Climatic controls on surface albedo and mass balance of Urumqi Glacier No. 1 in the last decade, 2014-2024.

Glaciers in Central Asia are retreating; we analyze how climate controls surface albedo on Urumqi Glacier No. 1 (UG1) using Landsat-8/9 (2014-2024) and ERA5-Land datasets integrated with the Google Earth Engine platform. Multiple linear regression and generalized additive models (GAMs) relate albedo to air temperature, snowfall, and aerosols. UG1's albedo declines significantly (-0.006 years-1), concurrent with sustained mass loss (-0.019 m w.e. yr-1). Air temperature and snowfall are dominant controls: linear models explain 34.6% of variance, while GAMs increase explained variance to 45.7% and reveal nonlinear threshold behavior near 0°C. Seasonal skill varies widely (7.7-37.7%). Integrating 45 mass-balance stakes shows spatially heterogeneous albedo-mass-balance correlations (-0.5 to +1.0); glacier-wide correlation is weak (r = -0.075, p = 0.191), with positive correlations in upper zones consistent with temperature-albedo feedback. Roughly 54% of variance remains unexplained, highlighting roles for non-meteorological processes. Results support nonlinear parameterizations of albedo in glacier mass-balance models and improve future projections for a warming Central Asia.