Previous studies have shown that methane emissions from East Africa are globally significant, yet their environmental controls remain poorly constrained. In this study, we use satellite observations of enhanced vegetation index (EVI), land surface temperature (LST), and rainfall to interpret methane emission estimates inferred from Greenhouse gases Observing SATellite (GOSAT) observations over 2010-2020, with a focus on the anomalous emission pulse during 2018-2020. We identify a pronounced seasonal contrast in the relationships between methane emissions and environmental variables. During the long rains, methane emissions are strongly correlated with contemporaneous EVI, LST, and rainfall across both the regional and basin scales. In contrast, during the short rains, cumulative EVI explains more variability in methane emissions ( r = 0.74) than seasonal EVI ( r = 0.35), indicating the importance of lagged ecosystem processes such as vegetation senescence and organic matter decomposition. We interpret this delayed emission peak as likely reflecting the combined influence of vegetation-driven carbon inputs and catchment-scale hydrological transport, whereby biomass production and water redistribution during the long rains contribute to enhanced methane emissions in the subsequent season. Periods of anomalously high EVI (2018-2020) coincide with elevated methane emissions, particularly over the Juba, Tana, Nile, and Rift basins. While these relationships are empirical, our results suggest that vegetation dynamics provide a stronger large-scale empirical indicator of methane variability than rainfall alone, and highlight the importance of accounting for lagged biogeochemical processes when interpreting wetland methane emissions.

Urban greenness is celebrated for its environmental and health benefits, yet its association with ozone (O-3)-related health risk via biogenic volatile organic compound (BVOC) emissions remains underexplored. Integrating WRF-CMAQ model simulations and population-based epidemiological analyses using a case time-series study design, we examined the paradoxical link between greenness and O-3 formation and associated mortality risk in Zhejiang, China, from 2013 to 2020. The association between per IQR increase in O-3 and all-cause mortality risk was significantly stronger in subdistricts with a high greenness exposure level (mortality increase in 4.8%, 95% CI: 3.8-5.7) compared to subdistricts with a low greenness exposure level (1.0%, 95% CI: 0.2-1.7; P for difference < 0.001). This observational association is supported by the WRF-CMAQ model simulations, which estimated a median greenness contribution to O-3 formation of 13.76% (IQR: 5.68%) in summer and 1.16% (0.85%) in winter. Similar difference in mortality risk was also observed for cardiovascular mortality and respiratory mortality. This effect modification was consistent across age and sex subgroups and more pronounced during the cold season and in rural areas. By highlighting the counterintuitive role of greenness in O-3-related health risks, our study offers critical insights for urban greening, emphasizing the need to balance ecological ambitions with public health imperatives for a sustainable environment.

Abstract Extensive observations highlight the significance of terpenoids in urban air quality. However, emission frameworks fail to capture their magnitude and spatial variability, limiting air quality management and atmospheric modeling. Here we reconcile observed and modeled urban terpenoid emissions and uncover intense, underrepresented anthropogenic sources in Greater Los Angeles by combining improved emission inventories, airborne flux measurements, and ground-based mobile measurements. We demonstrate that an enhanced biogenic inventory significantly reduces the discrepancy between modeled and observed isoprene fluxes, and the remaining gap in monoterpenoids is explained predominantly by anthropogenic sectors. We identify that these emissions originate from previously overlooked food industry and commercial businesses, which exhibit a biogenic-like temperature dependence. These anthropogenic sources account for 44% of the top 10% observed monoterpenoid flux hotspots and contribute 17% of the total flux. Given their ubiquity in urban and industrialized regions worldwide, these poorly characterized sources may influence air quality beyond Los Angeles.

Surface ozone concentrations typically increase with temperature, but emerging evidence indicates a decline at extreme-high temperatures. This reversal challenges the prevailing expectation of monotonic ozone increase under global warming and highlights a narrow temperature range in which peak ozone is most likely to occur, which is crucial for predicting extreme pollution events. Despite its importance, the global extent and underlying mechanism of this ozone-temperature reversal remain unclear. Using thousands of monitoring sites across the Northern Hemisphere, we demonstrate that this reversal is widespread and robust. Once temperatures exceed a threshold peak-ozone temperature, enhanced buoyancy-driven convection clears ozone and its precursors, thereby decreasing surface ozone. We show that this peak-ozone temperature corresponds closely to the theoretically derived onset of thermally driven convection and is governed by atmospheric thermodynamic conditions. Climate projections indicate that the peak-ozone temperature will rise due to enhanced convection inhibition under global warming. This upward shift delays the onset of ozone reversal, resulting in higher peak ozone levels and longer periods of elevated ozone exposure during heatwaves. Future mitigation efforts must anticipate the growing risk of co-occurrence of ozone pollution and heat extremes as surface ozone will peak both at higher temperatures and with higher values.

To evaluate the potential of an upcoming large-swath satellite for estimating surface methane (CH₄) fluxes at a weekly scale, we report the results from a series of observing system simulation experiments (OSSEs) that use an established modeling framework that includes the GEOS-Chem 3D atmospheric transport model and an ensemble Kalman filter. These experiments focus on the sensitivity of CH₄ flux estimates to systematic errors (μ) and random errors (σ) in the column average methane (XCH4) measurements. Our control test (INV_CTL) demonstrates that with median errors (μ = 1.0 ± 0.9 ppb and σ = 6.9 ± 1.6 ppb) in XCH₄ measurements over a 1000 km swath, global CH4 fluxes can be estimated with an accuracy of 5.1 ± 1.7%, with regional accuracies ranging from 3.8% to 21.6% across TransCom sub-continental regions. The northern hemisphere mid-latitudes show greater reliability and consistency across varying μ and σ levels, while tropical and boreal regions exhibit higher sensitivity due to limited high-quality observations. In σ-sensitive regions, such as the North American boreal zone, expanding the swath width from 1000 km to 3000 km significantly reduces discrepancies, while such adjustments provide limited improvements for μ-sensitive regions like North Africa. For TanSat-2 mission, with its elliptical medium Earth orbit and 1500 km swath width, the global total estimates achieved an accuracy of 3.1 ± 2.2%. Enhancing the swath width or implementing a dual-satellite configuration is proposed to further improve TanSat-2 inversion performance.

In support of decarbonization goals, utilizing existing natural gas pipelines as a means of storing, transmitting, and distributing renewable hydrogen is being considered. The addition of hydrogen to delivered pipeline gas could alter the performance and pollutant emission rates of end-use equipment used for power generation, industry, and residential and commercial applications. The magnitude of performance and emission changes are determined by many factors including the volume of hydrogen within the gas mixture, the equipment type and operating parameters. Furthermore, changes in emissions can impact regional air quality including the levels of ozone and fine particulate matter (PM2.5). Here, in-lab testing and numerical modeling are used to quantify changes in NOx emissions for industrial burner configurations with different fuel compositions, and then the changes in emissions from hydrogen and methane mixtures derived from these experiments are further used as input to a suite of air quality modeling tools to determine the impact on regional air quality and public health in California for a 20 % by volume hydrogen pipeline gas supply. Assuming the system is managed to ensure end- use equipment is favorable to mixtures of natural gas and hydrogen, projected impacts on state-wide NOx could range from a 6 % decrease to a 4 % increase. During high secondary pollutant formation periods in California, the emission changes for a managed scenario could improve ground-level ozone by 2.4 ppb in July 2035 and PM2.5 by 1.8 mu g/m3 in January 2035. Conversely, increases in ozone could reach 1.6 ppb (July 2035) and PM2.5 levels are enhanced up to 0.63 mu g/m3 (January 2035) under the least favorable assumptions. While changes in air quality are shown to potentially benefit or worsen public health including reducing or increasing the incidence of air pollution related mortality and morbidity, the results demonstrate that a carefully managed transition could offer air quality and public health co-benefits.

The California Air Resources Board's 2022 Scoping Plan (SP) outlines California's pathway to achieve carbon neutrality by 2045, targeting deep reductions in greenhouse gas and criteria air pollutant emissions. Understanding the regional implications of this statewide plan is critical for combating environmental inequalities. In this study, we evaluate air quality, health, and equity cobenefits of the 2022 SP in the South Coast Air Basin (SoCAB), using a high-resolution (1 km × 1 km) chemical transport model. Achieving carbon neutrality could reduce annual basin-wide PM2.5 by 1.72 μg/m3 in 2035 and 3.46 μg/m3 in 2045 under the SP scenario relative to the reference scenario, avoiding approximately 5211 and 12,222 premature deaths among 16.8 million residents. The monetary health benefits are US$49.5 and US$116.2 billion, which exceed the costs estimated within the SP for GHG abatement. While the SP benefits disadvantaged communities─with 41.4% of gains accruing to 42.0% SoCAB residents, several heavily impacted communities adjacent to the Ports of Long Beach still experience higher PM2.5 exposure relative to basin-average levels under SP. Overall, we highlight the considerable benefits of deep decarbonization in reducing ambient PM2.5 levels and safeguarding human health while also demonstrating the need for source-oriented emission regulations for disadvantaged communities.

Flood risk assessment is an effective tool for disaster prevention and mitigation. As land use is a key factor influencing flood disasters, studying the impact of different land use patterns on flood risk is crucial. This study evaluates flood risk in the Chang-Zhu-Tan (CZT) urban agglomeration by selecting 17 socioeconomic and natural environmental factors within a risk assessment framework encompassing hazard, exposure, vulnerability, and resilience. Additionally, the Patch-Generating Land Use Simulation (PLUS) and multilayer perceptron (MLP)/Bayesian network (BN) models were coupled to predict flood risks under three future land use scenarios: natural development, urban construction, and ecological protection. This integrated modeling framework combines MLP’s high-precision nonlinear fitting with BN’s probabilistic inference, effectively mitigating prediction uncertainty in traditional single-model approaches while preserving predictive accuracy and enhancing causal interpretability. The results indicate that high-risk flood zones are predominantly concentrated along the Xiang River, while medium-high- and medium-risk areas are mainly distributed on the periphery of high-risk zones, exhibiting a gradient decline. Low-risk areas are scattered in mountainous regions far from socioeconomic activities. Simulating future land use using the PLUS model with a Kappa coefficient of 0.78 and an overall accuracy of 0.87. Under all future scenarios, cropland decreases while construction land increases. Forestland decreases in all scenarios except for ecological protection, where it expands. In future risk predictions, the MLP model achieved a high accuracy of 97.83%, while the BN model reached 87.14%. Both models consistently indicated that the flood risk was minimized under the ecological protection scenario and maximized under the urban construction scenario. Therefore, adopting ecological protection measures can effectively mitigate flood risks, offering valuable guidance for future disaster prevention and mitigation strategies.

Accurate quantification of anthropogenic CO2 emissions is crucial for mitigating climate change and verifying emission reduction policies. This study conducts a comparative analysis of China's anthropogenic CO2 emissions for the period between 2000 and 2023 based on six widely used bottom-up inventories at their latest version (ODIAC2023, EDGAR2024, MEIC-global-CO2 v1.0, CAMS-GLOB-ANT v6.2, GEMS v1.0, and CEADs). The national total CO2 emissions increase from 3.43 (3.21–3.63) Gt yr−1 in 2000 to 12.03 (11.35–12.98) Gt yr−1 in 2023, with three growth periods: rapid growth (2000–2013, 0.56 ± 0.013 Gt yr−1), near-stagnation (2013–2016, −0.07 ± 0.022 Gt yr−1), and renewed growth (2016–2023, 0.30 ± 0.016 Gt yr−1). Emissions are dominated by the electricity and heat production, and the industry and construction (78 % of total emissions), with the former replacing the latter as the largest source after 2017. EDGAR consistently reports the highest national CO2 emissions, while MEIC provides the lowest, contributing to the large deviations after 2012. EDGAR and MEIC report different spatial distributions of the transport sector. EDGAR concentrates emissions along major roads and MEIC distributes them more diffusely. Extreme outliers (> 105 t CO2 km−2 yr−1, against an average of 102 t CO2 km−2 yr−1) in these inventories arise from discrepancies in point source data in the Carbon Monitoring for Action (CARMA) versus the China Power Emissions Database (CPED). Overall, the uncertainty of total national anthropogenic CO2 emissions is within 5 % (1σ), and the uncertainties are about 10 %–50 % (1σ) at the provincial level. Our study underscores that the improved spatial proxies, consistent regional inventories, and continued methodological updates are essential for improving the robustness of China's CO2 emission assessments and supporting mitigation planning.

Tropospheric ozone (O3) is a ubiquitous pollutant that is detrimental to human health and ecosystems. The Sichuan Basin (SCB), one of the most populous city clusters in China, has experienced more-intense O3 pollution episodes and longer O3 season with more-frequent stagnant conditions over the past decade. In 2022, the prolonged O3 season featured extremely high levels of O3 and region-wide O3 events were observed, posing significant threats to public health. However, it remains unclear to what extent meteorological fields could contribute to O3 anomaly and to the adverse health impacts from extreme O3 season. Here, we investigate the drivers of extreme summer O3 pollution in 2022 over the SCB using a high-resolution Community Multiscale Air Quality (CMAQ) model in conjunction with surface air quality measurements. Further, the health effects of exposure to high levels of O3 are quantified using the Environmental Benefits Mapping and Analysis Program (BenMAP). Both meteorological reanalysis data and the Weather Research and Forecasting (WRF) modeling revealed extreme heat featured by persistent heatwaves in the study period, which significantly perturbed daytime photochemical reactions and primed the landscape for elevating O3. Sensitivity experiments with fixed anthropogenic emissions indicate that unfavorable meteorology and subsequent enhancements in biogenic emissions substantially contributed to O3 anomaly. Importantly, this unprecedented O3 season resulted in 48285 all-cause deaths due to long-term exposure, which is 8064 higher than the same period in 2019 and significantly overtake previous recognition. CMAQ simulations point to that O3 elevation could be partially offset by concurrent 50 % emission reductions on nitrogen oxides (NOx) and volatile organic compounds (VOCs), leading to avoided deaths of 2660. This work highlights the underestimated O3-related mortality burden and pinpoints the necessity of stringent emission regulations toward O3 mitigation in basin topography.

Understanding anthropogenic CO2 emissions in megacities is important for mitigating climate change. In this study, we analyze tower-based carbon dioxide (CO2) mole fraction measurements at Beijing (BJ), Xianghe (XH), and Xinglong (XL) during two periods: November 2018 to March 2019 and November 2019 to March 2020. We compare the in-situ measurements with the simulated CO2 mole fractions from four emission inventories (ODIAC, EDGAR, MEIC, and CAMS) using the Stochastic Time-Inverted Lagrangian Transport model (STILT). Our results show that the STILT model can generally simulate the daily anthropogenic fluctuations in CO2 mole fractions at all three sites (R > 0.6, P < 0.01), with consistent results across the two periods. At BJ, the ODIAC-based simulations have the smallest bias (-4.4 ppm) and RMSE (33.0 ppm) compared to the other three inventories. Emissions within a radius of 5-25 km around the BJ measurement site account for 33.45 %-82.99 % of the CO2 mole fractions enhancements. At XH and XL, the four inventories show a similar performance. The sensitivity of atmospheric CO2 background estimates is evaluated by comparing results from CarbonTracker and the XL background, and the two sets of background estimates differ in bias within 2.3 ppm. Moreover, we quantify the contribution of extremely strong emission sources (>3 sigma) in ODIAC, and find that they only contribute less than 5 % to CO2 mole fraction enhancements, depending on the distance between these extremely strong emissions and the measurement location.

Assessing the impact of biomass burning (BB) emissions on tropospheric ozone is critical for understanding air pollution and climate interactions. BB emission inventories like Global Fire Emissions Database and Global Fire Assimilation System, typically based on sun-synchronous satellite observations, report emissions on daily, weekly or longer timescales with empirically derived factors generally used to overlay diurnal variations. To explore the sensitivity of tropospheric ozone to diurnal variability, we incorporated day-specific hourly BB variations inferred from geostationary satellite data into the GEOS-Chem atmospheric chemistry transport model. The simulations were compared with those using established inventories and evaluated against in situ and satellite observations. Simulations with real hourly-resolved emissions produce comparable surface ozone biases (−1.54 to +9.09 ppbv vs. −1.58 to +9.13 ppbv) and marginally higher correlations with TROPOMI nitrogen dioxide (r=0.80–0.89) and OMI ozone (r=0.80–0.94). Although the statistical improvements are limited, the geostationary-driven approach reveals pronounced regional ozone differences and mechanistic insights into the role of diurnal fire variability. Data-driven diurnal BB variations across Africa cause significant surface ozone changes (−8.57 to +21.88 ppbv) and alter tropospheric ozone columns by −0.41 to 1.09 DU, particularly in regions with intense fire activity like Angola and Zambia. These changes propagate globally, shifting regional OH concentrations by −4.4 % to +51.7 %. These findings emphasize the critical role of accurately describing diurnal BB variations in atmospheric models to better quantify its impacts on atmospheric composition, providing insights for Earth system model development and the use of geostationary-derived BB emissions datasets.

This paper presents a multivariable linear regression calibration method for non-dispersive infrared (NDIR) CO2 sensors in a low-cost carbon monitoring network. We test this calibration method with data collected in a temperature- and pressure-controlled laboratory and evaluate the calibration method with long-term observational data collected at the Xinglong Atmospheric Background Observatory. Compared to data collected by a high-accuracy cavity ring-down spectrometer (Picarro), the results show that a multivariable linear regression approach incorporating temperature, pressure, and relative humidity can reduce the mean absolute bias from 5.218 ppm to 0.003 ppm, with root mean square errors (RMSE) within 2.1 ppm after calibration. For field observations, the RMSE is reduced from 8.315 ppm to 2.154 ppm, and the bias decreases from 39.170 ppm to 0.018 ppm. The calibrated data can effectively capture the diurnal variation of CO2 mole fraction. The test of the number of reference data shows that about 10 days of co-located reference data are sufficient to obtain reliable measurements. Calibration windows taken from winter or summer provide better results, suggesting a strategy to optimize short-term calibration campaigns.

Satellite-based monitoring of atmospheric column-averaged dry-air mole fraction (XCH4) is essential for quantifying methane (CH4) emissions, yet uncharacterized spatially varying biases in XCH4 observations can cause misattribution in flux estimates. This study assesses the potential of the upcoming TanSat-2 satellite mission to estimate China’s CH4 emission using a series of Observing System Simulation Experiments (OSSEs) based on an Ensemble Kalman Filter (EnKF) inversion framework coupled with GEOS-Chem on a 0.5° × 0.625° grid, alongside an evaluation of current TROPOMI-based products against Total Carbon Column Observing Network (TCCON) observations. Assuming a target precision of 8 ppb, TanSat-2 could achieve an annual national emission estimate accuracy of 2.9% ± 4.2%, reducing prior uncertainty by 84%, with regional deviations below 5.0% across Northeast, Central, East, and Southwest China. In contrast, limited coverage in South China due to persistent cloud cover leads to a 26.1% discrepancy—also evident in pseudo TROPOMI OSSEs—highlighting the need for complementary ground-based monitoring strategies. Sensitivity analyses show that satellite retrieval biases strongly affect inversion robustness, reducing the accuracy in China’s total emission estimates by 5.8% for every 1 ppb increase in bias level across scenarios, particularly in Northeast, Central and East China. We recommend expanding ground-based XCH4 observations in these regions to support the correction of satellite-derived biases and improve the reliability of satellite-constrained inversion results.

Evaluating the efficacy of climate mitigation measures requires quantifying urban greenhouse gas (GHG) emissions. Both anthropogenic and biogenic GHG fluxes are important in urban systems, and disaggregation is necessary to understand urban GHG fluxes. In urban environments one common source of biogenic carbon dioxide (CO2) fluxes is turfgrass. We use CO2 fluxes measured using eddy covariance over a cemetery (less managed) and golf course (more managed) to investigate the contribution of turfgrass lawns to biogenic CO2 fluxes in Indianapolis, IN. We assess the ability of a simple light-use efficiency model, the Vegetation Photosynthesis and Respiration Model (VPRM), commonly used to create prior fluxes necessary for determining urban carbon dioxide (CO2) fluxes via inversion modeling, to represent daily and seasonal patterns in turfgrass CO2 fluxes. Our results show that the existing VPRM Plant Functional Types (PFTs) cannot capture observed daily and seasonal fluxes at either location. We then use data from these sites to create a new turfgrass PFT for the VPRM. We find that less-managed lawns like cemeteries are best represented by different parameters than heavily managed lawns like golf courses, and seasonally changing parameters best match the observed fluxes. We then use the new turfgrass PFT within the VPRM to explore daily and seasonal variability in turfgrass fluxes and their impact, integrated across the city, on urban ecosystem CO2 fluxes. This study illustrates the importance of representing turfgrass as a unique PFT when quantifying urban GHG fluxes and the biases resulting from misrepresentation.

Fine particulate matter (PM2.5) pollution is a critical air quality concern which poses threats to public health. Despite strict air pollution control measures implemented in China since 2013, PM2.5 exceedances and region-wide PM2.5 episodes are still frequently observed in the Sichuan Basin (SCB) located in southwestern China. Here, we examine ambient PM2.5 pollution within the SCB from 2013 to 2020, focusing on emission sources, trends, and health outcomes. By integrating ambient measurements, emission inventories, and the health impact model, our findings reveal a notable decrease in PM2.5 levels across the basin, with the Chengdu Plain showing a significant reduction of 56 μg/m3 in 2020 compared to 2013. Despite these improvements, it is still challenging for densely populated cities to attain the national air quality standards. We highlight a 46.8 % reduction in PM2.5 emissions from 2013 to 2020, driven largely by decreased emissions from residential and industrial sources, which accounted for an average of 38.6 % and 50.3 % of total reduced emissions, respectively. In contrast, the decreases of NOx emissions (26.0 %) were less pronounced compared to PM2.5 due to modest reductions from industrial and transportation sectors. Health impact assessments at 1 km × 1 km using the GEMM model attributes 157,637 deaths to long-term PM2.5 exposure in the SCB for 2017, with stroke and ischemic heart disease identified as leading causes. Further analysis indicates that significant variations in population density could greatly amplify the health impacts of long-term PM2.5 exposure, highlighting the need to prioritize PM2.5 reduction strategies specifically targeting megacities to maximize health benefits. These findings underscore the critical need for ongoing emission reduction efforts and the implementation of targeted pollution control measures to further improve air quality and reduce mortality burden in the SCB.

In recent years, wildfires in California have increased in frequency and intensity due to climate change and prolonged drought. The air pollutants released by wildfires cause significant health consequences, among which polycyclic aromatic hydrocarbons (PAHs) are particularly toxic. Estimating PAH emissions from wildfires is challenging due to variability in vegetation types. In this study, we estimate PAH emission rates across California at a high resolution, based on laboratory-measured PAH emission rates from 22 different vegetation types and detailed vegetation mapping. By combining these estimates with biomass burning data from the NCAR Fire Inventory, the Community Multiscale Air Quality Modeling System simulates PAH concentrations for the 2017 fire season. The modeling results compare favorably to measurements from three PAH monitoring sites in California. The peak PAH emissions from wildfire events are up to be 80 times higher in the gas phase and 32 times higher in the particle phase compared to a case without fire emissions. The population-weighted PAH concentrations from the fire case (0.053 µg/m3) are 47 % higher compared to a non-fire case (0.036 µg/m3) in the particle phase and 11 % higher in the gas phase (9.82 ppt compared to 8.83 ppt) during the study period. While highly depended on the meteorological condition, the simulated spatial distribution indicates that gas-phase PAHs are less likely to travel long distances from the fire source and are prone to aging into the particle phase during transport. Consequently, populations are more likely to be exposed to particle-phase PAHs during wildfire events. This finding has important implications for understanding the health impacts of wildfire-induced PAH concentrations, as particle-phase PAHs may have different toxicological effects compared to gas-phase PAHs.

Eddy-covariance (EC) flux measurements in Indianapolis were used to quantify the impact of the COVID-19 lockdown on CO and CO2 emissions from a highway and a suburban neighborhood. CO2 fluxes were measured for 6 weeks pre-lockdown (January 22, 2020–March 3, 2020) and during lockdown (March 25, 2020– May 5, 2020) using EC instrumentation at 41 m AGL. Fossil fuel CO2 emissions (CO2ff) were estimated by calculating eddy diffusivity to obtain CO flux and then scaling by the CO:CO2ff emissions ratio (RCO). Flux measurements segregated by wind direction were compared to hourly emissions from the 2020 Hestia inventory model. The lockdown CO2ff average weekday emissions from the highway estimated by EC decreased by 51.5 ± 10.9% (11.2 ± 2.2 µmol m−2 s−1) compared to pre-lockdown, similar to Hestia’s estimate 56 ± 7% (12 ± 1 µmol m−2 s−1). The EC measurements detected a significant (2.2 ± 0.7 µmol m−2 s−1) but smaller magnitude decrease in CO2ff emissions from the suburban neighborhood. The daily cycles of CO2ff emissions were significantly correlated with Hestia estimates from the highway but not from the suburbs. This study demonstrates that EC flux towers and high-resolution inventory models in regions with mixed and spatially heterogeneous sources can quantify abrupt changes in sector- and source-specific CO2 fluxes.

Improving the mitigation of global greenhouse effects requires drastic reductions in carbon emissions. Effective carbon management at different spatial scales is therefore crucial to a more sustainable path of economic development. Regional and local carbon management currently rely on the traditional bottom-up method of emission reports, which are known for their opaque and complex procedures. Novel top-down carbon monitoring methods including space-based, aerial and ground-based observations are tantalising alternatives to more reliable data sources of carbon emissions as well as sinks. Unmanned Aerial Vehicles (UAVs) based carbon monitoring is an upcoming field because of its lower costs, higher spatial-temporal resolution and operability at a wider range of weather conditions compared to satellite observations. UAVs equipped with Non-dispersive Infrared (NDIR) carbon dioxide (CO2) sensors can measure in the air with faster response and can intelligently navigate themselves within areas of greater interest. With the help of lighter NDIR payloads and multiple UAVs flying in synchrony, larger cross sections or areas for maximising information acquisition of emission plumes can be covered. In this article, we first introduce the design of a miniaturised NDIR payload, which is compatible with a lightweight UAV to achieve a single system weighing less than 4kg. Then, we test our design via comparisons against another NDIR payload to determine its accuracy (-0.82ppm and -0.44ppm) and precision (0.12ppm and -0.03ppm) in measurements of two vertical concentration profiles. Finally, we deployed four UAVs on two major anthropogenic emission source monitoring campaigns with coordinated flight patterns. To fully understand the benefit of multi-UAV carbon monitoring, we attempted two flight patterns and computed power plant emission rates via the cross-sectional flux method. While all flights successfully captured the concentration enhancements downwind of the power plants, the single-sectional flight pattern measured an emission rate of 10ktCO2/day which is 44% less than that reported monthly and the multi-sectional flight pattern yielded relatively varied results (8.5ktCO2/day on average, which is 54% less) for the respective sections due to wind underestimations, plume section undersampling and wind fluctuations.