As a measure of the accumulated heat deficit during the growing season transition, cooling degree days (CDDs) play a crucial role in regulating vegetation phenology and ecosystem dynamics. However, systematic analyses of CDD trends and their driving mechanisms remain limited, particularly in high-altitude regions where climate variability is pronounced. This study investigated the spatiotemporal variability in CDDs from 1982 to 2022 in alpine grasslands on the Qinghai–Tibetan Plateau (TP) and quantified the contributions of key climatic factors. The results indicate that lower CDD values (600 °C-days) were concentrated in colder, wetter areas. Temporally, area-averaged CDDs exhibited a significant decline, decreasing from 490.9 °C-days in 1982 to 495.8 °C-days in 2022 at a rate of 3.8 °C-days per year. Elevation plays a critical role in shaping CDD patterns, displaying a nonlinear relationship: CDDs decrease as elevation increases up to 4300 m, beyond which they increase, suggesting a transition from global climate-driven warming at lower elevations to local environmental controls at higher elevations, where snow–albedo feedback, topographic effects, and atmospheric circulation patterns regulate temperature dynamics. Tmax was identified as the dominant climatic driver of CDD variation, particularly above 4300 m, while radiation showed a consistent positive influence across elevations. In contrast, precipitation had a limited and spatially inconsistent effect. These findings emphasize the complex interactions between elevation, temperature, radiation, and precipitation in regulating CDD trends. By providing a long-term perspective on CDD variations and their climatic drivers, this study enhances our understanding of vegetation–climate interactions in alpine ecosystems. The results offer a scientific basis for modeling late-season phenological changes, ecosystem resilience, and land-use planning under ongoing climate change.

Soil moisture is a critical regulator in Earth’s energy cycle, exerting substantial influence on hydrological, ecological, and thermal processes. However, the spatio-temporal heterogeneity of soil moisture poses significant challenges for remote-sensing inversion, and conventional methods are constrained under complex surface conditions. In this study, Sentinel-1 and Sentinel-2 imagery were integrated with terrain data, and machine learning algorithms were employed to develop a soil moisture estimation model for the agro-pastoral transition zone. The primary objectives were to evaluate the accuracy of machine learning approaches for remote-sensing estimation of soil moisture under complex surface conditions, and to assess the suitability of different machine learning algorithms for this application. The results showed that ensemble methods, specifically boosting approaches, delivered low inversion errors and high R² values, whereas linear models showed limited accuracy. The vertically polarized microwave backscattering coefficients emerged as the parameters most sensitive to soil moisture variations, with optical bands and topographic factors also making substantial contributions. The overall inversion yielded a root mean square error (RMSE) of 0.066 cm³.cm⁻³. This study provides an effective approach for soil moisture inversion on complex surfaces, simplifies the modeling process, and offers a tool for soil moisture monitoring across large spatial scales.