significant increase trend from upstream to downstream ( P  < 0.05). Sal ranged from 0.00 to 1.30, with an average of 0.31 (Fig.  3 f). From the upper reaches to the lower reaches, there was a significant increase.

TN ranged from 1.81 to 57.70 mg/L, with an average value of 10.86 mg/L (Fig.  3 g). Similarly, there was a significant increase from the upper reaches to the lower reaches. TP ranged from 0.00 to 2.52 mg/L, and the average value was 0.45 mg/L (Fig.  3 h). In the waste water, the TP concentration was higher than others. The variable range of V w was 0.00–1.20 m/s (Fig.  3 i). In the upstream, the V w value was higher than that in the downstream. The Q value ranged from 0.00 to 0.80 m 3 /s (Fig.  3 j). In the grassland, the Q was higher than that in other land use types.

Seasonal variation of greenhouse gas emissions

The range of p CO 2 varied from 442.54 to 13,056.85 ppm, with a four-month average of 2230.65 ppm, which was almost five times that in air (the average in the air was 402.00 ppm) (Fig.  4 ). To better display the spatial variation of p CO 2 , Fig.  4 shows the variation of p CO 2 along the river. The highest value of p CO 2 appeared in sand land (11,937.33 ± 1,017.37 ppm), followed by swamp (5,089.54 ± 2,397.81 ppm), wastewater (2,048.93 ± 660.43 ppm), and pond (1,486.46 ± 673.71 ppm) land uses. p CO 2 in grassland, reservoir, and lake types were normally below 1000 ppm. Under different land use types, p CO 2 showed different seasonal characteristics. In the swamp, grassland, and wastewater, p CO 2 had the highest value in August. At all sampling sites, the average p CO 2 was 1,991.77 ± 2,890.53 ppm in April, 2,247.53 ± 2,882.77 ppm in June, 2,991.71 ± 3,587.52 ppm in August, and 1,872.35 ± 2,299.81 ppm in October.

Temporal variation of p CO 2 ( a ) and F CO 2 ( b ) in the Xilin River in 2018.

The range of F CO 2 varied from 0.00 to 6.09 mol m −2 d −1 , and the four-month average was 0.70 mol m −2 d −1 . Figure  4 shows the spatial change of F CO 2 under different land use types. The highest value of F CO 2 appeared in sand land (4.88 ± 0.73 mol m −2 d −1 ), followed by pond (1.39 ± 1.98 mol m −2 d −1 ), swamp (1.22 ± 0.92 mol m −2 d −1 ), waste water (0.41 ± 0.35 mol m −2 d −1 ) and grassland (0.28 ± 0.16 mol m −2 d −1 ); the values of F CO 2 in reservoir and lake types were within 0.02. Under the different land use types, F CO 2 showed different seasonal changes. F CO 2 and p CO 2 showed different trends, and in sand land, swamp and grassland, the F CO 2 value was highest in June. Due to the shortage of water in June, F CO 2 could not be measured in the factory area. For all sampling points, the average value of F CO 2 was 0.52 ± 1.06 mol m −2 d −1 in April, 1.42 ± 1.86 mol m −2 d −1 in June, 0.63 ± 1.09 mol m −2 d −1 in August, and 0.37 ± 0.95 mol m −2 d −1 in October.

The p CO 2 and F CO 2 values of grassland were at a low level and were only higher than those of the Xilin River Reservoir and the Xilinhot artificial lake. The p CO 2 and F CO 2 values of sand land were at a high level.

The range of p CH 4 varied from 2.92 to 1,800.73 ppm, with average value of 81.55 ppm. The highest value of p CH 4 appeared in the pond (477.83 ± 764.20 ppm), which was much higher than the atmospheric background value, followed by waste water (136.52 ± 90.50 ppm), lake (113.86 ± 100.40 ppm), and swamp (104.26 ± 69.88 ppm), and in the remaining region p CH 4 values were within 50 ppm (Fig.  5 a). Under different land use types, p CH 4 value showed different seasonal changes. Except for the abnormal p CH 4 value of ponds in October, the value of p CH 4 in June was highest affected by human activities. However, areas such as swamp, sand land, and grassland, had higher CH 4 values in April.

Temporal variation of p CH 4 ( a ) and F CH 4 ( b ) in the Xilin River in 2018 ( p CH 4 ( c ) and F CH 4 ( d ) except those points in the blue dotted box).

The range of F CH 4 was 0.06 to 105.19 mmol m −2 d −1 , and the average value was 7.76 mmol m −2 d −1 (Fig.  5 b). The value of F CH 4 was largest in the pond. There was no significant difference in the F CH 4 values. Under different land use types, there were no significant seasonal differences in F CH 4 .

The variation range of p N 2 O was 0.31–12.44 ppm, the mean value was 0.73 ppm (Fig.  6 ). In the factory area, the p N 2 O value was abnormal, and the p N 2 O values in April and October were higher than those in June and August. There were no obvious differences in p N 2 O in other areas.

Temporal variation of p N 2 O ( a ) and F N 2 O ( b ) in the Xilin River in 2018 ( p N 2 O ( c ) except those points in the blue dotted box).

The range of F N 2 O values was − 12.60 to 224.04 μmol m −2 d −1 , and the average value was 24.32 μmol m −2 d −1 . There was an** obvious spatial change in F N 2 O. The F N 2 O value of the factory area was largest, followed by sand land and grassland, yet the values of F N 2 O in the lake area was negative.

Spatial variation of greenhouse gas emissions

On the main stream, the p CO 2 values from upstream to downstream first decreased and then increased; they reached the lowest value in the middle stream (Fig.  7 ). The middle part of the tributaries was cut off, and the p CO 2 value was higher at the source of the river. The spatial tendency of F CO 2 value was the same as that of p CO 2 .

Spatial variation of p CO 2 ( a ) and F CO 2 ( b ) in the Xilin River in 2018 (generate by Arcgis 10).

On the main stream of the Xilin River, from the upstream to the downstream, the p CH 4 value was higher in the river source area and increased with the flow direction of the river (Fig.  8 ). The value of p CH 4 gradually increased in tributaries. The variation trend of F CH 4 value was not obvious, and it had the lowest and negative value in the downstream of the main stream.

Spatial variation of p CH 4 (a) and F CH 4 (b) in the Xilin River in 2018 (generate by Arcgis 10).

The value of p N 2 O did not change in the tributaries but first remained stable and then increased from the upstream to the downstream of the main stream (Fig.  9 ). The value of F N 2 O fluctuated near zero and increased in the downstream area.

Spatial variation of p N 2 O ( a ) and F N 2 O ( b ) in the Xilin River in 2018 (generate by Arcgis 10).

The global warming potential of CH 4 is 25 times larger than that of CO 2 , and the global warming potential of N 2 O is 296 times larger than that of CO 2 30 . For the hydrosystem of the Xilin River Basin, CO 2 emissions accounted for 63.35% of the three GHG emissions, whereas CH 4 and N 2 O emissions accounted for 35.98% and 0.66%, respectively.

In the swamp area, CO 2 emissions accounted for 20.88% of the emissions of the Xilin river, and CH 4 accounted for 6.14% (Table 2 ). In sand land, CO 2 emissions accounted for 8.03% of the emission in the Xilin river, and CH 4 and N 2 O emissions accounted for 0.45% and 0.02%, respectively. In the pond type, CO 2 emissions accounted for 9.52% in the Xilin river, and CH 4 and N 2 O emissions accounted for 13.46% and 0.01%, respectively. In waste water, CO 2 emissions accounted for 7.08% of the Xilin river, and CH 4 emission accounted for 10.38%. The hydrosystem of the Xilin River Basin showed as a source of carbon dioxide and methane; at the same time, the nitrous oxide in the lake region showed as a sink.

Impacts of water quality parameters on GHG

Pearson’s correlation analysis was used to analyze the correlation between eight water chemical factors with p GHG and F GHG (Fig.  10 ).

Relationship between p GHG and water quality parameters.

The main influencing factors of p CO 2 were pH and DO, and p CO 2 was negatively correlated with Alk, Tw, Sal, TDS, pH and DO. There was a significant negative correlation with pH (R = − 0.804, P  < 0.01) and DO (R = − 0.505, P  < 0.01). p CO 2 was positively correlated with TN, TP and V w . The main influence factors of F CO 2 and p CO 2 were the same; however, the flow velocity (V w ) had positive correlation with F CO 2 , Q (R = − 0.274, P  < 0.05) had a significant negative correlation with F CO 2 . p CH 4 had a significant correlation with TP (R = 0.365, P  < 0.01). F CH 4 had no significant correlation with all factors. The main influencing factors of p N 2 O were DO (R = 0.429, P  < 0.01) and TP (R = 0.437, P  < 0.01). p N 2 O was negatively correlated with T w . There was a positive correlation of F N 2 O with below factors: the most important were with Sal (R = 0.661, P  < 0.01), Alk (R = 0.374, P  < 0.01), TDS (R = 0.639, P  < 0.01), TP (R = 0.696, P  < 0.01), TN (R = 0.589, P  < 0.01) and DO (R = 0.361, P  < 0.01), which had a significant positive correlation with F N 2 O.

Previous studies have shown that water temperature is one of the factors that affect river p CO 2 and F CO 2 because the solubility of CO 2 decreases with the rise of temperature; this has been found in many river studies around the world 31 , 32 . Other studies have also found that the photosynthesis of plankton has a great influence on the changes of p CO 2 and F CO 2 in rivers 33 . In this study, the water temperatures in June and August are higher than in April and October, and the values of p CO 2 in June and August are higher than in April and October.

The temperature directly influences the production of methane by influencing the activity and structure of microflora 34 . In our study that temperature has no significant correlation with CH 4. When the temperature rises, the dissolved oxygen concentration in the water decreases, which is more conducive to the production of methane. However, the concentration of methane in the Xilin River decreases with increasing temperature, which may be due to the significant increase in the activity of oxidizing bacteria in methane due to the increase in temperature, thus resulting in an increase in methane consumption.

Because of the strong correlation between flow rate and velocity, river sections with large flow rate usually have higher flow velocity. Higher velocity helps to increase the degree of surface turbulence and fragmentation, increase the area of contact between water and air, and accelerate the gas exchange rate at the water–air interface 35 . The flow velocity ranges from 0 to 1.2 m s −1 at all sampling points, which is a very low level, but the amount of carbon dioxide released from the Xilin River is considerable. Because of the large amount of carbon in the environment around the channel, it can be washed into the river network system by runoff or groundwater 9 , 36 .

The total alkalinity of the Xilin River is low, and there is less plankton in the water, which indicates that the river carbon dioxide mainly comes from terrestrial organic carbon rather than inorganic carbon. A large part of terrestrial organic carbon comes from net primary productivity.

The influence of nutrient on river CO 2 affects the production and consumption of in situ CO 2 in the river mainly through changing the balance of primary production and ecosystem respiration 36 , 37 . The Xilin River is a low-pollution region. Studies have shown that the increase of nutrient salt mainly promotes the growth of phytoplankton in the water, enhances photosynthesis, and then reduces the concentration of CO 2 in the water due to carbon restriction in low-pollution rivers 38 .

Influence of land use on GHG

The p CO 2 value (> 10,000 ppm) in sand land is much higher than that of other land use types (< 10,000 ppm) due to groundwater recharge which had high CO 2 content 39 In large river systems, the p CO 2 value in the groundwater system is approximately ten times higher than that in the surface water system. Except for the areas affected by human activities, the river alkalinity in sand land areas is the largest 40 . A large amount of carbon ions (HCO 3 - , CO 3 2- ) and dissolved CO 2 is released from groundwater recharge surface water under the action of photosynthesis and weathering 41 . The carbon ion reaction in surface water is released into the atmosphere, the variation range of subsurface temperature is small, and the dissolved CO 2 exists stably in groundwater. However, after the groundwater recharge surface water is exposed to the river, the temperature in the atmosphere changes greatly, and the solubility of CO 2 varies with the change of temperature 35 .

The Xilin River source area is a swamp area, and groundwater is one of the sources of river runoff in swamp area and is rich in dissolved carbon dioxide; a small amount of groundwater recharge thus also provides sufficient carbon dioxide for river water 39 . At the same time, the bog type in the Xilin River source area is low-lying bog, the initial stage of bog development with a low-lying surface, which often becomes the place where surface runoff and groundwater collect and pool. Water supply is mainly groundwater, and there are many minerals and nutrients in the water. The pH of water and peat is acidic to neutral. The results of the Xilin River source area are the same as the report where p CO 2 was much higher than the downstream water 42 . The CH 4 of swamp in the Xilin River is 8.14 mmol m −2 d −1 , which is higher than the CH 4 value of the Tibetan Plateau (4.19 mmol m −2 d −1 ) 43 . In the study of CH 4 emissions of the Yukon River basin, both main stream and tributary showed that the upstream concentration of CH 4 was lower than that of the midstream and downstream 38 , 44 . The emissions of CH 4 from Xilin River also showed a similar distribution pattern.

In the Xilin River, the vegetation coverage in grassland is relatively high, similar to that in swamp. Carbon dioxide release from swamps coincides with the growth cycle of plants, and terrestrial organic carbon related to plants is thus one of the important sources of carbon dioxide in river water bodies 17 . Because of the low carbon density in the soil carbon pool, the soil carbon pool provides less carbon to the river carbon pool than the swamp. There are relatively few inundated-vegetation areas in grassland, and aquatic vegetation thus provides less organic carbon to rivers. On the other hand, the altitude of grassland cover areas is low, the water level of groundwater in mountain bodies is higher, and groundwater also acts as the source of river runoff 45 . The net primary productivity of the Xilin River Basin begins in April and reaches its peak in August, and the average annual net primary productivity from 1926 to 2017 is 185.38 gC m −2  a −1 ; to some extent, this shows that there is a strong relationship between river carbon and net primary productivity in the Xilin River Basin, and a part of the carbon dioxide in the river is provided by plant carbon. In the Xilin River Basin, the CH 4 emission of grassland is 31.25 μmol m −2 d −1 , and the CH 4 emission of rivers is approximately 100 times that of grassland 46 .

Influence of human activities on GHG

The CH 4 of river water is mainly produced by methanogens in sedimentary layers after a series of fermentation processes in the anaerobic environment with acetate or CO 2 /H 2 as a substrate 34 . An increase in temperature would stimulate the activity of soil/sediment methanogenic bacteria as well as promote a higher rate of organic matter degradation, which in turn would provide more substrates for methanogens to produce CH 4 47 , 48 . The small Pond (Wolongquan) is mainly used for raising fish. Because of the large amount of breeding and artificial feed, it contains a large number of nutrients, which promotes the reproduction and growth of algae 49 . The growth process of plankton produces a large amount of fresh organic carbon, which stimulates the production of CH 4 . The proliferation of planktonic plants and animals leads to the reduction of oxygen concentration in the deep water layer, which creates an anaerobic environment for the production of CH 4 and reduces CH 4 oxidation 50 . The dissolved oxygen value is very low, which promotes the growth of anaerobes, and the river is in a eutrophication state, which produces more CH 4 under the anoxic conditions 51 , 52 .

The p CO 2 value of the Xilin Reservoir is 670.15 ± 114.48 ppm, which is higher than the background value of atmospheric carbon dioxide, and its F CO 2 value (0.02 ± 0.02 mol m −2 d −1 ) is similar to that of an artificial lake. The construction of a reservoir changes the biogeochemical cycle of carbon and nitrogen in the basin; the flow rate slows down after the river enters the reservoir, which causes the deposition of plankton debris and other granular organic matter in the water. These changes will affect the production and release of greenhouse gas. Compared with rivers, reservoirs have longer hydraulic retention times, which is conducive to the accumulation of pollutants 53 . Phytoplankton debris, that is, endogenous organic matter, provides a rich source of easily degradable carbon for the accumulation of organic matter at the bottom of the reservoir and enhances the anaerobic conditions at the bottom of the reservoir, which thus provides a material basis and environmental conditions for the production of CH 4 . The flow velocity of water in the Xilin Reservoir is approximately zero, and there is no large fluctuation.

The Xilin River flows through an artificial lake in Xilinhot City, but the River flows intermittently to its downstream because of the impoundage of the Xilin Reservoir. The discharge amount of water in the Xilin Reservoir to the downstream is very small, which has little impact on p CO 2 in the downstream. The artificial lake in Xilinhot City is composed of river water and reclaimed water. The artificial lake flow rate is close to zero, and its water–gas interface gas release rate is extremely low, and the F CO 2 value is 0.02 ± 0.01 mol m −2 d −1 . There is no obvious plankton in the lake, and reeds grow around the artificial lake. The p CO 2 value in the artificial lake is higher than that in the atmosphere 54 . The CO 2 produced by the respiration of aquatic plants inside the water body is basically the same as the CO 2 fixed by the photosynthesis of aquatic plants.

The lower reaches of the Xilin River are located in the factory area and are greatly influenced by human activities, mainly by power stations and dairy farms; the power station extracts groundwater for production operations. After secondary treatment, sewage is discharged into the river and mixed with the Xilin River. The water discharged from the dairy farms also contains a large number of microorganisms, and the river contains a large number of organic substances as well as nitrogen and phosphorus compounds, which makes algae and microorganisms grow and reproduce; a large number of them exist in the river and promote river oxygen metabolism. The respiration of algae and microorganisms releases a small amount of carbon dioxide but microorganisms produce more CO 2 and CH 4 in the absence of oxygen 34 . The values of p CO 2 (2,048.93 ± 660.43 ppm) and F CO 2 (0.41 ± 0.74 mol m −2 d −1 ) in factory areas are higher than in other areas affected by human activities. The value of N 2 O in the factory area is higher, and the total nitrogen and total phosphorus in the river are positively correlated with the concentration of N 2 O 55 . The sewage discharged from the factory contains a large amount of nutrients, which makes the microbial activity produce a large amount of CH 4 and N 2 O. With the discharge of sewage into the channel, the release of CH 4 and N 2 O in the channel and downstream of the channel is indirectly affected 17 .

Comparison with other rivers

Our estimated results for CO 2 emissions in the Xilin River were greater than in most of the reported rivers, such as the Wuding River 56 , the Daning River 57 , and inland water in Africa 17 (Table 3 ). The emissions of CH 4 and N 2 O in the Xilin River were higher than the inland water in Africa. Because there are many land types in the Xilin River, the range of greenhouse gas emissions was larger than that of other rivers at home and abroad.

The carbon dioxide emissions from the swamp of the Xilin River were close to those of the Phragmites marsh, but the methane emissions were higher than those from the Phragmites marsh. Additionally, saltmarshes are sinks of carbon dioxide, and the values of methane emissions were between those of swamp and Phragmites marsh 58 , 59 . Due to the lower height of grassland vegetation, photosynthesis is stronger. The type of plant in the marsh area affects greenhouse gas emissions of water.

Carbon dioxide emissions were lower for the Min River than for the Xilin River, but the methane emissions of the Min River were higher than those of the Xilin River. The emissions of CH 4 of Min river was higher than the Xilin river, because the drainage of the Wolongquan pond was lower than the Min river. The drainage of the Min river significantly enhance CH 4 emissions 29 . The Wolongquan pond of the Xilin River is mainly used for farming fish with great artificial intervention, which proves that the addition of nutrients has a great influence on the greenhouse gas emissions of water bodies.

In the Xilin river reservoir, the emissions of CO 2 and CH 4 were equal to the Three Gorges and Shasta reservoir 53 , 60 . For reservoir, the flow velocity of surface water is slow, which causes the emissions of greenhouse were less. And the deep water created well-oxygenated conditions, resulting in lower methane emissions.

Conclusions

In this study we estimated emissions of greenhouse gas from the Xilin River, which is characterized by different land-use types and various degrees of human impacts. The results showed that the hydrological drainage network of the Xilin River was oversaturated in GHG (CO 2 , CH 4 and N 2 O) with respect to the atmospheric concentrations. For the hydrosystem of the Xilin River Basin, CO 2 emissions accounted for 63.35% of the three GHG emissions, whereas CH 4 and N 2 O emissions accounted for 35.98% and 0.66%, respectively. GHG emissions from the Xilin river were dominated by CO 2 emissions and were interpreted as being supplied by terrestrial carbon transportation and groundwater replenishment and by wastewater discharges. In future work, sampling should cover more sites with a greater frequency to better quantify the magnitude of CO 2 , CH 4 and N 2 O emissions at diurnal and monthly scales before upscaling them to annual estimates. Comparing the differences in greenhouse gas emissions after the cut-off, it is possible to predict the total greenhouse gas emissions after global river drying.

Arnell, N. W. & Reynard, N. S. The effects of climate change due to global warming on river flows in Great Britain. J. Hydrol. 183 , 397–424 (1996).

Article   ADS   Google Scholar  

Richey, J. E., Devol, A. H., Wofsy, S. C. & Victoria, R. Biogenic gases and the oxidation and reduction of carbon in Amazon River and floodplain waters. Limnol. Oceanogr. 33 , 551–561 (1988).

Article   ADS   CAS   Google Scholar  

Lingli, L. & Greaver, T. L. A review of nitrogen enrichment effects on three biogenic GHGs: the CO 2 sink may be largely offset by stimulated N 2 O and CH 4 emission. Ecol. Lett. 12 , 1103–1117 (2010).

Google Scholar  

Solomon, S. et al. Climate Change 2007: Synthesis Report. Contribution of Working Group I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change . Summary for Policymakers (2007).

Zhan, M. A., Hai, L. I., Yue, B., Gao, Q. X. & Dong, L. Study on emission characteristics and correlation of GHGs CH 4 and CO 2 in MSW landfill cover layer. J. Environ. Eng. Technol. 5 , 399–405 (2014).

Wang, G. et al. Nitrogen removal rates in a frigid high-altitude river estimated by measuring dissolved N 2 and N 2 O. Sci. Total Environ. 645 , 318–328 (2018).

Article   ADS   CAS   PubMed   Google Scholar  

Marx, A. et al. A review of CO 2 and associated carbon dynamics in headwater streams: a global perspective. Rev. Geophys. 55 , 560–585 (2017).

Cao, Y., Yuan, R., Jiao, S., Zhang, Q. & Deng, F. Distribution characteristics of partial pressure of carbon dioxide during thermal stratification in summer in Guangzhao Reservoir. Environ. Sci. Technol. 6 , 15–21 (2018).

Yao, G. et al. Dynamics of CO 2 partial pressure and CO 2 outgassing in the lower reaches of the Xijiang River, a subtropical monsoon river in China. Sci. Total Environ. 376 , 255–266 (2007).

Ran, L. et al. Long-term spatial and temporal variation of CO 2 partial pressure in the Yellow River China. Biogeosciences 12 , 921–932 (2015).

Beaulieu, J. J. et al. Nitrous oxide emission from denitrification in stream and river networks. Proc. Natl. Acad. Sci. USA 108 , 214–219 (2011).

Khalil, M. A. K. & Rasmussen, R. A. Sources, sinks, and seasonal cycles of atmospheric methane. J. Geophys. Res. Oceans 88 , 5131–5144 (1983).

Algesten, G. et al. Contribution of sediment respiration to summer CO 2 emission from low productive boreal and subarctic lakes. Microb. Ecol. 50 , 529–535 (2005).

Article   PubMed   Google Scholar  

Beaulieu, J., Arango, C., Hamilton, S. K. & Tank, J. L. The production and emission of nitrous oxide from headwater streams in the Midwestern United States. Glob. Change Biol. 14 , 878–894 (2010).

Garnier, J., Billen, G. & Cébron, A. Modelling nitrogen transformations in the lower Seine river and estuary (France): impact of wastewater release on oxygenation and N 2 O emission. Hydrobiologia 588 , 291–302 (2007).

Article   CAS   Google Scholar  

Yang, H., Zheng, Y. F., Rong-Jun, W. U., Yin, J. F. & Xia, S. Nitrous oxide flux at the water–air interface of the rivers in Nanjing during summer. Huan jing ke xue Huanjing kexue 35 , 348–355 (2014).

Borges, A. V. et al. Globally significant greenhouse-gas emissions from African inland waters. Nat. Geosci. 8 , 637–642 (2015).

Thi Phuong Quynh, L. et al. CO 2 partial pressure and CO 2 emission along the lower Red River (Vietnam). Biogeosciences 15 , 4799–4814 (2018).

Lauerwald, R., Hartmann, J., Ludwig, W. & Moosdorf, N. Assessing the nonconservative fluvial fluxes of dissolved organic carbon in North America. J. Geophys. Res. Biogeosci, 117 , G01027 (2012).

Xue, L. Y., Wu, J. K., Zhang, M. Q., Ding, Y. J. & Ye, B. S. Spatial and temporal distribution of trace elements in surface water in the Xilin River Basin. Environ. Sci. 33 , 3417–3425 (2012).

Lu, J. et al. Dissolved inorganic carbon (DIC) contents in middle and lower reaches of Lancang River: related to water environments and dams. J. Water Resour. Prot. 09 , 1132–1144 (2017).

Chen, S. P., Bai, Y. F., Lin, G. H., Liang, Y. & Han, X. G. Effects of grazing on photosynthetic characteristics of major steppe species in the Xilin River Basin, Inner Mongolia China. Photosynthetica 43 , 559–565 (2005).

Article   Google Scholar  

Wei, Z., Wu, S., Zhou, S. & Chen, L. Installation of impervious surface in urban areas affects microbial biomass, activity (potential C mineralisation), and functional diversity of the fine earth. Soil Res. 51 , 59–67 (2013).

Wu, Y., Hu, Z., Yang, L., Graham, B. & Kerr, P. G. The removal of nutrients from non-point source wastewater by a hybrid bioreactor. Bioresour. Technol. 102 , 2419–2426 (2011).

Article   CAS   PubMed   Google Scholar  

Lewis, E., Wallace, D. & Allison, L. J. Program developed for CO 2 system calculations. Office of Scientific & Technical Information Technical Reports (1998).

Penman, D. E., Hoenisch, B. R., Rasbury, E. T., Hemming, N. G. & Spero, H. J. Boron, carbon, and oxygen isotopic composition of brachiopod shells: intra-shell variability, controls, and potential as a paleo-pH recorder. Chem. Geol. 340 , 32–39 (2013).

Tortell, P. D., Bittig, H. C., Körtzinger, A., Jones, E. M. & Hoppema, M. 1 Biological and physical controls on N 2 , O 2 and CO 2 distributions in 2 contrasting Southern Ocean surface waters. Glob. Biogeochem. Cycles 29 , 994–1013 (2015).

Li, Z. et al. Soil–air greenhouse gas fluxes influenced by farming practices in reservoir drawdown area: a case at the Three Gorges Reservoir in China. J. Environ. Manag. 181 , 64–73 (2016).

Yang, P. et al. Fluxes of carbon dioxide and methane across the water-atmosphere interface of aquaculture shrimp ponds in two subtropical estuaries: the effect of temperature, substrate, salinity and nitrate. Sci. Total Environ. 635 , 1025–1035 (2018).

Fengqin, L., Yupeng, Z., Hong, L. & Dawen, G. Long-term harvesting of reeds affects greenhouse gas emissions and microbial functional genes in alkaline wetlands. Water Res. 164 , 114936 (2019).

Hope, D., Palmer, S. M., Billett, M. F. & Dawson, J. J. C. Variations in dissolved CO 2 and CH 4 in a first-order stream and catchment: an investigation of soil–stream linkages. Hydrol. Process. 18 , 3255–3275 (2004).

Parkin, T. B. & Kaspar, T. C. Temperature controls on diurnal carbon dioxide flux. Soil Sci. Soc. Am. J. 67 , 1763–1772 (2003).

Linn, D. M. Effect of water-filled pore space on carbon dioxide and nitrous oxide production in tilled and non-tilled soils. Soil Sci. Soc. Am. J. 48 , 647–653 (1984).

Stanley, E. H. et al. The ecology of methane in streams and rivers: patterns, controls, and global significance. Ecol. Monogr. 86 , 146–171 (2016).

Alin, S. R. et al. Physical controls on carbon dioxide transfer velocity and flux in low-gradient river systems and implications for regional carbon budgets. J. Geophys. Res. 116 , G01009 (2011).

Li, S. et al. Daily CO 2 partial pressure and CO 2 outgassing in the upper Yangtze River basin: a case study of the Longchuan River China. J. Hydrol. 466 , 141–150 (2012).

Peng, X., Wang, B., Liu, C., Liu, X. & Wang, F. Diurnal variations of pCO 2 in relation to environmental factors in the cascade reservoirs along the Wujiang River China. Chin. J. Geochem. 31 , 41–47 (2012).

Sawakuchi, H. O. et al. Methane emissions from Amazonian Rivers and their contribution to the global methane budget. Glob. Change Biol. 20 , 2829–2840 (2014).

Abril, G. et al. Amazon River carbon dioxide outgassing fuelled by wetlands. Nature 505 , 395–398 (2013).

Article   ADS   PubMed   CAS   Google Scholar  

Deirmendjian, L. & Abril, G. Carbon dioxide degassing at the groundwater-stream-atmosphere interface: isotopic equilibration and hydrological mass balance in a sandy watershed. J. Hydrol. 558 , 129–143 (2018).

Maher, D. T., Santos, I. R., Golsby-Smith, L., Gleeson, J. & Eyre, B. D. Groundwater-derived dissolved inorganic and organic carbon exports from a mangrove tidal creek: the missing mangrove carbon sink?. Limnol. Oceanogr. 58 , 475–488 (2013).

Teodoru, C. R., Giorgio, P. A. D., Prairie, Y. T. & Camire, M. Patterns in pCO 2 in boreal streams and rivers of northern Quebec Canada. Glob. Biogeochem. Cycles 23 , 1–11 (2009).

Wei, D. et al. Revisiting the role of CH 4 emissions from alpine wetlands on the Tibetan Plateau: evidence from two in situ measurements at 4758 and 4320 m above sea level. J. Geophys. Res. Biogeosci. 120 , 1741–1750 (2015).

Striegl, R. G., Dornblaser, M. M., Mcdonald, C. P., Rover, J. R. & Stets, E. G. Carbon dioxide and methane emissions from the Yukon River system. Glob. Biogeochem. Cycles 26 , 1–11 (2012).

Deirmendjian, L. et al. Importance of the vegetation-groundwater-stream continuum to understand transformation of biogenic carbon in aquatic systems - A case study based on a pine-maize comparison in a lowland sandy watershed (Landes de Gascogne, SW France). Sci. Total Environ. 661 , 613–629 (2019).

Wang, Y., Chen, Z. Z. & Ojima, D. Preliminary results of a study on CH4 flux in Xilin River Basin steppe under different grazing intensitles. Acta Phytoecol. Sin. 24 , 693–696 (2000).

Vizza, C., West, W. E., Jones, S. E., Hart, J. A. & Lamberti, G. A. Regulators of coastal wetland methane production and responses to simulated global change. Biogeosciences 14 , 1–29 (2017).

Wang, H. et al. Temporal and spatial variations of greenhouse gas fluxes from a tidal mangrove wetland in Southeast China. Environ. Sci. Pollut. Res. 23 , 1873–1885 (2016).

Yang, J. et al. Effect of salinity on soil respiration in relation to dissolved organic carbon and microbial characteristics of a wetland in the Liaohe River estuary, Northeast China. Sci. Total Environ. 642 , 946–953 (2018).

Franz, D., Koebsch, F., Larmanou, E., Augustin, J. & Sachs, T. High net CO 2 and CH 4 release at a eutrophic shallow lake on a formerly drained fen. Biogeosciences 13 , 3051–3070 (2016).

Holgerson, M. A. & Raymond, P. A. Large contribution to inland water CO 2 and CH 4 emissions from very small ponds. Nat. Geosci. 9 , 222–226 (2016).

Ping, Y., Lai, D. Y. F., Jia, F. H. & Tong, C. Effect of drainage on CO 2 , CH 4 , and N 2 O fluxes from aquaculture ponds during winter in a subtropical estuary of China. J. Environ. Sci. China 65 , 72–82 (2018).

Soumis, N., Duchemin, E., Canuel, R. & Lucotte, M. Greenhouse gas emissions from reservoirs of the western United States. Glob. Biogeochem. Cycles 18 , 1–11 (2004).

Maberly, S. C., Barker, P. A., Stott, A. W. & De Ville, M. M. Catchment productivity controls CO2 emissions from lakes. Nat. Clim. Change 3 , 391–394 (2012).

Stiles, W. A. V., Rowe, E. C. & Dennis, P. Nitrogen and phosphorus enrichment effects on CO 2 and methane fluxes from an upland ecosystem. Sci. Total Environ. 618 , 1199–1209 (2018).

Ran, L. et al. Riverine CO 2 emissions in the Wuding River catchment on the Loess Plateau: Environmental controls and dam impoundment impact. J. Geophys. Res. Biogeosci. 122 , 1439–1455 (2017).

Ni, M., Li, S., Luo, J. & Lu, X. CO2 partial pressure and CO 2 degassing in the Daning River of the upper Yangtze River China. J. Hydrol. 569 , 483–494 (2019).

Olsson, L. et al. Factors influencing CO 2 and CH 4 emissions from coastal wetlands in the Liaohe Delta Northeast China. Biogeosciences 12 , 3469–3503 (2015).

Wilson, B. J., Mortazavi, B. & Kiene, R. P. Spatial and temporal variability in carbon dioxide and methane exchange at three coastal marshes along a salinity gradient in a northern Gulf of Mexico estuary. Biogeochemistry 123 , 329–347 (2015).

Zhao, Y., Wu, B. F. & Zeng, Y. Spatial and temporal patterns of greenhouse gas emissions from Three Gorges Reservoir of China. Biogeosciences 10 , 1219–1230 (2013).

Download references

Acknowledgements

This study was funded by the National Key Research and Development Program of China (Grant No.2016YFC0500508), National Natural Science Foundation of China (Grant Nos. 51939006, 51869014), Science and Technology Major Project on Lakes of Inner Mongolia (Grant No. ZDZX2018054), Open Project Program of 'Ministry of Education Key Laboratory of Ecology and Resources Use of the Mongolian Plateau'. The authors are grateful to Dr. Xinyu Liu, Mingyang Tian, Yuanrong Su and Lishan Ran for their constructive discussions. Data were from field measurements.

Author information

Authors and affiliations.

Inner Mongolia Key Laboratory of River and Lake Ecology, School of Ecology and Environment, Inner Mongolia University, Hohhot, China

Xue Hao, Yu Ruihong, Zhang Zhuangzhuang, Qi Zhen & Lu Xixi

Key Laboratory of Mongolian Plateau Ecology and Resource Utilization Ministry of Education, Hohhot, China

Department of Geography, National University of Singapore, Lower Kent Ridge Road, Singapore, Singapore

Inner Mongolia Water Resource Protection and Utilization Key Laboratory, Water Conservancy and Civil Engineering College, Inner Mongolia Agricultural University, Hohhot, China

Liu Tingxi & Gao Ruizhong

You can also search for this author in PubMed   Google Scholar

Contributions

X.H., Y.R., L.X., L.T. and G.R. wrote the main manuscript text, Z.Z., Q.Z. prepared figure 1, and figures 7-9.

Corresponding authors

Correspondence to Yu Ruihong or Liu Tingxi .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Hao, X., Ruihong, Y., Zhuangzhuang, Z. et al. Greenhouse gas emissions from the water–air interface of a grassland river: a case study of the Xilin River. Sci Rep 11 , 2659 (2021). https://doi.org/10.1038/s41598-021-81658-x

Download citation

Received : 26 October 2020

Accepted : 30 December 2020

Published : 29 January 2021

DOI : https://doi.org/10.1038/s41598-021-81658-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Contrasting emissions of carbon-based greenhouse gases from two paddy soils under submerged conditions as affected by biochar addition.

• Binh Thanh Nguyen
• Hao Phu Dong
• Nam Van Thai

Environmental Earth Sciences (2023)

Impact of microplastics on riverine greenhouse gas emissions: a view point

• Pooja Upadhyay
• Sanjeev Kumar Prajapati

Environmental Science and Pollution Research (2022)

The effects of the climatic change on daily maximum and minimum temperatures along 102 years (1917–2018) recorded at the Fabra Observatory, Barcelona

• M. D. Martínez

Theoretical and Applied Climatology (2022)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Greenhouse Effect

Eighth-warmest December on record helps 2020 end as second-hottest year on record

Despite the influence of La Niña, December 2020 was the eighth-hottest December on record helping 2020 become the second-hottest year on record. 

November 2020: The year is ending as it began, on a hot streak

November 2020 was the second-warmest November on record, which increases 2020's chance of becoming the warmest year on record to over 50%.

A warm pool in the Indo-Pacific Ocean has almost doubled in size, changing global rainfall patterns

The large, warm pool of ocean water in the Indian and west Pacific Oceans has been growing warmer and expanding in size since 1900, impacting the Madden Julian Oscillation and regional rainfall. 

Climate.gov tweet chat: Talk with education experts on climate change education

Join  three education experts to talk about how engaging, empowering and educating others can lead to powerful climate action.

Doesn't carbon dioxide in the atmosphere come from natural sources?

Yes, there are, but the only new process on Earth that has been identified that can account for the significant tipping of Earth's carbon balance is human activity, including deforestation, biomass burning, cement production, and—especially—fossil-fuel emissions.

Climate.gov tweet chat: Talk with experts about ongoing NASA/NOAA field campaign to study smoke from wildfires and agricultural burning

If you missed our August 29 tweet chat, here's the transcript. Read what the fire and smoke experts had to say about the FIREX-AQ field campaign and its mission to study what's in the smoke from wildfires and agricultural burning. 

Great Lakes ice cover decreasing over last 40 years

Ice cover on the Great Lakes has been decreasing since the 1970s,  affecting everything from fishing to shipping. 

January 2017 brings the heat down under in Australia

February 10, 2017

Monthly carbon dioxide levels hit new milestone

For 800,000 years before the twentieth century, carbon dioxide levels in Earth’s atmosphere never exceeded 300 parts per million. In March 2015, the monthly average went above 400 ppm for the first time.

Science in space: New satellite could bring climate DSCOVRies

In addition to its primary mission of observing space weather, the Deep Space Climate Observatory (DSCOVR) satellite is carrying two instruments that are important to climate science: the NISTAR radiometer and the EPIC camera. 

Greenhouse gases - Mini Case Study for Climate Detectives

This mini case study is dedicated to the topic of climate change and greenhouse gases and students will investigate how greenhouse gas emissions and concentrations in the atmosphere evolved in their country in the last decades. 

The Mini Case Studies for Climate Detectives are intended to help Climate Detectives teams to identify a climate related topic that they can investigate and that applies to real world situations.  

In this mini case study, students will try to answer the research questions: “How have greenhouse gas emissions evolved in your country in the last decades? What are the past and present concentrations of greenhouse gases in the atmosphere? For that, teachers will find in the resource suggestions of different types of data that students could collect and analyse. The suggestions are not exhaustive, and the teachers may decide on their own specific focus within a given research area.  

The mini case study should be used in conjunction with the Climate Detectives  teacher guide  and not as a standalone document. 

Subject: Science; technology 

Age range: 12-17 years old 

Keywords: Climate; Climate Change; Earth Observation; Greenhouse Gases; Science; Technology 

Download : mini case study

This resource is part of a set of resources developed by ESA’s Education Office in collaboration with ESEROs to support the Climate Detectives project. 

Thank you for liking

You have already liked this page, you can only like it once!

November 3, 2023

Earth Reacts to Greenhouse Gases More Strongly Than We Thought

Climate scientists, including pioneer James Hansen, are pinning down a fundamental factor that drives how hot Earth will get

By Chelsea Harvey & E&E News

A 'Blue Marble' image of the Earth taken from the VIIRS instrument aboard NASA's Earth-observing satellite – Suomi NPP.

NASA/NOAA/GSFC/Suomi NPP/VIIRS/Norman Kuring

CLIMATEWIRE |  Climate scientist James Hansen is frustrated. And he’s worried.

For nearly 40 years, Hansen has been warning the world of the dangers of global warming. His testimony at a groundbreaking 1988 Senate hearing on the greenhouse effect helped inject the coming climate crisis into the public consciousness. And it helped make him one of the most influential climate scientists in the world.

Hansen has spent several decades as director of NASA’s Goddard Institute for Space Studies, and now at 82, he directs Columbia University’s  Climate Science, Awareness and Solutions program .

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing . By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

In the years since his seminal testimony, many of Hansen’s basic scientific predictions about the Earth’s climate future have come true. Greenhouse gas emissions have grown, and global temperatures have continued to rise. The world’s glaciers and ice sheets are melting and sea level rise is accelerating.

But Hansen has been disappointed with the scientific community’s response to some of his more recent projections about the future of the warming Earth, which some researchers have characterized as unrealistically dire.

In particular, he was discouraged by the response to a paper he published in 2016, suggesting catastrophic ice melt in Greenland and Antarctica, with widespread global effects, may be possible with relatively modest future warming.

Many researchers said such outcomes were unlikely. But Hansen described the paper as some of his most important work and a warning about the need for more urgent action.

Now he’s bracing himself for a similar reaction to his  latest paper , published Thursday morning.

“I expect the response to be characterized by scientific reticence,” he said in an email to E&E News.

The new paper, published in the research journal  Oxford Open Climate Change , addresses a central question in modern climate science: How much will the Earth warm in response to future carbon emissions? It’s a metric known as “climate sensitivity,” or how sensitive the planet is to greenhouse gases in the atmosphere.

Hansen’s findings suggest the planet may warm faster than previous estimates have indicated. And while some experts say it’s possible, others suggest that he’s taken the results too far.

In studies, scientists often tackle the climate sensitivity question by investigating how much the Earth would warm if atmospheric carbon dioxide concentrations doubled their preindustrial levels. Prior to the industrial era, global CO2 levels hovered around 280 parts per million, meaning a doubling would land around 560 ppm.

Today’s CO2 levels have already climbed above 400 ppm, giving the question a growing relevance.

Climate sensitivity is a difficult metric to estimate. It hinges on a wide variety of feedback loops in the Earth’s climate system, which can speed up or slow down the planet’s warming.

As the Earth’s reflective glaciers and ice sheets melt, for instance, the planet can absorb more sunlight and warm at a faster rate. Forests and other natural ecosystems may absorb different amounts of carbon as the planet warms. Different types of clouds can both speed up or slow down global warming, and it’s still unclear how they will change as the Earth heats up.

The uncertainties around these factors have made it challenging for scientists to pin down an exact estimate for climate sensitivity. But they’ve chipped away at it in recent years.

For decades, studies generally suggested that the Earth should experience anywhere from 1.5 to 4.5 degrees Celsius of warming with a doubling of CO2. But a  2020 paper narrowed the range  to between 2.6 and 3.9 C, using multiple lines of evidence including climate models, the Earth’s response to recent historical emissions and the Earth’s ancient climate history.

The latest assessment report from the U.N.’s Intergovernmental Panel on Climate Change adopted a similar estimate, suggesting a likely range of 2.5 to 4 C with a central estimate around 3 C.

Hansen’s new paper, published with an international group of co-authors, significantly ups the numbers. It suggests a central estimate of around 4.8 C, nearly 2 degrees higher than the IPCC’s figure.

The paper relies largely on evidence from Earth’s ancient climate history. One reason? It’s unclear whether current climate models accurately represent all the relevant feedback effects that may affect climate sensitivity, Hansen and his co-authors argue. The planet’s past provides a clearer view of how the Earth has responded to previous shifts in atmospheric carbon dioxide concentrations.

The paper also suggests that global warming is likely to proceed faster in the near term than previous studies have suggested.

Under the international Paris climate agreement, world leaders are striving to keep global warming well below 2 C and below 1.5 C if at all possible. The new paper warns that warming could exceed 1.5 C by the end of the 2020s and 2 C by 2050.

A gradual global decline in air pollution, driven by tightening environmental regulations, is part of the reasoning. Some types of air pollution are known to have a cooling effect on the climate, which may mask some of the impact of greenhouse gas emissions. As these aerosols decline in the atmosphere, some research suggests, this masking effect may fall away and global temperatures may rise at faster rates.

Hansen and his co-authors argue that better accounting for the declines in global aerosols should accelerate estimates of near-term global warming. Studies suggest that warming between 1970 and 2010 likely proceeded at around 0.18 C per decade. Post-2010, the new paper argues, that figure should rise to 0.27 C.

The findings should motivate greater urgency to not only cut greenhouse gas emissions but to eventually lower global temperatures closer to their preindustrial levels, Hansen suggests. That means using natural resources and technological means to remove carbon dioxide from the atmosphere.

Hansen also suggests that a controversial form of geoengineering, known as solar radiation management, is likely warranted. SRM, in theory, would use reflective aerosols to beam sunlight away from the Earth and lower the planet’s temperatures. The practice has not been tested at any large scale, and scientists have raised a variety of concerns about its ethics and potential unintended side effects.

Yet Hansen believes scientists and activists “should raise concerns about the safety and ethics of NOT doing SRM,” he said by email.

Climate change, caused by human greenhouse gas emissions, is in itself a form of planetary geoengineering, he added.

“My suggestion is to reduce human geoengineering of the planet,” he said.

Yet some scientists say the new paper’s findings — again — are overblown.

The paper “adds very little to the literature,” said Piers Forster, director of the Priestly International Centre for Climate at Leeds University in the U.K. and a lead chapter author of the IPCC’s latest assessment report, in an email to E&E News.

It presents high-end estimates of climate sensitivity based on ancient climate records from the Earth’s past — but those findings aren’t necessarily new, he said. Forster also suggested that some of the methods the new paper used to arrive at those high estimates were “quite subjective and not justified by observations, model studies or literature.”

Forster also took issue with the new paper’s treatment of previous climate sensitivity estimates, including the widely cited 2020 study, which the authors suggested were far too low. The 2020 study presented a careful analysis, using multiple lines of high-quality evidence, Forster said. And yet the authors of the new paper “dismiss it, on spurious grounds.”

Michael Oppenheimer, a climate scientist and director of the Center for Policy Research on Energy and Environment at Princeton University, said the uncertainties around the effects of declining aerosols were important to pay attention to. And he suggested that the new paper’s climate sensitivity estimates were possible.

But added that he regards them as “a worst-worst-case” scenario.

“I think it’s perfectly legitimate to have a worst-worst-case out there,” he added. “They help people think about what the boundaries of the possible are, and they are necessary for risk management against the climate problem.”

But there are still so many uncertainties about the kinds of feedback factors affecting the Earth’s climate sensitivity, he said, that “you can’t really nail it down with the kind of precision that [Hansen’s] provided.”

But Hansen says the new paper’s lines of evidence are based on the most up-to-date research on the Earth’s ancient history.

“[T]here is no basis whatever for the claim that our results are ‘unlikely,’” he said by email. “It is the IPCC sensitivity that is unlikely, less than 1 percent chance of being right, as we show quantitatively in our (peer-reviewed) paper.”

Hansen and 'scientific reticence'

Hansen has been into the deep end of climate debates for much of his career.

In 1988, at the time of his Senate testimony, scientists were still discussing whether the fingerprint of human-caused global warming could yet be detected above the “noise” of the Earth’s natural climate variations.

“When I first got into this, and when Jim and I were testifying, we were arguing about whether there's a global signal,” said Oppenheimer, the Princeton scientist, who testified alongside Hansen in 1988. “All the information we had was about global mean temperature, global mean sea level. We couldn’t talk in the language of things that people cared about.”

But even with the limitations of climate science at the time, the scientists warned the world of the dangers to come.

Hansen has co-authored dozens of papers on climate change in the years since, many of which have been highly regarded by the scientific community.

“Over time, he’s got a pretty damn good track record of turning out to be right about things that other people thought differently about,” Oppenheimer said.

Forster, the Leeds University scientist, agreed that “some of Hansen’s papers are brilliant and his work and deeds helped establish this IPCC in the first place.”

But he added that he still thought the new paper misses the mark.

The reception is similar to a major paper Hansen published in 2016, widely known as the  “Ice Melt” paper.

The Ice Melt paper, published in the journal  Atmospheric Chemistry and Physics , provided a grim, sweeping vision of the Earth’s climate future, focused on the consequences of the melting Greenland and Antarctic ice sheets. Drawing largely on ancient climate data — similar to the new paper — it warned of rapid melting and sea-level rise on the order of several meters within the next century.

It also suggested that the rapid influx of cold, fresh meltwater into the sea could affect ocean circulation patterns and even cause a giant Atlantic current to shut down. That’s a controversial prediction  deemed unlikely by the IPCC , one that would have severe impacts on global weather and climate patterns if it actually happened.

The paper received mixed reactions from other climate scientists upon publication. Some praised the paper, while many suggested the findings were unrealistic.

Another 2016 paper , published by a different group of scientists, later found that the likelihood of an Atlantic current shutdown was relatively small and suggested that Hansen’s paper relied on “unrealistic assumptions.”

In his new paper, Hansen referred to that study as an “indictment” of Ice Melt. He also noted that the IPCC’s latest assessment report did not include Ice Melt’s predictions, an omission he likened in the new paper to a form of censorship.

“Science usually acknowledges alternative views and grants ultimate authority to nature,” the new paper states. “In the opinion of our first author (Hansen), IPCC does not want its authority challenged and is comfortable with gradualism. Caution has merits, but the delayed response and amplifying feedbacks of climate make excessive reticence a danger.”

Responding to critiques of his new paper, Hansen again suggested that “scientific reticence” — or a kind of resistance to new findings — is at play. He pointed to a  1961 paper by sociologist Bernard Barber  suggesting that scientists themselves can be resistant to scientific discovery.

Claims that his new findings are unrealistic, Hansen said, are “a perfect example of the category of scientific reticence that Barber describes as ‘resistance to discovery.’ It takes a long time for new results to sink into the community.”

Resistance to scientific findings is nothing new to Hansen. His 1988 testimony initially shook the political establishment — yet decades later, global climate action is still proceeding too slowly to meet the Paris climate targets.

When he first testified to Congress in the 1980s, Oppenheimer said, he expected that world governments would have started meaningful emissions reduction programs by the year 2000 or so.

“We didn’t get ahead of the impacts,” he said. “And that’s probably because people weren't willing to support strong governmental action in most countries … until they were getting clobbered by unusual and highly damaging, and in some cases unprecedented, climate events.”

He regards the current state of global climate action now with a mix of skepticism and optimism.

“We’re in the process of muddling through — we’re in a period where climate change is gonna be painful for a while, it’s gonna hurt a lot of people in a lot of places, but we can get out the other side,” he said. “I think we can get there. But will we?”

Hansen echoed his sentiments in starker terms.

He wrote that he’s been surprised by “the increase of anti-science no-nothing thinking in our politics.”

“That's why I focus on young people,” he added. “They need to understand the situation and take control.”

Reprinted from E&E News with permission from POLITICO, LLC. Copyright 2022. E&E News provides essential news for energy and environment professionals.

• Share full article

The Science of Climate Change Explained: Facts, Evidence and Proof

Definitive answers to the big questions.

Credit... Photo Illustration by Andrea D'Aquino

Supported by

By Julia Rosen

Ms. Rosen is a journalist with a Ph.D. in geology. Her research involved studying ice cores from Greenland and Antarctica to understand past climate changes.

• Published April 19, 2021 Updated Nov. 6, 2021

The science of climate change is more solid and widely agreed upon than you might think. But the scope of the topic, as well as rampant disinformation, can make it hard to separate fact from fiction. Here, we’ve done our best to present you with not only the most accurate scientific information, but also an explanation of how we know it.

How do we know climate change is really happening?

• How much agreement is there among scientists about climate change?
• Do we really only have 150 years of climate data? How is that enough to tell us about centuries of change?
• How do we know climate change is caused by humans?
• Since greenhouse gases occur naturally, how do we know they’re causing Earth’s temperature to rise?
• Why should we be worried that the planet has warmed 2°F since the 1800s?
• Is climate change a part of the planet’s natural warming and cooling cycles?
• How do we know global warming is not because of the sun or volcanoes?
• How can winters and certain places be getting colder if the planet is warming?
• Wildfires and bad weather have always happened. How do we know there’s a connection to climate change?
• How bad are the effects of climate change going to be?
• What will it cost to do something about climate change, versus doing nothing?

Climate change is often cast as a prediction made by complicated computer models. But the scientific basis for climate change is much broader, and models are actually only one part of it (and, for what it’s worth, they’re surprisingly accurate ).

For more than a century , scientists have understood the basic physics behind why greenhouse gases like carbon dioxide cause warming. These gases make up just a small fraction of the atmosphere but exert outsized control on Earth’s climate by trapping some of the planet’s heat before it escapes into space. This greenhouse effect is important: It’s why a planet so far from the sun has liquid water and life!

However, during the Industrial Revolution, people started burning coal and other fossil fuels to power factories, smelters and steam engines, which added more greenhouse gases to the atmosphere. Ever since, human activities have been heating the planet.

Where it was cooler or warmer in 2020 compared with the middle of the 20th century

Global average temperature compared with the middle of the 20th century

+0.75°C

–0.25°

30 billion metric tons

Carbon dioxide emitted worldwide 1850-2017

Rest of world

Other developed

European Union

Developed economies

Other countries

United States

E.U. and U.K.

We are having trouble retrieving the article content.

Please enable JavaScript in your browser settings.

Thank you for your patience while we verify access. If you are in Reader mode please exit and  log into  your Times account, or  subscribe  for all of The Times.

Thank you for your patience while we verify access.

Already a subscriber?  Log in .

Want all of The Times?  Subscribe .

Advertisement

Loading metrics

Open Access

Peer-reviewed

Research Article

Global greenhouse gases emissions effect on extreme events under an uncertain future: A case study in Western Cape, South Africa

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – original draft

* E-mail: [email protected]

Affiliation Department of Civil and Environmental Engineering, Vanderbilt University, Nashville, Tennessee, United States of America

Roles Supervision, Writing – review & editing

Affiliation Drinking Water Justice Lab, Vanderbilt University, Nashville, Tennessee, United States of America

• Bowen He, 
• Ke Jack Ding

• Published: January 20, 2023
• https://doi.org/10.1371/journal.pclm.0000107
• Reader Comments

The growing effect of CO2 and other greenhouse gas (GHG) emissions on the extreme climate risks in the Western Cape, South Africa, calls for the need for better climate adaptation and emissions-reduction strategies to protect the region’s long-term social-economic benefits. This paper presents a comprehensive evaluation of changes in the future extreme events associated with drought and heatwave under three different greenhouse gas (GHG) emissions scenarios: Representative Concentration Pathway (RCP) 2.6, RCP 4.5, and RCP 8.5, from moderate to severe, respectively. Various diagnostic indices were used to determine how future heatwaves and drought will respond to each different RCP climate scenario in Western Cape based on Max Planck Institute-Earth System Model/REMO (MPI-ESM/REMO). The projected simulation results revealed that drought and heatwave extreme climate indices suggest strong relationships between future extreme climate risks and GHG emissions for Western Cape, South Africa. Anthropogenic activities and growing GHG emissions will lead to severer extreme climate stress in terms of drought and the duration, frequency, and magnitude of heatwave stresses. As a result, we believe that reducing the GHG emissions to alleviate future extreme climate stress becomes a practical solution to protect the local’s socio-economic system and further maintain the region’s economic prosperity.

Citation: He B, Ding KJ (2023) Global greenhouse gases emissions effect on extreme events under an uncertain future: A case study in Western Cape, South Africa. PLOS Clim 2(1): e0000107. https://doi.org/10.1371/journal.pclm.0000107

Editor: Fredolin Tangang, Universiti Kebangsaan Malaysia, MALAYSIA

Received: September 13, 2021; Accepted: December 15, 2022; Published: January 20, 2023

Copyright: © 2023 He, Ding. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data and scripts in this study are available at Github ( https://github.com/bowenhesteven/extreme_climate_indices_Western_Cape ).

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Extreme climate events, such as heatwaves and widespread drought, have a strong potential to cause extensive damage and impact cities worldwide. Studies regarding heatwaves revealed that the increasing frequency and intensity of heatwaves were seriously affecting both developed and developing countries [ 1 ]. For example, in 2003, an extreme heatwave event occurred in Europe and caused nearly 20,000 deaths [ 2 ]. A study regarding droughts based on historical records reported that droughts reduced global crop production by 10% from 1963 to 2007 [ 3 ] and global climate models have predicted that drought conditions will intensify in major breadbaskets of wheat and maize globally [ 4 ]. Besides, recent research led by Cooperative Institute for Research in Environmental (CIRES) found that in the southern plains and southwest of the U.S., as soil moisture vanishes during severe droughts, cooling by evapotranspiration is more severely curtailed during droughts, suggesting strong interactions exist between heatwaves and drought, and occurrence of one may induce and even amplify the other [ 5 ]. Thus, when evaluating a region’s future extreme climate risks, it is necessary to consider both heatwaves and drought events [ 5 ]. Furthermore, these climate-related extreme events are increasing in frequency and magnitude due to anthropogenic climate change, and there is increased potential for greater impacts and damage due to the location of urbanization and the ongoing expansion of urban centers and infrastructures [ 6 ]. Thus, it is critical that we understand how and when these extreme events might occur and how they respond to different anthropogenic activities.

South Africa has always been one of the most vulnerable regions to climate change-induced extreme weather events. The region’s mean annual temperature has increased by at least 1.5 times the observed global average of 0.65°C during the past five decades [ 7 , 8 ]. Furthermore, other studies have shown that mean temperature across the subtropics, and central tropical Africa are rising at about double the global rate [ 9 ]. Changes in the features of extreme weather and climate events have also been observed over recent decades in South Africa [ 10 ]. Previous studies have shown the region’s averages of heatwaves frequency, duration, and intensity are increasing in association with the increasing mean temperature [ 11 ]. However, that may not necessarily be the case on a regional scale, as temperature varies from place to place depending on factors such as latitude, sea-level elevation, or prevailing weather conditions [ 11 ]. For Western Cape Province, the lack of rainfall, decreasing storage levels in major reservoirs, and increasing water demand driven by rapid population growth and urban expansion, combined with problematic and ineffective water management practices, make the region particularly vulnerable to extreme weather events induced by global warming and climate change [ 12 , 13 ].

Many studies that focused on the regional future climate risks such as drought only used simple climate characteristics such as precipitation and maximum surface air temperature to evaluate climate risk, overlooking the role of extreme events’ impact on the region’s social-economic system. For example, He and Ding [ 14 ] used a high-resolution GCM-RCM Coordinated Regional Downscaling Experiment Simulations (CORDEX) model chain to highlight the impact of GHG emissions on Western Cape’s local climate system. They stressed that efficient water-management practices and GHG emissions reduction strategies are vital to mitigate more severe droughts such as the “Day-Zero” crisis in 2018, especially for the City of Cape Town and several other coastal regions within the Overberg and Eden district. While their study highlights the projections of a drying and high-heat South Africa region, they only analyzed the region’s future climate risks from a series of basic climate metrics such as precipitation, near-surface air temperature, daily maximum surface air temperature, ignoring the combined effects of these climate signals that can cause extreme events such as drought and heatwaves. Molina et al. [ 15 ] employed Euro-CORDEX simulations to fully assess future heatwaves in the Mediterranean region. They highlighted that forcing global models and emissions scenarios play a significant role in future heatwaves in the Mediterranean region. While their study adopted comprehensive extreme events indices such as Warm Spell Duration Index (WSDI) to investigate future heatwaves, they only considered two diagnostic methods, including WSDI and Heatwave Magnitude Index daily (HWM) to study the future heatwave features from limited aspects. Besides, they only considered two emissions scenarios, RCP 4.5 and RCP 8.5. The article could benefit from including more climate scenarios in lower GHG concentrations, such as RCP 2.6, to provide a more comprehensive GHG mitigation strategy that can help alleviate future extreme climate stress. Sillmann and Roeckner [ 16 ] calculated comprehensive indices for temperature and precipitation extremes based on the GCM-RCM model chain Coupled Climate model consisting of the Atmospheric general circulation Model and the Max-Planck-Institute Ocean Model (ECHAM5/MPIOM) simulations. They concluded that in the climate projections for the twenty-first century, all considered temperature-based indices show a significant increase worldwide [ 16 ]. Although their research delivered integrated knowledge of future extreme events from all possible perspectives, their global-level results can hardly be downscaled to inform regional hotspots of climate change, such as Cape Town, South Africa.

The overarching goal of this study is to quantify the effect of climate change on future extreme events in Western Cape Province at the sub-regional level under three GHG emissions scenarios: RCP 2.6, RCP 4.5, RCP 8.5.

Representative concentration pathway (RCP) 2.6 is a “very stringent” and optimistic pathway that requires that CO 2 emissions start to decline by 2020 and go to zero by 2100 [ 17 ]. The scenario requires negative CO 2 emissions that can be achieved by influential policy approaches and innovative Direct Air Capture (DAC) technologies such as liquid solvent systems [ 18 ].

RCP 4.5 is a scenario of long-term, global emissions of greenhouse gases, short-lived species, and land-use-land-cover stabilizing radiative forcing at 4.5 W/m 2 up to the year 2100 [ 19 ]. RCP 4.5 is a stabilization scenario that assumes that climate policies are invoked to achieve the goal. Like RCP 2.6, RCP 4.5 requires negative CO 2 emissions that can be achieved by shifting to lower emissions energy technologies and deploying carbon capture and geologic storage technology [ 19 ].

RCP 8.5 is a scenario that does not include any specific climate mitigation strategy. People are conducting business as usual, leading to a radiative forcing of 8.5 W/m 2 by the end of this century [ 20 ]. Hence, it represents the upper bound of the RCPs’ set system, which can also be called the business-as-usual scenario [ 20 ].

By conducting this research, we aim to answer the following research questions: (1) How do different GHG emission scenarios impact future extreme events such as drought and heatwaves in the Western Cape region? (2) How will different districts within Western Cape respond to the 3 GHG emission scenarios in terms of future extreme events such as drought and heatwaves? Various diagnostics methods were applied to project features of these extreme events in the future (2021–2100). Therefore, we can better understand how these extreme weather events will change by the end of the 21 st century under different emissions scenarios. The paper is structured as follows: Section 2 describes the data and methods used in the study, Section 3 presents the results and discussions, and Section 4 delivers the conclusions and implications of the study.

Study region

We chose Western Cape Province in South Africa as our study region, which is in the southernmost section of Southern Africa ( Fig 1 ). It is surrounded by Northern Cape Province and Eastern Cape Province, as well as the Atlantic Ocean in the west and the Indian Ocean in the south. It ranges from 15.0° E and 25.0° E longitudinally and 30.0° S to 35.0° S latitudinally. The Western Cape accounts for 12% of South Africa’s total agricultural area, provides 20% of the nation’s total agricultural production outputs, and nurtures a world-famous wine appellation [ 21 , 22 ]. The climate conditions across the region are the temperate Mediterranean, with warm, dry summers and mild, moist winters, rendering it favorable for cereal farming such as wheat, oats, barley, and viticulture [ 23 – 25 ]. Average summer temperatures range from 5°C to 27°C, while winter temperatures range from 5°C to 22°C [ 26 ]. The Western Cape is one of South Africa’s driest regions, with approximately 350 mm of annual precipitation, well below the national yearly average of 500 mm precipitation [ 27 ]. Precipitation is also highly heterogeneous and varies greatly, from semi-arid areas to relatively wet areas on the windward slope of mountains [ 28 ].

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

The study domain: ( a ) Six main districts (City of Cape Town, West Coast, Cape Winelands, Overberg, Cape Winelands, Central Karoo) in the Western Cape region (South Africa). The blue solid line delineates the political boundary of each main district in the Western Cape. ( b ) The Western Cape locates in the southernmost part of South Africa. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g001

Western Cape is the fourth largest of the nine provinces and the third most populous province, with an estimated 7 million inhabitants in 2020 [ 29 ]. About two-thirds of these inhabitants live in the metropolitan area of Cape Town, which is also the provincial capital [ 29 ]. Western Cape is also the second-largest contributor to the country’s total GDP and one of the fastest-growing economies in the county (Statistics South Africa, 2020).

In this study, we obtained the CORDEX “Phase 1” simulation data from the Earth System Grid Federation (ESGF). The CORDEX models have been proven to correctly capture the spatial distribution of major climate variables over the Western Cape region and reproduce the essential climatic features in the observed temperature and moisture fields [ 23 ]. Thus, they are reliable models to predict future extreme events over the Western Cape region. “Phase 1” data were made available at the daily temporal resolution, 0.44-degree spatial resolution, by far the most comprehensive GCM-RCM downscaled data available. We downloaded the data under AFR-44, which indicates the African continent with 0.44-degree downscaling. We selected several variables, including pr (precipitation), tasmax (daily maximum near-surface air temperature), and tasmin (daily minimum near-surface air temperature) as key input variables to calculate a variety of extreme events indices such as WSDI (heatwave), Consecutive Dry Days (CDD, drought) in Western Cape. RCP 2.6, RCP 4.5, and RCP 8.5 were selected as experiment configurations, and daily data were downloaded. We selected MPI-M-MPI-ESM-LR as the driving model because it is currently the only driving model available for all three RCP scenarios. Moreover, its overall performance is better than its predecessor ECHAM5/MPIOM model, based on a modified Reichler-Kim standardized error due to improvements in the extratropical circulation [ 30 ]. Furthermore, many previous studies have verified the credibility and advantage regarding the MPI-ESM-LR-REMO (GCM-RCM) chain on the projection of climate change signals over different CORDEX regions [ 31 , 32 ]. Complete information regarding three downscaled GCM-RCMs input simulation data in this study is summarized in Table 1 .

https://doi.org/10.1371/journal.pclm.0000107.t001

Daily spatial climate data of the African continent for 3 RCP scenarios were imported and analyzed in the open-source program RStudio [ 33 ]. The aim of using RStudio at this step was to quickly retrieve the climate variable data of interest as input simulation data for further calculation. Specifically, we used the “ncdf4” package in RStudio to retrieve the climate characteristics data such as precipitation, daily maximum near-surface air temperature, and so forth for each RCP scenario for the whole African continent. We used the “ncks” command from NetCDF Operators (NCO) to downscale the data from the African continent to Western Cape, South Africa, to focus on our study area. The downscaled data were then imported as input simulation data into the ClimPACT2 package to be further analyzed and calculated to obtain a series of extreme events indices associated with drought and heatwave. ClimPACT2 is a powerful R software package that calculates the Expert Team on Sector-Specific Climate Indices (ET-SCI) and additional climate extremes events indices from data stored in text or network Common Data Form (netCDF) files.

In this study, we used CDD and Standardized Precipitation Evapotranspiration Index (SPEI) with a scale of 12 months to assess droughts. The CDD index has been extensively used in the climate literature to assess dryness and its impacts on the coupled human-natural systems [ 34 , 35 ]. For instance, Marengo et al. [ 36 ] used CDD to review, evaluate, and predict the drought situation in Northeast Brazil for the past, present, and future, respectively. The standardized precipitation-evaporation index (SPEI) is a comprehensive index that measures drought using precipitation and temperature data with the advantage of combining multiscalar characters to include the effects of temperature variability on drought assessment [ 37 ]. The index is an extension of the widely used Standardized Precipitation Index (SPI) that takes potential evapotranspiration (PET) into account [ 37 ]. A previous study by Nail and Abiodun [ 23 ] has confirmed the superiority of SPEI over standardized precipitation index (SPI) that SPI projections may underestimate the influence of global warming on drought because it doesn’t account for the effect of potential evapotranspiration (PET).

Besides, we adopted the Warm Spell Duration Index (WSDI) and five different heatwave indices (HWI) to evaluate heatwaves from several various aspects, including amplitude (HWA), duration (WSDI, HWD), frequency (HWF), magnitude (HWM), and quantities (HWN). WSDI is a popular extreme climate index that measures the duration aspect of heatwaves in many previous studies. For instance, Chen and Sun [ 38 ] adopted WSDI to evaluate heatwaves in China. They concluded that the risks of unprecedented heatwaves would be devastating for the country under the background of increasing GHG emissions and global warming. HWA is an index that measures the hottest day (amplitude) of the hottest event, and it has been widely adopted in previous studies [ 39 , 40 ]. HWD is an index that measures the length of the longest heatwave event. HWF is an index that measures the total number of days satisfying the heatwave criteria. HWM is an index that measures the mean heatwave events’ intensity, calculated by averaging the temperature from all participating heatwave days. HWN is an index that quantifies the total number of defined heatwave events. Complete information regarding the ET-SCI extreme events indices investigated in this study is summarized in Table 2 .

https://doi.org/10.1371/journal.pclm.0000107.t002

It should be noted that all HWI in this study are custom percentile-based threshold indices that user needs to specify the threshold on their own. The reference percentiles used in calculating percentile-based indices were computed over a base period of 1961–1990 for long-term climate change assessments as recommended by the World Meteorological Organization [ 41 ]. To exhibit the spatial-temporal patterns of these extreme events indices, we averaged each extreme climate index every 20 years for each 0.44 by 0.44-degree inside the study area. The averaged extreme events indices data then were mapped for four two-decade spans from 2021 to 2100 for each of the three climate scenarios.

Results and discussions

Spatial-temporal patterns of drought indices.

In this section, each extreme weather index from Table 2 under three different GHG emissions was spatially evaluated using outputs from three downscaled GCMs ( Table 1 ). The results are well aligned with the previous studies by Naik and Abiodun [ 23 ], of which the projected changes in drought characteristics over the Western Cape show a robust drying signal under the RCP 8.5 emission scenario. Furthermore, He and Ding [ 14 ] also demonstrated that great potential of reducing climate risks and vulnerability exists in lowering GHG emissions for Western Cape region. Besides, by directly investigating extreme weather indices such as drought indices like CDD and heatwave indices like WSDI, this study better illustrates how emission reduction will protect the region’s socio-economic systems by alleviating future extreme climate stress.

The CDD value ranges from 11 to 79 days for the whole Western Cape area, with the inland districts such as West Coast experiencing the longest dry days and coastal districts such as Overberg experiencing the shortest ( Fig 2 ). There is no significant variability for CDD value distribution under the scenario of the RCP 2.6 and the RCP 4.5 since few dry spots can be found from 2081 to 2100 compared to that of 2021–2041. However, the whole region is expected to experience much longer dry days under the RCP 8.5 scenario as opposed to a more heterogenous response in lower emission scenarios with much longer dry days for the inland districts such as West Coast and Central Karoo, and relatively shorter dry days for the coastal regions such as Overberg and Eden. This indicates that GHG emissions scenarios and geospatial characteristics, such as the distance to the ocean, might play an essential role in determining the duration aspect of drought for a region. In addition, the SPEI model projections indicate that the drought intensity increases across the Western Cape for all emissions scenarios, but the drying patterns are not homogeneous across the region ( Fig 3 ). Projections of the SPEI drought intensity under the RCP 2.6 and the RCP 4.5 suggest that a more intense drying signal might occur towards the inland districts such as West Coast and Central Karoo, which aligns well with the previous CDD distribution pattern ( Fig 2 ). Nonetheless, the drought intensity situation will worsen for the whole Western Cape region under the RCP 8.5 emissions scenario, with the whole area experiencing much more severe drought intensity and not many differences between the inland and coastal regions. Combined with the previous CDD spatial distribution, we consider that spatial location, such as the distance to the ocean, might have less significance in determining the intensity aspect of drought compared to drought duration for the region. Thus, the emissions scenario becomes the dominant factor affecting the region’s drought intensity.

Spatial-temporal patterns of CDD in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the CDD projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g002

Spatial-temporal patterns of 12-month scale SPEI in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the SPEI projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g003

Spatial-temporal patterns of heatwaves indices

The projections from three climate scenarios reveal that WSDI values in Western Cape range from 1 to 211 days ( Fig 4 ). There is no significant variability for WSDI value distribution under the RCP 2.6 and the RCP 4.5 since few noticeable differences are found from 2081 to 2100 compared to that of 2021–2041 ( Fig 4 ). However, the whole region will experience a much longer period of hot days under the scenario of the RCP 8.5 compared to lower emission scenarios with much longer hot days for the coastal districts such as the City of Cape Town and Overberg, and relatively shorter hot days for the inland regions such as the West Coast and Central Karoo. The strongest increase in WSDI occurs in urban regions such as the City of Cape Town underscores the anthropogenic influence, such as population growth and urban heat island effect on the high-temperature extreme climate events for the Western Cape area.

Spatial-temporal patterns of WSDI in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the WSDI projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g004

The observed projections of HWA range from 21°C to 49°C ( Fig 5 ). The variabilities between the three emissions scenarios are smaller than drought-related indices such as the CDD and the SPEI, especially for the differences between the RCP 8.5 and the lower emissions scenarios ( Fig 5 ). However, the extremely high temperature is not homogeneous across the region, with the inland areas such as the Cape Winelands and the Central Karoo likely to experience days with higher extreme hot temperatures in the future.

Spatial-temporal patterns of HWA in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the HWA projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g005

HWD projections range from 3 to 35 days with an approximately homogeneous distribution across the region under the RCP 2.5 and the RCP 4.5 scenarios ( Fig 6 ). Besides, the variabilities between the three emissions scenarios are small, with limited regions near the coast, such as the West Coast and the Overberg, likely to experience an extreme long period of heatwave events during the last 20 years of the 21 st century only under RCP 8.5 ( Fig 6 ). Fig 7 displays that HWM ranges from 20°C to 43°C. The mean heatwave events’ intensity exhibits a heterogeneous spatial distribution that the inland regions such as the Cape Winelands and the West Coast will experience more extreme heatwave events. Fig 7 also shows a non-significant variability across the time dimension with similar mean heatwave events’ intensity distribution between 2021–2040 and 2081–2100 for all three emissions scenarios.

Spatial-temporal patterns of HWD in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the HWD projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g006

Spatial-temporal patterns of HWM in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the HWM projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g007

HWF ranges from 2 to 105 days ( Fig 8 ). The projections exhibit a homogeneous spatial distribution across the region for the RCP 2.6 and the RCP 4.5 scenarios and a non-significant variability across the time dimension. However, HWF patterns prediction shows that there will be significant increases in the total number of days categorized as heatwaves for the Western Cape by the end of the 21 st century under the RCP 8.5 scenario. Besides, Fig 8 displays that some inland regions, such as the Cape Winelands, the northeastern part of the Central Karoo, as well as limited areas close to the coast in the southern part of the Overberg district, will likely experience more hot days by the end of the 21 st century under the RCP 8.5 emission scenario. This finding aligns well with the HWD pattern displayed in Fig 6 .

Spatial-temporal patterns of HWF in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the HWF projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g008

HWN is an index that quantifies the total number of defined heatwave events. The HWN projections reveal that inland districts such as the northern part of the Cape Winelands, the northeastern part of the Central Karoo as well as the southern part of the Overberg will likely experience more heatwave events ( Fig 9 ), which aligns well with the HWF patterns shown in Fig 8 . There is a strong increasing trend of HWN across the region for the study period under the RCP 8.5, with more heatwave events occurring in the last 2081–2100 compared to the beginning.

Spatial-temporal patterns of HWN in Western Cape, South Africa (for 2021–2100) for 3 emission scenarios: (a-l) exhibit the HWN projection pattern for the each 20-year span from 2021–2100 under the 3 different GHG emission scenarios. Sources: Esri.

https://doi.org/10.1371/journal.pclm.0000107.g009

Regional trends of heatwave indices

We used the 10-year moving average time series of 5 heatwave indices for each district in Western Cape to assess the region’s overall temporal trend of heatwave conditions. Fig 10 displays the 10-year moving average time series of 5 heatwave indices (HWA, HWD, HWF, HWM, HWN) in 6 districts of Western Cape, South Africa, under three emissions scenarios. All five indices show a distinct pattern of increasing trends for all aspects of heatwaves for each district in the Western Cape region, from lower GHG emissions to higher GHG emissions scenarios. However, the magnitude of the increase for each heatwave index may vary tremendously for a different district. For example, Fig 10 exhibits that the amplitude (HWA) of heatwave events in the City of Cape Town is projected to increase by 9.6%. In comparison, the duration (HWD) of heatwave events in the district is projected to increase by over 300%. This indicates that different districts have various sensitivities to GHG emissions associated with heatwave features such as intensity and duration in Western Cape, South Africa. One possible reason that heatwave events will be severer for the City of Cape Town under the RCP 8.5 scenario than the other regions is that high buildings in urban areas cause urban heat island effect and generation of heat, making the urban center several degrees warmer than its surrounding areas.

10-year moving average time series of heatwave indices in 6 districts of Western Cape, South Africa (for 2021–2100) under 3 emission scenarios: (a-o) exhibit the 10-year moving average time series in 6 districts of Western Cape, Africa for each the heatwave index from 2021–2100 under the 3 different GHG emission scenarios.

https://doi.org/10.1371/journal.pclm.0000107.g010

Impact of GHG emissions on socio-economic system

Based on the results shown above, it is more likely to expect severer droughts and heatwaves across the Western Cape region when doing business as usual, as opposed to taking serious actions and making policies to reduce GHG emissions. Such impact will be realized on natural and human systems, such as agricultural production, and the wellbeing of the people living in the region. For communities with limited financial, managerial, and technical resources to support climate adaptation, like the Western Cape area, the effect of GHG emissions on the local’s socio-economic system can be further exacerbated [ 42 ]. Numerous studies have demonstrated that extreme climate events such as droughts can devastate smallholder farmers, directly impacting local food security [ 42 ]. And on top of the drought, heatwaves further accelerate the moisture evaporation of soil and crops, ultimately putting a toll on the crop’s yield [ 43 ].

Impact of drought and heatwaves on agriculture production

Previous studies have proven that crop production needs to adapt to the agricultural environment in the background of climate change [ 44 ]. As a region to produce over 20% of the nation’s total agricultural production outputs mainly consists of wheat and oats, it is necessary to maintain the sustainability of the region’s agriculture production, including irrigation and soil microbial community. Nonetheless, the business-as-usual emissions scenario can prevent the sustainability of the region’s agriculture production from prolonging. For instance, although wheat is generally a cool season crop and does not require much water, it does need between 12 and 15 inches of rain over a growing season to produce a good crop which cannot be satisfied under the scenario of RCP 8.5 by the end of the 21 st century [ 45 ]. Oats need more water than most other grains. Previous studies have proved that severe drought spell during the sensitive period of oat production season has a quantifiable negative effect on yield, and the drought is one of the most essential key causes of interannual oat yield variability [ 46 ]. In addition, because of dry and heat interactions, extreme heat was much more damaging in arid than in normal conditions for crops like maize, soybeans, and wheat. Based on Matiu et al. [ 47 ] study, drought significantly decreased global yields of maize, soybeans, and wheat by 11.6%, 12.4%, and 9.2% respectively combined with the effects of high temperatures.

SPEI is proven to be a sensitive indicator of crop yield. A negative SPEI indicating a mild drought can reduce the crop yield from 20% to 70% based on the location and type [ 48 ]. Based on our study results, the West Coast district will suffer extremely dry under the RCP 8.5 scenario based on the SPEI at the end of this century ( Fig 3(L)) . As one of the main wheat production places in the Western Cape, it is reasonable to predict that the impact of the future drought on the district’s wheat production will be significant. Moreover, combined with Fig 10(I) and 10(O) that predicts over 50% increase in heatwave frequencies and events, the additive effect of extreme heat will make the future crop yield in the West Coast district alarming if we keep the business as usual.

In addition to sufficient water to grow crops, soil function and microbial diversity are critical to maintaining a region’s agricultural efficiency. Previous studies have demonstrated that combined heat-drought climate stresses may induce different microbial responses that may damage the soil function to high yields compared to those observed individual extreme climate events such as heatwaves or drought alone [ 49 ]. Bindi [ 50 ] also illustrated that to maintain agricultural sustainability in the background of global warming, it is necessary to consider the adaptation options with the multifunctional role of agriculture, and extreme climate stresses. From another perspective, climate change-induced extreme climate stresses are increasingly affecting the global agricultural system from all possible aspects.

Regional extreme climate events including heat waves and prolonged droughts under high carbon emission scenarios are one of the biggest challenges the wine industry faces today [ 51 ]. As home to most of the South African wine industry, and the country’s most famous wine regions, Stellenbosch and Paarl, wine production serves as one of the most vital economic activities. Many previous studies found that extreme climate events such as heatwaves and drought caused by high GHG emissions will severely challenge the ability to grow grapes adequately and produce quality wine in the region [ 24 , 52 ]. As Araujo [ 24 ] demonstrated that the impact of drought on grape yields highly coincides with the intensity of droughts, and high temperatures and low rainfall during summer and winter can cause additional stress on the grapevines through increased diseases, inadequate chill units, and other physiological stress resulting from inadequate water uptake. The future is not optimistic for the district of Cape Winelands as the most crucial district to produce grapes and quality wine. Fig 3(L) projects that the extreme dry will be a considerable concern for the wine industry. Combined with Fig 10(F) , 10(I) and 10(O) which shows over 50% increase in heatwave duration, frequency, and events, the impacts of drought on grapes could be further amplified. Reducing the GHG emissions to alleviate future extreme climate stress could effectively reduce the risk of climate change on the local wine industry and further maintain the prosperity of the local economy.

Impact of heatwaves on human health

Heatwave events can be dangerous to human health. Prolonged exposure to extreme heat can trigger various health conditions, such as heat stroke, heat exhaustion, heat cramps, and even death. Besides, higher temperatures and respiratory problems are also linked since higher temperatures contribute to the build-up and spread of harmful air pollutants [ 53 ].

Studies found that heatwaves contribute to the increase in mortality rate across the world. Older people are more vulnerable to the effects of extreme heat through a range of physiological and physical factors. For example, Arbuthnott and Hajat [ 54 ] found that the 1995 heatwave in the U.K. was estimated to have caused an increase in mortality of 8.9% over England and Wales, and a 16% increase in the Greater London area. They further informed that in 2013, there was an estimated total of 195 excess deaths across all heatwave days in those older than 65 years, with an excess of 10 deaths per heatwave day [ 54 ]. Based on a U.S. Centers for Disease Control and Prevention (CDC) study, extreme heat can be blamed for an average of 688 deaths each year in the U.S. Thus, we are confident to speculate the mortality rate in the Western Cape region will also increase significantly in the RCP 8.5 scenario. Taking the City of Cape Town as an example, Fig 10(F) predicts that both the heatwave duration and frequency in the City of Cape Town will increase by over 300% under the RCP 8.5 scenario compared to the RCP 2.6 scenario. For a major municipal city with over 4 million population and 6.2% elderly people, the impact of heatwaves on the city’s mortality will be significant if we keep business as usual. Thus, we consider lowering the GHG emissions will be a practical solution for alleviating the mortality rate in the City of Cape Town.

Conclusions

In this study, Co-ordinated Regional Climate Downscaling Experiment (CORDEX) data were used to examine the potential impacts of various greenhouse gas emissions (RCP 2.6, RCP 4.5, RCP 8.5) on the future extreme climate stress in six districts within the Western Cape, South Africa. The global simulations have been downscaled with REMO for the future 80 years of the 21 st century. The CDD and SPEI were used to predict the future drought condition in the Western Cape region. In addition, the WSDI, HWA, HWD, HWF, HWM, HWN were used to predict the various aspects of future heatwave conditions for the region. Both spatial-temporal analysis and regional-average analysis suggest the following:

• Both drought-related extreme climate indices and heatwave-related extreme climate indices indicated a solid relationship between the future extreme climate risks and the GHG emissions for Western Cape, South Africa. Anthropogenic activities and the growing GHG emissions will lead to severer extreme climate stress in terms of every aspect of drought and heatwave.
• The sensitivities between the different aspects of extreme climate events on GHG emissions may vary. In terms of drought and heatwave, the duration aspect is generally more sensitive to the intensity aspect in Western Cape, South Africa.
• Climate change poses a localized effect on the different regions in Western Cape. The spatial location factor, such as the distance to the ocean, plays a critical role in determining the impact on various aspects of extreme climate risks from GHG emissions. In terms of drought, the inland regions will suffer from longer and severer drought risks. For heatwaves, both the inland regions and the coastal regions will experience more hot days in terms of intensity and duration.
• Under the influence of climate change and global warming, the co-existence of different extreme climate events, such as heatwaves and drought, may exhibit amplifying effects. Efficient water-management practices and greenhouse gas emissions reduction strategies are urgently needed to avoid more severe droughts such as the “Day-Zero” crisis in 2018, especially for the City of Cape Town and several other coastal regions within the Overberg and Eden district. Besides, local government should consider local conditions such as the spatial location and economic development when making decisions to maintain long-term sustainability.

This study highlights the importance of the impact of GHG emissions scenarios (RCP 2.6, RCP 4.5, and RCP 8.5) on the regional extreme climate events simulated by the GCM-RCM model chain MPI-ESM/REMO. The results sufficiently demonstrated the great potential of mitigating extreme climate risks and vulnerability under lower GHG emissions scenarios for the Western Cape region. Since this is the only GCM-RCM model combination available, future studies need more available data products from other downscaled climate models to consider uncertainties of the models, so an integrated comprehensive evaluation of possible extreme climate events signals can be conducted in the region. When more downscaled climate simulation data across different RCPs are available for Africa, it will enable uncertainty analyses and quantification of future extreme climate risks using a large ensemble of simulations.

• 1. Field CB, Barros V, Stocker TF, Qin Dahe. Managing the risks of extreme events and disasters to advance climate change adaptation: special report of the intergovernmental panel on climate change. Cambridge: Cambridge University Press; 2012.
• View Article
• PubMed/NCBI
• Google Scholar
• 4. Seneviratne S, Nicholls N, Easterling D, Goodess C, Kanae S, Kossin J, et al. Changes in climate extremes and their impacts on the natural physical environment. In: Field CB, Barros V, Stocker TF, Qin Dahe. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC). Cambridge: Cambridge Univ. Press; 2012. pp. 109–230.
• 21. Directorate StatisticsAnalysis, Economic. Abstract of Agricultural Statistics: 2013. Department of Agriculture, Forestry and Fisheries, Republic of South Africa (DAFF), Pretoria. 2013. Available from: https://www.daff.gov.za/docs/statsinfo/Abstact2013.pdf .
• 22. Metelerkamp L. Commercial agriculture in the Swartland: Investigating emerging trends towards more sustainable food production. Doctoral. Thesis, University of Stellenbosch. 2011. Available from: https://core.ac.uk/download/pdf/37326554.pdf .
• 25. Midgley GF, Chapman RA, Hewitson B, Johnston P, De Wit M, Ziervogel G, et al. A status quo, vulnerability and adaptation assessment of the physical and socio-economic effects of climate change in the Western Cape. Stellenbosch: CSIR Environmentek; June 2005 [cited 2021 Aug 28]. 170 p. Report No.: ENV-S-C 2005–073. Available from: https://www.africaportal.org/documents/9677/05Midgeley-etal_CC_in_Western_Cape.pdf .
• 26. Hurry L, Van Heerden J. Southern Africa’s Weather Patterns: A Guide to the Interpretation of Synoptic Maps. 1st ed. Cape Town: Via Afrika; 1982.
• 29. Statistics South Africa. Mid-Year Population Estimates. Department: Statistics South Africa. 2020 [cited 2021 July 2]. Available from: http://www.statssa.gov.za/publications/P0302/P03022020.pdf .
• 33. R. Studio. RStudio: integrated development environment for R. RStudio Inc. Boston, MA. 2012.
• 41. World Meteorological Organization. WMO guidelines on the calculation of climate normals. WHO. 2017 [cited 2021 Sep 16]. Available from: https://library.wmo.int/doc_num.php?explnum_id=4166 .

Open Access is an initiative that aims to make scientific research freely available to all. To date our community has made over 100 million downloads. It’s based on principles of collaboration, unobstructed discovery, and, most importantly, scientific progression. As PhD students, we found it difficult to access the research we needed, so we decided to create a new Open Access publisher that levels the playing field for scientists across the world. How? By making research easy to access, and puts the academic needs of the researchers before the business interests of publishers.

We are a community of more than 103,000 authors and editors from 3,291 institutions spanning 160 countries, including Nobel Prize winners and some of the world’s most-cited researchers. Publishing on IntechOpen allows authors to earn citations and find new collaborators, meaning more people see your work not only from your own field of study, but from other related fields too.

Brief introduction to this section that descibes Open Access especially from an IntechOpen perspective

Want to get in touch? Contact our London head office or media team here

Our team is growing all the time, so we’re always on the lookout for smart people who want to help us reshape the world of scientific publishing.

Home > Books > Air Quality - Models and Applications

Variation of Greenhouse Gases in Urban Areas-Case Study: CO2, CO and CH4 in Three Romanian Cities

Submitted: 25 October 2010 Published: 05 July 2011

DOI: 10.5772/17985

Cite this chapter

There are two ways to cite this chapter:

From the Edited Volume

Air Quality - Models and Applications

Edited by Dragana Popovi?

To purchase hard copies of this book, please contact the representative in India: CBS Publishers & Distributors Pvt. Ltd. www.cbspd.com | [email protected]

Chapter metrics overview

2,942 Chapter Downloads

Impact of this chapter

Total Chapter Downloads on intechopen.com

Total Chapter Views on intechopen.com

Overall attention for this chapters

Author Information

Iovanca haiduc *.

• Babeş-Bolyai University, Faculty of Environmental Science,Cluj-Napoca, Romania

Mihail Simion Beldean-Galea

*Address all correspondence to:

1. Introduction

The natural equilibrium of atmospheric gases has been maintained for millions of years, but with the beginning of the industrial age, it became more fragile due to human activity. In the Intergovernmental Panel on Climate Change Report (IPCC) named “Climate Change 2007” ( IPCC-AR4, 2007 ) it is specified that “ the keep going emissions of the greenhouse gases (GHG) at/over current rate, will cause in the future global warming and will induce more global climate changes in the 21st century than those of 20th century” . More than, in the coming IPCC Report named “Carbon cycle including ocean acidification (CCT)” ( IPCC-AR5, 2010 ) is stipulated that ocean acidification will be a further critical and direct consequence of increasing atmospheric GHG concentrations.

In 1886, the chemist Svante Arrhenius (Nobel prize for Chemistry in 1903) calculated for the first time the CO 2 contribution (from fossil fuel combustion) to climatic changes and used for the first time the term of “greenhouse effect”. Almost 100 years were necessary for the confirmation of Arrhenius predictions about the evolution of global climatic factors, and the fact that CO 2 is the main greenhouse gas, with a contribution of 55% to Global Warming Effect.

The first IPCC Report (IPCC-FAR, 1990) draws the conclusion about the possible existence of a global warming phenomenon. The second IPCC Report ( IPCC-SAR, 1995 ) shows the contribution of humans to global warming and predicts a major warming in the 21st century. The third IPCC Report ( IPCC-TAR, 2001 ) affirms a very probable (60% - 90%) warming for the next century. In the IPCC-AR4 Report ( IPCC-AR4, 2007 ) adds for the understanding of the impact of climate changes over the vulnerability and the adaptation of the environment, the most relevant scientific, technical and socio-economical information from more than 1500 scientific papers. This report accepts with a probability of over 90% that the emission of greenhouse gases and not the environmental conditions gives the global warming effect.

The IPCC Guide from 2006 makes an inventory of gases from atmosphere and distinguishes between:

gases with GWP (Global Warming Potential) listed in IPCC 2001: CO 2 , CH 4 , N 2 O, hydro fluorocarbons, per fluorocarbons, SF 6 , NF 3 , SF 5 CF 3 , halogenated ether, C 4 F 9 OC 2 H 5 , CHF 2 OCF 2 OC 2 F 4 OCHF 2 , CHF 2 OCF 2 OCHF 2 and other halocarbons CF 3 I, CH 2 Br2, CHCl 3 , CH 3 Cl, CH 2 Cl 2 .

gases without GWP: C 3 F 7 C(O)C 2 F 5 , C 7 F 16 , C 4 F 6 , C 5 F 8 and C 4 F 8 O.

As a follow-up of these reports, the scientific community had started a cycle of research programs having as scientific goal the complex study (emission sources, consumption sources, the balance of changes between the components of environment etc.) of these gases and the effect produced by them (Projects CARBOEUROPE, AEROCARD, CHIOTTO, Global Carbon Project, IGOS, NACP etc).

The CO 2 it is an important green-house gas and it level in the atmosphere has significantly increased from 280 ppm in the pre-industrial era to current 380 ppm. The first increase of 50 ppm occurred during a period of ca. 200 years, starting with the beginning of the industrial revolution until 1973. Between 1973 and 2006 the concentration of CO 2 increased with another 50 ppm.

The Global Warming Potential (GWP) for CO 2 is conventionally choose as 1, i.e. the atmospheric residence time between 50 and 200 years, and its contribution to the greenhouse effect is ca. 52 %.

According to IPCC Report ( IPCC-AR4, 2007 ), the contribution of anthropic CO 2 is predominant, and the main anthropic sources are:

the energetic sector 30 %

industrial processes 20 %

fuels used in transportation 20 %

burning of biomass 9.1 %

processing and distribution of fossil fuels 8.4 %

other sources 12.5%

According to IPCC estimations, the increase of CO 2 concentration in the atmosphere leads to climate changes and will produce a global warming of the planet through the greenhouse effect.

The monitoring of the global levels of CO 2 is the concern of American government since about 30 years. The National Oceanic & Atmospheric Administration - Earth System Research Laboratory (NOAA-ESRL) performs measurements for main greenhouse gases in about 68 locations spread all over the world. According to NOAA data, the concentration of CO 2 has an ascending trend. Figure 1 shows the CO 2 variations in Mauna Loa for the period of 2003-2008 according to NOAA-ESRL source.

Variation of the monthly average of CO 2 in Mauna Loa

At the European level, the Carboeurope-Clusters Program (CarboEurope) carries out measurements of atmospheric CO 2 concentrations in 61 locations in 17 countries. This program contains eight different projects working together to contribute to the understanding of the carbon cycle at the European level. The projects involved in this program are: AEROCARB, CAMELS, CARBOAGE, CARBODATA, CARBOEUROFLUX, CARBOEUROPE GHG; CARBOINVENT, CHIOTTO, CARBODATA, TACOS, EUROSIBERIAN CARBONFLUX, FORCAST, GREENGRASS, RECAB, TACOS-INFRASTRUCTURE, and TCOS SIBERIA. Two other projects, CARBOMONT and SILVISTRAT are associated to this program.

In 1998 Idso and co-workers ( Idso et al., 2001 ) introduced the term ““urban DOME” for the persistence of CO 2 over the urban cities as a result of anthropogenic contribution to CO 2 budget. After that, several individual studies regarding to CO 2 variation in urban areas have been reported ( Day et al., 2002 ; Idso et al., 2002 ). These studies showed that the concentration of CO 2 in urban area is higher that the CO 2 concentration in rural area and this fact is a consequence of human activities. A literature review about these results is presented in this chapter.

Methane (CH 4 ) is another important greenhouse gas, with GWP = 25 and a residence time of more than 100 years. It is produced both naturally and through human activities. The global mixing ratios of CH 4 in the atmosphere have more than doubled since the pre-industrial period, rising from around 750 ppb (parts per billion) in 1800 (Simpson et al., 2002; Dlugokencky et al., 2003) to the current level of around 1770 ppb (NOAA-ESRL). The global trend of the methane concentration in the air is ascending, with a rate of increase of 5–10 ppb/year, and for the period 1984-2004 this tendency is shown in Figure 2 .

The variation of the global average concentration of CH 4 . (source NOAA-ESRL)

The main natural source of methane are dominated by wetlands. The primary way for CH 4 transformation is its destruction in the atmosphere by hydroxyl radicals (Prinn et al., 1995, 2001). Some CH 4 is also oxidized by microorganisms (called methanotrophs), which use CH 4 as a source of carbon and energy. Tropospheric CH 4 is eventually oxidized to carbon dioxide; its atmospheric lifetime is estimated to be 8–12 years (NOAA-ESRL; Cunnold et al., 2002 ; IPCC-AR4, 2007 ).

It is estimated that the contribution of methane to the greenhouse effect is 18%, and the most important sources are:

Residual agricultural products 40%

Processing and distribution of fossil fuels 29.6%

Storage and processing of domestic waste 18.1%

Burning of biomass and grazing 6.6%

Other sources 4.8%

In the IPCC Report (IPCC-TAR, 2001) is suggested that the natural sources account for ca. 40% of total methane sources. Aikawa (Aikawa, 2006 ) indicates that the transportation contributes 1.1% of CH 4 emissions in Japan and suggests that there is a small influence of methane emissions from mobile sources on the concentrations in ambient air. The possible evolution of anthropogenic methane emissions at global level has been discussed by Cofala (Cofala et al., 2007 ) who predicts an increase of CH 4 emissions from 250 Tg/year in 1990 to 420 Tg/year in 2030.

Carbon monoxide (CO) is the most significant polutant. It has a short life time in the atmosphere due to its reaction with other atmospheric components, such as hydroxyl radicals. It has an indirect radiative effect by increasing the concentration of methane and troposperic ozone. In urban areas, CO reacts photochemically with aldehydes, to produce peroxy radicals. These radicals react with nitrogen oxide, to form nitrogen dioxide, which is the main responsible for the formation of the fotochemical smog.

Through natural processes, the CO can be oxidized to CO 2 , thus contributing to the increase of the later in the atmosphere. Thus, through resulting products the anthropogenic CO can indirectly contribute to the greenhouse effect and to the global warming.

Anyway, even if CO 2 and CH 4 are the main greenhouse gases, with a contribution of more than 70% (CO 2 55%, CH 4 15%) on global warming ( IPCC-AR4, 2007 ), for a good estimation of these two gases to the Global Warming Potential (GWP) in urban areas it is necessary to take into account the indirect contribution of CO which is not a greenhouse gas but changes the atmospheric chemistry and the abundance of other greenhouse gases. CO is a key air pollutant which can be utilized like tracer in the separation of CO 2 and CH 4 from biogenic and anthropogenic sources ( Daniel & Solomon, 1998 ).

Different studies ( Daniel & Solomon, 1998 ; Fuglesvedt et al., 1996 ; Prather, 1996 cited in IPCC-TAR, Chapter 4, 2001 ) estimate the indirect GWP of the CO due to O 3 production and to feedbacks on the CH 4 . This approach was made using a box model and estimate the indirect GWP of CO for time horizons of 20, 100, and 500 years. The indirect GWP value due to CO is gave in table 1 .

Estimated Indirect Global Warming Potentials of CO for time horizons of 20, 100, and 500 years, (source IPCC-TAR, 2001 )

Regarding to long time measurement of CO 2 concentration in urban area as well as the variation of meteorological parameters allow to understanding the rule of the ecosystem functioning and meteorological parameters over inter-annual variation in carbon fluxes. The inter-annual variation of CO 2 fluxes has been typically studied either by modeling approaches ( Higuchi et al., 2005 ; Ito et al., 2005 ; Bergeron & Strachan, 2011 ) or by correlation coefficient analyses together with meteorological parameters ( Aubinet et al., 2002 ; Aurela et al., 2004; Wohlfahrt et al., 2008 ). These studies showed that, the CO 2 flux is strongly influenced by biological vegetation cycle and the variation of meteorological parameters. Thus the maximum value of CO 2 is registered during the cold season while the minimum value of CO 2 was registered during the summer. Another study ( Sottocornola & Kiely, 2010 ), show that the wet conditions favored the CO 2 uptake by the ecosystem in autumn and in winter, while the warmer and dryer weather reduce the sequestration of CO 2 in the ecosystem. A study performed in urban and sub-urban area of Montreal ( Bergeron & Strachan, 2011 ) showed that the CO 2 flux is also influenced by the anthropic activity. According to this study, “Lower emissions at the suburban site are attributed to the large biological uptake in summer and to its relatively low population density inducing low anthropogenic emissions. Higher emissions at the urban site are partly associated with its greater population and building density, promoting higher emissions from vehicular traffic and heating fuel combustion. Vehicular traffic CO 2 emissions influenced the diurnal cycle of CO 2 fluxes throughout the year at the urban site. At the suburban site, summer CO 2 fluxes were dominated by vegetation sources and sinks as daytime CO 2 uptake occurred and CO 2 fluxes responded to incoming light levels and air temperature in a fashion similar to natural ecosystems. To a lesser extent, the vegetation component also helped offset CO 2 emissions from other sources in summer at the urban site.”

This chapter will present the results of a case study of CO 2 , CH 4 and CO variations during one year in three selected cities from Romania with different anthropic activity. In order to identify the influence of biogenic and anthropogenic sources to the budget of mentioned greenhouse gases the 13 CO 2 and 13 CH 4 isotopic composition have been determinate. Experimental results were finally correlated with meteorological parameters.

2. CO 2 in urban area

2.1. trends of co 2 variation in urban areas. a literature review.

The problem of urban carbon dioxide came into the attention of scientists in the year 1998, with the discovery and characterization of the urban CO 2 dome of Phoenix, Arizona, USA by Idso and col. ( Idso et al., 2001 ). Early work found that under certain meteorological conditions, urban CO 2 concentrations could be as high as 550-600 ppm (some 200 ppm higher than the surrounding countryside) ( Idso et al., 2002 ). Soegaard and Moller-Jensen ( Soegaard & Moller-Jensen, 2003 ) studied the urban CO 2 dome of Copenhagen and indicated that "traffic is the largest single CO 2 source in the city," and demonstrate that "emission rates range from less than 0.8 g CO 2 m -2 h -1 in the residential areas up to a maximum of 16 g CO 2 m -2 h -1 along the major entrance roads in the city center."

Following these studies, research regarding the urban CO 2 domes has been performed in many other parts of the world ( Table 2 ). The results obtained from studies conducted in several cities from all over the world, show several commonalities. Thus, anthropogenic CO 2 emissions are the primary source of the urban CO 2 dome; the dome is generally stronger in city centers, in winter, on weekdays, at night, under conditions of heavy traffic, close to the ground, with little to no wind, and in the presence of strong temperature inversions. These conclusions are in agreement with the data provided by Commonwealth Scientific and Industrial Research Organisation (CSIRO) which indicate that typical concentrations of CO 2 in urban areas is situated between 350 and 600 ppm and depend on meteorological parameters and urban agglomeration.

Overview of urban CO 2 measurements

Taking in to account the concentration of CO 2 from the cities, some of researches were focused on the impact of local CO 2 emissions over local temperature. Thus, Balling et al. running the CO 2 concentration though a radiation model calculated that local CO 2 emissions modify the local temperature with more than one-tenth of one degree Celsius. In fact, Balling et al. sugest that this increasing of temperature is insignificant by comparing it to the overall urban heat island in Phoenix which typically adds 5 to 10 degrees C. ( Balling, Jr., et al., 2001 ).

Recently, Jacobson (Jacobson, 2010 ) found that domes form above cities more than a decade ago, cause local temperature increases that in turn increase the amounts of local air pollutants, raising concentrations of health-damaging ground-level ozone, as well as particles in urban air. Also, this study has shown that “CO 2 dome” that develops over urban areas is a local problem, which creates much more health problems than in rural areas.

The conclusions of Jacobson about the human health effects of CO 2 created many controversies; therefore, more research is necessary on the measurement of CO 2 variation over the urban areas.

2.2. CO 2 variation in three Romanian cities. Case study

The available literature contains no information about the variation of CO 2 concentrations in Romania in urban areas. However, the American system of global monitoring of CO 2 (NOAA-ESRL) has a station of continuous measurement of the main parameters of air quality placed in Constanta, which also measures CO 2 concentrations.

According to NOAA-ESRL data, the variation of CO 2 concentrations at the Constanţa measurement station has an ascending trend; the yearly average values are between 365 ppm in 1995 and 395 ppm in 2007. According to the same source the concentration of CO 2 has a seasonal variation with maxima in the cold season and minima in the warm season.

In order to study the influence of anthropic activity upon the CO 2 level the study was performed in three different Romanian cities from Cluj County, as follows:

a.Cluj-Napoca city, 400 000 inhabitants,

b.Turda city, 59 600 inhabitants

c.Huedin town, 10 000 inhabitants

The case study has been carried out during one year (four seasons) from July 2008 to June 2009.

The measurements were performed monthly (during 8 hours from 8.30 am to 15.30 pm) in all the three selected locations using a NDIR CO 2 analyzer model EMG-4. The measurements of CO 2 levels were performed in a portable meteorological shelter.

For the estimation of the anthropogenic contribution over the CO 2 budget in all three studied areas a reference point situated outside of the cities has been chosen ( Roba et all, 2009 ).

For the Cluj-Napoca city,three measurement points have been selected: one situated in a zone with intense traffic (Piaţa Mărăşti), one with moderate traffic (Cartier Grigorescu) and a reference point located in a periphery location (meteorological station in Cartier Gruia).

The results of the measurements recorded in the three locations in the city of Cluj-Napoca show that during the year the CO 2 level is strongly influenced by the amplitude of the anthropic activities. The highest levels were recorded in the location with intense anthropic activity (Piata Marasti) and the lowest levels in the reference point located outside of the city. The recorded values are comprised between 380 and 530 ppm (Mărăşti), 376-456 (Grigorescu) and 373-444 (reference point).

It was also found that the highest values were recorded during the months of October and November (during the final period of the biological cycle of plants) and during the winter. It was also found that starting with the months of March; the CO 2 concentrations decrease and become comparable with the summer values ( Figure 3 ).

For the Turda city two measurement points have been selected: one in the city (Potaisa School) and another one located outside of the city (Meteo Station).

The results of the measurements in the two locations indicate a concentration difference between 15 and 20 ppm, depending of the period of measurements. The largest values of CO 2 concentration were obtained in the months of October and November, at the end of the biological cycle of plants and in the winter. The concentrations measured are in the range 380-502 ppm at Potaisa site and 371-450 ppm at the reference point.

In the town of Huedin the measurements were carried out simultaneously in two locations: one in the interior of the city (Liceul Octavian Goga) and a reference point in the peripheral area (Meteo Station). The results show a slight difference between the two locations. It was also observed again that the largest values are recorded in the months of October and November, i.e. at the end of the biological cycle of plants, and during the winter.

Starting with the month of March, the CO 2 concentrations decrease to normal values, comparable to those of the summer. The values recorded are in the range 355- 433 ppm (Octavian Goga site) and 350-411 ppm (reference point).

A comparison of the measured values carried out in different locations shows that the CO 2 concentrations depend on the size of the town, the highest values being recorded in Cluj-Napoca city, followed by Turda and Huedin ( Figure 3 ).

Annual variation of CO 2 in urban studied areas

Diurnal variations were also observed; the highest values were measured in the morning and the lowest values at the astronomic midday, both in summer and in winter seasons ( Figure 4 ).

Diurnal variation of CO 2 in urban studied areas

Taking into account the results of this case study we can conclude that the variation of CO 2 in studied urban areas is in agreement with other results reported in the scientific literature, which report that in urban areas the CO 2 levels are situated between 350 and 600 ppm, depending on meteorological parameters and urban agglomeration, with the observation that in the absence of rainfall the CO 2 level increases both in urban areas and outside.

3. CH 4 in urban area

3.1. trends of ch 4 variation in urban areas. a literature review.

Regarding to CH 4 variation in urban areas there are only a few studies about the level of methane in urban atmosphere, and these results are reported starting after the year 1980. For Romania such data are available only after 1995.

According to data from CSIRO the common concentrations of CH 4 in urban areas have values between 1700-2500 ppb and are influenced by meteorological parameters and urban agglomeration. Kuc and co-workers ( Kuc et al., 2003 ) show that, the CH 4 level in urban areas is comprised between the natural level (1650 ppb) and 4200 ppb. Ito and co-workers ( Ito et al. 2000 ) compared the atmospheric CH 4 concentrations recorded in Nagoya with the values measured at Mauna Loa Observatory in Hawaii (USA) and estimated that the excess concentration of CH 4 in the urban atmosphere of Nagoya was 170 ppb in 1988 and 150 ppb in 1997. A selective bibliography on the subject in presented in Table 3 .

3.2. CH 4 variation in three Romanian cities. Case study

In Romania, the available literature provides no information about the variation of methane concentrations in urban areas and no such studies are reported. However, the American System of Global Monitoring of Air Quality (NOAA-ESRL), has a station of continuous measurements of atmospheric methane concentrations at Constanţa. According to NOAA data, the concentration of CH 4 has an ascending trend, with average values comprised between 1880 ppb in 1995 and 1980 ppb in 2006. According to the same source, the methane concentration shows a seasonal variation, with maxima in the summer months and minima in the autumn and spring.

Overview of urban CH 4 measurements

Our case study has been focused on the measurement of the CH 4 variation in three urban areas from Cluj county as described in sub-chapter 2. The study was carried out during one year (four seasons) from July 2008 to June 2009. The measurements were performed monthly in all selected areas at the astronomic midday (in Romania at 12.30 h). In all three areas a measurement point located in the city and a reference point located outside has been selected.

The samples were collected in Cluj-Napoca, in Marasti location in the city and as reference point at Gruia location. In Turda the measurements in the city were made at Liceul Potaisa, and the reference point at the Meteo station. In Huedin the measurements were made at Liceul Octavian Goga in the town and the reference point was the meteo station.

For atmospheric CH 4 measurements, the air samples where collected by the flask sampling method and analysed by gas chromatography technique (GC) coupled with a flame ionisation detector (FID) ( Cristea et all., 2009 ).

The results show a signficant variation of atmospheric methane, depending on the season and the urban aglomeration degree. Thus, the highest values were recorded in the city of Cluj-Napoca (11.5 ppm in April 2009) and the lowest values were recorded in the month of August 2008 (2.5 ppm).

In Turda the concentrations of atmospheric methane were comprised between 2.2 and 8.6 ppm, with the lowest values measured in August 2008 (2.2 ppm) and the highest values recorded in April 2009 (8.6 ppm).

In Huedin the concentrations of methane were measured between 1.4 and 7 ppm, with the lowest values recorded in July 2008 (1.4 ppm) and the highest ones in April 2009 (7 ppm).

Significant differences are also bserved between te methane concentrations in the interior of the cities and the reference points located outside. These differences were recorded throughout the experiments, which leads to the conclusion that the anthropic activities, the automobil traffic in particular, are an important source of methane in the urban atmosphere.

The analysis of the methane concentrations in the three areas investigated indicates a similar profile for the measurements in the interior of the cities, with minima in the summer months and maxima during the spring. This may be attributed to the absence of rains in the spring (March-April). In May 2009, when the precipitations started, the concentration of atmospheric methane became closer to the values measured at the reference points.

For the reference points, the values of the atmospheric methane concentrations are in the range 2.1-4.2 ppm in Cluj-Napoca, 1.7-3.5 in Turda and 1.4-2.9 ppm in Huedin. As for the measurements in the city, a slight increase of the concentrations were observed in the spring period of 2009, due to the lack of precipitations.

The variation of methane concentration in the urban areas of Cluj county locations.

The results of our measurements indicate that the atmospheric CH 4 level in urban areas is strongly influenced by the size of the urban aglomeration as well as by the meteorological parameters. These results is in agreement with other results from scientific literatures.

4. CO in urban areas

4.1. trends of co variation in urban areas. a literature review.

According to World Health Organization data ( WHO, 2000 ), in the main European cities the average atmospheric CO concentration is situated under 2 mg/m 3 air, with a maximum concentration lower than 6.0 mg/m 3 air. At global level, the concentration of CO is composed between 0.05 and 0.12 ppm in the air. This concentration is an average between the values measured in urban and rural areas. In rural areas the CO concentration is due mainly to natural processes, but in urban areas it is strongly influenced by anthropic activities.

The CO concentration in the air of urban zones depends upon the density of combustion sources, the topography of the measurements location, the meteorological conditions and from the distance between the measurement point and the auto traffic routes.

The monitoring of CO in USA is carried out since 1980; currently, there are 243 measurement stations distributed all over the USA territory. According to EPA Reports ( EPA, 2009 ), the concentration of CO decreased in the period 1980-2006 from 14 ppm to 3 ppm.

In Europe, the monitoring of CO in urban areas is 20 years old. Several European projects were in action, to evaluate the exposure of the population to CO, and the measurements were carried out both with fixed and mobile stations. According to WHO data ( WHO, 2000 ) in large European cities the CO concentrations (during 8 hours measurements) are situated bellow 20 mg/m 3 in the air, and the maxima are not higher than 10 mg/m 3 in the air.

The first network for the measurement of pollutants resulted from anthropic activities was created in France in 1979 under the name AIRPARIF and measures the daily, monthly and annual concentrations of NOx, SO 2 , O 3 , PM, CO and of some organic compounds. According to this source, the CO concentration in the Paris region decreased from 4000 μg/m 3 air in 1994 to 1200 μg/m 3 air in 2006. In 2004 started measuring background measurements. The variation of annual average decreased from 500 de μg/m3 air in 2003 to 400 μg/m 3 air in 2006.

In Great Britain, the measurement of the concentrations of atmospheric pollutants dates from 1973; currently, there are more than 100 station in urban zones for continuous monitoring of the air quality parameters. In London, the quality of air is monitorised by as many as 30 stations. The network was created in 1993 under the name London Air Quality Network (LAQN), and since 1997 this network also measures the evolution of daily CO concentrations.

At the European level functions the European Environment Agency (EEA) with 32 members: all the 27 EU member countries, also Island, Liechtenstein, Norway, Switzerland and Turkey. Under the coordination of EEA was created the European Environment Information and Observation Network (EIONET), with the role of processing and validating the data from the stations of the member countries connected to this network. The information is available as Reports to interested users. Among the workstations connected to EIONET, 163 measure the concentrations of CO. According to EEA data, the concentration of CO at the European level decreased from 1 mg/m 3 air in 1995 to 0.5 mg/m 3 air in 2005.

In addition to the data from the monitoring stations there are numerous studies about the determination of CO concentrations in urban zones all over the world. A synthesis of these results is given in Table 4 .

Overview of urban CO measurements

4.2. Case study Cluj-Napoca city

In Romania, the monitoring of air quality is done by the Agencies for Environment Protection and follows the concentrations of nitrogen oxides, sulfur dioxide, ozone, BTEX, material particles, etc. Of these, 53 monitoring stations are connected to the European EIONET System and 12 stations also measure the concentrations of CO in urban zones.

In the city of Cluj-Napoca, there are four stations for continuous monitoring of air quality, and beginning with August 2005 the Agency measures the CO concentrations in two locations in the city of Cluj-Napoca and one in the city of Dej. The measurements in Cluj county show that the concentration of CO in the atmosphere is much below the admitted level (10 mg/m 3 ) and varies between 0.09 and 0.4 mg/m 3 air.

In this case study the measurements were performed daily in Cluj-Napoca at the astronomic midday (in Romania at 12.30 h) using a NDIR CO analyzer Horiba model APMA-360. The results showed that the CO level in Cluj-Napoca is less than 1 mg/m 3 with a tendency of accumulation during the winter season. It is also observed a trend of accumulation during the spring months, due to the lack of precipitations. Beginning with the end of May, when the rain regime becomes normal, the values of CO concentrations are around 0.1 mg/m 3 air ( Figure 6 ).

Variation of CO concentrations in Cluj-Napoca in the period July 2008-June 2009

5. Anthropogenic contribution in CO 2 and CH 4 budgets

5.1. isotopic 13 co 2 measurements.

Knowledge of the terrestrial CO 2 cycle will help to understand the climate change phenomena and to predicting the future atmospheric CO 2 concentrations and global temperatures. By estimating the terrestrial CO 2 cycle, including such factors as emissions, storages and fluxes and combining this with the isotope compositions of atmospheric CO 2 will help to identify the contribution of different factors to the atmospheric CO 2 budget. More than the observation of the CO 2 isotopic composition provides important information about sources such as fossil fuel combustion and biogenic respiration. The determination of isotopic concentrations of 13 C enforces the analyze of different species of atmospheric CO 2 and CH 4 collected in situ in glass recipients of different measures (flask sampling) by mass spectrometry. Thus, Takahashi et al. (2001 , 2002 ) using CO 2 isotope compositions method for investigating the sources of atmospheric CO 2 observed carbon isotope compositions of Δ 14 C and δ 13 C in atmospheric CO 2 and estimated the contributions of fossil fuels and biogenic respiration, while Pataki (Pataki et al., 2003 , 2006a , b ) observing δ 13 C and δ 18 O isotope compositions in atmospheric CO 2 has reported the contribution of natural gas combustion, gasoline combustion and biogenic respiration over the total atmospheric CO 2 budget.

To identify sources of carbon and to quantify these sources it is used the Keeling plot ( Pataki et al., 2003 ). The equation used in the Keeling plot is derived from the basic assumption that the atmospheric concentration of a substance (CO 2 , CH 4 ) in air reflects the combination of some background amount of the substance that is already present in the atmosphere and some amount of substance that is added or removed by sources or sinks:

where C T , C A , and C S are the concentrations of the substance in air, in the background of atmosphere, and that contributed by sources, respectively. Isotope ratios of these different components can be expressed by a simple mass balance equation:

where δ T , δ A , and δ S represent the isotopic composition of the substance in the atmosphere, in the background, and of the sources, respectively. By combine Eqs. 1 and 2 it is obtained equation 3 :

This is a linear relationship with a slope of C A (δ A –δ S ) and an intercept at the δ S value of the net sources/sinks in the atmosphere.

According to Pataki ( Pataki et al., 2006 ) the mixing ratios originating from local sources (C S ) is composed from the CO 2 mixing of natural gas combustion (C N ) and CO 2 mixing of gasoline combustion (C G ):

Using the values of measured concentrations of CO 2 (C T ) in the same time and in the same air as the measurements of δ T , and know the δ N , δ G and C N it is possible to estimate C G . The isotopic mass balance equation in this case is:

where δ T and δ G represent the isotopic composition of the CO 2 results from natural gas combustion and δ G is the isotopic composition of the CO 2 results from gasoline combustion.

In order to study the effects of the emission and diffusion of CO 2 from fossil fuel combustion most of the recent studies have focused on urban areas where CO 2 is mainly the product of power plants and transportation. The results of these measurements were correlated with variations in carbon isotopic composition and its show that while the natural level of δ 13 C value is −8.02‰, in urban areas the δ 13 C values is down to −12‰ for atmospheric CO 2 . This difference is given by the increasing input of CO 2 derived from fossil fuel ( Clark- Thorne and Yapp, 2003 , Lichtfouse et al., 2003 ; Widory and Javoy, 2003 , Newman S et al. 2008 , Wada et al 2010 ).

5.2. Isotopic 13 CH 4 measurements

The carbon isotopic composition ( 12 C, 13 C and 14 C) of atmospheric methane is used to estimate the local CH 4 sources contribution over the CH 4 budget in a local areas (Moriizumi et al., 1998)

According to (Miller et al., 2003 cited by Cuna et al. 2008 ) the methane mixing ratio in air, [CH 4 ], and its isotopic ratio, δ 13 C CH4 , may be derived from three main sources: methane produced by microbial, [CH 4 ] micr , fossil methane, [CH 4 ] ff , and methane produced from biomass burning, [CH 4 ] bmb

In equation (6) [CH 4 ] bg is defined as the smoothed marine boundary layer (MBL) at the latitude of interest (Dlugokencky et al., 1994 cited by Cuna et al. 2008 ).

Each of these emissions has a more-or-less distinct isotopic signature with bacterial methane δ 13 C micr ≈ 60‰, thermogenic methane δ 13 C ff ≈ 40‰, and biomass burning methane δ 13 C bmb ≈ 25‰ (Quay et al., 1999 cited by Cuna et al. 2008 ).

Separating CH 4 sources using isotopic signatures is complicated by enrichment during uptake processes such as bacterial CH 4 oxidation, or methanotrophy (Chanton et al., 2005 cited by Cuna et al. 2008 ). Thus, the methane mixing ratio changes over time according to Eq. (8) , where [CH 4 ] S is the sum of all sources and τ is the lifetime of methane with respect to its destruction by OH and addition from other processes (Montzka et al., 2000; Hein et al., 1997 cited by Cuna et al. 2008

The δ 13 C of methane measured in an air sample results from several different sources, such that

In Eq. (9) δ 13 C S is the flux-weighted isotopic ratio of all sources expressed in δ notation, δ 13 C bg is the atmospheric background isotopic ratio and ε is the average isotopic fractionation associated with these processes (Cantrell et al., 1990 cited by Cuna et al. 2008 ).

Using the values of measured concentrations of methane [CH 4 ] T in the same time and in the same air as the measurements of δ 13 , it is possible to estimate the contribution of all sources at the atmospheric methane budget. Thus, Moriizumi (Moriizumi et al., 1998) analyzing the CH 4 in Nagoya, Japan found that “the contribution of fossil CH 4 to local CH 4 released from the urban area was calculated to be 102±8%, and its δ 13 C was −40.8±3.0‰. In a suburban area of Nagoya fossil, CH 4 contributed to less than 10% of local release and the calculated value of δ 13 C for non-fossil CH 4 was approximately −65‰, which is within the range of reported values of δ 13 C for CH 4 derived from bacterial CH 4 sources such as irrigated rice paddies”. Kuc (Kuc et al. 2003 ), measuring the CH 4 in Krakow found that “The linear regression of δ 13 C values of methane plotted versus reciprocal concentration yields the average δ 13 C signature of the local source of methane as being equal to −54.2‰. This value agrees very well with the measured isotope signature of natural gas being used in Krakow (−54.4±0.6‰) and points to leakages in the distribution network of this gas as the main anthropogenic source of CH 4 in the local atmosphere”. Nakagawa (Nakagawa et al., 2005 ), using the stable carbon and hydrogen isotopic compositions (δ 13 C and δD) of methane quantified the contribution of automobile exhaust to local CH 4 budget. The authors estimated that for local sources, automobile exhaust in Nagoya, Japan, contribute significant amounts (up to 30%) of CH 4 to the troposphere in the studied area.

Studies performed in wetlands showed that the isotopic signature δ 13 C of methane is situated between -67.4 and -53.3‰ with lower values in the summer and higher values in the winter ( Cuna et al., 2008 ; Tarasova et. al., 2006). These values confirm that in the wetlands, the biogenic CH 4 is the main source of atmospheric CH 4.

5.3. Isotopic 13 CO 2 measurements in Cluj county. Case study

In order to study the role of CO 2 resulted from anthropic activities in the urban atmosphere of Cluj county, the variation of CO 2 concentrations and the corresponding δ 13 C values, in samples collected in the three areas (Cluj-Napoca, Turda, Huedin) were measured (by flask sampling) during a whole year (July 2008-June 2009). For each area two points of measurements were selected, one in the city and one reference outside of the city. The determination of CO 2 concentrations was done with an infrared gas analyzer, and the isotopic ratios were measured with a DELTA V Advantage, Thermo Finnigan mass spectrometer. A graphic representation of the isotopic ratios as a function of 1/ [CO 2 ] gives a Keeling plot and the value of the intercept of Keeling slope provide information about the isotopic signature of the source. Depending on the climatic conditions and the size of the urban agglomeration, correlations between δ 13 C values and corresponding CO 2 concentrations were between -11 ‰ for Cluj-Napoca, -10.0‰ for Turda and -9.0‰ for Huedin ( Tables 5 -7). If we consider δ 13 C = -8‰ the isotopic composition of natural CO 2 , the anthropogenic contribution for CO 2 budget is higher for Cluj-Napoca and near the natural level for small town Huedin. As the data in tables show, for the locations in the interior of the cities, the isotopic values are displaced from the average values by 0.5-1.5 ‰ compared with the reference points, which suggests that the CO 2 source in the urban location is composed from the zone with δ 13 C = -8‰, with clean air, and an anthropic source which can be CO 2 resulted from burning fossil fuels (mainly gasoline and methane) plus a biogenic source of CO 2 resulted from the respiration of the local vegetation. The largest difference occurred in the municipality of Cluj-Napoca (1.376), where the average values of δ 13 C = -

Values of CO 2 (ppm) and δ 13 C PDB (‰) in Cluj-Napoca city

Values of CO 2 (ppm) and δ 13 C PDB (‰) in Turda city

Values of CO 2 (ppm) and δ 13 C PDB (‰) in Huedin town

10.266‰ were obtained in the city center, compared with the δ 13 C = -8.890‰ for the reference point. For the other locations studied (Turda and Huedin) the isotopic concentrations are close to the atmospheric background, namely -9,291‰ for Turda and -8,895‰ for Huedin, comparable with the average values for the reference points (-8.874‰ for Turda and -8.327‰ and Huedin) suggesting that anthropic CO 2 is not contributing to the pollution.

5.4. Isotopic 13 CH 4 measurements in Cluj county. Case study

For the evaluation of the role of CH 4 resulted from anthropic activities in the urban atmosphere in the Cluj county, the variation of CH 4 concentrations and the corresponding δ 13 C values were measured in air samples collected in three areas: Cluj-Napoca (Piaţa Mărăşti), Turda (Potaisa School) and Huedin (O. Goga High School) during the period between January-June 2009, using the flask sampling. Again, the δ 13 C values were measured with a ThermoFinnigan Delta V Advantage mass spectrometer. The methane concentrations in the same samples were measured with a gas chromatograph equipped with a FID detector.The observed variation of methane concentrations ( Table 8 ) is rather large, between 4.7 and 11.5 ppm. The values measured are above the atmospheric background, which suggests that there is an anthropic source of CH 4 in all investigated areas. The average value of δ 13 C = - 40 ‰ suggests that the source of methane in the atmosphere is the gas fuel network of the urbane zone investigated.

Simultaneous value of CH 4 (ppm) and δ 13 C (‰) in the Cluj district.

6. Correlation between CO 2 trend variation in urban area and variation of meteorological parameters

6.1. case study cluj-napoca city.

For this study a daily measurements of the CO 2 concentration and the main meteorological parameters (temperature, relative humidity and wind velocity) were recorded for a whole calendar year, beginning with July 2008 until June 2009. The measurements were carried out in the centre of Cluj-Napoca city, the time of midday (12.30), at the selected latitude. The results of measurements revealed a daily variation of CO 2 concentrations correlated with the meteorological factors and biological cycles of plants. Thus, the largest values of CO 2 were recorded in the fall and winter in the absence of vegetation, and the lowest values in the summer months, when the biologic cycle of plants is at the maximum.

The graphic representation of the CO 2 values as a function of meteorological parameters indicates a direct correlation with temperature and an inverse correlation with the wind velocity and relative humidity.

A computation of linear regression slopes of CO 2 versus air temperature for two months from winter (January and February) and two months from summer (July and August) gives positive slopes for both seasons with a highest correlation factor (0.666) in winter in the absence of photosynthesis ( figure 7 ).

The same representation for CO 2 and for relative humidity shows a negative slope in summer and a positive slope in the winter. The computation of linear regression slopes for CO 2 versus wind velocity show a negative slopes both in summer and in winter ( figure 7 ).

The results of this case study show that in urban area it is difficult to estimate by correlation coefficient analyses the influence of meteorological parameters over the CO 2 variation. In order to correlate the variation of CO 2 concentrations with the variation of meteorological parameters a statistic analysis of data is necessary. For the statistical approach the regression analysis and principal component analysis (PCA) has been used.

Corelation between CO 2 variation and meteorological parameters in two seasons

6.1.1. The regression analysis

The regression analysis was performed with the aid of Curve estimation model of SPSS statistics program and the regression coefficients were calculated, having the CO 2 concentration as independent variable and the meteorological parameters as dependent variables. For analysis the squares of regression coefficients and the regression curves were used. The regression coefficients R 2 were computed for the most frequently used types of regression, namely linear, logarithmic, polynomial and exponential ( Table 8 ).

The analysis of regression coefficients leads to some important conclusions.

The only meteorological parameter which correlated with the CO 2 concentration over the 0.6 threshold is the air temperature. Both the air humidity and the wind velocity have very low regression coefficients, suggesting that there is a low probability that the variation of the CO 2 is influenced by these meteorological factors. However, there are singular situations, when the correlation coefficients are close to the 0.6 threshold. Thus, for the relative

The Estimation of regression coefficients (R 2 ). Note: The types of regression R are: Li – linear, Lo – logaritmic, Po – polynomial, Ex – exponential

humidity the highest regression coefficient was 0.52, for the type polynomial in June 2009. For the wind velocity the largest coefficient was 0.56 in May 2009, for the linear regression.The regression between CO 2 concentration and temperature reveals two interesting aspects. The 0.6 threshold of the correlation coefficient was overfull filed in the three months of winter, at the end of spring and beginning of the summer, in May and June. The lowest value of the correlation coefficient was observed in August when the measurements were made at high temperatures over 30 Celsius degree. Large values, above 0.8 were observed in January and May. To illustrate the correlations we present in table 9 all the situations when the correlation coefficient was higher than 0.6.

6.1.2. The analysis of main components (PCA)

This type of analysis was necessary, because we wanted to see, which is the weight of meteorological parameters and CO 2 concentrations in the explanation of total variations. The PCA (Principal Component Analysis) method without factor rotation was used and the results are shown in Table 10 .

Explanation of total variations. Extraction Method: Principal Component Analysis

After the value of 0.5 was selected for the Eigenvalue, three components were extracted which together explain 93.139% of the total variation. The first component explains 44.002 % of the variation, the second 26.649 and the third 22.5 %. The difference to 100 % is due to a fourth component, which was eliminated because it had an Eigenvalue of only 0.274.The matrix of components ( Table 11 ) allows the identification of each factor. Thus, the first factor is very well correlated with the air temperature (inverse correlation) and with the air humidity (direct correlation). The second component is very well correlated with the wind velocity. The concentration of CO 2 is very well correlated with the third component.

Matrix of components

7. Conclusions

The monitoring of the CO 2 levels in urban area could estimate the contribution of anthropic activities over the global CO 2 level. This contribution is essential in order to establish the presence of CO 2 dome over the cities. The results of presented case study confirm the presence of a CO 2 dome over the urban studied area. More than that, it confirms that the anthropogenic CO 2 emissions are the primary source of the urban CO 2 . Taking into account the obtained results it can be observed that the level of CO 2 in urban areas is influenced by the size of the city and by the amplitude of anthropic activities. Thus the highest values of CO 2 were obtained in the biggest city Cluj-Napoca (between 380 and 530 ppm at Mărăşti Square, 376-456 at Grigorescu and 373-444 at reference point) followed by Turda (380-502 ppm at Potaisa School and 371-450 ppm at the reference point) and Huedin (355- 433 ppm at O. Goga High School and 350-411 ppm at the reference point. It is also observed that the concentration of urban CO 2 has an annual variation with the lower value in the summer and the highest value in the autumn and spring. Regarding the daily CO 2 variation it also observed that it is dominated by the photosynthesis.

The results of the atmospheric methane measurements show a signficant variation depending on the season and the urban aglomeration degree. Thus, the methane concentrations in the three investigated areas indicate a similar profile for the measurements carried out in the cities, with minima in the summer months and maxima during the spring. The highest values were recorded in the city of Cluj-Napoca (11.5 ppm in April 2009) and the lowest values were recorded in the month of August 2008 (2.5 ppm). In Turda the concentrations of atmospheric methane were comprised between 2.2 and 8.6 ppm, with the lowest values measured in August 2008 (2.2 ppm) and the highest values recorded in April 2009 (8.6 ppm) while in Huedin the concentrations of methane varied between 1.4 and 7 ppm, with the lowest values recorded in July 2008 (1.4 ppm) and the highest ones in April 2009 (7 ppm). Significant differences are also observed between te methane concentrations in the interior of the cities and the reference points located outside. These differences prove that the anthropic activities, in particular the automobile traffic, are an important source of methane in the urban atmosphere.

The carbon isotopic composition measurement of CO 2 and CH 4 is the best way to establish the biogenic and anthropic contribution at CO 2 and CH 4 budget in urban areas. Regarding the case study performed in three Romanian cities the results of 13 CO 2 show that the value of δ 13 C is depending on the size of the urban agglomeration. Thus, the lower value -11 ‰ were obtained for Cluj-Napoca, followed by Turda (-10.0‰) and Huedin (-9.0‰). If we consider δ 13 C = -8‰ the isotopic composition of natural CO 2 , the anthropogenic contribution for CO 2 budget is higher for Cluj-Napoca and near the natural level for small town Huedin. More than, the recorded data show a difference of 0.5-1.5 ‰ between the measurements city points and reference points which suggests that the CO 2 source in the urban location is composed from the zone with δ 13 C = -8‰, with clean air, and an anthropic source which can be CO 2 resulted from burning fossil fuels (mainly gasoline and methane) plus a biogenic source of CO 2 resulted from the respiration of the local vegetation. The largest difference occurred in the municipality of Cluj-Napoca (1.376), where the average values of δ 13 C = -10.266‰ were obtained in the city center, compared with the δ 13 C = -8.890‰ for the reference point. For the other locations studied (Turda and Huedin) the isotopic concentrations are close to the atmospheric background, namely -9,291‰ for Turda and -8,895‰ for Huedin, comparable with the average values for the reference points (-8.874 for Turda and -8.327 and Huedin).

Regarding the 13 CH 4 measurements the results obtained in the case study are above - 40 ‰ which suggests that there is an anthropic source of CH 4 in all investigated areas. We think that these values are a consequence of methane resulted from gas fuel network of the urbane investigated areas.

Regarding the correlation between CO 2 variation in urban area and variation of meteorological parameters the results of the case study indicate a direct correlation of CO 2 level with temperature and an inverse correlation with the wind velocity and relative humidity. Although, these correlations are poor and the analysis of regression coefficients showed that only the air temperature is correlated with the CO 2 concentration over the 0.6 threshold. The 0.6 threshold of the correlation coefficient was overfull filed in the three months of winter, at the end of spring and beginning of the summer, in May and June. The lowest value of the correlation coefficient was observed in August when the measurements were made at high temperatures, over 30 degrees Celsius. Large values, above 0.8 were observed in January and May. The air humidity and the wind velocity have very low regression coefficients, suggesting that in the urban areas studied, there is a low probability that the variation of the CO 2 is influenced by these meteorological factors. Thus, for the relative humidity the highest regression coefficient was 0.52, for the type polynomial in June 2009 while for the wind velocity the largest coefficient was 0.56 in May 2009, for the linear regression.

Taking into account the results obtained, the present case study shows that the variation of CO 2 in urban area is in agreement with other results reported in the scientific literature. Thus, according to the carbon isotopic composition measurements of CO 2 the anthropogenic CO 2 emissions are the primary source of the urban CO 2 dome; the dome is generally stronger in city centers, in winter, under conditions of heavy traffic, with little or no wind, and in the presence of strong temperature inversions.

Acknowledgments

The case study has been supported by the Romanian National Authority for Scientific Research, Project no. 213-1/2007

• 2. AIRPARIF: Available from: http://www.airparif.asso.fr/index.php?setlang=en
• 6. CarboEurope, Available from: http://carboeurope.org/
• 16. CSIRO, Available from: http://www.csiro.au/science/Climate-Change.html
• 21. DEQ-Oregon, 2007 Available from: http://www.deq.state.or.us/aq/toxics/index.htm
• 23. EEA: Available from: http://www.eea.europa.eu
• 24. EIONET: Available from: http://www.eionet.europa.eu
• 26. EPA, 2009 Available from: http://www.epa.gov/oar/airtrends/carbon.html
• 35. Idso, C.D.; Idso, S.B. & Balling, R.C. ( 2001 ). An intensive two-week study of an urban CO 2 dome. Atmospheric Environment , Vol. 35, Iss.6, pp. 995-1000.
• 36. Idso, S.B.; Idso, C.D. & Balling, R.C. ( 2002 ). Seasonal and diurnal variations of near-surface atmospheric CO 2 concentrations within a residential sector of the urban CO 2 dome of Phoenix, AZ, USA. Atmospheric Environment , Vol. 36, pp. 1655-1660.
• 37. IPCC-AR1, 1992 Available from: http://www.ipcc.ch/ipccreports/far/IPCC_1990_ and_1992_Assessments/English/ipcc-90-92-assessments-overview.pdf
• 38. IPCC-AR4, 2007 Available from: http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr.pdf
• 39. IPCC-AR5, 2010 Available from: http://www.ipcc-wg2.gov/AR5/ar5.html
• 40. IPCC-SAR, 1995 Available from: http://www.ipcc.ch/pdf/climate-changes-1995/ipcc-2nd-assessment/2nd-assessment-en.pdf
• 41. IPCC-TAR, 2001 Available from: http://www.grida.no/publications/other/ipcc_tar/
• 51. LAQN: Available from: http://www.londonair.org.uk
• 64. NOAA-ESRL, Available from: http://www.esrl.noaa.gov/
• 88. WHO , 2000 : Available from: http://www.euro.who.int/__data/assets/pdf_file/0005/74732/E71922.pdf

© 2011 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike-3.0 License , which permits use, distribution and reproduction for non-commercial purposes, provided the original is properly cited and derivative works building on this content are distributed under the same license.

Continue reading from the same book

Air quality.

Edited by Dragana Popovic

Published: 05 July 2011

By Piero Lovreglio,Maria Nicolà D’Errico, Silvia Fus...

2939 downloads

By Ramirez-Sanchez HU, Meulenert-Peña AR, Garcia-Gua...

2353 downloads

By Aceves-Fernandez Marco Antonio, Sotomayor-Olmedo A...

3045 downloads

Reports FAR Climate Change: Scientific Assessment of Climate Change Chapters Graphics

8: detection of the greenhouse effect in the observations, download (1.8 mb).

FAR Climate Change: Scientific Assessment of Climate Change

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Sensors (Basel)

Developing a Modern Greenhouse Scientific Research Facility—A Case Study

Davor cafuta.

1 Department of Information Technology and Computing, Zagreb University of Applied Sciences, 10000 Zagreb, Croatia; [email protected] (I.C.); [email protected] (T.K.)

2 Multimedia, Design and Application Department, University North, 42000 Varaždin, Croatia

Ivica Dodig

Tin kramberger, associated data.

Not applicable.

Multidisciplinary approaches in science are still rare, especially in completely different fields such as agronomy science and computer science. We aim to create a state-of-the-art floating ebb and flow system greenhouse that can be used in future scientific experiments. The objective is to create a self-sufficient greenhouse with sensors, cloud connectivity, and artificial intelligence for real-time data processing and decision making. We investigated various approaches and proposed an optimal solution that can be used in much future research on plant growth in floating ebb and flow systems. A novel microclimate pocket-detection solution is proposed using an automatically guided suspended platform sensor system. Furthermore, we propose a methodology for replacing sensor data knowledge with artificial intelligence for plant health estimation. Plant health estimation allows longer ebb periods and increases the nutrient level in the final product. With intelligent design and the use of artificial intelligence algorithms, we will reduce the cost of plant research and increase the usability and reliability of research data. Thus, our newly developed greenhouse would be more suitable for plant growth research and production.

1. Introduction

Advances in computing technologies based on embedded systems with the recent development in smart sensors are leading to cost-effective solutions for the Internet of Things (IoT). The Internet of Things is an essential component of smart home systems, smart transportation, healthcare, and smart agronomy. In any production environment, especially in agronomy, Internet of Things devices enable efficient planning and resource allocation, providing economic benefits and increasing competitiveness in the market [ 1 , 2 ].

The continuous fusion of computing and agronomy science opened a new field called precision agriculture, leading to higher crop yield within the greenhouse facility [ 3 ]. An innovative approach using IoT as a data source and deep learning as a decision maker can optimize the greenhouse environment such as temperature, humidity and nutrients [ 4 ]. By monitoring the growing process in the greenhouse, better quality of food, cosmetic products and medicinal substances can be achieved by increasing the plant nutrient levels [ 5 ].

According to related work in greenhouse design, the sensors and their location inside the greenhouse are essential components since some parts of the greenhouse contain microclimate pockets. The sensors are organized in several combinations of horizontal, vertical and hybrid arrangements to detect and eliminate microclimate pockets [ 6 ]. Additionally, camera positioning system should be flexible enough to allow precise and diverse image acquisition for successful deep learning model training. Image quality, especially noise levels, can reduce the deep learning model precision [ 7 ].

In this paper, we present the system architecture and design of a modern scientific greenhouse research facility for the purpose of Croatian Science Foundation’s Project Urtica—BioFuture. Several sensor nodes are proposed in different locations: nutrient solution, environmental (inside and outside the greenhouse), and sensor nodes for energy efficiency and power supply. They are connected via a dedicated central node. As a main contribution, we propose a novel system architecture concept for automated sensor positioning using suspended platform concept to measure accurate environmental data in any available position to achieve the best possible automated hybrid arrangement for microclimate pocket isolation. Using this measurement microclimate pockets will be detected and isolated. The proposed positioning system enables precise image acquisition from multiple angles, thus resulting in image data diversity. Image diversity plays an important role in deep learning model regularization.

Sensor sensing techniques and communication technologies are also considered in this paper. Precise sampling techniques are used resulting in Big Data due to the scientific nature of this data acquisition. To enable a steady flow of this data, a constant power supply and uninterrupted connection are essential. The data is stored locally and continuously synchronized with the cloud service.

The cloud will provide plant health calculations according to sensor data analysis. Opposite to calculation, we propose a methodology to use a deep learning method that uses RGB camera images, chlorophyll leaf images, and thermal camera images to estimate plant health. Such methodology can lead to equivalently precise, yet more affordable solutions applicable in the production. Common benefits of deep learning models and Big Data mining are proactive alerting and monitoring systems or autonomous decision making, which are particularly useful in smart agriculture [ 8 ]. Additionally, combining visual data such as images and using sensors to train the corresponding deep neural network model based on visual information proves essential for building an affordable smart agriculture system [ 9 ]. Visual information analysis reduces the monitoring complexity and overall price while maintaining the precision achieved with the sensor cluster.

This calculation of plant health is used in the project to optimize ebb timing periods. The decision when to make a phase change is a key issue in the project. The goal is to achieve extended ebb periods for higher plant nutrient levels while avoiding plant wilting. This is the main challenge to be addressed in the upcoming project experiment.

We wrote this paper as part of Croatian Science Foundation’s Project Urtica— BioFuture [ 10 ]. The project focuses on the development of a modern greenhouse research facility as a quality basis for future research at the Faculty of Agricultural Sciences, University of Zagreb, Croatia, with the support in computer sciences from Zagreb University of Applied Sciences. This project focuses on the nutritional and functional Urtica Dioica (common nettle) values in modern hydroponic cultivation techniques [ 10 ].

This paper is organized as follows. Related work on existing greenhouse solutions is discussed in Section 2 . Then, key highlights of the system architecture and design are presented in Section 3 . In this section, a detailed description of the sensors and data acquisition follows, highlighting the greenhouse layout where a new sensor data positioning is proposed to capture all microclimate pockets. Later, a cloud communication and storage is described. Finally, the cost of the system is approximated. In Section 4 we presented an experiment with a model of suspended platform. The paper is concluded in Section 5 , where the advantages of our proposed system and suspended platform are discussed, and finally some future research directions are given.

2. Related Work

The sensor system is a crucial element of smart agriculture. In greenhouse cultivation, especially in the laboratory environment, any value in an experiment can be significant. Majority of the current greenhouse solutions use sensors in different stages of farming for information gathering, effective monitoring and decision making. The main drawback of these greenhouse solutions is the lack of sensory diversification.

Wei et al. [ 11 ] presented a review of the current development of technologies and methods in aquaponics. In the greenhouse environment, water quality, environmental data and nutrient information are involved in intelligent monitoring and control. The paper summarizes intelligent, intensive, accurate and efficient aquaponics concepts that we used as a start point for our greenhouse development.

2.1. Sensors

In modern scientific greenhouse research experiments, a vast number of different sensors must be used to reduce the possibility of inadequate research results. The significant number of sensors is used to reduce the influence factors on different greenhouse locations and to detect different influence factors in the plant growth. Due to the nature of any scientific development, it is of great importance to keep the expenses within the project limits. Therefore, experimenting with expensive and complicated sensors may be uneconomical in such projects. Additionally, it can be challenging to apply such an environment to production facilities [ 12 ]. Various sensors are essential for science-based approaches to smart and precision agriculture. These sensors include environmental, power supply (for energy efficiency), nutrient solution sensors, and sensors that determine the chlorophyll content of plants [ 13 ].

Almost all environmental variables (temperature, humidity, amount of light in common and individual spectral regions, atmospheric pressure and air quality) in the greenhouse system can be used as sensed data. Due to the specific requirements of the greenhouse experiment, different types of environmental variables need to be monitored, and thus different values of sensors need to be measured [ 13 ]. Many different combinations are sampled based on experience and experimental parameters: Temperature, humidity, CO 2 concentration, illumination, illuminance (limited to a specific part of the spectrum). Other sensors include barometric pressure, specific gas concentration (oxygen, nitrogen, ozone) [ 13 , 14 ].

In addition to environmental sensors, there are other sensors that are used to increase the environmental and energy consumption awareness of the project (green-it solutions), resulting in an advantage in economic costs. For this purpose, power supply sensors are used to determine the energy footprint of the greenhouse. The building strategy of the modern greenhouse is focused on equipment, sensors and processes that are energy efficient. Bersani et al. [ 15 ] wrote an article on precision and sustainable agriculture approaches that focuses on the current advanced technological solution to monitor, track and control greenhouse systems to make production more sustainable. Pentikousis et al. [ 16 ] discusses the communication environment of the sensors to transmit their data and propose server-side data aggregation methods. In addition, the article presents sustainable approaches to achieve near-zero energy consumption while eliminating water and pesticide use.

In production greenhouses, environmental and power supply sensors are used as part of monitoring control processes to delay or accelerate decisions about opening windows, blinds, or switching thermal processes such as cooling or heating. An example of the monitoring and control system is presented in [ 17 , 18 ]. The collected data can be processed using hybrid AI methods [ 19 ] or by applying mathematical models [ 20 ]. With the usage of the monitoring and control system, a zero-energy footprint can be achieved. In addition, the power supply sensors can be used as an alert medium for a power outage warning, which may cause irreparable damage and loss of scientific research data. As presented in [ 21 ], in case of main power failure, adaptive power management can be implemented to extend backup power supply lifespan.

In greenhouses, power supply is used for nutrient delivery to the plants and maintenance of proper level of nutrient solutions in the floating system. Therefore, solution level sensor is used to monitor the level of solution in the floating system [ 11 ]. Nutrient solution sensors are used to determine the properties of the nutrient solution. The most common properties measured in the nutrient solution are temperature, dissolved oxygen, total dissolved solids (TDS), and hydrogen strength (pH) values [ 11 ].

In the hydroponic floating system, the root of the plant is partially immersed or sprayed in the nutrient solution and in most cases lies in a growing medium. This growing medium draws moisture from the nutrient solution. The moisture content can be measured with a soil hygrometer ( humidity detection sensor) [ 22 ] which is inserted into the growing medium. The sensor consists of an EC probe and a soil resistance metric. It is used to measure the electrical resistance of the soil, which is an indicator of soil salinity. The salt concentration of the nutrient solution can change over time, affecting the sensor reading. Therefore, the differential values of the sensor over time are more relevant than directly measured results [ 23 ].

A well-balanced plant nutrient growing solution results in a healthier plant. The plant health can be observed by monitoring the visual physiognomy of the plant, and this system can also be used to analyze and detect plant diseases or crop damage [ 24 ]. Nutrient solution should be inspected and changed frequently to enhance the elimination of phytopathogens [ 25 ].

Visual monitoring ranges from custom-made devices such as LeafSpec [ 26 , 27 ], the use of a normal camera combined with a microcontroller, a processor board [ 28 , 29 , 30 ] or a smartphone camera [ 31 , 32 , 33 ]. Papers propose monitoring plants with different types of cameras: standard spectral camera, infrared camera, thermal imaging camera, or color component camera.

A hyperspectral and spectroscopy system camera is used [ 34 , 35 ] to obtain better results. There is also an example of a custom-made system used in [ 36 ]. The camera can observe the plant as a whole or just a part of it, such as the leaves. The context is also distinguished by image precision. The image can be taken in a precise position with little background noise, or from a distance with somewhat unpredictable background and viewing conditions.

Opposite to camera systems, RGB color sensors are used in [ 37 , 38 ]. A RGB color sensor or infrared sensor provides a direct numerical value for a specific detail on the captured image.

2.2. Data Acquisition

Different greenhouse segments are subject to a specific microclimate pocket, usually caused by the greenhouse orientation, external shading, materials used, materials or other causes. Therefore, sensor positioning and sampling time frames are critical to data acquisition in a modern greenhouse. Specific microclimate pockets affect plant growth and will affect the data if not included in the calculation of the experiment. Therefore, sensors must provide normalized data and microclimate data specific to the position in the greenhouse. Normalized data is collected by using specific models that estimate or interpolate sensor data across the greenhouse [ 39 , 40 ].

Kochhar et al. [ 6 ] classified fixed sensor positioning as horizontal, vertical, and hybrid. This type of positioning is not sufficient to capture all microclimate data [ 41 ]. Wu et al. [ 41 ] proposed a sensor placement model to maximize target coverage without occlusion. As an alternative to fixed positioning, multiple papers propose mobile sensor placement in greenhouses [ 13 , 42 , 43 , 44 , 45 ].

When using an autonomous sensor carrier vehicle, significant attention must be paid to layout optimization for rapid and safe navigation [ 45 ]. In paper [ 46 ], an obstacle detection system using Kinect sensor is proposed. These sensors are connected to a robotic vehicle that drives around the greenhouse [ 43 ]. On a robotic vehicle, an arm can be placed for further reach [ 42 ].

In the previous papers, sensors are moved through the greenhouse to detect and measure microclimate pockets. In contrast to the movement of sensors, the plant delivery system is proposed to eliminate the influence of microclimate on plant growth [ 44 ]. This complex solution still leaves the influence of microclimate on sensor data. Other works propose the use of drones, especially in plant fields [ 13 ].

When using a variable sensor layout, a large amount of data is collected and processed locally or sent to the cloud. This data can be very complex to analyze due to the added component of its locality of acquisition. Data reduction can be achieved by removing repetitive results using sensor data sampling techniques. Similar measurements of the neighboring locality can be excluded if the difference is below the context-specific threshold, which depends on the required data quality. Another approach in the sampling procedure assumes a small hysteresis around the last measurement result. If the result remains within the given frame, it is discarded since no change is detected [ 47 ]. There is also a proposal that small anomalies can be discarded [ 6 ]. By using the algorithm proposed by Kochhar et al. the sensor frequency sampling can be maximized to capture specific events and redundant data is discarded [ 6 ].

Data acquisition, processing, and sampling require computational power in the form of data processing and storage. Computing power board equipped with microcontroller or processor with an operating system is essential to link sensor data and the database. The database can be available on-site or through a connection to a remote database in the cloud. Depending on the requirements, each system can be based on microcontrollers, a processor board with an operating system, or a hybrid system.

Microcontrollers provide better connection interface options with sensors. Most of them are equipped with multiple connection interfaces such as I2C, SPI or UART. The most commonly used microcontrollers are based on Arduino. The most popular Arduino compatible boards include Arduino UNO, Arduino Yun, Arduino Nano, Arduino Mega, ESP8266, ESP32, Intel Galileo Gen 2, Intel Edison, Beagle Bone Black and Electric Imp 003 [ 48 ].

Microcontrollers provide direct analogue input interfaces as they are equipped with analogue-to-digital converters. However, they lack storage, multithreading and multiprocessing capabilities. Rabadiya et al. [ 49 ] proposed a system implemented using ESP8266 and Arduino support. There are also multiple papers using Arduino boards for data processing in greenhouses [ 13 , 50 , 51 ].

Another approach opposite to microcontrollers is the processor boards with the operating system. The most common operating systems are specific Linux distributions without graphical interface. In such environments there is the possibility of local database storage with multi-thread and multiprocessing capabilities. The most popular processor boards that include the operating system are Raspberry Pi, Orange Pi, Banana Pi, Odroid. However, these boards have a smaller number of pins than microcontroller boards. They have I2C, SPI, and UART interfaces, but they lack analogue input pins that are equipped with analogue to-digital converters. These types of boards usually have larger power requirements and dimensions. There are hybrid solutions based on a microcontroller board with a tiny OS (e.g., RTOS, MicroPython) [ 23 ].

In multiple papers, a combination of microcontrollers and processor boards is proposed to reduce power requirements and provide multiple analogue interface sensors. Systems with lower power requirements are usually based on solar or battery powered concepts [ 52 ].

In a combination system, a node consists of a set of microcontrollers that provide sensor interfaces to processor boards that aggregate and send data to the cloud. Each node collects data from multiple sensors connected via interfaces. The nodes can be connected to power, battery or be solar powered. In a combined system, a central node based on the processor board node is required [ 52 ].

There is a need for interconnections between the nodes to enable communication. These connections can be classified into wireless and wired. There are multiple wireless standards available for IoT devices. The proposed wireless connection depends on the availability of the microcontroller or processor board interface, the required power requirement, the required connection bandwidth, the communication distance, and the common obstacles in the communication [ 53 ]. The connection protocols vary from Bluetooth and WiFi to GSM, radio (NRF) or ZigBee [ 6 , 54 ].

There are also mobile network protocols such as GPRS, 3G, 4G and 5G [ 55 ]. A particular protocol can be invented, but it is not a standard solution for use due to incompatibility with other systems. When wireless communication is used, more power node consumption is expected.

In contrast, a wired connection may use a connecting wire to supply power. The most known protocol is power over ethernet. However, there are other options that are not standardized. The wired connection provides an uninterruptible power supply (UPS), which ensures system availability in the event of a power failure. The UPS also provides information about a power failure or low UPS battery to the nodes. This information can be used to gracefully shut down all nodes and alert maintenance personnel in a timely manner.

Each communication is composed of a physical link layer and a logical link layer. The physical link layer can be used as a wired or wireless link. Above the physical layer is a logical layer in the form of a communication protocol. In most simple solutions, a specific protocol can be programmed specifically for the solution at hand. In most cases, standard networking protocols and addressing are used, such as Internet Protocol (IP). IoT devices have standardized specific protocols. The most commonly used specific protocol is Message Queuing Telemetry Transport (MQTT) [ 50 ]. Despite the specific IoT protocols, standard web service communication protocols such as HTTP, HTTPS, and SOAP are common.

When working with publicly available services, it is necessary to pay attention to security. In any network architecture, there is a risk of cybersecurity threats. To make a system more secure, Astillo et al. [ 56 ] proposed a lightweight specification-based distributed detection to efficiently and effectively identify the misbehavior of heterogeneous embedded IoT nodes in a closed-loop smart greenhouse agriculture system.

2.3. Big Data Collection and Deep Learning

The data received from the greenhouse sensor system is stored in the cloud. The cloud allows data to be displayed in time frames and complex analysis to predict greenhouse behavior. The collected data stored in the cloud can be processed by different algorithms in the complex model or fed as training data for a neural network [ 3 ]. Kocian et al. [ 57 ] predict plant growth in greenhouses using Bayesian network model. Plant growth can be predicted using simple algorithms such as linear regression [ 58 ]. Ready decisions or inferences can be used as triggers in other systems, such as smart home implementations as described by Chen et al. [ 59 ].

Complex deep neural networks are becoming an indispensable tool for Big Data analysis in a variety of scientific fields, including smart agriculture [ 60 , 61 , 62 ]. Harnessing the vast amount of data collected over a long period of time enables the training of complex deep neural models. Deep neural network models are one of the crucial approaches used in computer vision. A deep neural model with many parameters can be used for crop classification, yield prediction, and early detection of stress and disease. A considerable amount of computer vision-based work in smart agriculture focuses on plant stress detection, either as disease early detection [ 63 ] or water stress detection [ 64 , 65 , 66 , 67 , 68 ].

Plant classification is another important research direction, as it enables the detection and elimination of weeds [ 69 ], leading to fully automated cropping systems. Fruit counting [ 70 , 71 , 72 ] using deep neural networks and computer vision significantly improves yield prediction and automated harvesting. Object detection can be used to detect obstacles in greenhouses, leading to autonomous vehicle passage.

Deep Learning improves weather prediction [ 73 , 74 ], a key to successfully predict weather hazards (storms or floods) that can cause severe damage to the greenhouse. Plant feature recognition as part of plant phenotyping [ 75 ] has recently benefited from deep learning models that replace manual work, improving efficiency and effectiveness in precision architecture.

In modern greenhouse research, image analysis using computer vision drastically reduces the need for various sensors and even enables low-cost solutions with few to multiple image acquisition instances [ 34 , 35 ]. Deep learning can assist in clorophile fluorescence estimation, as presented in [ 76 ]. To successfully train a deep neural network model, a reliable verification model is crucial. A carefully designed sensor layout is necessary for the successful training and validation of the computer vision neural model. Specific sensors can be used to provide numerical data in correlation with the obtained images [ 37 , 38 ].

3. System Design and Architecture

The system design and architecture is presented in the Figure 1 . The figure describes the overall architecture of the proposed greenhouse system, and as such it is discussed in subsections throughout this chapter. The design and architecture are described in detail in this section as follows. First, the sensor acquisition is described, then the sensor placement is proposed and discussed. Data acquisition methodology is presented in the third part of this chapter, and finally data acquisition and data storage are presented and described.

System design and physical architecture scheme. The image describes the organization of major greenhouse nodes with short descriptions. All nodes are interconnected through the local area network and communicate with cloud via the wide area network.

3.1. Sensor Selection

According to related work, there are a variety of sensors for greenhouse monitoring in agronomy. There are sensors that directly provide data describing the condition of the plants or the nutrient solution state. Values in greenhouse cultivation such as temperature, humidity, light in common and single spectral ranges, air pressure, air quality, soil moisture, soil pH and oxygen saturation can be efficiently monitored with sensors. This wide range of sensors differs in terms of their sensing techniques, electrical characteristics, communication technologies, power requirements, and precision and range. Sensors assembled according to related work can be classified according to their localization in measurement:

• Energy efficiency and power supply unit (PSU) validity sensor node
• External environment sensor node
• Internal environment and leaf sensor node
• Nutrient sensor node emerged in the prepared solution
• Nutrient sensor node emerged in the floating system

The energy efficiency sensor node is based on monitoring the power supply unit. The monitored values are voltage level, current level, power factor, power output and power consumption. We propose to use the digital power meter for measuring voltage, current, power and power values in real time. The power values can be used to estimate the maximum power during the day which is defined as voltage and current in the given time. The power consumption is calculated in a desired time frame and defines the energy required during the selected time period. These two values can define optimal parameters for alternative energy sources. In addition, it allows us to monitor all specific processes in the greenhouse to make them more energy efficient. For energy measurement we propose PZEM-004T electric power meter [ 77 ] sensor connected to a smart device via serial interface.

We propose the classification of power consumption in the greenhouse into monitoring, heating/cooling and cultivation processes. The monitoring process allows us to monitor plant growth using several different sensors and processes. The measurements obtained from these sensors provide the information that leads to a decision on the parameters of the nutrient solution and serve as input for other greenhouse processes. The energy requirements of this system depend on the number of sensors, their location, sampling rates, and the technologies used to collect data. The energy consumption monitoring system is essential for the research phase, while in the production environment the greenhouse should have a predictable energy footprint.

The heating-cooling process allows for constant temperature and humidity parameters within the greenhouse. This process is very energy consuming and plays a significant role in plant growth. In the laboratory environment, the maximum allowable temperature and humidity deviations can range from a minimum to no deviation limit.

The cultivation process consists of nutrient solution preparation, water level estimation processes, and transfer of nutrients from storage to a floating system. In this process, the monitoring of the power supply unit is mainly focused on the error message, because the power consumption should not fluctuate significantly. Power failures should be detected to minimize the interruption of nutrient solution levels in floating systems.

External environmental sensor nodes outside the laboratory greenhouse measure meteorological data. This data is used to estimate the energy efficiency of the greenhouse by comparing the energy consumption for heating or cooling the greenhouse to the desired temperature and humidity. This node additionally provides readings on the intensity of the light spectrum and the general air quality. The sensor node consists of CO 2 , temperature, humidity, pressure, multichannel gas sensor, ultra-violet (UV) and visible light, and sensor for visible light with IR cut filter. Sensor selection, measurement range and accuracy were estimated from previous data measured manually in the greenhouse.

The internal ambient and leaf sensor node is mounted above the floating system. The collected data is used to control the internal greenhouse processes. Internal greenhouse processes are heating, cooling, opening windows, ventilating and blocking out external light. They are used to set the preferred temperature, humidity, CO 2 level and light intensity in the IR, visible and UV spectral range. This sensor node consists of similar set of sensors similar to external sensor node, additionally equipped with RGB and thermal camera, and RGB color sensor.

The measured data are used to assess the plant environment and thus influence plant health. Due to the microclimate behavior of the greenhouse, the position for the internal sensor node should be accurately determined according to related work. The internal sensor node is equipped with a leaf sensor node, which contains a thermal imaging camera and a visible camera without IR-blocking filter. The camera images are used to detect the chlorophyll and nutrient content in the leaf expressed in numerical values. The position of the camera sensor is crucial to provide higher quality images without noise. The internal sensor node must be positioned over the plant or next to the growing plant to produce images from different angles.

The sensor node is equipped with an RGB color sensor to accurately detect the color of the leaf when it is illuminated from above, according to the related work [ 37 , 38 ]. The obtained sensor data is used as training data to build an AI model that estimates the data from images only. In the later stage, the sensor data is used to verify the model predictions.

Nutrient sensor nodes created in the prepared solution and nutrient sensor nodes created in a floating system provide information about the state of the nutrient solution. Hydroponic system sensors include temperature, levels of PH, dissolved oxygen, total dissolved solids (TDS) sensor, and moisture sensor inserted into the growing media. A level sensor is used to monitor and alarm about the level of nutrient solution in the floating system. A laser range sensor is used to accurately monitor the level of the nutrient solution in a low light environment.

3.2. Sensor Placement

Sensor placement represents how the sensors are arranged in the greenhouse. In the literature, sensor placement is often referred to as layout or greenhouse layout. Placement focuses on the physical location of the sensors rather than the topology of the system, which describes the flow of information between sensors, microcomputers, and the cloud.

Sensor placement is a major factor that needs to be implemented carefully, as described in related work. The inside of a greenhouse is a dynamic environment where temperature differences during the plant growth cycle or air flow adjustments can affect the outcome of the sensors. A large greenhouse may have several microclimate pockets that may vary in location or intensity over periods of time.

According to related work, there is a well-known conventional horizontal and vertical sensor positioning system [ 6 ]. Besides horizontal and vertical positioning, there are also hybrid solutions such as shelves, boxes, tier-based and master-slave solutions. These solutions try to eliminate the microclimate effect by excluding it from the experiment (plants near the greenhouse walls are not included in the measurement results) or by measuring the microclimate effect in each position [ 13 , 42 , 43 , 44 , 45 ].

An automated robotic vehicle equipped with environmental sensors is proposed to provide data in different parts of the greenhouse [ 45 ]. The advantage of this solution is a horizontal coverage of the greenhouse. The disadvantage is a measurement of a certain vertical plane near the greenhouse floor. In case of table experiments, vehicle sensor plane and camera angle may become useless. Even with dynamic vertical positioning, vertical and horizontal positioning is limited due to the inaccessible hover system and plant growth areas. This approach is also not feasible in greenhouses without level ground, as the vehicle can be problematic to navigate.

Other approaches propose the use of a drone (rotorcraft) that can be flown autonomously or manually [ 13 ]. Integrating sensors into an unmanned drone system can introduce multiple sources of bias and uncertainty if not properly accounted for [ 78 ]. For example, a measurement may be incorrect due to drone thrust, temperature, humidity, and gas levels. Measurements can be mathematically adjusted in a laboratory setting with additional experiments. The drone system poses an additional safety risk, as people or plants in the greenhouse could potentially be damaged during flight. If continuous sampling is required, drones (especially heavily equipped ones) consume a lot of energy, so flight time and battery charging time can become an issue.

To mitigate shortcomings of the classic horizontal and vertical sensor placement, different automated robotic vehicle concepts, or even sensor equipped drone techniques, we propose a solution to implement a suspended platform with the sensor node. With this approach, we eliminate the problem of uneven greenhouse ground or other obstacles which can appear on greenhouse floor such as water piping or other infrastructural objects. Moreover, with constant power supply, battery duration is not an issue, compared to autonomous vehicle or drones. Side effect of positioning is minimal opposite to drones which generate air turbulence and affect the measurements. The concept of suspended platform is inspired by the mechanical design of a CNC machine table or a 3D printer. This design is very rigid, and it may affect the sunlight of the plant by blocking it. It is more difficult to assemble due to the lightweight construction rods of the greenhouse.

To solve these problems, a new concept of a hanging 3D positioning system is proposed based on a novel approach to large-scale 3D printing [ 79 ]. This concept allows the suspended platform to be suspended with sensor nodes and controlled by attached wires. To enable 3D oriented positioning, wires are attached from the suspended platform to the ceiling and diagonally to the angles of the greenhouse. The system is shown in Figure 2 .

The proposed suspended platform concept. The suspended platform uses a six-degree-of-freedom cable-suspended robot for positioning. Cable-positioning systems can be easily applied in different greenhouse layouts since they provide large ranges of motion.

Suspended sensor node allows the sensor node to be placed in any possible position above the floating system by manipulating the X (width), Y (length), or Z (height) coordinate. The experimental system can be programmed to automatically position the sensor node over horizontal and vertical positions to obtain results from different microclimate pockets. Using the results data in time intervals, specific microclimate pockets can be identified, and their variations estimated.

Internal environmental and leaf sensor nodes in the laboratory greenhouse are placed together on a suspended platform. The suspended platform is used to detect microclimate pockets as their position changes, and cameras simultaneously capture images of the plants. Using an automatic guidance system for the suspended platform, plant images are captured in time frames and uploaded to the cloud. At the same time, the real data is measured and linked to the images in a database. This technique can be used for data preparation for the AI learning process and later as a verification technique. Additionally, microclimate pockets can be discovered by analyzing this data.

The nodes of the external environmental sensors outside the greenhouse should be placed in an optimal position, e.g., above the roof or in a more remote location without the influence of internal factors of the greenhouse. In our case, one external environmental node is sufficient because the greenhouse is directly exposed to the sun without any obstacles. If the greenhouse has a specific orientation or obstacles that partially block part of the greenhouse during the day, multiple sensor nodes would be a mandatory solution.

The nutrient sensor node that has emerged in the prepared solution is placed on the floating platform inside the holding tank. Nutrient sensor nodes that have emerged in the nutrient solution for plant growth are placed on the floating platform within the floating system. Nutrient sensors require special treatment due to sediment formation on the probes. pH and oxygen probes should not be continuously immersed in the nutrient solution. After successful measurement, the probes must be removed from the nutrient solution and immersed in clean distilled water before used in the same or a different nutrient solution. The cleaning process of the probes can be done manually or automatically using the robotic arm concept. We propose using high-quality probes that can be immersed in the nutrient solution for extended periods of time without negatively affecting the measurement results.

3.3. Data Sampling

The data sampling procedure is used in plant analysis, where a predetermined number of data points are taken from a more comprehensive set of observations. The sampling procedure is very specific to the type of sensor and its interface. To properly document changes in the parameters sampled, sampling should be done at optimal time intervals. The limitation of the sampling frequency is determined by the interface type or the specific sensor technology.

The interface type determines the connection speed, but this is limited by the sensor technology or the common bus throughput when multiple sensors are connected. For example, the direct digital interface, analogue-to-digital converter, serial interface, I2C, and SPI interface have different data flow speed limitations. For multiple devices, the speed is divided by several devices on a common bus. The datasheet is analyzed for each sensor and interface, and the maximum sampling speed is presented in Table 1 . Additionally, the average sensor cost is presented in table. There is an additional time limit for the first measurement in the case of a pH or dissolved oxygen sensor. These limitations are presented in Table 1 .

Used sensors according to related work.

Each sensor node has its own computing power for data analysis and local data storage. Proposed computing power is a Raspberry Pi with MySQL database equipped with additional scripts. The scripts enable interaction between the sensor interface and the database. They are also responsible for the communication between the local storage and the cloud [ 52 ]. The local sensor node database defines the sampling interval, the location of the script, the location of the local database, the deviation range, and additional sensor data, which are presented in Figure 3 . The system is run from a central execution script written in Python that runs multiple scripts for each available sensor.

ER model of the local sensor node database.

An exception to storing in the database are images which are stored in the local file system. Images are not stored in the database because the database engine is not capable of handling a large blob. Path and name are placed in the local database table instead of result data to track images stored in the file system.

Script queries sensors and stores result in local database along with current timestamp. Nodes are synchronized with atomic clock daily to ensure accurate timestamp. All scripts are adjusted to discard values that deviate significantly from the estimated threshold during test measurement periods. Repetitive values are not recorded because they take up space in the database and would slow down query execution. Their absence from the database does not affect the final result, as the system assumes that the value has not changed during the queried period.

Smaller deviations can be caused by sensor fluctuation, which is common with analogue-to-digital converters due to the specific measurement process. Fluctuation can also be caused by sensor-specific measurement techniques or properties of the media being sensed, such as sensor purity, water movement, air flow, or light reflection. These fluctuations do not need to be stored in the database as they have no direct influence on the plant growth process. The fluctuation limit must be carefully estimated from the sensor data sheet and the empirical measurement process.

A high deviation means that an alarm must be triggered for sensor inspection. These deviations can be caused by contamination of the sensor, movement (out of the medium or out of range of the sun) or technical malfunctions. Reported alarms are automatically processed in the cloud and forwarded to maintenance. Due to the potentially significant impact on the plant growth process, a quick response is required in some cases (nutrient solution level or temperature).

The sensor nodes need to communicate efficiently with the cloud. This process introduces a compression algorithm with or without data loss to reduce the data flow to the central database. For a large amount of data, a NoSQL database [ 94 ] is recommended. In our greenhouse model, a SQL database is used as a local buffer to provide accurate alerting due to limited storage capacity. The cloud database will be based on NoSQL due to the large amount of data. A warehouse model of the collected data can additionally be built for specific time periods.

3.4. Data Collection

Each sensor node has local database storage, file system storage, and processing computing power. All nodes are connected to the wired Internet. The wired Internet is used to ensure continuous connectivity, as a wireless connection has a higher interference rate. A wired interconnect cable is used to provide power through a method known as power over ethernet. This method uses four wires that are not used in a standard 100 Mbps ethernet connection. The non-standard power supply voltage is used (12 V) to power the computing node, sensors, and motor system of the suspended platform. The available voltage (12 V) is rectified within the node into other required voltages according to the data sheet of the sensors. In this way, connectivity and power are provided simultaneously through a cable connection with a central power supply. The proposed sensors have low power requirements and do not require a high current throughput cable (large cross section). The use of batteries or solar cells is not practical, even in combination with a microcontroller and sensor sleeping functions, since the motors of the suspended platform require a significant amount of energy to wind the cables.

A sensor node is provided as a central node. Based on the position of the nodes, the central position node is the energy efficiency and power supply node. This node is closest to the wired wide area link and power supply and has an additional sensor to check the availability of the main power supply. The entire system is connected to the main cable via the uninterruptible power supply (UPS), which has a serial interface to communicate with the energy efficiency and power supply sensor node. In the event of a main power supply failure, the system operates without interruption for a certain time frame defined according to the UPS capacity. The UPS uses its battery power instead of the main power supply and sends information to the energy efficiency and power supply node in case of a power failure. When a power failure is detected, the energy efficiency and power supply node alerts the maintenance staff to verify the reason for the power failure. In our case study, the proposed time frame is eight hours to enable timely maintenance response.

The UPS informs the energy efficiency and power supply node to start shutdown requests that propagates to other sensor nodes as soon as the battery power decreases. Since all nodes are equipped with the operating system, local database, local memory, and scripts on the SD board, a graceful shutdown is expected. In the event of an immediate power failure, there is a possibility that the file system will be corrupted and thus the operating system will not boot. Each node acknowledges the orderly shutdown request and starts the shutdown process. After losing network connectivity with the sensor node, the energy efficiency and power node knows that a graceful shutdown has been completed on a sensor node. After determining that all nodes have completed the shutdown process, the energy efficiency and power supply node will shut down. This process must begin in time before the complete power failure of UPS to complete successfully. The shutdown period must be extended as the batteries of UPS lose capacity over time.

Energy efficiency and power nodes inform maintenance personnel with alerts of the following priorities: fatal, technical, and anomaly. Fatal faults such as power supply failure are immediately sent to maintenance personnel. Technical and anomaly faults are collected and presented to cloud users upon connecting. Technical faults are associated with technical system architecture and maintenance. Anomaly faults are linked to outlier sensor readings. Low priority errors may increase. For example, an incorrect sensor reading is an anomaly fault. If multiple anomaly warnings are repeatedly detected within a short period of time, an anomaly fault is elevated to a technical fault. If an anomaly is detected over an extended period of time, it is upgraded to a major fault because it may indicate equipment failure and require intervention.

All sensor data is collected and stored in the local database for each sensor node. The energy efficiency and power supply node hold information about other sensor nodes and local sensor data. This data needs to be transferred to the cloud for detailed analysis.

3.5. Cloud Data Storage and Analysis

Cloud-based data storage is an obvious requirement for any potentially distributed system configured to collect data in short time intervals. This is especially true for images, where on-site storage can quickly become insufficient, limiting scalability. Today, the price of cloud storage makes such data storage affordable for almost any budget, guaranteeing data availability and the necessary infrastructure support for low-latency data access.

The cloud receives the data through a publicly accessible web service point protected by a standard authentication mechanism and a whitelist for IP addresses. Data is transmitted as simple JSON and stored in the NoSQL data store due to direct compatibility with JSON format. Images are uploaded in RAW format, which is referenced in the JSON data and stored in the cloud blob storage. The local sensor nodes organize the data and upload it to the cloud immediately using the sampling process. In case of possible network failure or server problems, the data is stored locally for a longer period of time to avoid data loss. The proper period for local storage is empirically estimated and depends on the sampling process and hard disc capacity. The received data is analyzed in the cloud to determine the state of the system. The high-level system architecture is presented in Figure 4 .

The high-level system architecture.

The data obtained from the greenhouse is organized and summarized to analyze the dependent and independent variables of the process. The deep neural network model acts as a high order nonlinear function that determines plant health based only on simple camera images. This may include a deep neural network model based on a thermal image and multiple dependent basic color (RGB) images of the camera without infrared filters. In addition, a data warehouse solution is available to support the need for recurring data reports for specific time periods.

Plant health will decide when the end of the ebb period is reached, as plant health deteriorates with prolonged ebb periods. The decision made in this way should extend the ebb periods as much as possible and thus, according to previous research in agronomy, provide a plant with better nutritional values [ 95 ]. There will be other floating systems with fixed ebb periods that will act as an experimental control group during this experiment. Plant health will also be calculated for them.

Plant health will be determined in two different processes. The first process involves a deep neural network model that estimates plant health by analyzing greenhouse images. The second process estimates plant health mathematically based on sensor readings provided by the greenhouse. The data obtained from the second process is used as a correction factor for training and fitting the deep learning model. Calculating plant health only from multiple statically placed camera devices significantly reduces the implementation cost of the greenhouse.

Even without the sensor node system, it is possible to detect a malfunction of the system based on the calculation of plant health over a period of time. During this period, sudden deviations in the plant health calculation will alert the researchers because there is a possible problem with the proposed calculation model or serious problems within the greenhouse system, such as nutrient solution level, temperature, or artificial light error. Ultimately, the images processed with the deep neural network model should be sufficient to replace the sensor node system for determining plant health in the production greenhouse.

3.6. Deep Neural Network Model

Deep learning models usually contain a considerable number of trainable parameters that take a long time to train. In the context of computer vision, inference can also be the bottleneck. Although there is significant development in edge computing and optimizing such models to run in the field and even on embedded devices, for optimal results, a high-end computing device should be used to achieve real-time or near real-time inference speed. Even with a fast CPU, deep learning models can take a significant amount of time to evaluate, so GPU computing units that support a high degree of parallelism and are optimized for running complex deep learning models are needed. Large-scale smart farming systems typically do not require real-time processing. Nevertheless, the cloud solution enables cost-effective on-premises sensor and camera equipment and provides the ability to simultaneously support multiple distributed deployments with centralized AI analysis nodes. Once the deep-learning-based model processing is complete, the data is stored and made available for any further data processing. In fact, the system is designed to retrain the model with a larger amount of data when enough new data is collected, increasing the efficiency and precision of the model.

Supervised learning is a simple approach in the given system, mainly due to high availability and a large amount of ground truth data—plant health value—calculated from a reliable sensor source. For image processing, the deep learning model consists of a backbone based on convolutional neural networks using one of the proven backbone architectures such as ResNet [ 96 ], Inception [ 97 ], DenseNet [ 98 ] or an efficient concept of backbone network scaling [ 99 ]. Since the plant health value is a single number, the model contains a regression head with MSE loss function. Due to catastrophic forgetting, small periodic model updates are not easy to achieve. Therefore, we tended to use large periodic updates over a longer period of time. We leave a detailed analysis of the model update time frame to future work.

3.7. Implementation Cost Analysis

For the described smart greenhouse architecture to be competitive in the market, cost estimate should be included. The expenses can be divided into setup expense and operational cost. Setup or installation cost includes the sensor set cost, RGB and thermal camera, Internet connection installation (if missing) and suspended platform mounting. Table 1 shows the estimated cost breakdown per sensor. The sensor cost can be reduced by using the AI module to estimate the sensor values based on RGB plant images. Operational cost includes the Internet connection rates, data storage and compute cost, and GPU processing cost for AI image analysis. Depending on the data retention and level of sampling the storage and compute cost can be somewhat adjusted to specific needs. GPU processing in pay-as-you-go pricing models would require approx. 200–300 ms GPU processing time per image analyzed. Image analysis frequency can also be reduced if measurements follow a predictable pattern or high precision of not of essence. The cost of model training is not included as it is performed once during the research, and henceforth the trained model will be used only for inference.

4. Experimental Findings

In every greenhouse the temperature and the humidity are measured. These two sensors form the minimum measurement setup, although each specific greenhouse might require a specific set of sensors. Each sensor from set provides specific values that depend on the element measured. Very often, the elements measured depend on the measurement position and corresponding spatial variations. For instance, the temperature next to a window or door, next to a glass wall or in a corner in the shade will report different results. The measurement differences acquired this way form microclimate pockets.

Therefore, the sensor positioning within the greenhouse is extremely important, since our primary goal is to locate and isolate the microclimate pockets. Variations measured in the microclimate pockets affect plant health and should be included in the calculations and data analysis. This requires automated sensor positioning as opposed to the horizontal or vertical fixed positioning. Related research focuses on autonomous vehicles, conveyors and drones to find and isolate microclimate pockets. We have proposed the suspended platform architecture that allows flexible spatial positioning, covering all three spatial dimensions as it can be seen in Figure 5 . Additionally, the flexible positioning concept is essential to ensure diverse plant image acquisition to improve deep learning model applicability to new and unseen environments.

Suspended platform model. View from above and below on mounted internal sensor node. Suspended platform model during experimental positioning—test of cameras and platform stability during image acquire.

The suspended platform is equipped with internal sensor node sensors. The sensor node consists of CO 2 , temperature, humidity, pressure, multichannel gas sensor, ultra-violet (UV) and visible light, sensor for visible light with IR cut filter and RGB color sensor to detect leaf color. Additionally, to collect images, RGB and thermal camera are attached. During installation, sensors are mounted to prevent the influence on the camera’s field of view. Due to the suspended platform positioning concept, sensor node heat output or sunlight blockage is not an issue since positioning in single location is short. This makes our proposed suspended platform very flexible and precise while not being invasive for plant or plant environment.

In previous articles, we found mainly targeted measurements of microclimatic points based on the specific orientation of the greenhouse or some specific parts such as curtains, blurred windows, or a more densely placed structure. We suggest another approach to divide the plant growing area into the grid of 50 × 50 cm squared zones. Decreasing the square size, positioning system requires frequent calibrations and the time to visit the entire grid increases and becomes non-viable, especially for larger greenhouses. For certain types of sensors, the measurement itself does not occur momentarily, but a certain amount of time must pass before the value stabilizes (e.g., temperature). Overly granulated grid can lead to inaccurate data because the measurement times for different squares are not visited often enough. The size of the grid square should not be too large, otherwise microclimatic pockets might not be precisely isolated.

For a proposed square size of 50 × 50 cm, we can conclude that the allowable deviation of the suspended platform positioning is equal to half the side of the square: 25 cm. The suspended platform is implemented as a cable-driven parallel robot. Essentially, it is a set of at least 6 cables that are wound and unwound by winches and connect a frame and a platform. By synchronously adjusting the length of the various cables, the load can be moved smoothly over a wide area of the footprint, with control and stability in all 6 degrees of freedom.

To confirm the concept and determine the variations in positioning, we propose an experiment to build a model of the suspended platform and to test its positioning abilities. Three laser pointers are mounted on the platform and the printer is guided through wires by hand to specific position. Each laser pointer covers one axis: the display on the right wall, the display on the end wall, and the display on the ground. Each time the suspended platform was moved, we marked the previous point and measured the deviation of the new position from the previous point. Through several cycles of guiding, we reduced the results to acceptable average of 2.7% ± 2% deviation in positioning after full grid positioning cycle and before next calibration. The measurement provided allows for grid size slightly over 800 cm between opposite grid sectors. Additionally, in our laboratory surroundings we tested the possibility of positioning in diverse location, especially near the corners of the laboratory. Precise height positioning of the suspended platform is also satisfactory to be able to provide a closer leaf inspection. With the model experiment we established that it is possible to cover the laboratory ground except for corners.

As a comparison, positioning deviation for the similar process in the field of 3D printing spans up to 1.5%, with isolated outlier of 9.4% [ 79 ]. We believe that after motorization with additional calibration, through test experience the better results can be achieved.

The network cable provides power and secures the network connection to the internal sensor node located on the platform. Although we have designed the cable to be flexible, it is obvious during positioning that it affects the balance of the suspended platform, since the results are slightly improved in the experiment without the ethernet cable. After a few tests we abandoned the use of ethernet cable and decided to use a wireless connection instead. In the presented figure only power cable provided as connection to suspended platform.

Our experimental testing on model also showed that the flax cable is more suitable than thin rope as a wire guide. There is a possibility to power the internal sensor node through two wires from which the platform is suspended. This solution would be tested later by replacing two wires with thin steel cable.

This device to be able to reset its positioning should have a zero point. The edge zero point is extremely impractical for the zero point. For this reason, we propose to install a 3-axis sensor and secure a point in the greenhouse where the unique position of the suspended platform can be confirmed. The sensor could be implemented via ultrasound or laser. In addition, such a sensor could detect obstacles during the movement of the platform and stop the operation of the device, i.e., avoid the obstacle by positioning it via the second axis. With this experimental finding we can conclude that suspending platform can be used to detect microclimate pockets and to provide diverse and high-quality close-up plant images for deep learning model training as presented in Figure 5 .

5. Conclusions

With a higher market for organic food production, there are demands for greenhouse growing in sterile environments, pesticide and fertilizer free, which is hard to find in our surroundings. The integration of IoT devices into non-computational domains provides the opportunity to obtain Big Data analytics of every measurable section of an internal greenhouse process. Such analysis with deep learning models provides valuable insights and scientific knowledge [ 100 ].

The main goal of this paper was to present a state-of-the-art scientific greenhouse research facility that can be used during and after Project Urtica-BioFuture. In this paper, we have analyzed related work to gain knowledge about the most commonly used sensors and greenhouse equipping projects in precision agriculture. A detailed sensor node system architecture to cover all internal greenhouse processes and to obtain Big Data, which is subsequently analyzed in the cloud, is presented. The system architecture is presented to describe the design of the components and their interconnection.

The collected data is synchronized with the cloud in real time, which enables additional calculation in the cloud. A deep neural model will be trained on sensor data to estimate plant health from RGB camera images only. This is one of the primary Project Urtica-BioFuture goals. The trained model can be used as a replacement for the sensor system to make the greenhouse system more energy and cost efficient in the production environment.

Microclimatic influences can become a problem in measurement evaluation. To detect microclimates, different layouts for sensor organization are proposed. In this paper, we propose an automated hybrid sensor layout based on a suspended platform to detect microclimate pockets. The proposed layout covers the greenhouse area and allows precise positioning throughout the greenhouse. In addition, it allows camera positioning above the plants thus enabling better plant coverage.

The automated hybrid layout with suspended platform offers the advantage of positioning the sensor node above the plant growing area in all axes. With the introduction of the system, we eliminate problems with fixed horizontal and vertical layouts, problems with expensive conveyor systems, problems with floor leverage and obstacles with automated robotic vehicles, and sensor compensation by drone propulsion. In addition, the proposed suspended platform is powered by wires, eliminating the concept of battery replacement and recharging.

To validate the concept, we conducted a simple experiment by building a model of the suspended platform. In this experiment, we verified positioning errors to confirm the use of the system according to the proposed grid system over the plant growing area in the greenhouse. During the experiment, we also identified raised problems and made suggestions for them. We believe that this paper will enable us to collect better plant images for AI and detect microclimate pockets and enabling their elimination. This would make the proposed greenhouse system more effective and provide a novel starting point for the Urtica-BioFuture project.

For future work, we propose several possible avenues. A detailed analysis of the microclimate pockets in the greenhouse to obtain a mathematical model describing their influence in the surrounding areas of the greenhouse is worth considering. With analyses of the collected data, the sensor system can be further optimized by eliminating or introducing an additional sensor to replace the sensor group. The deep neural network model can be further optimized to provide exact mathematical model for plant health calculations by collecting additional training data from multiple greenhouses and different plant crops. With this approach a sensor data network simplification with the introducing of a deep neural network model will be achieved.

Acknowledgments

This paper is a part of the Project Urtica-BioFuture. The authors would like to thank all project participants for their cooperation and support.

Author Contributions

Conceptualization, D.C. and I.D.; methodology, D.C. and I.D.; software, I.C. and T.K.; validation, I.D., T.K. and I.C.; formal analysis, D.C. and I.C.; investigation, D.C., I.C. and I.D.; resources, D.C., I.D., I.C. and T.K.; data curation, T.K. and I.C.; writing—original draft preparation, D.C. and I.D.; writing—review and editing, I.C. and T.K.; visualization, T.K. and D.C.; supervision, D.C.; project administration, D.C.; funding acquisition, D.C., I.D., I.C. and T.K. All authors have read and agreed to the published version of the manuscript.

This research was funded by Croatian Science Foundation, grant number IP-2019-04. The APC was funded by University of Applied Sciences, Zagreb, Croatia.

Institutional Review Board Statement

Informed consent statement, data availability statement, conflicts of interest.

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Greenhouse Effect

Global warming describes the current rise in the average temperature of Earth’s air and oceans. Global warming is often described as the most recent example of climate change.

Earth Science, Meteorology, Geography

Loading ...

Global warming describes the current rise in the average temperature of Earth’s air and oceans. Global warming is often described as the most recent example of climate change . Earth’s climate has changed many times. Our planet has gone through multiple ice ages , in which ice sheets and glaciers covered large portions of Earth. It has also gone through warm periods when temperatures were higher than they are today. Past changes in Earth’s temperature happened very slowly, over hundreds of thousands of years. However, the recent warming trend is happening much faster than it ever has. Natural cycles of warming and cooling are not enough to explain the amount of warming we have experienced in such a short time—only human activities can account for it. Scientists worry that the climate is changing faster than some living things can adapt to it. In 1988, the World Meteorological Organization and the United Nations Environment Programme established a committee of climatologists , meteorologists , geographers , and other scientists from around the world. This Intergovernmental Panel on Climate Change (IPCC) includes thousands of scientists who review the most up-to-date research available related to global warming and climate change. The IPCC evaluates the risk of climate change caused by human activities. According to the IPCC’s most recent report (in 2007), Earth’s average surface temperatures have risen about 0.74 degrees Celsius (1.33 degrees Fahrenheit) during the past 100 years. The increase is greater in northern latitudes . The IPCC also found that land regions are warming faster than oceans. The IPCC states that most of the temperature increase since the mid-20th century is likely due to human activities. The Greenhouse Effect Human activities contribute to global warming by increasing the greenhouse effect. The greenhouse effect happens when certain gases—known as greenhouse gases —collect in Earth’s atmosphere . These gases, which occur naturally in the atmosphere, include carbon dioxide , methane , nitrogen oxide, and fluorinated gases sometimes known as chlorofluorocarbons (CFCs). Greenhouse gases let the sun’s light shine onto Earth’s surface, but they trap the heat that reflects back up into the atmosphere. In this way, they act like the insulating glass walls of a greenhouse. The greenhouse effect keeps Earth’s climate comfortable. Without it, surface temperatures would be cooler by about 33 degrees Celsius (60 degrees Fahrenheit), and many life forms would freeze . Since the Industrial Revolution in the late 1700s and early 1800s, people have been releasing large quantities of greenhouse gases into the atmosphere. That amount has skyrocketed in the past century. Greenhouse gas emissions increased 70 percent between 1970 and 2004. Emissions of carbon dioxide, the most important greenhouse gas, rose by about 80 percent during that time. The amount of carbon dioxide in the atmosphere today far exceeds the natural range seen over the last 650,000 years. Most of the carbon dioxide that people put into the atmosphere comes from burning fossil fuels such as oil , coal , and natural gas . Cars, trucks, trains, and planes all burn fossil fuels. Many electric power plants also burn fossil fuels. Another way people release carbon dioxide into the atmosphere is by cutting down forests . This happens for two reasons. Decaying plant material, including trees, releases tons of carbon dioxide into the atmosphere. Living trees absorb carbon dioxide. By diminishing the number of trees to absorb carbon dioxide, the gas remains in the atmosphere. Most methane in the atmosphere comes from livestock farming , landfills , and fossil fuel production such as coal mining and natural gas processing. Nitrous oxide comes from agricultural technology and fossil fuel burning. Fluorinated gases include chlorofluorocarbons, hydrochlorofluorocarbons , and hydrofluorocarbons. These greenhouse gases are used in aerosol cans and refrigeration. All of these human activities add greenhouse gases to the atmosphere, trapping more heat than usual and contributing to global warming. Effects of Global Warming Even slight rises in average global temperatures can have huge effects. Perhaps the biggest, most obvious effect is that glaciers and ice caps melt faster than usual. The meltwater drains into the oceans, causing sea levels to rise and oceans to become less salty. Ice sheets and glaciers advance and retreat naturally. As Earth’s temperature has changed, the ice sheets have grown and shrunk, and sea levels have fallen and risen. Ancient corals found on land in Florida, Bermuda, and the Bahamas show that the sea level must have been five to six meters (16-20 feet) higher 130,000 years ago than it is today. Earth doesn’t need to become oven-hot to melt the glaciers. Northern summers were just three to five degrees Celsius (five to nine degrees Fahrenheit) warmer during the time of those ancient fossils than they are today. However, the speed at which global warming is taking place is unprecedented . The effects are unknown. Glaciers and ice caps cover about 10 percent of the world’s landmass today. They hold about 75 percent of the world’s fresh water. If all of this ice melted, sea levels would rise by about 70 meters (230 feet). The IPCC reported that the global sea level rose about 1.8 millimeters (0.07 inches) per year from 1961 to 1993, and 3.1 millimeters (0.12 inches) per year since 1993. Rising sea levels could flood coastal communities, displacing millions of people in areas such as Bangladesh, the Netherlands, and the U.S. state of Florida. Forced migration would impact not only those areas, but the regions to which the “ climate refugees ” flee . Millions more people in countries like Bolivia, Peru, and India depend on glacial meltwater for drinking, irrigation , and hydroelectric power . Rapid loss of these glaciers would devastate those countries. Glacial melt has already raised the global sea level slightly. However, scientists are discovering ways the sea level could increase even faster. For example, the melting of the Chacaltaya Glacier in Bolivia has exposed dark rocks beneath it. The rocks absorb heat from the sun, speeding up the melting process. Many scientists use the term “climate change” instead of “global warming.” This is because greenhouse gas emissions affect more than just temperature. Another effect involves changes in precipitation like rain and snow . Patterns in precipitation may change or become more extreme. Over the course of the 20th century, precipitation increased in eastern parts of North and South America, northern Europe, and northern and central Asia. However, it has decreased in parts of Africa, the Mediterranean, and parts of southern Asia. Future Changes Nobody can look into a crystal ball and predict the future with certainty. However, scientists can make estimates about future population growth, greenhouse gas emissions, and other factors that affect climate. They can enter those estimates into computer models to find out the most likely effects of global warming. The IPCC predicts that greenhouse gas emissions will continue to increase over the next few decades . As a result, they predict the average global temperature will increase by about 0.2 degrees Celsius (0.36 degrees Fahrenheit) per decade. Even if we reduce greenhouse gas and aerosol emissions to their 2000 levels, we can still expect a warming of about 0.1 degree Celsius (0.18 degrees Fahrenheit) per decade. The panel also predicts global warming will contribute to some serious changes in water supplies around the world. By the middle of the 21st century, the IPCC predicts, river runoff and water availability will most likely increase at high latitudes and in some tropical areas. However, many dry regions in the mid-latitudes and tropics will experience a decrease in water resources. As a result, millions of people may be exposed to water shortages . Water shortages decrease the amount of water available for drinking, electricity , and hygiene . Shortages also reduce water used for irrigation. Agricultural output would slow and food prices would climb. Consistent years of drought in the Great Plains of the United States and Canada would have this effect. IPCC data also suggest that the frequency of heat waves and extreme precipitation will increase. Weather patterns such as storms and tropical cyclones will become more intense. Storms themselves may be stronger, more frequent, and longer-lasting. They would be followed by stronger storm surges , the immediate rise in sea level following storms. Storm surges are particularly damaging to coastal areas because their effects (flooding, erosion , damage to buildings and crops) are lasting. What We Can Do Reducing our greenhouse gas emissions is a critical step in slowing the global warming trend. Many governments around the world are working toward this goal. The biggest effort so far has been the Kyoto Protocol , which was adopted in 1997 and went into effect in 2005. By the end of 2009, 187 countries had signed and ratified the agreement. Under the protocol , 37 industrialized countries and the European Union have committed to reducing their greenhouse gas emissions. There are several ways that governments, industries, and individuals can reduce greenhouse gases. We can improve energy efficiency in homes and businesses. We can improve the fuel efficiency of cars and other vehicles. We can also support development of alternative energy sources, such as solar power and biofuels , that don’t involve burning fossil fuels. Some scientists are working to capture carbon dioxide and store it underground, rather than let it go into the atmosphere. This process is called carbon sequestration . Trees and other plants absorb carbon dioxide as they grow. Protecting existing forests and planting new ones can help balance greenhouse gases in the atmosphere. Changes in farming practices could also reduce greenhouse gas emissions. For example, farms use large amounts of nitrogen-based fertilizers , which increase nitrogen oxide emissions from the soil. Reducing the use of these fertilizers would reduce the amount of this greenhouse gas in the atmosphere. The way farmers handle animal manure can also have an effect on global warming. When manure is stored as liquid or slurry in ponds or tanks, it releases methane. When it dries as a solid, however, it does not. Reducing greenhouse gas emissions is vitally important. However, the global temperature has already changed and will most likely continue to change for years to come. The IPCC suggests that people explore ways to adapt to global warming as well as try to slow or stop it. Some of the suggestions for adapting include:

• Expanding water supplies through rain catchment , conservation , reuse, and desalination .
• Adjusting crop locations, variety, and planting dates.
• Building seawalls and storm surge barriers and creating marshes and wetlands as buffers against rising sea levels .
• Creating heat-health action plans , boosting emergency medical services, and improving disease surveillance and control.
• Diversifying tourism attractions, because existing attractions like ski resorts and coral reefs may disappear.
• Planning for roads and rail lines to cope with warming and/or flooding.
• Strengthening energy infrastructure , improving energy efficiency, and reducing dependence on single sources of energy.

Barking up the Wrong Tree Spruce bark beetles in the U.S. state of Alaska have had a population boom thanks to 20 years of warmer-than-average summers. The insects have managed to chew their way through 1.6 million hectares (four million acres) of spruce trees.

Disappearing Penguins Emperor penguins ( Aptenodytes forsteri ) made a showbiz splash in the 2005 film March of the Penguins . Sadly, their encore might include a disappearing act. In the 1970s, an abnormally long warm spell caused these Antarctic birds' population to drop by 50 percent. Some scientists worry that continued global warming will push the creatures to extinction by changing their habitat and food supply.

Shell Shock A sudden increase in the amount of carbon dioxide in the atmosphere does more than change Earth's temperature. A lot of the carbon dioxide in the air dissolves into seawater. There, it forms carbonic acid in a process called ocean acidification. Ocean acidification is making it hard for some sea creatures to build shells and skeletal structures. This could alter the ecological balance in the oceans and cause problems for fishing and tourism industries.

Audio & Video

Worksheets & handouts, media credits.

The audio, illustrations, photos, and videos are credited beneath the media asset, except for promotional images, which generally link to another page that contains the media credit. The Rights Holder for media is the person or group credited.

Illustrators

Educator reviewer, last updated.

October 19, 2023

User Permissions

For information on user permissions, please read our Terms of Service. If you have questions about how to cite anything on our website in your project or classroom presentation, please contact your teacher. They will best know the preferred format. When you reach out to them, you will need the page title, URL, and the date you accessed the resource.

If a media asset is downloadable, a download button appears in the corner of the media viewer. If no button appears, you cannot download or save the media.

Text on this page is printable and can be used according to our Terms of Service .

Interactives

Any interactives on this page can only be played while you are visiting our website. You cannot download interactives.

Related Resources

• Games & Quizzes
• History & Society
• Science & Tech
• Biographies
• Animals & Nature
• Geography & Travel
• Arts & Culture
• On This Day
• One Good Fact
• New Articles
• Lifestyles & Social Issues
• Philosophy & Religion
• Politics, Law & Government
• World History
• Health & Medicine
• Browse Biographies
• Birds, Reptiles & Other Vertebrates
• Bugs, Mollusks & Other Invertebrates
• Environment
• Fossils & Geologic Time
• Entertainment & Pop Culture
• Sports & Recreation
• Visual Arts
• Demystified
• Image Galleries
• Infographics
• Top Questions
• Britannica Kids
• Saving Earth
• Space Next 50
• Student Center

• What does Earth look like?
• When did science begin?
• Where was science invented?
• How does global warming work?
• Where does global warming occur in the atmosphere?

greenhouse effect

Our editors will review what you’ve submitted and determine whether to revise the article.

• National Geographic - Education - Greenhouse Effect
• Natural Resources Defense Council - Greenhouse Effect
• UCAR Center for Science Education - The Greenhouse Effect
• University of California Museum of Paleontology - Understanding Global Change - Greenhouse effect
• Library of Economics and Liberty - Greenhouse Effect
• greenhouse effect - Children's Encyclopedia (Ages 8-11)
• greenhouse effect - Student Encyclopedia (Ages 11 and up)

greenhouse effect , a warming of Earth ’s surface and troposphere (the lowest layer of the atmosphere ) caused by the presence of water vapour, carbon dioxide , methane , and certain other gases in the air. Of those gases, known as greenhouse gases , water vapour has the largest effect.

The origins of the term greenhouse effect are unclear. French mathematician Joseph Fourier is sometimes given credit as the first person to coin the term greenhouse effect based on his conclusion in 1824 that Earth’s atmosphere functioned similarly to a “hotbox”—that is, a heliothermometer (an insulated wooden box whose lid was made of transparent glass) developed by Swiss physicist Horace Bénédict de Saussure , which prevented cool air from mixing with warm air. Fourier, however, neither used the term greenhouse effect nor credited atmospheric gases with keeping Earth warm. Swedish physicist and physical chemist Svante Arrhenius is credited with the origins of the term in 1896, with the publication of the first plausible climate model that explained how gases in Earth’s atmosphere trap heat . Arrhenius first refers to this “hot-house theory” of the atmosphere—which would be known later as the greenhouse effect—in his work Worlds in the Making (1903).

The atmosphere allows most of the visible light from the Sun to pass through and reach Earth’s surface. As Earth’s surface is heated by sunlight , it radiates part of this energy back toward space as infrared radiation . This radiation, unlike visible light, tends to be absorbed by the greenhouse gases in the atmosphere, raising its temperature. The heated atmosphere in turn radiates infrared radiation back toward Earth’s surface. (Despite its name, the greenhouse effect is different from the warming in a greenhouse , where panes of glass transmit visible sunlight but hold heat inside the building by trapping warmed air.)

Without the heating caused by the greenhouse effect, Earth’s average surface temperature would be only about −18 °C (0 °F). On Venus the very high concentration of carbon dioxide in the atmosphere causes an extreme greenhouse effect resulting in surface temperatures as high as 450 °C (840 °F).

Although the greenhouse effect is a naturally occurring phenomenon, it is possible that the effect could be intensified by the emission of greenhouse gases into the atmosphere as the result of human activity. From the beginning of the Industrial Revolution through the end of the 20th century, the amount of carbon dioxide in the atmosphere increased by roughly 30 percent and the amount of methane more than doubled. A number of scientists have predicted that human-related increases in atmospheric carbon dioxide and other greenhouse gases could lead by the end of the 21st century to an increase in the global average temperature of 3–4 °C (5.4–7.2 °F) relative to the 1986–2005 average. This global warming could alter Earth’s climates and thereby produce new patterns and extremes of drought and rainfall and possibly disrupt food production in certain regions.

Knowledge is power

Stay in the know about climate impacts and solutions. Subscribe to our weekly newsletter.

By clicking submit, you agree to share your email address with the site owner and Mailchimp to receive emails from the site owner. Use the unsubscribe link in those emails to opt out at any time.

Yale Climate Connections

What’s the deal with terms like “greenhouse effect,” “global warming,” “climate change,” and “the climate emergency”?

Share this:

• Click to share on Facebook (Opens in new window)
• Click to share on X (Opens in new window)

Almost two decades ago, Al Gore’s 2006 documentary “ An Inconvenient Truth ” raised awareness of the problems associated with what was then commonly called “global warming.” Although most people had moved away from referring to the heating of our planet as the “greenhouse effect,” we were still a few years from adopting the term “climate change,” a more accurate though less evocative label, as the leading descriptor of our profound environmental challenges.

The words we use to characterize our climate concerns can influence how we view the issue, and as a consequence, the actions we take. As data scientists, we recently studied the evolution in terminology by examining 79,134 articles that mentioned key climate terms in the New York Times, the Wall Street Journal, the Washington Post, and USA Today between 1980 and 2023. This review allows us to not only understand the terms most commonly used in the media but also to see whether their usage tracks the terminology favored by scientific experts and the general public.

We were struck by the long-term evolution in the relative prominence of “greenhouse effect,” “global warming,” and “climate change.” But what really stood out was the sudden emergence in late 2018 of more dramatic language like “climate crisis,” “climate emergency,” and even “climate apocalypse.” Looking beyond the media, we found that this surge of alarmist language parallels the framing of environmental issues by experts but has not yet become commonplace among the broader public — at least not as measured by internet search data.

How terms changed over time

Although the first mention of “global warming” was in a 1975 scientific study , throughout the 1980s, “greenhouse effect” was still more commonly used by the four newspapers we analyzed. It wasn’t until 1989 that “global warming” became the term of choice in these mainstream outlets, as illustrated in the graphic below.

By 2009, “climate change” had surpassed “global warming” in the four newspapers. This terminological transition was championed as early as 2005 by the National Academy of Sciences as a more accurate way to describe the phenomenon. It may also have reflected the increasingly politicized nature of climate issues, driven in part by the oil and gas industry’s intensive campaigning . A 2011 study demonstrated that using the term “climate change” rather than “global warming” at the time resulted in a 16% increase in Republicans endorsing the phenomenon as real.

Interestingly, it wasn’t until the summer of 2015 that Google searches for “climate change” outpaced those for “global warming,” as shown in the graphic below. This suggests that the media do not have an immediate, overwhelming effect on the terms used by the public. Indeed, it can take years for a new term to seep into public consciousness and be reflected in everyday language.

The sudden rise of “climate crisis”

The media is much quicker to adopt new terminology. In late 2018, “climate crisis” and related terms, including “climate emergency,” “climate catastrophe,” “climate apocalypse,” “climate breakdown,” and “planetary emergency,” suddenly appeared in the media. These emotive terms are qualitatively different from the phrases used since the 1980s. Why did they break through at that time?

Global activism likely contributed to this sharp rise. In August 2018, Greta Thunberg launched her Fridays for Future campaign. In the U.S., organizations such as the Sunrise Movement grabbed headlines through high-profile protests that drew sustained attention. They helped to frame the issue as an emergency requiring immediate action. As Thunberg said in her December 2018 COP24 speech, “We cannot solve a crisis without treating it as a crisis.”

She was not alone in this view. In February 2019, climate journalist David Wallace-Wells penned an opinion piece for the New York Times entitled, “Time to Panic,” in which he argued that “climate change is a crisis precisely because it is a looming catastrophe that demands an aggressive global response, now.” Other journalists also embraced this perspective, especially in the U.K., where the Guardian changed its house style guide in May 2019 to favor terms like “climate crisis.”

This rapid evolution in newspaper language closely tracks expert discourse. In addition to global environmental movements, late 2018 saw the publication of a special U.N. Intergovernmental Panel on Climate Change report that, for the first time, quantified the time we have left to act to avoid catastrophic damage. Scopus , a prominent database containing a diverse array of scientific studies, shows a year-on-year doubling between 2018 and 2019 in scholarship that used “climate crisis” and “climate emergency,” and then a quadrupling of such terms between 2019 and 2020.

The convergence of a push from activists and a shift in expert views turned awareness of what seemed to be distant consequences into a time-sensitive crisis. But that doesn’t mean that the public has adopted these terms as their own. Google searches for “climate change” still outpace those for “global warming” by a significant degree. As the graphic below shows, even when comparing the relatively less common “global warming” to “climate crisis,” there are strikingly few people searching for the more urgent term.

If the lag in general usage of “climate change” compared to “global warming” is any indication, we may simply have to wait a few more years for “climate crisis” and its analogues to become common search terms. Use of these more dire words will likely affect public perception of the issue and the resulting sentiment on climate policy. The question is how.

Climate communications scholars are divided on the impact. Some studies point to the use of terms like “crisis” and “emergency” leading to distrust of news sources. Other studies, however, find no impact of the terms on a person’s willingness to engage in climate action. Alternatively, framing climate concerns as an existential threat has also been shown to generate strong emotions that motivate people to act on behalf of the environment.

This is critical because by any tangible metric, the planet is indeed experiencing a climate crisis. Carbon dioxide levels are unprecedented in the modern era, dozens of species are going extinct every day, and damages from increasingly common weather disasters are rising. The 10 hottest years on record all occurred in the last decade.

In ancient Greek, a “crisis” meant a turning point — one that might spur people to action. By embracing the “climate crisis” in this spirit, we just might provide an impetus that leads us to a better future.

Erik Bleich is the Charles A. Dana Professor of Political Science at Middlebury College, where he directs the Media Portrayals of Minorities Project lab that uses data science techniques to analyze contemporary issues. Eli Richardson is a 2024 Middlebury College graduate who specializes in climate topics. Noah Rizika is a 2024 Middlebury College graduate with experience in conservation research and carbon footprint analysis. Both recent graduates are members of the Media Portrayals of Minorities Project lab.

Tom Toro  is a cartoonist and writer who has published over 200 cartoons in The New Yorker since 2010.

We help millions of people understand climate change and what to do about it. Help us reach even more people like you.

more like this

Resources for debunking common solar and wind myths 

Venezuelan-American journalist brings climate change awareness to both Spanish- and English-speaking audiences

How to talk to a climate doomer (even if that doomer is you)

The Greenhouse Metaphor and the Greenhouse Effect: A Case Study of a Flawed Analogous Model

• Conference paper
• Cite this conference paper

• Xiang Chen 3  

Part of the book series: Studies in Applied Philosophy, Epistemology and Rational Ethics ((SAPERE,volume 2))

1748 Accesses

3 Citations

26 Altmetric

Metaphors are double-edge swords. By connecting an abstract and unknown phenomenon to a tangible and familiar one, a metaphor also creates a new reality. For example, we frequently use a metaphor to describe global warming – the atmosphere works like a greenhouse and CO 2 traps heat as panes of glass in a greenhouse do. However, this greenhouse metaphor leads to an ontological assumption that conceptualizes heat as a material-like object, a series of ideas that ignore the roles of the ocean in the process of thermal transfer within the climate system, and an underestimation of the time delay effect in climate change. By producing an illusion that the climate system will respond instantly at the moment when CO 2 level is reduced, the greenhouse metaphor is ultimately responsible for the wait-and-see approach to climate change.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Unable to display preview.  Download preview PDF.

Baillargeon, R., Spelke, E., Wasserman, S.: Object permanence in 5-month-old infants. Cognition 20, 191–208 (1985)

Article   Google Scholar  

Boyes, E., Stanisstreet, M.: Children’s models of understanding of the two major global environmental issues (Ozone layer and greenhouse effect). Research in Science & Technological Education 15, 19–28 (1997)

Carey, S.: The origin of concepts. Oxford University Press, Oxford (2009)

Book   Google Scholar  

Chen, X.: The object bias and the study of scientific revolutions: Lessons from developmental psychology. Philosophical Psychology 20, 479–503 (2007)

Chen, X.: A different kind of revolutionary change: Transformation from object to process concepts. Studies in History and Philosophy of Science 41, 182–191 (2010)

Chen, X.: Why do people misunderstand climate change? Heuristics, mental models and ontological assumptions. Climatic Change 108, 31–46 (2011)

Chi, M.: Three types of conceptual change: Belief revision, mental model transformation, and categorical shift. In: Vosniadou, S. (ed.) International Handbook of Research on Conceptual Change, pp. 61–82. Routledge, New York (2008)

Google Scholar  

Fiddaman, T.: Feedback complexity in integrated climate-economy models. Ph.D Thesis, MIT Sloan School of Management, Cambridge MA 02142 (1997)

Fleming, J.: Historical perspectives on climate change. Oxford University Press, New York (1998)

Keil, F.: Semantic and conceptual development: An ontological perspective. Harvard University Press, Cambridge (1979)

Khalid, T.: Pre-service high school teachers’ perceptions of three environmental phenomena. Environmental Education Research 9, 35–50 (2003)

Lakoff, G., Johnson, M.: Metaphors we live by. University of Chicago Press, Chicago (1980)

Moxnes, E.: Overexploitation of renewable resources: The role of misperceptions. Journal of Economic Behavior & Organization 37, 107–127 (1998)

Shepardson, D., Niyogi, D., Choi, S., Charusombat, U.: Seventh grade students’ conceptions of global warming and climate change. Environmental Education Research 15, 549–570 (2009)

Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K., et al.: Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge (2007)

Sterman, J., Booth Sweeney, L.: Understanding public complacency about climate change: Adults’ mental models of climate change violate conservation of matter. Climate Change 80, 213–238 (2007)

The Ocean Project, America, the ocean, and climate change: New research insights for conservation, awareness, and action (2009), http://theoceanproject.org/resources.php

US Environmental Protection Agency (2009), Climate change: Kids site, http://epa.gov/climatechange/kids/greenhouse.html

Vosniadou, S.: Analogical reasoning as a mechanism in knowledge acquisition: a developmental perspective. In: Vosniadou, S., Ortony, A. (eds.) Similarity and Analogical Reasoning, pp. 413–437. Cambridge University Press, New York (1989)

Chapter   Google Scholar  

Download references

Author information

Authors and affiliations.

Department of Philosophy, California Lutheran University, Thousand Oaks, CA, 91360, USA

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Xiang Chen .

Editor information

Editors and affiliations.

, Department of Philosophy, Universita Pavia, Piazza Botta 6, Pavia, 27100, Italy

Lorenzo Magnani

, Department of Philosophy, Sun Yat-sen University, XinGang Xi Lu 135, Guangzhou, 510275, China, People's Republic

Rights and permissions

Reprints and permissions

Copyright information

© 2012 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper.

Chen, X. (2012). The Greenhouse Metaphor and the Greenhouse Effect: A Case Study of a Flawed Analogous Model. In: Magnani, L., Li, P. (eds) Philosophy and Cognitive Science. Studies in Applied Philosophy, Epistemology and Rational Ethics, vol 2. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-29928-5_5

Download citation

DOI : https://doi.org/10.1007/978-3-642-29928-5_5

Publisher Name : Springer, Berlin, Heidelberg

Print ISBN : 978-3-642-29927-8

Online ISBN : 978-3-642-29928-5

eBook Packages : Humanities, Social Sciences and Law Philosophy and Religion (R0)

Share this paper

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research