With the acceleration of global warming and urbanization, the impact of the urban heat island effect on residents' livelihoods and urban development has gradually intensified. This study utilizes MODIS data (MOD11A2) from NASA covering the period from 2010 to 2020 to analyze the spatial-temporal variations in land surface temperature (LST) across counties in Xinjiang through trend analysis and spatial autocorrelation methods. Additionally, we investigate the relationship between LST and its influencing factors in 106 counties of Xinjiang using a geographical detector model. The findings reveal that: (1) On a temporal scale, interannual variations in LST among Xinjiang counties have exhibited a warming trend over the past decade, with an average annual climate tilt rate of 0.0792 degrees C/a. The average annual LST in southern Xinjiang is significantly higher than that observed in northern Xinjiang as well as for all of Xinjiang. (2) On a spatial scale, Altay Region has experienced the most rapid interannual increase in LST over recent years, while decreasing trends are primarily noted along both sides of the Tianshan Mountains and within marginal areas of the Tarim Basin. Furthermore, county-level LST across Xinjiang demonstrate positive spatial autocorrelation alongside significant spatial agglomeration. (3) Among various influencing factors affecting LST, topographic factors (DEM) exhibit the strongest explanatory power, followed by human factors (LULC), whereas socio-economic factors show comparatively weaker explanatory capacity. This research holds substantial scientific significance for elucidating mechanisms underlying climate change impacts within typical arid oasis urban agglomerations and for detecting trends related to LST changes.