Covariation Research Articles

Support Carbon Neutral Goal with a High-resolution Carbon Monitoring System in Beijing

Abstract China has committed to peak its carbon emissions before 2030 and achieve carbon neutrality before 2060, known as the Dual Carbon Goals (DCG). To support the DCG, monitoring and assessment of progresses in emissions reduction have become a central research focus. Here we describe the Beijing-Tianjin-Hebei (JJJ) Carbon Monitoring System, established over the last six years through a collaborative effort of 16 research institutions and universities. The monitoring capability includes ground-based high-precision reference stations, a dense network of medium-accuracy low-cost sensors, carbon satellites, vertical profile samplers, and mobile platforms. Observations using the network have shown CO2 variations from diurnal, synoptic to seasonal-interannual timescales, as well as dramatic on-road changes during the COVID-19 lockdown. High-resolution modeling and data assimilation at km-scale reveal major deficiencies in global carbon emissions datasets at regional/city scale. The inversion found the total fossil fuel CO2 emissions of the City of Beijing to be 117 megatonne (Mt), but much lower than the national-data-based inventories (152 MtCO2). The results have been used in emissions assessment by national and local environmental protection agencies. Similar systems are now being developed in other Chinese cities.

Eutrophication and Dissolved Organic Matter Exacerbate the Diel Discrepancy of CO2 Emissions in China's Largest Urban Lake.

The large variability in the emissions of carbon dioxide (CO2) from urban lakes remains a challenge for partitioning these sources at meaningful spatial and temporal scales. Dissolved organic matter (DOM) governs the spatial and temporal variations in CO2, yet relationships of the CO2 concentration (cCO2) and emission flux (FCO2) with DOM in urban lakes have rarely been reported. In this study, we monitored levels of cCO2, FCO2, and the composition of DOM over a 24 h period at three sites during the dry and wet seasons in China's largest urban lake, Tangxun Lake. Our study found the ratio of day/night FCO2 (millimoles per square meter per day) decreased from the dry season (0.79; 7.68/9.68) to the wet season (0.25; 6.05/24.16), averaging 0.42 (6.77/15.97), implying that accounting for nighttime CO2 emissions can increase regional estimates by 70%. This study revealed that eutrophication affected diurnal CO2 emissions with greater algal growth enhancing daytime CO2 uptake and subsequently increasing nighttime CO2 emissions via DOM degradation (larger protein-like DOM fraction). We anticipate that the relative magnitude of FCO2 between day and night from lakes is likely to increase due to urbanization and climate change, underscoring the importance of treating urban lakes as a distinct group and integrating DOM dynamics into carbon cycling in future research.

Indoor air quality assessment of an existing apartment located in Indian tropical city, Bhubaneswar

Introduction: A proper investigation on indoor pollution level in economical apartments is important, where the provision for ventilation is limited. A series of experiments were conducted in the houses within an existing economical apartment in Bhubaneswar, India to evaluate various pollutant levels. Materials and methods: Temperature, relative humidity, CO, CO2, Particulate Matter (PM10 and PM2.5), Formaldehyde and Total volatile organic compounds in the bedrooms and kitchens are measured by pollution meters. The experiments were conducted with doors and windows closed conditions. The readings were taken on a four consecutive working day in February 2023 at an interval of 3 h (9:00 am to 9:00 pm). Results: The day wise temperature and humidity variations inside the bedroom and kitchen shows a reverse trend. At the afternoon, the indoor temperature becomes high, while during the night time humidity becomes the highest. The day wise indoor CO2 and CO variation trend is pretty similar. Both CO2 and CO concentrations in bedrooms are the highest in the evening. In contrast to that CO2 and CO concentrations in kitchens becomes maximum during noon time. High particulate matter concentration at outdoor and indoor is observed at the evening time. Higher formaldehyde (HCHO) and Total Volatile Organic Compounds (TVOCs) concentration at the indoor is observed at noon and afternoon time. Conclusion: The results obtained were compared with the recommended values of World Health Organization (WHO) and National Ambient Air Quality Standards (NAQIS). The results revealed that all measured parameters are at a higher level than the recommended values except the indoor CO2 concentrations.

On-orbit spectral characteristics analysis of the greenhouse gases monitoring instrument: Spectral wavelength stability and instrumental line shape stability

The Greenhouse Gases Monitoring Instrument (GMI) is a carbon satellite payload developed based on the principle of spatial heterodyne spectroscopy technology, which is specifically designed for the global analysis of greenhouse gases (CO2 and CH4) “sources” and “sinks”. Due to the low concentration and minimal gradient variations of CO2 and CH4 in the atmosphere, higher precision is required for their retrieval. Vibrations during satellite launch and harsh space environments during on-orbit operation may cause changes in the performance of various payload components, resulting in decreased quality of spectrum products and retrieval precision. This paper proposes a set of on-orbit spectral characteristic evaluation methods tailored for the GMI to monitor the quality of its spectrum. It also develops an adaptive blind pixel detection and correction algorithm specifically for the GMI and optimizes the interferogram processing flow algorithm. On-orbit spectral calibration is performed, and methods to evaluate spectral characteristic parameters have been established. The analysis focuses on the variations of the spectral characteristics in the CO2-1 bands (1.575 μm) of the GMI during one-year of on-orbit calibration spectrum. The initial spectral wavenumber offset is 0.133 cm−1 after entering orbit, and the average spectral wavenumber offset is 0.0313 cm−1 during stable operation. During the one-year period, the maximum variation in the instrumental line shape function of the GMI is 0.006 cm−1. The observed spectrum obtained in nadir observation mode, underwent to accuracy verification. The results demonstrate consistency between the spectral peak wavenumbers and the relative trend changes of the observed and theoretical spectrum, with an average relative residual of 0.987%.

Effects of CO2 on the Nitrogen Isotopic Composition of Marine Diazotrophic Cyanobacteria

AbstractBiological N2 fixation has been crucial for sustaining early life on Earth. Very negative δ15N values detected in Archean sediments, which are not observed in present‐day environments, have been attributed to the low efficiency of proto‐nitrogenases. Alternatively, variations in early atmospheric CO2 may also play a role. Here we examine the effects of CO2 concentrations on the biomass δ15N signatures of the diazotrophs Trichodesmium erythraeum and Crocosphaera watsonii, which utilize Mo‐Fe nitrogenase (the most common form of the enzyme). Our results show that these organisms produce biomass with δ15N values up to ∼3‰ lower under both decreased and elevated CO2 concentrations compared to modern levels (∼380 μatm). These deviations from modern CO2 levels reduce nitrogenase enzyme efficiency, leading to increased organismal isotopic fractionation during N2 fixation. This study offers an alternative explanation for the observed fluctuations in geological δ15N records and provides new insights into the past nitrogen cycle on Earth.

CO2 and CH4 Concentrations in Headwater Wetlands Influenced by Morphology and Changing Hydro-Biogeochemical Conditions

AbstractHeadwater wetlands are important sites for carbon storage and emissions. While local- and landscape-scale factors are known to influence wetland carbon biogeochemistry, the spatial and temporal heterogeneity of these factors limits our predictive understanding of wetland carbon dynamics. To address this issue, we examined relationships between carbon dioxide (CO2) and methane (CH4) concentrations with wetland hydrogeomorphology, water level, and biogeochemical conditions. We sampled water chemistry and dissolved gases (CO2 and CH4) and monitored continuous water level at 20 wetlands and co-located upland wells in the Delmarva Peninsula, Maryland, every 1–3 months for 2 years. We also obtained wetland hydrogeomorphologic metrics at maximum inundation (area, perimeter, and volume). Wetlands in our study were supersaturated with CO2 (mean = 315 μM) and CH4 (mean = 15 μM), highlighting their potential role as carbon sources to the atmosphere. Spatial and temporal variability in CO2 and CH4 concentrations was high, particularly for CH4, and both gases were more spatially variable than temporally. We found that groundwater is a potential source of CO2 in wetlands and CO2 decreases with increased water level. In contrast, CH4 concentrations appear to be related to substrate and nutrient availability and to drying patterns over a longer temporal scale. At the landscape scale, wetlands with higher perimeter:area ratios and wetlands with higher height above the nearest drainage had higher CO2 and CH4 concentrations. Understanding the variability of CO2 and CH4 in wetlands, and how these might change with changing environmental conditions and across different wetland types, is critical to understanding the current and future role of wetlands in the global carbon cycle.

Atmospheric CO2 fertilization effect on cereal yields in Morocco using the CARAIB dynamic vegetation model

Climate change and rising atmospheric CO2 levels are critical factors influencing agricultural productivity, particularly in Morocco, where cereal crops are essential for food security. The primary objective of this study is to evaluate the combined effects of atmospheric CO2 variations and climatic changes on cereal yields up to 2099 using the CARAIB dynamic vegetation model. This evaluation is driven by four future scenarios based on the Euro-CORDEX initiative's regional climate models under the Representative Concentration Pathway 8.5. As part of this evaluation, the study also validates the CARAIB model for Morocco’s major cereal crops: soft wheat, durum wheat, and barley, in order to ensure the model's accuracy in simulating crop responses under projected environmental conditions. The CARAIB model effectively simulated historical cereal yield trajectories across major farming regions in Morocco from 2000 to 2016, demonstrating its robust predictive capability. Our future projections suggest that elevated CO2 levels might initially sustain cereal yields at approximately 90 % of current levels until 2050. This trend indicates that the increase in atmospheric CO2 may exert a moderating influence on the negative impacts of other environmental stressors on crop yields. However, despite this initial buffering effect, the overall yield trend from the present until 2099 indicates a decrease for most combinations of crop, zone, and climate model, even with the CO2 fertilization effect, except in some cases, the model exhibits slight increases or stabilization in yields. Additionally, the CARAIB model predicts potential yield shocks of 10–35 % below current levels from the 2080 s onwards, primarily due to periodic droughts. This variation underscores the complexity of the interplay between CO2 fertilization and climatic changes, emphasizing the urgent need for Morocco to develop adaptive agricultural strategies for long-term sustainability in the face of climatic challenges.

Metabolic processes drive spatio-temporal variations of carbon sink/source in a karst river

Riverine carbon dioxide (CO2) exchange is a crucial component of the global carbon cycle. However, the changes in the CO2 sink/source in karst rivers caused by differences in lithological features and climate, hindered the resolution of the spatio-temporal heterogeneity of global inland water carbon emissions. Here, we use hydrochemical data and CO2 gas isotopic data to reveal the spatio-temporal variations of CO2 sink/source in karst rivers and their controlling mechanisms. Fifty-two monitoring transects were set up in the subtropical Lijiang River in southwest China in June and December 2019. Our results indicated that the CO2 flux across the water-air interface (FCO2) in the Lijiang River basin ranged from −43.77 to 519.67 mmol/(m2·d). In June, the Lijiang River acted as an atmospheric carbon source due to higher water temperatures (Twater). However, driven by hydrodynamic conditions and the metabolism of aquatic photosynthesis, the river shifts from being an atmospheric carbon source in June to an atmospheric carbon sink in December. The stable isotopes of CO2 (δ13C-CO2) show significant differences in the spatio-temporal variations of CO2 sink/source. In December, the transects of the Lijiang River basin with a negative CO2 flux are significantly negatively correlated with dissolved oxygen (DO) and chlorophyll-a (Chl-a) concentration (p < 0.05). This confirms that the enhancement of aquatic photosynthesis efficiency increased water DO concentrations, which resulted in the positive movement of water δ13C-CO2 and a decrease in the partial pressure of CO2 (pCO2) and FCO2. Comparative analysis with global river FCO2 indicates that under the combined driving forces of metabolic processes of aquatic photosynthetic organisms and hydrodynamic conditions, rivers tend to act more frequently as CO2 sinks, particularly in subtropical and temperate rivers. In conclusion, this study represents a new example focusing on CO2 dynamics to address the spatio-temporal heterogeneity of carbon emissions in inland waters on a global scale.

The response of mesophyll conductance to short-term CO2 variation is related to stomatal conductance.

The response of mesophyll conductance (gm) to CO2 plays a key role in photosynthesis and ecosystem carbon cycles under climate change. Despite numerous studies, there is still debate about how gm responds to short-term CO2 variations. Here we used multiple methods and looked at the relationship between stomatal conductance to CO2 (gsc) and gm to address this aspect. We measured chlorophyll fluorescence parameters and online carbon isotope discrimination (Δ) at different CO2 mole fractions in sunflower (Helianthus annuus L.), cowpea (Vigna unguiculata L.), and wheat (Triticum aestivum L.) leaves. The variable J and Δ based methods showed that gm decreased with an increase in CO2 mole fraction, and so did stomatal conductance. There were linear relationships between gm and gsc across CO2 mole fractions. gm obtained from A-Ci curve fitting method was higher than that from the variable J method and was not representative of gm under the growth CO2 concentration. gm could be estimated by empirical models analogous to the Ball-Berry model and the USO model for stomatal conductance. Our results suggest that gm and gsc respond in a coordinated manner to short-term variations in CO2, providing new insight into the role of gm in photosynthesis modelling.

Impact of different CO2 price paths on the development of the European electricity system

In energy system transformation studies, the assumption of linearly rising CO2 prices is common and based on the decreasing issued European allowances (EUA). The actual auction price for allowances is influenced by market dynamics, resulting in non-linear variations in CO2 prices due to the possibility of banking allowances. This study explores the impact of non-linear CO2 price trajectories on the European electricity market. Various scenarios, such as increasing prices with the market interest rate, early price peaks, collapse, or fluctuations, are examined until 2030. The study quantifies the effects on power plant investments, profitability, consumer costs, and government revenues. The findings reveal that early and high price signals lead to an earlier transition with 7 % more wind power plants but at 4 % higher costs. Conversely, lower initial prices delay power plant investments in wind power plants by 6 % and result in 21 % higher emissions. Compensation for climate impact costs can yield a positive cost-benefit analysis in scenarios with early and high price signals when emissions are avoided through mechanisms like the Market Stability Reserve (MSR).

Different sensitivity of postglacial vegetation to climate and human disturbance between Tibetan Plateau and its surrounding regions

Characterizing the ecosystems' response to environmental changes on Tibetan Plateau (TP), as well as elucidating the regional disparities of these changes across the plateau's surface and its neighboring regions, holds significance for safeguarding the ecological security of Asia. Here, an improved pollen-based biomization procedure, incorporating weighted taxa, was conducted to illustrate the spatial and temporal differences of vegetation changes in TP, Xinjiang and southwestern China over the past 21,000 years. Results showed that the vegetation changes in TP and Xinjiang were characterized by alternations among herbaceous biomes, while the southwestern China were featured by changes among various woody biomes. Asynchronous vegetation change rates (ROCs) were discovered among TP, Xinjiang and southwestern China. The ROCs in TP were generally low during the Last Glacial Maximum but rose abruptly and peaked in the late deglacial period, whereas Xinjiang and southwestern China had high and increasing ROCs during Holocene. Those trajectories might relate to different sensitivities of vegetation to climate change and human disturbances. Habitat limitation on TP should be pretty intense during glacial period, leading to rapid vegetation establishment after glacier retreat. While in Xinjiang, water availability was the most obvious limitation, making the vegetation change sensitive to moisture and atmospheric CO2 variations. The forest in Yunnan behaved the most stability but was characterized by vertical migration responding to deglacial warming. Holocene acceleration of ROCs aligned with intensified human activities in Xinjiang and southwestern China. To sum up, environmental conditions have been always important in determining post glacial vegetation changes in west China, but the ROCs were asynchronous due to different sensitivities of vegetation to climate change and human activity. Although the plateau's vegetation was quite stable during Holocene, the plateau terrestrial ecosystem may subject to increasing risk under accelerating global warming and intensifying human activities.

Generating Daily High-Resolution Regional XCO2 by Deep Neural Network and Multi-Source Data

CO2 is one of the primary greenhouse gases impacting global climate change, making it crucial to understand the spatiotemporal variations of CO2. Currently, commonly used satellites serve as the primary means of CO2 observation, but they often suffer from striping issues and fail to achieve complete coverage. This paper proposes a method for constructing a comprehensive high-spatiotemporal-resolution XCO2 dataset based on multiple auxiliary data sources and satellite observations, utilizing multiple simple deep neural network (DNN) models. Global validation results against ground-based TCCON data demonstrate the excellent accuracy of the constructed XCO2 dataset (R is 0.94, RMSE is 0.98 ppm). Using this method, we analyze the spatiotemporal variations of CO2 in China and its surroundings (region: 0°–60° N, 70°–140° E) from 2019 to 2020. The gapless and fine-scale CO2 generation method enhances people’s understanding of CO2 spatiotemporal variations, supporting carbon-related research.

Development of a mixed chamber experimental and CFD database of indoor environments

Recent studies reveal that individuals spend the majority of their time indoors, and indoor air quality (IAQ) plays a crucial role in occupant health and productivity. However, current research primarily focuses on specific environments with complex variables and challenging boundary conditions, limiting our understanding of IAQ and thermal comfort. Many studies use fixed experimental chambers, restricting control over boundary conditions. To address these limitations, we propose a novel data-driven smart indoor environmental control framework. This study advocates for a standardized database that integrates diverse IAQ and thermal comfort metrics to enhance indoor environmental conditions and presents the initial phase of this endeavor, detailing the development of a chamber experimental database. This phase involves examining six standardized experiments and over 96 distinct case studies, resulting in a comprehensive dataset of 528,000 entries, rigorously analyzed. Simultaneously, this research enhances its empirical foundation by developing and validating computational fluid dynamics (CFD) models. These models complement and refine the database, increasing its utility and robustness. Notably, experimental findings regarding CO2 and temperature variations indicate a significant 32 % and 65 % decrease in average relative CO2 concentration when the inlet air speed increased from 1.7 to 2.1 and 2.5 m/s, respectively. Spatial variations were also observed, with areas near the inlet surface having lower pollution concentrations and higher air speeds. In summary, this research contributes significantly to the establishment of a comprehensive database, combined with insights from rigorous experimentation and computational modeling, paves the way for advancements in smart indoor environmental control strategies.

Influence of deep stratosphere-to-troposphere transport on atmospheric carbon dioxide and methane at the Mt. Cimone WMO/GAW global station (2165 m a.s.l., Italy): A multi-year (2015–2022) investigation

This work was dedicated to provide a first systematic assessment of the role played by deep stratosphere-to-troposphere transport events (denoted as “stratospheric intrusions–SIs”) in influencing the CO2 and CH4 variability at a high-elevation site in Italy (Monte Cimone, CMN, 2165 m a.s.l.). To select SI events, we used a methodology based on the analysis of the temporal variability of in-situ observed stratospheric tracers (O3, CO and RH), modeled PV along 5-day back-trajectories starting at the measurement site, and satellite measurements of total column ozone (TCO). The analysis of two case studies revealed clear fingerprints of the stratospheric air masses on the atmospheric composition at CMN: CO2 and CH4 decreases were observed in winter, while CO2CH4 increased (decreased) in summer. For 2015–2022, 9.8% of the hourly data were selected as influenced by SI events. The systematic analysis of the SI effect on CO2 confirmed differences between vegetative and non-vegetative seasons: during May–September, air masses from the stratosphere were richer in CO2 (+1.2 ± 0.1 ppm, average ±σ/N). On the other hand, during winter, air masses from higher altitudes were characterised by a lower amount of CO2 (−1.5 ± 0.1 ppm). Furthermore, SI events were usually characterised by an overall decrease in CH4 compared to the remaining data throughout the solar year (with average values ranging from −7.6 to −16.1 ppb as a function of the seasonal period). However, based on the analysis of seasonal anomalies of residuals, it looks that SI events did not play a major role in determining the interannual variability of CO2 and CH4 at CMN.

Differences in stomatal sensitivity to CO2 and light influence variation in water use efficiency and leaf carbon isotope composition in two genotypes of the C4 plant Zea mays.

The ratio of net CO2 uptake (Anet) and stomatal conductance (gs) is an intrinsic measurement of leaf water use efficiency (WUEi) however its measurement can be challenging for large phenotypic screens. Measurements of leaf carbon isotope composition (δ13Cleaf) may be a scalable tool to approximate WUEi for screening because it in part reflects the competing influences of Anet and gs on the CO2 partial pressure (pCO2) inside the leaf over time. However, in C4 photosynthesis the CO2 concentrating mechanism complicates the relationship between δ13Cleaf and WUEi. Despite this complicated relationship, several studies have shown genetic variation in δ13Cleaf across C4 plants. Yet there has not been a clear demonstration of whether Anet or gs are the causal mechanisms controlling WUEi and δ13Cleaf. Our approach was to characterize leaf photosynthetic traits of two Zea mays recombinant inbred lines (Z007E0067 and Z007E0150) which consistently differ for δ13Cleaf even though they have minimal confounding genetic differences. We demonstrate that these two genotypes contrasted in WUEi driven by differences in the speed of stomatal responses to changes in pCO2 and light that lead to unproductive leaf water loss. These findings provide support that differences in δ13Cleaf in closely related genotypes do reflect greater WUEi and further suggests that differences in stomatal kinetic response to changing environmental conditions is a key target to improve WUEi.

Temperature-dependent carbon isotope fractionation in coccolithophores

Introduction: The stable carbon isotope ratio of long-chain alkenones produced by marine haptophyte phytoplankton has often been used to estimate past variations in atmospheric CO2 throughout the Cenozoic. However, previous experimental studies and surveys of alkenones from surface sediment and suspended particulate matter document additional environmental and physiological influences on carbon isotopic fractionation in alkenones.Methods: To clarify the non-CO2 effects on the alkenone carbon isotope fractionations, an important alkenone producer, Gephyrocapsa oceanica, was cultured in laboratory. To separate effects of different environment parameters, G. oceanica was grown in continuous cultures under a matrix of environmental conditions in order to explore the influence of temperature independently of CO2(aq). Through careful manipulation of the media carbon system, we can control the variation of the media CO2(aq) independently of temperature solubility. Carbon isotope fractionations from alkenones, coccolith, and particulate organic carbon were measured from this steady state system.Results and Discussion: We find εp in alkenones and particulate organic carbon inversely correlates with temperature, and temperature affects εp more strongly than CO2(aq). The magnitude of the temperature effect can be explained by higher growth rates at warmer temperatures with a similar growth rate dependence as observed in previous cultures in which growth rate was regulated by other factors. Where the past temperature influence on growth rate could be constrained using the UK’37 alkenone index in the same samples, our finding offers an approach to deconvolve an important physiological factor affecting ancient alkenones εp, and may therefore improve past pCO2 estimates.

Carbon cycling during the India-Asia collision revealed by δ26Mg−δ66Zn−δ98Mo evidence from ultrapotassic volcanoes in NW Tibet

India-Asia continental collision−induced volcanic gas emissions are thought to have played an important role in driving Cenozoic atmospheric CO2 variations, yet the details of how the deep carbon cycle may influence volcanic CO2 degassing are not understood. We present a novel study employing Mg-Zn-Mo isotopic compositions of Cenozoic ultrapotassic lavas from NW Tibet. The negative Mg-Zn isotope correlation (δ26Mg = −0.39‰ to −0.19‰; δ66Zn = +0.27‰ to +0.36‰), bolstered by petrographic analysis of mantle-derived xenoliths from these lavas, demonstrates that the ultrapotassic magmas originated from a lithospheric mantle source that had been enriched by recycled carbonate-bearing sediments rich in calcite and dolomite. Very low δ98Mo values (−0.78‰ to 0‰) relative to the average continental crust (δ98Mo = +0.10‰ to +0.35‰) further indicate that the sedimentary components were derived from the subducted Indian continental crust after its dehydration. Monte Carlo modeling estimates that the input flux of carbon (elemental C) from such sediments into the lithospheric mantle is ∼5.6 Mt/yr, with a predicted CO2 emission rate of ∼15.5 Mt/yr. We suggest that the still ongoing subduction of the Indian tectonic plate has played a crucial role in introducing substantial quantities of carbonate-rich sediments into the Tibetan lithospheric mantle, leading to the sequestration of large amounts of CO2 via carbonatite metasomatism. Hence, partial melting of such a carbon-rich mantle reservoir in an orogenic setting provides the positive feedback mechanism that can explain the high flux of volcanic CO2 during India-Asia collision. These findings not only highlight the importance of continental subduction, sediment recycling, and mantle metasomatism by carbon-rich melts/fluids in the generation of Tibetan ultrapotassic volcanism, but they also show how the deep carbon cycle influences volcanic CO2 degassing.

Spatiotemporal Modeling of Carbon Fluxes over Complex Underlying Surfaces along the North Shore of Hangzhou Bay

Urban areas contribute to over 80% of carbon dioxide emissions, and considerable efforts are being undertaken to characterize spatiotemporal variations of CO2 (carbon dioxide) at a city, regional, and national level, aiming at providing pipelines for carbon mission reduction. The complex underlying surface composition of urban areas makes process-based and physiology-based models inadequate for simulating carbon flux in this context. In this study, long short-term memory (LSTM), support vector machine (SVM), random forest (RF), and artificial neural network (ANN) were employed to develop and investigate their viability in estimating carbon flux at the ecosystem level. All the data used in our study were derived from the long-term chronosequence observations collected from the flux towers within urban complex underlying surface, along with meteorological reanalysis datasets. To assess the generalization ability of these models, the following statistical metrics were utilized: coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). Our analysis revealed that the RF model performed the best in simulating carbon flux over long time series, with the highest R2 values reaching up to 0.852, and exhibiting the smallest RMSE and MAE values at 0.293 μmol·m−2·s−1 and 0.157 μmol·m−2·s−1. As a result, the RF model was chosen for simulating carbon flux at spatial scale and assessing the impact of urban impervious surfaces in the simulation. The results showed that the RF model performs well in simulating carbon flux at the spatial scale. The input of impervious surface area index can improve the performance of the RF model in simulating carbon flux, with R2 values of 84.46% (with the impervious surface area index in) and 83.74% (without the impervious surface area index in). Furthermore, the carbon flux in Fengxian District, Shanghai, exhibited significant spatial heterogeneity: the CO2 flux in the western part of Fengxian District was less than in the eastern part, and the CO2 flux gradually increased from the west to the east. In addition, we creatively introduced the diurnal impervious surface area index based on the Kljun model, and clarified the influence of impervious surface on the spatiotemporal simulation of CO2 flux over the complex urban underlying surface. Based on these findings, we conclude that the RF models can be effectively applied for estimating carbon flux on the complex underlying urban surface. The results of our study reduce the uncertainty in modeling carbon cycling in terrestrial ecosystems, and make the variety of models for the carbon cycling of terrestrial ecosystems more diverse.

In situ measurements of dissolved gases in xylem sap as tracers in plant physiology

Abstract Trees transport gases from the ground into the atmosphere through the process of transpiration. Tracing gases transported through this mechanism continuously and under field conditions remains an experimental challenge. Here we measured gases dissolved in the tree sap in situ and in real time, aiming to simultaneously analyse the transport of several gases (He, Ar, Kr, N2, O2 and CO2) from the soil, through the trees, into the atmosphere. We constructed and inserted custom-made semi-permeable membrane probes in the xylem of a fir tree and measured gas abundances at different heights using a portable gas equilibrium membrane-inlet mass spectrometer (‘miniRUEDI’). With this method, we were able to continuously measure the abundances of He, Ar, Kr, N2, O2 and CO2 in sap over several weeks. We observed diurnal variations of CO2 and O2 concentrations that reflected tree physiological activities. As a proof of the concept that trees do uptake dissolved gases in soil water, we irrigated the tree with He-enriched water in a tracer experiment and were able to determine upward sap flow velocity. Measurements of inert gases together with reactive species, such as CO2 and O2, allowed separation of the physical transport and exchange of gases derived from the soil or atmosphere from biological reactions. We discuss the opportunities that our technique provides for continuous in situ measurements of gases in the tree sap.

Stream CO2 emissions are overestimated without consideration of diel water and atmospheric CO2 variability

AbstractStreams are significant emitters of carbon dioxide (CO2) to the atmosphere that are influenced by diel CO2 dynamics. However, we know little about diel CO2 variability within streams, the diel dynamics of CO2 in the air above streams, and the consequences for emission calculations. We studied five pre‐alpine streams by equipping three sites per stream in close proximity (~ 1 km apart) with automatic logging stations that continuously recorded water and air CO2 partial pressures (pCO2) for 2–4 d. All streams and sites showed increased pCO2 at night and decreased pCO2 during the day, however, with fourfold higher diel amplitudes for atmospheric pCO2 compared to the water. Calculating diffusive CO2 fluxes with fixed compared to dynamic measured atmospheric CO2 resulted in negligible to 431% lower estimates. We might thus currently overestimate fluvial CO2 emissions and should include diel water and air CO2 variability to more accurately assess stream CO2 emissions.