Some sloping peatlands in northern regions often develop surface microtopographic patterns to maintain their water balance and ecosystem functioning. However, we do not know whether and how spatial patterning would influence the water balance and peat formation of permafrost-affected peatlands in relatively dry regions. Here we used data from the field observations and Unmanned Aerial Vehicle (UAV) survey of a slope peatland at an elevation of around 4800 m in the hinterland of the Qinghai-Tibetan Plateau (QTP) to document and understand the topographic controls of water balance and vegetation growth. Our terrain analysis result shows that the peatland-located on the middle of a hillslope-has a gentle slope of 5.6 degrees +/- 2.5 degrees, while the non-peatland upper has a steep slope of 12 degrees +/- 4.5 degrees. The great upstream catchment area and the presence of shallow impermeable permafrost likely create a saturated condition for peat formation. Our UAV results show obvious spatial patterning of abundant pools and ridges across this peatland, and pool sizes and ridge abundance increase with increasing slopes, suggesting that slope-controlled water flow gradient is the main driver of ridge formation and that ridges is to slow down the runoff. UAV-derived greenness values show a positive relationship with the total pool extent locally (R2 = 0.60) and decrease with increasing distance from the individual pools, suggesting sensitive responses of vegetation growth to surface moisture. Thus, enhanced vegetation growth and likely resultant great peat accumulation immediately around pools potentially further differentiate surface micro-topography, strengthening the pool stability. We conclude that the local slope gradient, surface patterning (pools and ridges) and permafrost interact together to regulate water flow and maintain water balance, which in turn regulate the vegetation growth, peat accumulation and peatland stability. Our study implies that the delicate water balance maintained partly by microtopography is sensitive to climate change-especially potential extreme hydroclimate events-and natural and human-induced disturbances that may modify the surface patterning and weaken the peatland's stability, affecting the carbon sequestration ability of this type of peatlands.

The growth of vegetation on the Qinghai Tibet Plateau (QTP) is experiencing significant changes due to climate change. There is still a lack of high -precision simulation methods for alpine grassland cover (AGC), and the climate feedback mechanisms of AGC remain unclear, which poses challenges for the production of highprecision AGC products and the formulation of ecological conservation policies. In this study, a transferable stacking deep learning (Stacking -DL) model is proposed based on a CNN, a DNN, and a GRU for AGC time series simulation. The applicability of deep learning models for AGC simulation is evaluated based on long time series of measured data, MODIS data, and environmental factors. Finally, the AGC spatiotemporal changes and controlling environmental factors in the alpine region were analyzed based on Sen 's slope and structural equation modeling (SEM). The results showed that feature selection and parameter optimization improved the applicability of the deep learning models in AGC simulations, and the DNN (R 2 = 0.899, RMSE = 0.078) model performed best among the base deep learning models. The Stacking -DL model combines the advantages of multiple models and achieves high transfer accuracy. In the YRSR, the AGC increase area (20.34 %) is greater than the AGC decrease area (3.34 %), the increase area is mainly located in the northeast, and the decrease area is mainly located in the southwest. AGC changes in the YRSR are mainly controlled by permafrost and climate. This study provides a high -precision and transferable vegetation monitoring model for alpine mountain regions based on advanced deep learning models and clarifies the response mechanism of AGC under climate change.