The accurate estimation of urban extreme precipitation is essential for urban design and risk management, which is hard for developing countries, due to the fast urbanization and sparse rain gauges. Satellite precipitation products (SPPs) have emerged as a promising solution. Not only near real-time SPPs can provide critical information for decision making, but post-processed SPPs can also offer essential information for climate change adaption, risk management strategy development, and related fields. However, their ability in urban extreme precipitation estimation has not been examined in detail. This study presents a comprehensive evaluation of four recent SPPs that are post-processed, including IMERG, GSMaP_Gauge, MSWEP, and CMFD, for their ability to capture urban extreme precipitation in mainland China at the national, city, and inner-city scales. The performance of the four SPPs was assessed using daily observations from the 821 urban gauges from 2001 to 2018. The assessment includes: (1) the extreme precipitation estimates from the four SPPs in the total urbanized areas of mainland China were evaluated using correlation coefficients (CC), absolute deviation (AD), relative deviation (RB), and five extreme precipitation indices; (2) The extreme precipitation estimates over 21 Chinese major cities were assessed with the two most important extreme indices, namely the 99th percentile of daily precipitation on wet days (R99) and total precipitation when daily precipitation exceeding R99 (R99TOT); and (3) Bivariate Moran’s I (BMI) was adopted to assess the inner-city spatial correlation of R99 and R99TOT between SPPs and gauge observations in four major cities with most gauges. The results indicate that MSWEP has the highest CC of 0.79 and the lowest AD of 1.61 mm at the national scale. However, it tends to underestimate urban precipitation, with an RB of −8.5%. GSMaP_Gauge and IMERG performed better in estimating extreme values, with close extreme indices with gauge observations. According to the 21 major cities, GSMaP_Gauge also shows high accuracy in estimating R99 and R99TOT values, with the best RB and AD in these cities, while CMFD and MSWEP exhibit the highest CC values for R99 and R99TOT, respectively, indicating a strong correlation between their estimates and those obtained from gauge observations. At the inner-city scale, MSWEP shows advantages in monitoring the spatial distribution of urban extreme precipitation in most of cities. The study firstly provided the multiscale assessment of urban extreme precipitation by SPPs over mainland China, which is useful for their applications. Full article

Dust emission is a common catastrophic weather phenomenon in Northern China. This phenomenon not only causes environmental problems, such as air pollution, but also has an important impact on the global dust cycle and climate change. On the basis of the dust weather observation data of 44 surface meteorological stations in the Tarim Basin from 1989 to 2021, combined with the dust aerosol optical depth (DAOD), dust surface mass concentration (DUSMASS) and wind speed data, this paper analyses the spatial and temporal dust weather characteristics in the Tarim Basin over the past 33 years. Results show that the frequency of dust weather in the Tarim Basin has declined in the past 33 years. Dust weather mainly consisted of floating dust, followed by blowing dust and dust storm. This weather had a significant seasonal change, with more dust in spring and summer and less in autumn and winter. The dust weather was mainly distributed along the south edge of the Tarim Basin and the desert hinterland of Tazhong. The spatial distribution of the dust intensity (DI) index was basically consistent with the dust weather days. Moreover, the DAOD was obviously affected by dust weather and had a significant positive correlation with the number of dust weather days and the DI, suggesting the vertical concentration of dust particles to a certain extent. Wind is also one of the most important factors affecting the release of dust. The frequency of strong wind weather decreases from the northeast to the southwest, which corresponds to the distribution of the DUSMASS. Full article

Hyperspectral images (HSI) contain powerful spectral characterization capabilities and are widely used especially for classification applications. However, the rich spectrum contained in HSI also increases the difficulty of extracting useful information, which makes the feature extraction method significant as it enables effective expression and utilization of the spectrum. Traditional HSI feature extraction methods design spectral features manually, which is likely to be limited by the complex spectral information within HSI. Recently, data-driven methods, especially the use of convolutional neural networks (CNNs), have shown great improvements in performance when processing image data owing to their powerful automatic feature learning and extraction abilities and are also widely used for HSI feature extraction and classification. The CNN extracts features based on the convolution operation. Nevertheless, the local perception of the convolution operation makes CNN focus on the local spectral features (LSF) and weakens the description of features between long-distance spectral ranges, which will be referred to as global spectral features (GSF) in this study. LSF and GSF describe the spectral features from two different perspectives and are both essential for determining the spectrum. Thus, in this study, a local-global spectral feature (LGSF) extraction and optimization method is proposed to jointly consider the LSF and GSF for HSI classification. To increase the relationship between spectra and the possibility to obtain features with more forms, we first transformed the 1D spectral vector into a 2D spectral image. Based on the spectral image, the local spectral feature extraction module (LSFEM) and the global spectral feature extraction module (GSFEM) are proposed to automatically extract the LGSF. The loss function for spectral feature optimization is proposed to optimize the LGSF and obtain improved class separability inspired by contrastive learning. We further enhanced the LGSF by introducing spatial relation and designed a CNN constructed using dilated convolution for classification. The proposed method was evaluated on four widely used HSI datasets, and the results highlighted its comprehensive utilization of spectral information as well as its effectiveness in HSI classification. Full article

A major obstacle to mapping Ecosystem Services (ES) and the application of the ES concept has been the inadequacy of data at the landscape level necessary for their quantification. This study takes advantage of free satellite imagery to map and provide relevant information regarding ES and contribute to the sustainable management of natural resources in developing countries. The aim is to assess the flow of ES in mangrove ecosystem of Ungwana Bay, located on the northern coast of Kenya, by adopting the Land Use Land Cover (LULC) matrix approach. This study characterized LULC classes present in the study area, identified the most important ES, and collected data on expert opinions via a survey on ES flow supplied by the mangrove ecosystem. A qualitative and quantitative analysis of the expert scoring produced a LULC matrix which, when integrated with the LULC maps, showed the spatial distribution of ES flow. The assessment indicates very high flow (5.0) for the regulating and supporting services, high flow (4.0) for the cultural services, and medium flow (3.0) for the provisioning services as supplied by mangroves. In addition, the analysis indicates there are sixteen major ES supplied by the mangrove ecosystem of Ungwana bay as of the year 2021. This study highlights the importance of mangroves as a coastal ecosystem and how the visualization of the spatial distribution of ES flow using maps can be useful in informing natural resource management. In addition, the study shows the possibilities of using freely accessible satellite imagery and software to bolster the ES assessment studies lacking in developing countries. Full article

Basalts from Mare Orientale are representative of lunar flood volcanism, which sheds light on the lunar farside’s thermal and volcanic past. We use Chandrayaan’s Moon Mineralogy Mapper data to examine the spectral and chemical makeup of the volcanic units located in the Orientale basin; the analysis specifically focuses on three formations: Mare Orientale, Lacus Veris, and Lacus Autumni. The main assemblage in these basaltic units consists of calcic augite and ferroaugite. Pyroxenes in the Orientale volcanic units have an average chemical composition of En_35.53 Fs_34.11 Wo_30.35. The trend in the composition of pigeonites and augites suggests that the magma was fractionated as it crystallized. The pyroxene quadrilateral plot’s distinct chemical trends indicate that the Orientale Basin underwent a number of volcanic eruptions from heterogeneous magma sources during the Imbrium to Eratosthenian period. Full article

In this study, the presence of permafrost layer and its potential variation in the last three decades will be examined through the multitemporal analysis of satellite data in the area of the Deosai Plateau (Northern Pakistan). In the area, only global maps on the potential presence of permafrost layer are known. The results are based on the evaluation of variation of the number and water levels of the small lakes, and the changes of the extensions of the wetlands. The adopted methodology is based on the use of spectral indices and visual interpretation of a time-series data of Landsat images in the range 1990–2019, and on the processing of radar data from Sentinel 1 satellites, adopting new methods to extract the vertical displacement. The main findings are: (i) a high temporal dynamic of the number and surface areas of small lakes, and (ii) the evidence of a subduction in a wetland area (Black Hole), coherent with its extension, and suggesting the potential presence of a permafrost layer slowly degrading. This analysis can play a useful role on the management of the Deosai National Park (DNP), adopting careful measures for the human activities inside the park. Full article

This paper presents the development and application of a deep learning-based approach for semi-automated detection of tar production kilns using new Finnish high-density Airborne Laser Scanning (ALS) data in the boreal taiga forest zone. The historical significance of tar production, an important livelihood for centuries, has had extensive environmental and ecological impacts, particularly in the thinly inhabited northern and eastern parts of Finland. Despite being one of the most widespread archaeological features in the country, tar kilns have received relatively little attention until recently. The authors employed a Convolutional Neural Networks (CNN) U-Net-based algorithm to detect these features from the ALS data, which proved to be more accurate, faster, and capable of covering systematically larger spatial areas than human actors. It also produces more consistent, replicable, and ethically sustainable results. This semi-automated approach enabled the efficient location of a vast number of previously unknown archaeological features, significantly increasing the number of tar kilns in each study area compared to the previous situation. This has implications also for the cultural resource management in Finland. The authors’ findings have influenced the preparation of the renewal of the Finnish Antiquities Act, raising concerns about the perceived impacts on cultural heritage management and land use sectors due to the projected tenfold increase in archaeological site detection using deep learning algorithms. The use of environmental remote sensing data may provide a means of examining the long-term cultural and ecological impacts of tar production in greater detail. Our pilot studies suggest that artificial intelligence and deep learning techniques have the potential to revolutionize archaeological research and cultural resource management in Finland, offering promising avenues for future exploration. Full article

This work presents an algorithm based on a neural network (NN) for cloud detection to detect clouds and their thermodynamic phase using spectral observations from spaceborne microwave radiometers. A standalone cloud detection algorithm over the ocean and land has been developed to distinguish clear sky versus ice and liquid clouds from microwave sounder (MWS) observations. The MWS instrument—scheduled to be onboard the first satellite of the Eumetsat Polar System Second-Generation (EPS-SG) series, MetOp-SG A1—has a direct inheritance from advanced microwave sounding unit A (AMSU-A) and the microwave humidity sounder (MHS) microwave instruments. Real observations from the MWS sensor are not currently available as its launch is foreseen in 2024. Thus, a simulated dataset of atmospheric states and associated MWS synthetic observations have been produced through radiative transfer calculations with ERA5 real atmospheric profiles and surface conditions. The developed algorithm has been validated using spectral observations from the AMSU-A and MHS sounders. While ERA5 atmospheric profiles serve as references for the model development and its validation, observations from AVHRR cloud mask products provide references for the AMSU-A/MHS model evaluation. The results clearly show the NN algorithm’s high skills to detect clear, ice and liquid cloud conditions against a benchmark. In terms of overall accuracy, the NN model features 92% (88%) on the ocean and 87% (85%) on land, for the MWS (AMSU-A/MHS)-simulated dataset, respectively. Full article

Tropical cyclones (TCs) seriously threaten the safety of human life and property especially when approaching a coast or making landfall. Robust, long-lead predictions are valuable for managing policy responses. However, despite decades of efforts, seasonal prediction of TCs remains a challenge. Here, we introduce a deep-learning prediction model to make skillful seasonal prediction of TC track density in the Western North Pacific (WNP) during the typhoon season, with a lead time of up to four months. To overcome the limited availability of observational data, we use TC tracks from CMIP5 and CMIP6 climate models as the training data, followed by a transfer-learning method to train a fully convolutional neural network named SeaUnet. Through the deep-learning process (i.e., heat map analysis), SeaUnet identifies physically based precursors. We show that SeaUnet has a good performance for typhoon distribution, outperforming state-of-the-art dynamic systems. The success of SeaUnet indicates its potential for operational use. Full article

Tailings ponds’ failure and environmental pollution make tailings monitoring very important. Remote sensing technology can quickly and widely obtain ground information and has become one of the important means of tailings monitoring. However, the efficiency and accuracy of traditional remote sensing monitoring technology have difficulty meeting the management needs. At the same time, affected by factors such as the geographical environment and imaging conditions, tailings have various manifestations in remote sensing images, which all bring challenges to the accurate acquisition of tailings information in large areas. By improving You Only Look Once (YOLO) v5s, this study designs a deep learning-based framework for the large-scale extraction of tailings ponds information from the entire high-resolution remote sensing images. For the improved YOLOv5s, the Swin Transformer is integrated to build the Swin-T backbone, the Fusion Block of efficient Reparameterized Generalized Feature Pyramid Network (RepGFPN) in DAMO-YOLO is introduced to form the RepGFPN Neck, and the head is replaced with Decoupled Head. In addition, sample boosting strategy (SBS) and global non-maximum suppression (GNMS) are designed to improve the sample quality and suppress repeated detection frames in the entire image, respectively. The model test results based on entire Gaofen-6 (GF-6) high-resolution remote sensing images show that the F1 score of tailings ponds is significantly improved by 12.22% compared with YOLOv5, reaching 81.90%. On the basis of both employing SBS, the improved YOLOv5s boots the mAP@0.5 of YOLOv5s by 5.95%, reaching 92.15%. This study provides a solution for tailings ponds’ monitoring and ecological environment management. Full article

The visibility samples generated by the synthetic aperture interferometric radiometer (SAIR) under near-field observation conditions contain information about the distance from the target to the instrument. This requires a precise understanding of the target–instrument distance to guarantee imaging quality in near-field SAIR applications. In this paper, we introduce a novel distance estimate approach for near-field SAIR systems, which achieves satisfactory imaging performance in the absence of prior information on target–instrument distance. First, we reformulate the signal model of near-field SAIR from the fractional Fourier transform (FRFT) perspective. This formulation ties the distance that is variable to the visibility function in a straightforward manner, offering an efficient solution for image reconstruction in near-field SAIR. Subsequently, we present an iterative strategy for target–instrument distance estimation based on simulated annealing (SA). In each iteration, the modified average gradient (MAG) of images reconstructed within the FRFT framework is evaluated, and based on the Metropolis criterion, the estimated target–instrument distance is optimally updated iteratively. Finally, the validity and effectiveness of the proposed distance estimation method for near-field SAIR imaging systems are demonstrated through numerical simulation and real experiments. Full article

The geolocation accuracy of Synthetic Aperture Radar (SAR) images is crucial for their application in various industries. Five high-resolution SAR satellites, namely ALOS, TerraSAR-X, Cosmo-SkyMed, RadarSat-2, and Chinese YG-3, provide a vast amount of image data for research purposes, although their geometric accuracies differ despite similar resolutions. To evaluate and compare the geometric accuracy of these satellites under the same ground control reference, a validation field was established in Xianning, China. The rational function model (RFM) was used to analyze the geometric performance of the five satellites based on the Xianning validation field. The study showed that each image could achieve sub-pixel positioning accuracy in range and azimuth direction when four ground control points (GCPs) were placed in the corners, resulting in a root mean square error (RMSE) of 1.5 pixels. The study also highlighted the effectiveness of an automated GCP-matching approach to mitigate manual identification of points in SAR images, and results demonstrate that the five SAR satellite images can all achieve sub-pixel positioning accuracy in range and azimuth direction when four GCPs are used. Overall, the verification results provide a reference for SAR satellite systems’ designs, calibrations, and various remote sensing activities. Full article

Soil moisture (SM) and land surface temperature (LST) are entangled, and the retrieval of one of them requires a priori specification of the other one. Due to insufficient observational information, retrieval of LST and SM from passive microwave remote sensing data is often ill-posed, and the retrieval accuracy needs to be improved. In this study, a novel fully-coupled paradigm is developed to robustly retrieve SM and LST from passive microwave data, which integrates deep learning, physical methods, and statistical methods. The key condition of the general paradigm proposed by us is that the output parameters of deep learning can be uniquely determined by the input parameters theoretically through a certain mathematical equation. Firstly, the physical method is deduced based on the energy radiation balance equation. The nine unknowns require the brightness temperatures of nine channels to construct nine equations, and the solutions of the physical method equations are obtained by model simulation. Based on the derivation of the physical method, the solution of the statistical method is constructed using multi-source data. Secondly, the solutions of physical and statistical methods constitute the training and test data of deep learning, which is used to obtain the solution curve of physical and statistical methods. The retrieval accuracy of LST and SM is greatly improved by smartly utilizing the mutual prior knowledge of SM and LST and cross iterative optimization calculations. Finally, validation indicates that the mean absolute error of the retrieved SM and LST data are 0.027 m³/m³ and 1.38 K, respectively, at an incidence angle of 0–65°. A model-data-knowledge-driven and deep learning method can overcome the shortcomings of traditional methods and provide a paradigm for retrieval of other geophysical variables. The proposed paradigm not only has physical meaning, but also makes deep learning physically interpretable, which is a milestone in the retrieval of geophysical remote sensing parameters based on artificial intelligence technology. Full article

Long-time coherent integration works to significantly increase the detection probability for maneuvering targets. However, during the observation time, the problems that often tend to occur are range cell migration (RCM) and Doppler frequency cell migration (DFCM), due to the high velocity and acceleration of the maneuvering target, which can reduce the detection of the maneuvering targets. In this regard, we propose a new coherent integration approach, based on the product scale zoom discrete chirp Fourier transform (PSZDCFT). Specifically, by introducing the zoom operation into the modified discrete chirp Fourier transform (MDCFT), the zoom discrete chirp Fourier transform (ZDCFT) can correctly estimate the centroid frequency and chirp rate of the linear frequency-modulated signal (LFM), regardless of whether the parameters of the LFM signal are outside the estimation scopes. Then, the scale operation, combined with ZDCFT, is performed on the radar echo signal in the range frequency slow time domain, to remove the coupling. Thereafter, a product operation is executed along the range frequency to inhibit spurious peaks and reinforce the true peak. Finally, the velocity and acceleration of the target estimated from the true peak position, are used to construct a phase compensation function to eliminate the RCM and DFCM, thus achieving coherent integration. The method is a linear transform without energy loss, and is suitable for low signal-to-noise (SNR) environments. Moreover, the method can be effectively fulfilled based on the chirp-z transform (CZT), which prevents the brute-force search. Thus, the method reaches a favorable tradeoff between anti-noise performance and computational load. Intensive simulations demonstrate the effectiveness of the proposed method. Full article

We performed a joined multitemporal and multiscale analysis of ground vertical movements around the main seismogenic source of Ischia island (Southern Italy) that, during historical and recent time, generated the most catastrophic earthquakes on the island, in its northern sector (Casamicciola fault). In particular, we considered InSAR (2015–2019) and ground-levelling data (1987–2010), attempting to better define the source that caused the recent 2017 earthquake and interpret its occurrence in the framework of a long-term behavior of the fault responsible for the major historical earthquakes in Casamicciola. Our results unambiguously constrain the location and the kinematics of the 2017 rupture and further confirm the presence of a relatively large sliding area west of the 2017 surface break. Overall, the studied seismogenic fault reveals a complex dynamic, moving differentially and aseismically in the pre- and post-seismic event, in response to the long-term subsidence of the central sector of the island, dominated by Mt. Epomeo. The fault segment that slipped coseismically also is evidence of post-seismic viscous relaxation. The long-term differential vertical movement on the apparently creeping eastern sector of the Casamicciola fault provides an estimate of the slip rate occurring on the fault (0.82 mm/y⁻¹). The analysis of the time of occurrence and the magnitude of the known historical earthquakes reveals that this rate is consistent with the recurrence of the earthquakes that occurred during at least the past three centuries and suggests that the time to the next seismic event at Casamicciola might be a few decades. More generally, our findings provide evidence of the link between subsidence and earthquakes in volcanic areas indicating, in this case, a high hazard for the island of Ischia. Results might be also useful for characterizing capable faulting in similar volcano-tectonic settings worldwide. Full article