The Qaidam Basin is an alpine arid inland basin characterized by water resource shortages and fragile ecological environments. Studying the evapotranspiration and water consumption characteristics of the Qaidam Basin is important for regional water-resource management and environmental protection. Based on eddy covariance flux data for four underlying surfaces (Golmud cropland, shrubland, Nomuhong cropland, and alpine meadow) in the Qaidam Basin in 2020, the evapotranspiration variation characteristics for different underlying surfaces at different time scales were analyzed, the influence of different typical meteorological factors on actual evapotranspiration was explored, and water consumption characteristics of the different underlying surfaces were analyzed. The results showed that the evapotranspiration of each underlying surface was mainly concentrated in the growing season. The Golmud cropland and alpine meadow had the highest evapotranspiration in July, at 62.50 mm and 88.92 mm, respectively, while the shrubland and Nomuhong cropland had the highest evapotranspiration in August, at 40.47 mm and 100.02 mm, respectively. The average daily evapotranspiration of Golmud cropland, shrubland, Nomuhong cropland, and alpine meadow was 1.12 mm, 0.78 mm, 1.98 mm, and 1.79 mm, respectively. The half-hour evapotranspiration reached a maximum value from 14:00–15:00. The evapotranspiration of different underlying surfaces was strongly correlated with air temperature, followed by relative humidity, and weakly correlated with wind speed. Daily evapotranspiration was strongly correlated with the soil temperature of Golmud cropland, shrubland, and alpine meadow, and the soil volume water content of shrubland and alpine meadows. The water consumption variation characteristics indicated that each underlying surface was dominated by water consumption, accounting for 96.73%, 96.26%, 96.26%, 74.30% in Golmud cropland, shrubland, Nomuhong cropland, and alpine meadow, respectively. Among them, precipitation was the main factor affecting the water consumption of the different underlying surfaces. The purpose of this study was to explore the actual evapotranspiration characteristics of different underlying surfaces, the results of which can be used as a reference for studies of the water cycle in Qaidam Basin.

China’s coastal wetlands play a crucial role in maintaining biodiversity and providing essential ecosystem services. However, the absence of high-resolution wetland type maps poses substantial challenges for effective conservation and management. This study proposes a two-step classification framework that integrates pixel-based Random Forest algorithms with object-based hierarchical decision trees, utilizing Landsat-8 time-series imagery to generate a detailed wetland map comprising 10 wetland types and 5 non-wetland categories. The results reveal distinct spatial patterns along China’s coastline: freshwater wetlands and riverine systems dominate the northern regions, whereas southern coastal zones feature extensive tidal flats, aquaculture ponds, and mangrove ecosystems. The proposed method achieved an overall accuracy of 89.76% and a Kappa coefficient of 0.891, demonstrating its effectiveness for large-scale wetland mapping. This study provides robust technical support for the sustainable conservation and ecological management of coastal wetlands.

High-spatiotemporal-resolution precipitation data are essential for studies on regional hydrology, meteorology, and ecological conservation. Because the Qaidam Basin is a data-scarce region with a few ground stations and coarse-resolution remote sensing products, its utility in regional research is constrained. Therefore, high-resolution precipitation data are urgently needed. Here, longitude, latitude, the normalized difference vegetation index (NDVI), the digital elevation model (DEM), daytime and nighttime land surface temperature, slope, and aspect were selected as environmental variables. Four machine learning methods, Artificial Neural Network (ANN), Cubist, Random Forest (RF), and Support Vector Machine (SVM), were used to downscale Tropical Rainfall Measuring Mission (TRMM) precipitation data from 25 to 1 km in the Qaidam Basin and validated using ground observation stations. For annual downscaling, the accuracy ranked as Cubist > ANN > RF > SVM, and residual correction further improved performance. The Cubist model produced the best results, generating finer spatial patterns and reducing outliers in both annual and monthly products. Longitude, latitude, the DEM, and the NDVI were important contributors to the Cubist model. The resulting high-resolution dataset provides valuable support for hydrological and climate change research in the Qaidam Basin.