As the most abundant greenhouse gas in the atmosphere, CO₂ has a significant impact on climate change. Therefore, the determination of the temporal and spatial distribution of CO₂ is of great significance in climate research. However, existing CO₂ monitoring methods have great limitations, and it is difficult to obtain large-scale monitoring data with high spatial resolution, thus limiting the effective monitoring of carbon sources and sinks. To obtain complete Chinese daily-scale CO₂ information, we used OCO-2 XCO₂ data, Carbon Tracker XCO₂ data, and multivariate geographic data to build a model training data set, which was then combined with various machine learning models including Random Forest, Extreme Random Forest, XGBoost, LightGBM, and CatBoost. The results indicated that the Random Forest model presented the best performance, with a cross-validation R² of 0.878 and RMSE of 1.123 ppm. According to the final estimation results, in terms of spatial distribution, the highest multi-year average RF XCO₂ value was in East China (406.94 ± 0.65 ppm), while the lowest was in Northwest China (405.56 ± 1.43 ppm). In terms of time, from 2016 to 2018, the annual XCO₂ in China continued to increase, but the growth rate showed a downward trend. In terms of seasonal effects, the multi-year average XCO₂ was highest in spring (407.76 ± 1.72 ppm) and lowest in summer (403.15 ± 3.36ppm). Compared with the Carbon-Tracker data, the XCO₂ data set constructed in this study showed more detailed spatial changes, thus, can be effectively used to identify potentially important carbon sources and sinks.

NO₂ (nitrogen dioxide) is a common pollutant in the atmosphere that can have serious adverse effects on the health of residents. However, the existing satellite and ground observation methods are not enough to effectively monitor the spatiotemporal heterogeneity of near-surface NO₂ concentrations, which limits the development of pollutant remediation work and medical health research. Based on TROPOMI-NO₂ tropospheric column concentration data, supplemented by meteorological data, atmospheric condition reanalysis data and other geographic parameters, combined with classic machine learning models and deep learning networks, we constructed an ensemble model that achieved a daily average near-surface NO₂ of 0.03° exposure. In this article, a meteorological hysteretic effects term and a spatiotemporal term were designed, which considerably improved the performance of the model. Overall, our ensemble model performed better, with a 10-fold CV R² of 0.89, an RMSE of 5.62 µg/m³, and an MAE of 4.04 µg/m³. The model also had good temporal and spatial generalization capability, with a temporal prediction R² and a spatial prediction R² of 0.71 and 0.81, respectively, which can be applied to a wider range of time and space. Finally, we used an ensemble model to estimate the spatiotemporal distribution of NO₂ in a coastal region of southeastern China from May 2018 to December 2020. Compared with satellite observations, the model output results showed richer details of the spatiotemporal heterogeneity of NO₂ concentrations. Due to the advantages of using multi-source data, this model framework has the potential to output products with a higher spatial resolution and can provide a reference for downscaling work on other pollutants.

In the context of global warming, the Wuhan Urban Agglomeration is actively responding to China’s carbon peak and carbon neutrality goals and striving to achieve a reduction in carbon sources and an increase in carbon sinks. Therefore, it is critical to investigate carbon emissions from land use. This study uses the carbon emission coefficient method to calculate carbon emissions from land use in the Wuhan Urban Agglomeration, analyzes its temporal and spatial changes and differences in urban structure, and couples with the Markov–PLUS model to simulate and predict the carbon emissions of four scenarios of land use in 2035. The research found the following: (1) during the Wuhan “1+8” City Circle stage, carbon sources and emissions increased steadily, with average annual growth rates of 1.92% and 1.99%, respectively. Carbon sinks remained stable and then decreased, with an average annual growth rate of −0.46%. (2) During the Wuhan Metropolitan Area stage—except for 2020 and 2021, which were affected by COVID-19—carbon sources, sinks, and emissions continued to grow in general, and the average annual growth rates increased to 4.46%, 1.58%, and 4.51%, respectively. (3) In terms of urban structure differences, Wuhan is a high-carbon optimization zone; Xianning, Huangshi, and Huanggang are ecological protection zones; other cities, such as Ezhou, Xiaogan, and Xiantao are comprehensive optimization zones; and there is no low-carbon development zone. (4) The multi-scenario simulation results show that carbon sources and emissions are the highest under the economic development scenario, with values of 100.2952 and 9858.83 million tons, respectively, followed by cropland protection, natural development, and low-carbon development scenarios. Under low-carbon development, carbon sinks were the highest, with values of 1.9709 million tons, followed by natural development, economic development, and cropland protection scenarios. The research results are conducive to the formulation of carbon peak and neutrality goals as well as low-carbon development plans for the Wuhan Urban Agglomeration.

High concentrations of ground-level ozone (O₃) pose a significant threat to human health. Obtaining high-spatiotemporal-resolution information about ground-level O₃ is of paramount importance for O₃ pollution control. However, the current monitoring methods have a lot of limitations. Ground-based monitoring falls short in providing extensive coverage, and remote sensing based on satellites is constrained by specific spectral bands, lacking sensitivity to ground-level O₃. To address this issue, we combined brightness temperature data from the Himawari-8 satellite with meteorological data and ground-based station data to train four machine learning models to obtain high-spatiotemporal-resolution information about ground-level O₃, including Categorical Boosting (CatBoost), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LGBM), and Random Forest (RF). Among these, the CatBoost model exhibited superior performance, achieving a ten-fold cross-validation R² of 0.8534, an RMSE of 17.735 μg/m³, and an MAE of 12.6594 μg/m³. Furthermore, all the selected feature variables in our study positively influenced the model. Subsequently, we employed the CatBoost model to estimate averaged hourly ground-level O₃ concentrations at a 2 km resolution. The estimation results indicate a close relationship between ground-level O₃ concentrations and human activities and solar radiation.

Recently, the near-ground ozone pollution has become an important factor restricting economic development and ecological environment protection. Due to the aging equipment of satellite sensors and the limitations of spatial resolution, the current approach utilizing satellite remote sensing observation faces challenges in effectively monitoring small-scale areas with sufficient data. Taking the near-ground ozone concentration as the research object, this work combined multiple classical machine learning (ML) methods based on tree models, and developed a temporally ensemble model to achieve the estimation of near-surface ozone in the 1 km ² ; area of the Yangtze River Delta region in China. In the ensemble model, the coefficient of determination (R ² ) of the 10-fold cross-validation was 0.91, and the Root Mean Square Error (RMSE) was 9.21μg/m ³ . All evaluation indicators confirms that our approach was more accurate than some conventional ML models. The predicted spatial errors were evenly distributed, which indicated the superior spatial stationarity of the ensemble model. On the temporal scale, the ozone distribution predicted by the model agreed well with the results of ground-based meteorological station monitoring, both showing distinct seasonal trends. On the spatial scale, the model output reflected well the refined spatial variation of near-ground ozone at a small scale, and captured the “medium-high-low” trend of near-ground ozone concentration in Shanghai and the trend of “low-medium” in Hangzhou, China. In contrast, the satellite observation data cannot well reflect the differences in details. In the future, this model has good application potential in the refined monitoring of polluting gases across the country.

Groundwater depletion has emerged as a pressing global challenge, yet the low spatial resolution (0.25°) of Gravity Recovery and Climate Experiment (GRACE) satellite data limits its application in regional groundwater monitoring. In this study, based on 0.25° spatial resolution groundwater storage anomalies (GWSAs) data derived from GRACE satellite observations and GLDAS hydrological model outputs, supplemented with hydrological data, humanities data, and other geographic parameters, we constructed a Stacking-based ensemble machine learning model that achieved a 1 km spatial resolution of GWSAs distribution data across the contiguous United States (CONUS) from 2010 to 2020. The ensemble model integrates eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Categorical Boosting (CatBoost) models using an Attention-Based Dynamic Weight Allocation (ADWA) approach, along with a ridge regression model. The results indicate that our ensemble model outperforms individual machine learning (ML) models, achieving a coefficient of determination (R²) of 0.929, root mean square error (RMSE) of 25.232 mm, mean absolute error (MAE) of 19.125 mm, and Nash–Sutcliffe efficiency (NSE) of 0.936, validated by 10-fold cross-validation. In situ measurements indicate that, compared with the original data, approximately 61.7% of the monitoring wells (266 out of 431) exhibit a higher correlation after downscaling, with the overall correlation coefficient increasing by about 18.7%, which suggests that the downscaled product exhibits an appreciable improvement in accuracy. The ensemble model proposed in this study, by integrating the advantages of various ML algorithms, is better able to address the complexity and uncertainty of groundwater storage variations, thus providing scientific support for the sustainable management of groundwater resources.