Although Central Asia has a strong continental climate with a constant moisture deficit and low relative humidity, it is covered by thousands of lakes that are critical to the sustainability of ecosystems and human welfare in the region. Vulnerability to climate change and anthropogenic activities have contributed to dramatic inter-annual and seasonal changes of the lakes. In this study, we explored the high spatio–temporal dynamics of the lakes of Central Asia using the terraPulse™ monthly Landsat-derived surface water extent dataset from 2000 to 2015 and the HydroLAKES dataset. The results identified 9493 lakes and significant linear decreasing trends were identified for both the number (rate: −85 lakes/year, R²: 0.69) and area (rate: −1314.1 km²/year, R²: 0.84) of the lakes in Central Asia between 2000 and 2015. The decrease rate in lake area accounted for 1.41% of the total lake area. About 75% of the investigated lakes (7142 lakes), mainly located in the Kazakh steppe (especially in the north) and the Badghyz and Karabil semi-desert terrestrial ecological zones, experienced a decrease in the water area. Lakes with increasing water area were mainly distributed in the Northern Tibetan Plateau–Kunlun Mountains alpine desert and Qaidam Basin semi-desert zones in the east-south corner of Central Asia. The possible driving factors of lake decreases in Central Asia were explored for the Aral Sea and Tengiz Lake on yearly and monthly time scales. The Aral Sea showed the greatest decrease in the summer months because of increased evaporation and massive irrigation, while the largest decrease for Tengiz Lake was observed in early spring and was linked to decreasing snowmelt.

Landslide susceptibility mapping (LSM) is an important decision basis for regional landslide hazard risk management, territorial spatial planning and landslide decision making. The current convolutional neural network (CNN)-based landslide susceptibility mapping models do not adequately take into account the spatial nature of texture features, and vision transformer (ViT)-based LSM models have high requirements for the amount of training data. In this study, we overcome the shortcomings of CNN and ViT by fusing these two deep learning models (bottleneck transformer network (BoTNet) and convolutional vision transformer network (ConViT)), and the fused model was used to predict the probability of landslide occurrence. First, we integrated historical landslide data and landslide evaluation factors and analysed whether there was covariance in the landslide evaluation factors. Then, the testing accuracy and generalisation ability of the CNN, ViT, BoTNet and ConViT models were compared and analysed. Finally, four landslide susceptibility mapping models were used to predict the probability of landslide occurrence in Pingwu County, Sichuan Province, China. Among them, BoTNet and ConViT had the highest accuracy, both at 87.78%, an improvement of 1.11% compared to a single model, while ConViT had the highest F1-socre at 87.64%, an improvement of 1.28% compared to a single model. The results indicate that the fusion model of CNN and ViT has better LSM performance than the single model. Meanwhile, the evaluation results of this study can be used as one of the basic tools for landslide hazard risk quantification and disaster prevention in Pingwu County.

Soybean is one of the most important agricultural commodities in the world, thus making it important for global food security. However, widely used process-based crop models, such as the GIS-based Environmental Policy Integrated Climate (GEPIC) model, tend to underestimate the impacts of extreme climate events on soybean, which brings large uncertainties. This study proposed an approach of hybrid models to constrain such uncertainties by coupling the GEPIC model and extreme climate indicators using machine learning. Subsequently, the key extreme climate indicators for the globe and main soybean producing countries are explored, and future soybean yield changes and variability are analyzed using the proposed hybrid model. The results show the coupled GEPIC and Random Forest (GEPIC+RF) model (R: 0.812, RMSD: 0.716 t/ha and rRMSD: 36.62%) significantly eliminated uncertainties and underestimation of climate extremes from the GEPIC model (R: 0.138, RMSD: 1.401 t/ha and rRMSD: 71.57%) compared to the other five hybrid models (R: 0.365–0.612, RMSD: 0.928–1.021 and rRMSD: 47.48–52.24%) during the historical period. For global soybean yield and those in Brazil and Argentina, low-temperature-related indices are the main restriction factors, whereas drought is the constraining factor in the USA and China, and combined drought–heat disaster in India. The GEPIC model would overestimate soybean yields by 13.40–27.23%. The GEPIC+RF model reduced uncertainty by 28.45–41.83% for the period of 2040–2099. Our results imply that extreme climate events will possibly cause more losses in soybean in the future than we have expected, which would help policymakers prepare for future agriculture risk and food security under climate change.

The MODIS 8-day composite evapotranspiration (ET) product (MOD16A2) is widely used to study large-scale hydrological cycle and energy budgets. However, the MOD16A2 spatial resolution (500 m) is too coarse for local and regional water resource management in agricultural applications. In this study, we propose a Deep Neural Network (DNN)-based MOD16A2 downscaling approach to generate 30 m ET using Landsat 8 surface reflectance and temperature and AgERA5 meteorological variables. The model was trained at a 500 m resolution using the MOD16A2 ET as reference and applied to the Landsat 8 30 m resolution. The approach was tested on 15 Landsat 8 images over three agricultural study sites in the United States and compared with the classical random forest regression model that has been often used for ET downscaling. All evaluation sample sets applied to the DNN regression model had higher R2 and lower root-mean-square deviations (RMSD) and relative RMSD (rRMSD) (the average values: 0.67, 2.63 mm/8d and 14.25%, respectively) than the random forest model (0.64, 2.76 mm/8d and 14.92%, respectively). Spatial improvement was visually evident both in the DNN and the random forest downscaled 30 m ET maps compared with the 500 m MOD16A2, while the DNN-downscaled ET appeared more consistent with land surface cover variations. Comparison with the in situ ET measurements (AmeriFlux) showed that the DNN-downscaled ET had better accuracy, with R2 of 0.73, RMSD of 5.99 mm/8d and rRMSD of 48.65%, than the MOD16A2 ET (0.65, 7.18 and 50.42%, respectively).

Building height information plays an important role in many urban-related applications, such as urban planning, disaster management, and environmental studies. With the rapid development of real scene maps, street view images are becoming a new data source for building height estimation, considering their easy collection and low cost. However, existing studies on building height estimation primarily utilize remote sensing images, with little exploration of height estimation from street-view images. In this study, we proposed a deep learning-based method for estimating the height of a single building in Baidu panoramic street view imagery. Firstly, the Segment Anything Model was used to extract the region of interest image and location features of individual buildings from the panorama. Subsequently, a cross-view matching algorithm was proposed by combining Baidu panorama and building footprint data with height information to generate building height samples. Finally, a Two-Branch feature fusion model (TBFF) was constructed to combine building location features and visual features, enabling accurate height estimation for individual buildings. The experimental results showed that the TBFF model had the best performance, with an RMSE of 5.69 m, MAE of 3.97 m, and MAPE of 0.11. Compared with two state-of-the-art methods, the TBFF model exhibited robustness and higher accuracy. The Random Forest model had an RMSE of 11.83 m, MAE of 4.76 m, and MAPE of 0.32, and the Pano2Geo model had an RMSE of 10.51 m, MAE of 6.52 m, and MAPE of 0.22. The ablation analysis demonstrated that fusing building location and visual features can improve the accuracy of height estimation by 14.98% to 69.99%. Moreover, the accuracy of the proposed method meets the LOD1 level 3D modeling requirements defined by the OGC (height error ≤ 5 m), which can provide data support for urban research.

Urban greening distribution critically impacts residents’ quality of life and environmental sustainability. While the Green View Index (GVI), derived from street view imagery, is widely adopted for urban green space assessment, its limitation lies in the inability to capture non-street-area vegetation. Remote sensing imagery, conversely, provides full-coverage urban vegetation data. This study focuses on Beijing’s Third Ring Road area, employing DeepLabv3+ to calculate a street-view-based GVI as a predictor. Correlations between the GVI and Sentinel-2 spectral bands, along with two vegetation indices, such as the Normalized Difference Vegetation Index (NDVI) and Fractional Vegetation Cover (FVC), were analyzed under varying buffer radius. Regression and classification models were subsequently developed for GVI prediction. The optimal classifier was then applied to estimate green perception levels in non-street zones. The results demonstrated that (1) at a 25 m buffer radius, the near-infrared band, NDVI, and FVC exhibited the highest correlations with the GVI, reaching 0.553, 0.75, and 0.752, respectively. (2) Among the five machine learning regression models evaluated, the random forest algorithm demonstrated superior performance in GVI estimation, achieving a coefficient of determination (R²) of 0.787, with a root mean square error (RMSE) of 0.063 and a mean absolute error (MAE) of 0.045. (3) When evaluating categorical perception levels of urban greenery, the Extremely Randomized Trees classifier (Extra Trees) demonstrated superior performance in green vision perception level estimation, achieving an accuracy (ACC) score of 0.652. (4) The green perception level in non-road areas within Beijing’s Third Ring Road is 56.8%, which is considered relatively poor. Moreover, the green perception level within the Second Ring Road is even lower than that in the area between the Second and Third Ring roads. This study is expected to provide valuable insights and references for the adjustment and optimization of green perception distribution in Beijing, thereby supporting more informed urban planning and the development of sustainable, human-centered green spaces across the city.