Search

In the field of remote sensing image denoising, the current mainstream methods usually only consider using clean or noisy images to guide the network in the training phase. Most of them only apply to specific types of noise, and the denoising effect is not satisfactory enough, with problems such as artifacts and noise residues. In this article, we endeavor to deal with a wide range of noise types, preserving as much detailed information in the image as possible and aiming to address the relevant limitations. Inspired by contrastive learning, we propose a remote sensing image denoising framework based on contrastive learning, named RSID-CR, which constructs positive and negative sample pairs between clean, noisy, and denoised images. Then, we construct a joint loss function consisting of reconstruction loss and contrastive regularization as a guide signal to train the denoising network, such that the denoised image is pushed closer to the clean image and farther away from the noisy image in the feature space. We conduct extensive experiments on two public datasets for five types of noise often present in remote sensing images. In addition, we validate our method using two real noisy remote sensing datasets. The experimental results indicate that our proposed method achieves satisfactory outcomes.

Abstract To reduce the uncertainty estimation of clouds and improve the forecast of surface shortwave radiation (SSR) over the Tibetan Plateau, a new cloud assimilation system is proposed which is the first attempt to directly apply the four-dimensional local ensemble transform Kalman filter method to assimilate the cloud optical thickness (COT). The high-resolution spatial and temporal data assimilated from the next-generation geostationary satellite Himawari-8, with the high-assimilation frequency, realized an accurate estimation of the clouds and radiation forecasting. The COT and SSR were significantly improved after the assimilation by independent verification. The correlation coefficient (CORR) of the SSR was increased by 11.3%, and the root-mean-square error (RMSE) and mean bias error (MBE) were decreased by 28.5% and 58.9%, respectively. The 2-h cycle assimilation forecast results show that the overestimation of SSR has been effectively reduced using the assimilation system. These findings demonstrate the high potential of this assimilation technique in forecasting of SSR in numerical weather prediction. The ultimate goal that to improve the model forecast through the assimilation of cloud properties requires further studies to achieve.

The Environmental Trace Gases Monitoring Instrument (EMI-II) onboard the Chinese GaoFen-5B (GF5B) and DaQi-1 (DQ1) satellites is the successor of the previous EMI onboard the Chinese GaoFen-5 (GF5) satellite, and has a higher spatial resolution and a better signal-to-noise ratio. The GF5B and DQ1 were launched in September 2021 and April 2022, respectively. As part of China’s ultraviolet-visible hyperspectral satellite instrument series, the EMI-II aims to conduct network observations of pollution gases globally in the morning and early afternoon. In this study, NO2 data were retrieved from the EMI-II payloads on the GF5B and DQ1 satellites using the Differential Optical Absorption Spectroscopy (DOAS) algorithm. The two satellites were consistently compared, and the results showed strong consistency on various spatial and temporal scales (R2 > 0.8). In four representative regions worldwide, NO2 data from the EMI-II exhibited good spatial consistency with those from the TROPOMI. The correlation coefficient (R2) of the total vertical column density (VCD) between the EMI-II and TROPOMI exceeded 0.85, and that of the tropospheric NO2 VCD exceeded 0.57. Compared with single-satellite observations, the dual-satellite network of the GF5B and DQ1 can effectively increase the observation frequency. On a daily scale, dual-satellite observations can reduce the impact of cloud coverage by 6–8% compared to single-satellite observations, and there are two valid observations of nearly 50% of the world’s regions. Additionally, the differences between the two satellites can reflect the NO2 diurnal variations, which demonstrates the potential for studying pollutant gas diurnal variations.

Fugitive dust is an important source of total suspended particulate matter in urban ambient air. The existing segmentation methods for dust sources face challenges in distinguishing key and secondary features, and they exhibit poor segmentation at the image edge. To address these issues, this paper proposes the Dust Source U-Net (DSU-Net), enhancing the U-Net model by incorporating VGG16 for feature extraction, and integrating the shuffle attention module into the jump connection branch to enhance feature acquisition. Furthermore, we combine Dice Loss, Focal Loss, and Activate Boundary Loss to improve the boundary extraction accuracy and reduce the loss oscillation. To evaluate the effectiveness of our model, we selected Jingmen City, Jingzhou City, and Yichang City in Hubei Province as the experimental area and established two dust source datasets from 0.5 m high-resolution remote sensing imagery acquired by the Jilin-1 satellite. Our created datasets include dataset HDSD-A for dust source segmentation and dataset HDSD-B for distinguishing the dust control measures. Comparative analyses of our proposed model with other typical segmentation models demonstrated that our proposed DSU-Net has the best detection performance, achieving a mIoU of 93% on dataset HDSD-A and 92% on dataset HDSD-B. In addition, we verified that it can be successfully applied to detect dust sources in urban areas.

Ozone stands out as a crucial trace gas within the Earth’s atmosphere, exerting a substantial influence on climate change and air pollution. Tropospheric ozone plays an important role in the formation of photochemical smog, and its variations are associated with human activities. The utilization of satellite remote sensing technology for tropospheric ozone monitoring enables a quantitative analysis of its global and regional spatiotemporal characteristics. It also facilitates the investigation of the mechanisms involved in ozone formation within the troposphere. The significant progress in product accuracy and spatiotemporal resolution of ozone remote sensing products, including total ozone and vertical profiles, can be attributed to the extensive development of satellite remote sensing techniques. Nevertheless, the precision of tropospheric ozone products remains inadequate for contemporary scientific purposes, primarily because of faint signals in the lower atmosphere, the intricate nature of the underlying surface, and the existence of clouds and aerosols. This study places emphasis on the satellite remote sensing of tropospheric ozone, encompassing a comprehensive review of the advancements in satellite sensors and the characteristics and suitability of various retrieval algorithms. Moreover, this research delves into the possible utilization of satellite remote sensing for the provision of reliable tropospheric ozone observation data on a global and regional level.

Abstract The multi‐spectral satellite sensors such as MODIS have a large swath, high spatial resolution, and well onboard calibration, enabling aerosol retrievals with daily global coverage. Despite numerous available bands, MODIS aerosol algorithms over land typically only utilize measurements from 2 to 3 spectral wavelengths to retrieve Aerosol Optical Depth (AOD) based on prescribed aerosol models. To make full use of multi‐spectral measurements and prior information, we developed an aerosol algorithm based on physics‐informed deep learning (PDL) approach. With physical constraint from radiative transfer simulation, PDL can construct model functions between the whole spectral measurements and each retrieved aerosol parameter separately. AERONET validations in eastern China show that MODIS PDL algorithm can accurately retrieve AOD and fine AOD (R = 0.936) at 1 km resolution and has reliable performance in coarse AOD as well as notable sensitivity to aerosol absorption. The flexible and efficient PDL method provides a generalized algorithm for common multi‐spectral satellite measurements.

Abstract Daprodustat is an oral small molecule hypoxia‐inducible factor (HIF) prolyl hydroxylase inhibitor (PHI) approved in Japan and the United States for the treatment of anemia associated with chronic kidney disease. This phase 1, nonrandomized, 2‐period, crossover study in 6 healthy men characterized and quantified the metabolites generated after a microtracer IV infusion of 50 μg (125 nCi) [14C]‐daprodustat administered concomitantly with a nonradiolabeled therapeutic dose of a 6‐mg daprodustat tablet, followed by a single oral solution dose of 25 mg (62.5 μCi) [14C]‐daprodustat. High‐performance liquid chromatography (HPLC) coupled with radioactivity detection (TopCount or AMS) and HPLC‐tandem mass spectrometry (HPLC‐MSn) were used for quantitative measurement and structural identification of radioactive metabolites in plasma, urine, feces, and bile. Following oral administration of [14C]‐daprodustat, unchanged daprodustat was the principal circulating drug‐related component, accounting for 40% of plasma radioactivity. Predominant oxidative metabolites M2, M3, M4, and M13 individually represented 6–8% of the plasma radioactivity and together accounted for the majority of radioactivity in urine and feces (53% in both matrices; 12% and 41% of dose, respectively). Unchanged daprodustat was not detected in urine and was only 0.7% of total radioactivity in feces (

This study utilized an infrared spotlight Hyperspectral infrared Atmospheric Sounder (HIRAS) and the Medium Resolution Spectral Imager (MERSI) mounted on FY3D cloud products from the National Satellite Meteorological Center of China to obtain methane profile information. Methane inversion channels near 7.7 μm were selected based on the different distribution of methane weighting functions across different seasons and latitudes, and the selected retrieval channels had a great sensitivity to methane but not to other parameters. The optimization method was employed to retrieve methane profiles using these channels. The ozone profiles, temperature, and water vapor of the European Centre for Medium-Range Weather Forecasts (ECMWF) fifth-generation reanalysis data (ERA5) were applied to the retrieval process. After validating the methane profile concentrations retrieved by HIRAS, the following conclusions were drawn: (1) compared with Civil Aircraft for the Regular Investigation of the Atmosphere Based on an Instrument Container (CARIBIC) flight data, the average correlation coefficient, relative difference, and root mean square error were 0.73, 0.0491, and 18.9 ppbv, respectively, with lower relative differences and root mean square errors in low-latitude regions than in mid-latitude regions. (2) The methane profiles retrieved from May 2019 to September 2021 showed an average error within 60 ppbv compared with the Fourier transform infrared spectrometer (FTIR) station observations of the Infrared Working Group (IRWG) of the Network for the Detection of Atmospheric Composition Change (NDACC). The errors between the a priori and retrieved values, as well as between the retrieved and smoothed values, were larger by around 400–500 hPa. Apart from Toronto and Alzomoni, which had larger peak values in autumn and spring respectively, the mean column averaging kernels typically has a larger peak in summer.

This study employs an ensemble Kalman filter assimilation method to validate and update the pollutant emission inventory to mitigate the impact of uncertainties on the forecasting performance of air quality numerical models. Based on nationwide ground-level pollutant monitoring data in China, the emission inventory for the entire country was inverted hourly in 2022. The emission rates for PM2.5, CO, NOx, SO2, NMVOCs, BC, and OC updated by the inversion were determined to be 6.6, 702.4, 37.2, 13.4, 40.3, 3, and 18.2 ng/s/m2, respectively. When utilizing the inverted inventory instead of the priori inventory, the average accuracy of all cities’ PM2.5 forecasts was improved by 1.5–4.2%, especially for a 7% increase on polluted days. The improvement was particularly remarkable in the periods of January–March and November–December, with notable increases in the forecast accuracy of 12.5%, 12%, and 6.8% for the Northwest, Northeast, and North China regions, respectively. The concentration values and spatial distribution of PM2.5 both became more reasonable after the update. Significant improvements were particularly observed in the Northwest region, where the forecast accuracy for all preceding days was improved by approximately 15%. Additionally, the underestimated concentration of PM2.5 in the priori inventory compared to the observation value was notably alleviated by the application of the inversion.

O3 poses a significant threat to human health and the ecological environment. In recent years, O3 pollution has become increasingly serious, making it difficult to accurately control O3 precursor emissions. Satellite indicator methods, such as the FNR (formaldehyde-to-nitrogen dioxide ratio (HCHO/NO2 ratio)), provide an effective way to identify ozone pollution control areas on a large geographical scale due to their simple acquisition of datasets. This can help determine the primary factors contributing to O3 pollution and assist in managing it. Based on TROPOMI data from May 2018 to December 2022, combined with ground-based monitoring data from the China National Environmental Monitoring Centre, we explored the uncertainty associated with using the HCHO/NO2 ratio (FNR) as an indicator in ozone control area determination. We focused on the four representative regions in China: Jing-Jin-Ji-Lu-Yu (JJJLY), Jiang-Zhe-Hu-Wan (JZHW), Chuan-Yu (CY), and South China. By using the statistical curve-fitting method, we found that the FNR thresholds were 3.5–5.1, 2.0–4.0, 2.5–4.2, and 1.7–3.5, respectively. Meanwhile, we analyzed the spatial and temporal characteristics of the HCHO, NO2, and O3 control areas. The HCHO concentrations and NO2 concentrations had obvious cyclical patterns, with higher HCHO column densities occurring in summer and higher NO2 concentrations in winter. These high values always appeared in areas with dense population activities and well-developed economies. The distribution characteristics of the ozone control areas indicated that during O3 pollution periods, the urban areas with industrial activities and high population densities were primarily controlled by VOCs, and the suburban areas gradually shifted from VOC-limited regimes to transitional regimes and eventually reverted back to VOC-limited regimes. In contrast, the rural and other remote areas with relatively less development were mainly controlled by NOx. The FNR also exhibited periodic variations, with higher values mostly appearing in summer and lower values appearing in winter. This study identifies the main factors contributing to O3 pollution in different regions of China and can serve as a valuable reference for O3 pollution control.

Oil wells play an important role in the extraction of oil and gas, and their future potential extends beyond oil and gas exploitation to include the development of geothermal resources for sustainable power generation. Identifying and detecting oil wells are of paramount importance given the crucial role of oil well distribution in energy planning. In recent years, significant progress has been made in detecting single oil well objects, with recognition accuracy exceeding 90%. However, there are still remaining challenges, particularly with regard to small-scale objects, varying viewing angles, and complex occlusions within the domain of oil well detection. In this work, we created our own dataset, which included 722 images containing 3749 oil well objects in Daqing, Huatugou, Changqing oil field areas in China, and California in the USA. Within this dataset, 2165 objects were unoccluded, 617 were moderately occluded, and 967 objects were severely occluded. To address the challenges in detecting oil wells in complex occlusion scenarios, we propose the YOLOv5s-seg CAM NWD network for object detection and instance segmentation. The experimental results show that our proposed model outperforms YOLOv5 with F1 improvements of 5.4%, 11.6%, and 23.1% observed for unoccluded, moderately occluded, and severely occluded scenarios, respectively.

In this paper, a multi-focus image fusion algorithm via the distance-weighted regional energy and structure tensor in non-subsampled contourlet transform domain is introduced. The distance-weighted regional energy-based fusion rule was used to deal with low-frequency components, and the structure tensor-based fusion rule was used to process high-frequency components; fused sub-bands were integrated with the inverse non-subsampled contourlet transform, and a fused multi-focus image was generated. We conducted a series of simulations and experiments on the multi-focus image public dataset Lytro; the experimental results of 20 sets of data show that our algorithm has significant advantages compared to advanced algorithms and that it can produce clearer and more informative multi-focus fusion images.

Abstract: There are significant uncertainties in the retrieval accuracy of multilayer clouds with different phase states, leading to bias in the subsequent estimation of the surface downward shortwave radiation (DSR). Single-layer clouds are generally assumed for the retrieval of cloud optical and microphysical properties from satellite measurements, although multilayer clouds often occur in reality. In this article, the impact of multilayer clouds (thin ice clouds overlying lower-level water clouds) on the retrieval of cloud microphysical properties is simulated with the radiative transfer model RSTAR. The simulated results demonstrate the impact of double-layer clouds on the accuracy of retrieval of the cloud parameters and estimation of DSR. To understand the uncertainties of the input parameters, thorough sensitivity tests are simulated by RSTAR in the Results section. As compared with the retrieval results of single-layer clouds when the ice particle model of the upper-layer cloud is assumed to be ellipsoidal, the maximum relative bias in DSR is 0.63% when the COT for the ice cloud is 1.2 and for water cloud is 32.45. When the upper-layer ice cloud is assumed to be a hexagonal column, the maximum relative bias in DSR is 55.34% when the COT for the ice cloud is 2 and for the water cloud is 58.4. In addition, relative bias in DSR tends to increase both with radiance and ice cloud COT for a given radiance. This finding can provide a basis of reference for the estimation accuracy of radiative forcing in the IPCC report and the subsequent enhancement and improvement of retrieval algorithms.

To solve problems of brightness and detail information loss in infrared and visible image fusion, an effective infrared and visible image fusion method using rolling guidance filtering and gradient saliency map is proposed in this paper. The rolling guidance filtering is used to decompose the input images into approximate layers and residual layers; the energy attribute fusion model is used to fuse the approximate layers; the gradient saliency map is introduced and the corresponding weight matrices are constructed to perform on residual layers. The fusion image is generated by reconstructing the fused approximate layer sub-image and residual layer sub-images. Experimental results demonstrate the superiority of the proposed infrared and visible image fusion method.

The Chinese High-resolution Earth Observation System (CHEOS) program has successfully launched 7 civilian satellites since 2010. These satellites are named by Gaofen (meaning high resolution in Chinese, hereafter noted as GF). To combine the advantages of high temporal and comparably high spatial resolution, diverse sensors are deployed to each satellite. GF-1 and GF-6 carry both high-resolution cameras (2 m resolution panchromatic and 8 m resolution multispectral camera), providing high spatial imaging for land use monitoring; GF-3 is equipped with a C-band multipolarization synthetic aperture radar with a spatial resolution of up to 1 meter, mostly monitoring marine targets; GF-5 carried 6 sensors including hyperspectral camera and directional polarization camera, dedicated to environmental remote sensing and climate research, such as aerosol, clouds, and greenhouse gas monitoring; and GF-7 laser altimeter system payload enables a three-dimensional surveying and mapping of natural resource and land surveying, facilitating the accumulation of basic geographic information. This study provides an overview of GF civilian series satellites, especially their missions, sensors, and applications.

Crop residue open burning has significant adverse effects on regional air quality, climate change, and human health. Emissions from crop residue open burning estimated by satellites are underestimated in central China due to long-term cloud cover and the limitation of spatial-temporal resolution of satellites. In this study, we used a statistical-based method to investigate the crop residue open burning emissions in central China from 2012 to 2020. The open burning proportion (OBP) of residue, updated annually by the Visible Infrared Imaging Radiometer Suite (VIIRS) 375 m active fire product (VNP14IMG), and the latest observed emission factors (EFS) were used to improve the accuracy of the estimated emissions. Annual emissions of pollutants were allocated into 0.1° × 0.1° spatial grid cells using fire counts and land cover data. The results showed that the total emissions of black carbon (BC), organic carbon (OC), sulfur dioxide (SO2), nitric oxide (NOX), carbon monoxide (CO), carbon dioxide (CO2), fine particles (PM2.5), coarse particles (PM10), ammonia (NH3), methane (CH4) and non-methane volatile organic compound (NMVOC) were 34.84, 149.72, 41.06, 90.11, 2640.97, 78,094.91, 485.17, 481.05, 35.21, 246.38 and 499.59 Gg, respectively. The largest contributor of crop residue open burning was rice, followed by wheat, rapeseed and corn, with the contribution rates of 35.34–64.07%, 15.78–34.71%, 9.12–25.56%, and 5.69–14.06%, respectively. The pollutants emissions exhibit large annual variation, with the highest emissions in 2013 and a remarkable decrease from 2013 to 2015 under strict control measures. Since 2015, the emissions remained at a low level, which shows that air quality control policies play a role in recent years. The result indicates that using OBP updated by satellite active fire product in a statistical-based method can help to get more accurate and reliable multi-year emissions.

Multi-angle polarization measurement is an important technical means of satellite remote sensing applied to aerosol monitoring. By adding angle information and polarization measurements, aerosol optical and microphysical properties can be more comprehensively and accurately retrieved. The accuracy of aerosol retrieval can reflect the advantages and specific accuracy improvement of multi-angle polarization. In this study, the Multi-angle Imaging SpectroRadiometer (MISR) V23 aerosol products and the Polarization and Directionality of the Earth’s Reflectance (POLDER) GRASP “high-precision” archive were evaluated with the Aerosol Robotic Network (AERONET) observations over China. Validation of aerosol optical depth (AOD), absorbing aerosol optical depth (AAOD), and the Ångström exponent (AE) properties was conducted. Our results show that the AOD inversion accuracy of POLDER-3/GRASP is higher with the correlation coefficient (R) of 0.902, slope of 0.896, root mean square error (RMSE) of 0.264, mean absolute error (MAE) of 0.190, and about 40.71% of retrievals within the expected error (EE, ± 0.05+0.2×AODAERONET) lines. For AAOD, the performance of two products is poor, with better results for POLDER-3/GRASP data. POLDER-3/GRASP AE also has higher R of 0.661 compared with that of MISR AE (0.334). According to the validation results, spatiotemporal distribution, and comparison with other traditional scalar satellite data, the performance of multi-angle polarization observations is better and is suitable for the retrieval of aerosol properties.

Fine particulate matter (PM2.5) has attracted extensive attention due to its harmful effects on humans and the environment. The sparse ground-based air monitoring stations limit their application for scientific research, while aerosol optical depth (AOD) by remote sensing satellite technology retrieval can reflect air quality on a large scale and thus compensate for the shortcomings of ground-based measurements. In this study, the elaborate vertical-humidity method was used to estimate PM2.5 with the spatial resolution 1 km and the temporal resolution 1 hour. For vertical correction, the scale height of aerosols (Ha) was introduced based on the relationship between the visibility data and extinction coefficient of meteorological observations to correct the AOD of the Advance Himawari Imager (AHI) onboard the Himawari-8 satellite. The hygroscopic growth factor (f(RH)) was fitted site-by-site and month by month (1–12 months). Meanwhile, the spatial distribution of the fitted coefficients can be obtained by interpolation assuming that the aerosol properties vary smoothly on a regional scale. The inverse distance weighted (IDW) method was performed to construct the hygroscopic correction factor grid for humidity correction so as to estimate the PM2.5 concentrations in Sichuan and Chongqing from 09:00 to 16:00 in 2017–2018. The results indicate that the correlation between “dry” extinction coefficient and PM2.5 is slightly improved compared to the correlation between AOD and PM2.5, with r coefficient values increasing from 0.12–0.45 to 0.32–0.69. The r of hour-by-hour verification is between 0.69 and 0.85, and the accuracy of the afternoon is higher than that of the morning. Due to the missing rate of AOD in the southwest is very high, this study utilized inverse variance weighting (IVW) gap-filling method combine satellite estimation PM2.5 and the nested air-quality prediction modeling system (NAQPMS) simulation data to obtain the full-coverage hourly PM2.5 concentration and analyze a pollution process in the fall and winter.

Due to the advantage of geostationary satellites, Himawari-8/AHI can provide near-real-time air quality monitoring over China with a high temporal resolution. Satellite-based aerosol optical depth (AOD) retrieval over land is a challenge because of the large surface contribution to the top of atmosphere (TOA) signal and the uncertainty of aerosol modes. Here, by combining satellite TOA reflectance, sun-sensor geometries, meteorological factors and vegetation information, we propose a data-driven AOD detection algorithm based on a deep neural network (DNN) model for Himawari-8/AHI. It is trained by sample data of 2018 and 2019 and is applied to derive hourly AODs over China in 2020. By comparison with ground-based AERONET measurements, R2 for DNN-estimated AOD is up to 0.8702, which is much higher than that for the AHI AOD product with R2 = 0.4869. The hourly AOD results indicate that the DNN model has a good potential in improving the performance of AOD retrieval in the early morning and in the late afternoon, and the spatial distribution is reliable for capturing the variation of aerosol pollution on the regional scale. By analyzing different DNN modeling strategies, it is found that seasonal modeling can hardly increase the accuracy of AOD retrieval to a certain extent, and R2 increases from 0.7394 to 0.8168 when meteorological features, especially air pressure, are involved in the model training.

This paper presents an algorithm for the retrieval of nitrous oxide profiles from the Atmospheric InfraRed Sounder (AIRS) on the Earth Observing System (EOS)/Aqua using a nonlinear optimal estimation method. First, an improved Optimal Sensitivity Profile (OSP) algorithm for channel selection is proposed based on the weighting functions and the transmissions of the target gas and interfering gases, with 13 channels selected for inversion in this algorithm. Next, the data of the High-Performance Instrumented Airborne Platform for Environmental Research (HIAPER) Pole-to-Pole Observations (HIPPO) aircraft and the Earth System Research Laboratory (ESRL) are used to verify the retrieval results, including the atmospheric nitrous oxide profile and the column concentration. The results show that using AIRS satellite data, the atmospheric nitrous oxide profile between 300–900 hPa can be well retrieved with an accuracy of ~0.1%, which agrees with the corresponding Jacobian peak interval of selected channels. Analysis of the AIRS retrievals demonstrates that the AIRS measurements provide useful information to capture the spatial and temporal variations in nitrous oxide between 300–900 hPa.

The semantic segmentation of fine-resolution remotely sensed images is an urgent issue in satellite image processing. Solving this problem can help overcome various obstacles in urban planning, land cover classification, and environmental protection, paving the way for scene-level landscape pattern analysis and decision making. Encoder-decoder structures based on attention mechanisms have been frequently used for fine-resolution image segmentation. In this paper, we incorporate a coordinate attention (CA) mechanism, adopt an asymmetric convolution block (ACB), and design a refinement fusion block (RFB), forming a network named the fusion coordinate and asymmetry-based U-Net (FCAU-Net). Furthermore, we propose novel convolutional neural network (CNN) architecture to fully capture long-term dependencies and fine-grained details in fine-resolution remotely sensed imagery. This approach has the following advantages: (1) the CA mechanism embeds position information into a channel attention mechanism to enhance the feature representations produced by the network while effectively capturing position information and channel relationships; (2) the ACB enhances the feature representation ability of the standard convolution layer and captures and refines the feature information in each layer of the encoder; and (3) the RFB effectively integrates low-level spatial information and high-level abstract features to eliminate background noise when extracting feature information, reduces the fitting residuals of the fused features, and improves the ability of the network to capture information flows. Extensive experiments conducted on two public datasets (ZY-3 and DeepGlobe) demonstrate the effectiveness of the FCAU-Net. The proposed FCAU-Net transcends U-Net, Attention U-Net, the pyramid scene parsing network (PSPNet), DeepLab v3+, the multistage attention residual U-Net (MAResU-Net), MACU-Net, and the Transformer U-Net (TransUNet). Specifically, the FCAU-Net achieves a 97.97% (95.05%) pixel accuracy (PA), a 98.53% (91.27%) mean PA (mPA), a 95.17% (85.54%) mean intersection over union (mIoU), and a 96.07% (90.74%) frequency-weighted IoU (FWIoU) on the ZY-3 (DeepGlobe) dataset.

Fine particulate matter (PM2.5) threatens human health and the natural environment. Estimating the near-ground PM2.5 concentrations accurately is of great significance in air quality research. Statistical and deep-learning models are widely used for estimating PM2.5 concentration based on remotely sensed aerosol optical depth (AOD) products. Deep-learning models can effectively express the nonlinear relationship between AOD, parameters, and PM2.5. This study proposed a capsule network model (CapsNet) to address the spatial differences in PM2.5 concentration distribution by introducing a capsule structure and dynamic routing algorithm for the first time, which integrates AOD, surface PM2.5 measurements, and auxiliary variables (e.g., normalized difference vegetation index (NDVI) and meteorological parameters). Moreover, we examined the longitude and latitude of pixels as input parameters to reflect spatial location information, and the results showed that the introduction of longitude (LON) and latitude (LAT) parameters improved the model fitting accuracy. The coefficient of determination (R2) increased by 0.05 ± 0.01, and the root mean square error (RMSE), mean relative error (MRE), and mean absolute error (MAE) decreased by 3.30 ± 1.0 μg/m3, 8 ± 3%, and 1.40 ± 0.2 μg/m3, respectively. To verify the accuracy of our proposed CapsNet, the deep neural network (DNN) model was executed. The results indicated that the R2 values of the validation dataset using CapsNet improved by 4 ± 2%, and RMSE, MRE, and MAE decreased by 1.50 ± 0.4 μg/m3, ~5%, and 0.60 ± 0.2 μg/m3, respectively. Finally, the effects of seasons and spatial region on the fitting accuracy were examined separately from 2018 to 2020. With respect to seasons, the model performed more robustly in the cold season. In terms of spatial region, the R2 values exceeded 0.9 in the central-eastern region, while the accuracy was lower in the western and coastal regions. This study proposed the CapsNet model to estimate PM2.5 concentrations for the first time and achieved good accuracy, which could be used for the estimation of other air contaminants.

The rapid development of atmospheric satellite instruments since 1990s provides unprecedented large amount of observational datasets concerning global atmospheric pollutants. The continuous and long-term large-scale satellite products such as aerosol optical depth, tropospheric NO2 and SO2 enable effective and objective evaluation of air quality. Satellite columnar aerosol optical parameters can be used to indicate particle pollution near surface after correction. By contrast, satellite results of trace gas pollutants such as NO2 and SO2 from fossil fuel combustion with short lifetime around half one day are used to estimate anthropogenic emissions. It is shown that the overall anthropogenic emissions in China have largely declined since strict emission reduction policy implemented since 2013. However, coarse pixel resolution of the trace gases, limited information and retrieval bias of aerosol properties tend to hinder further application of satellite in air quality research. Recently launched satellite missions with advanced detection abilities will greatly enhance global atmospheric observations with much more datasets available.

Quantifying spatiotemporal polar ozone changes can promote our understanding of global stratospheric ozone depletion, polar ozone-related chemical processes, and atmospheric dynamics. By means of ground-level measurements, satellite observations, and re-analyzed meteorology, the global spatial and temporal distribution characteristics of the total column ozone (TCO) and ozone profile can be quantitatively described. In this study, we evaluated the ozone datasets from CrIS/NOAA20, AIRS/Aqua, and ERA5/ECWMF for their performance in polar regions in 2020, along with the in situ observations of the Dobson, Brewer, and ozonesonde instruments, which are regarded as benchmarks. The results showed that the ERA5 reanalysis ozone field had good consistency with the ground observations (R > 0.95) and indicated whether the TCO or ozone profile was less affected by the site location. In contrast, both CrIS and AIRS could capture the ozone loss process resulting from the Antarctic/Arctic ozone hole at a monthly scale, but their ability to characterize the Arctic ozone hole was weaker than in the Antarctic. Specifically, the TCO values derived from AIRS were apparently higher in March 2020 than those of ERA5, which made it difficult to assess the area and depth of the ozone hole during this period. Moreover, the pattern of CrIS TCO was abnormal and tended to deviate from the pattern that characterized ERA5 and AIRS at the Alert site during the Arctic ozone loss process in 2020, which demonstrates that CrIS ozone products have limited applicability at this ground site. Furthermore, the validation of the ozone profile shows that AIRS and CrIS do not have good vertical representation in the polar regions and are not able to characterize the location and depth of ozone depletion. Overall, the results reveal the shortcomings of the ozone profiles derived from AIRS and CrIS observations and the reliability of the ERA5 reanalysis ozone field in polar applications. A more suitable prior method and detection sensitivity improvement on CrIS and AIRS ozone products would improve their reliability and applicability in polar regions.

Continuous increases in the human population and human activities have resulted in remarkable changes in the composition of the atmosphere since the industrial revolution [...]

Climate change and air pollution are emerging topics due to their possible enormous implications for health and social perspectives. In recent years, tropospheric ozone has been recognized as an important greenhouse gas and pollutant that is detrimental to human health, agriculture, and natural ecosystems, and has shown a trend of increasing interest. Machine-learning-based approaches have been widely applied to the estimation of tropospheric ozone concentrations, but few studies have included tropospheric ozone profiles. This study aimed to predict the Northern Hemisphere distribution of Lower-Stratosphere-to-Troposphere (LST) ozone at a pressure of 100 hPa to the near surface by employing a deep learning Long Short-Term Memory (LSTM) model. We referred to a history of all the observed parameters (meteorological data of European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5), satellite data, and the ozone profiles of the World Ozone and Ultraviolet Data Center (WOUDC)) between 2014 and 2018 for training the predictive models. Model–measurement comparisons for the monitoring sites of WOUDC for the period 2019–2020 show that the mean correlation coefficients (R2) in the Northern Hemisphere at high latitude (NH), Northern Hemisphere at middle latitude (NM), and Northern Hemisphere at low latitude (NL) are 0.928, 0.885, and 0.590, respectively, indicating reasonable performance for the LSTM forecasting model. To improve the performance of the model, we applied the LSTM migration models to the Civil Aircraft for the Regular Investigation of the Atmosphere Based on an Instrument Container (CARIBIC) flights in the Northern Hemisphere from 2018 to 2019 and three urban agglomerations (the Sichuan Basin (SCB), North China Plain (NCP), and Yangtze River Delta region (YRD)) between 2018 and 2019. The results show that our models performed well on the CARIBIC data set, with a high R2 equal to 0.754. The daily and monthly surface ozone concentrations for 2018–2019 in the three urban agglomerations were estimated from meteorological and ancillary variables. Our results suggest that the LSTM models can accurately estimate the monthly surface ozone concentrations in the three clusters, with relatively high coefficients of 0.815–0.889, root mean square errors (RMSEs) of 7.769–8.729 ppb, and mean absolute errors (MAEs) of 6.111–6.930 ppb. The daily scale performance was not as high as the monthly scale performance, with the accuracy of R2 = 0.636~0.737, RMSE = 14.543–16.916 ppb, MAE = 11.130–12.687 ppb. In general, the trained module based on LSTM is robust and can capture the variation of the atmospheric ozone distribution. Moreover, it also contributes to our understanding of the mechanism of air pollution, especially increasing our comprehension of pollutant areas.

Aerosol distribution with fine spatial resolution is crucial for atmospheric environmental management. This paper proposes an improved algorithm of aerosol retrieval from 250-m Medium Resolution Spectral Image (MERSI) data of Chinese FY-3 satellites. A mixing model of soil and vegetation was used to calculate the parameters of the algorithm from moderate-resolution imaging spectroradiometer (MODIS) reflectance products in 500-m resolution. The mixing model was used to determine surface reflectance in blue band, and the 250-m aerosol optical depth (AOD) was retrieved through removing surface contributions from MERSI data over Guangzhou. The algorithm was used to monitor two pollution episodes in Guangzhou in 2015, and the results displayed an AOD spatial distribution with 250-m resolution. Compared with the yearly average of MODIS aerosol products in 2015, the 250-m resolution AOD derived from the MERSI data exhibited great potential for identifying air pollution sources. Daily AODs derived from MERSI data were compared with ground results from CE318 measurements. The results revealed a correlation coefficient between the AODs from MERSI and those from the ground measurements of approximately 0.85, and approximately 68% results were within expected error range of ±(0.05 + 15%τ).

Estimation of the number and geo-location of oil wells is important for policy holders considering their impact on energy resource planning. With the recent development in optical remote sensing, it is possible to identify oil wells from satellite images. Moreover, the recent advancement in deep learning frameworks for object detection in remote sensing makes it possible to automatically detect oil wells from remote sensing images. In this paper, we collected a dataset named Northeast Petroleum University–Oil Well Object Detection Version 1.0 (NEPU–OWOD V1.0) based on high-resolution remote sensing images from Google Earth Imagery. Our database includes 1192 oil wells in 432 images from Daqing City, which has the largest oilfield in China. In this study, we compared nine different state-of-the-art deep learning models based on algorithms for object detection from optical remote sensing images. Experimental results show that the state-of-the-art deep learning models achieve high precision on our collected dataset, which demonstrate the great potential for oil well detection in remote sensing.

The Cross-track Infrared Sounder (CrIS) aboard the Suomi National Polar-orbiting Partnership (S-NPP) satellite is a spaceborne Fourier transform infrared spectrometer. The study aims to retrieve carbon dioxide (CO2) information (the CO2 profile and column-averaged dry-air mole fraction of XCO2) from June 2018 to December 2019 based on the The National Oceanic and Atmospheric Administration (NOAA)-Unique Combined Atmospheric Processing System (NUCAPS) Cloud-Cleared Radiances (CCRs) via the CrIS. The CCRs products for the CrIS with 2223 channels have been available since 22 May 2018. Characteristics of the CO2 weighting functions inform the choice of multiple channels that are around 15 μm in size that differ by latitude and season to maximize retrieval sensitivity to CO2 and minimize sensitivity to other interfering atmospheric parameters. CO2 was retrieved from these channels using an adopted nonlinear optimization algorithm. The temperature, water vapor, and ozone profiles used in the inversion process were gathered from the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5). Validations of CO2 concentrations as retrieved from CrIS showed the following conclusions: (1) The relative error of the retrieved CO2 concentrations, as compared to Civil Aircraft for the Regular Investigation of the atmosphere Based on an Instrument Container (CARIBIC) in situ aircraft measurements, was less than 0.5%, and the root mean square errors (RMSE) were less than 0.7 ppmv (with correlation coefficients of 0.56–0.86); (2) the retrieved XCO2 from June 2018 to December 2019 correlated well with the ground-based Total Carbon Column Observing Network (TCCON) observations, and the differences were within ±0.2 ppmv. Further analysis of the temporal and spatial distribution of the retrieved CO2 at 300 hPa demonstrated a strong seasonal variation of CO2 in 0–60° N in the Northern Hemisphere with the maximum values in June–August and larger amplitudes of seasonal variation in the northeast of Asia and northeastern part of North America. The variations likely occurred due to larger sinks of atmospheric CO2 that are dominated by CO2 uptake in the summer. In the Southern Hemisphere, the CO2 displayed high concentration anomalies in the latitudinal range of 30–60° S in September–November and December–February, which probably occurred due to the lofted smoke plumes from the strong fire seasons in South America and Southern Africa.

The Visible Infrared Imaging Radiometer Suite (VIIRS) has been observing aerosol optical depth (AOD), which is a critical parameter in air pollution and climate change, for more than 7 years since 2012. Due to limited and uneven distribution of the Aerosol Robotic Network (AERONET) station in China, the independent data from the Campaign on Atmospheric Aerosol Research Network of China (CARE-China) was used to evaluate the National Oceanic and Atmospheric Administration (NOAA) VIIRS AOD products in six typical sites and analyze the influence of the aerosol model selection process in five subregions, particularly for dust. Compared with ground-based observations, the performance of all retrievals (except the Shapotou (SPT) site) is similar to other previous studies on a global scale. However, the results illustrate that the AOD retrievals with the dust model showed poor consistency with a regression equation as y = 0.312x + 0.086, while the retrievals obtained from the other models perform much better with a regression equation as y = 0.783x + 0.119. The poor AOD retrieval with the dust model was also verified by a comparison with the Moderate Resolution Imaging Spectroradiometer (MODIS) aerosol product. The results show they have a lower correlation coefficient (R) and a higher mean relative error (MRE) when the aerosol model used in the retrieval is identified as dust. According to the Ultraviolet Aerosol Index (UVAI), the frequency of dust type over southern China is inconsistent with the actual atmospheric condition. In addition, a comparison of ground-based Ångström exponent (α) values yields an unexpected result that the dust model percentage exceed 40% when α < 1.0, and the mean α shows a high value of ~0.75. Meanwhile, the α peak value (~1.1) of the “dust” model determined by a satellite retravel algorithm indicate there is some problem in the dust model selection process. This mismatching of the aerosol model may partly explain the low accuracy at the SPT and the systemic biases in regional and global validations.

The air quality in China has experienced dramatic changes during the last few decades. To improve understanding of distribution, variations, and main influence factors of air pollution in central China, long-term multiple satellite observations from moderate resolution imaging spectroradiometer (MODIS) and ozone monitoring instrument (OMI) are used to characterize particle pollution and their primary gaseous precursors, sulfur dioxide (SO2), and nitrogen dioxide (NO2) in Hubei province during 2005−2017. Unlike other regions in eastern China, particle and gaseous pollutants exhibit distinct spatial and temporal patterns in central China due to differences in emission sources and control measures. OMI SO2 of the whole Hubei region reached the highest value of ~0.2 Dobson unit (DU) in 2007 and then declined by more than 90% to near background levels. By contrast, OMI NO2 grew from ~3.2 to 5.9 × 1015 molecules cm−2 during 2005−2011 and deceased to ~3.9 × 1015 molecules cm−2 in 2017. Unlike the steadily declining SO2, variations of OMI NO2 flattened out in 2016 and increased ~0.5 × 1015 molecules cm−2 during 2017. As result, MODIS AOD at 550 nm increased from 0.55 to the peak value of 0.7 during 2005−2011 and then decreased continuously to 0.38 by 2017. MODIS AOD and OMI SO2 has a high correlation (R > 0.8), indicating that annual variations of SO2 can explain most changes of AOD. The air pollution in central China has notable seasonal variations, which is heaviest in winter and light in summer. While air quality in eastern Hubei is dominated by gaseous pollution such as O3 and NOx, particle pollutants are mainly concentrated in central Hubei. The high consistency with ground measurements demonstrates that satellite observation can well capture variations of air pollution in regional scales. The increasing ozone (O3) and NO2 since 2016 suggests that more control measures should be made to reduce O3-related emissions. To improve the air quality in regional scale, it is necessary to monitor the dynamic emission sources with satellite observations at a finer resolution.