A New Exploration of the Evaluation of Multisource Snow Data and Variations in Snow on the Qinghai–Tibet Plateau

Abstract

Snow is considered a climate indicator. The Qinghai–Tibet Plateau (QTP) is covered largely with typical alpine snow, influencing the local water–heat balance and the surrounding regional climate. However, there are large uncertainties in land surface snow data. A comprehensive, quantitative multimetric evaluation is an urgent need. In this study, five snow datasets are comprehensively evaluated on the basis of station-observed snow depth data. The temporal and spatial variations in snow depth over the QTP from 1979 to 2022 are analysed, and a new indicator is proposed to represent the overall snow variation on the QTP in a more reasonable way. The main conclusions are as follows: (1) In terms of snow depth on the QTP, the China Long Time Series Snow Depth dataset (CLSD) has the best performance, followed by ERA5-Land and NOAA (V3). The bias of the Northern Hemisphere Long Time Series Day-by-Day Snow Depth dataset (NHSD) is small compared with the observation. (2) The first EOF mode of snow depth on the QTP, in annual, autumn, winter and spring, shows a reversal spatial distribution between the main area of QTP and the northwestern Hengduan Mountains. The main area of QTP in snow depth has a decreasing trend, and the northwestern Hengduan Mountains in snow depth has an increasing trend. (3) After detrending, the main characteristic of EOF mode in snow depth on the QTP is consistent variations throughout the region. The variation indicator of snow depth on the QTP (QTPSDI) reflects the variation of the snow depth over the overall plateau. The QTPSDI has a significant 2–7 year periodicity. Therefore, we emphasise that the selection of accurate and applicable snow data is as important as the reasonable representation of snow cover variations.