Characterizing Spatiotemporal Patterns of Snowfall in the Kaidu River Basin from 2000–2020 Using MODIS Observations

Abstract

Characterizing spatiotemporal patterns of snowfall is essential for understanding cryosphere responses to warming climate stress. The changes in snowfall and topographic controls in mountain regions still need to be clarified. This study proposes a general parsimonious methodology to obtain the frequency of snowfall in mountainous areas. The methodology employed is easily transferable to any other mountain region. Utilizing daily MODIS observations from June 2000 to May 2020 and the snowfall event detection algorithm, we monitored the frequency of snowfall in a long time series in the Kaidu river basin. The results are as follows: (1) The method for detecting the frequency of snowfall has high accuracy. The annual detected results agreed with surface observations, with an R2 of 0.65 and RMSE of 3.39. (2) The frequency of snowfall events increased monotonically with elevation. The influence of slope angle on snowfall gradually decreased with increasing elevation. (3) The frequency of snowfall events in the Kaidu river basin was dominated by an increasing trend. The trends showed a pronounced topographic dependence. This study reveals the distribution characteristics and changing snowfall trends in mountain regions. The results provide a reference for snowfall research in mountainous areas.

Keywords

• snowfall
• spatial-temporal pattern
• mountain regions
• snow grain size
• MODIS
• Kaidu river basin