Warming amplification with both altitude and latitude in the Tibetan Plateau

Abstract

AbstractMany studies have examined elevation‐dependent warming in mountain regions of the world, yet little attempt has been undertaken to examine the relationship of the observed warming with both altitude and latitude on a regional scale. Here, based on the mean temperature series (1961–2015) of 105 stations in the Tibetan Plateau, the signals of warming amplification with these two factors are quantified using four groups of methods. The results, highly consistent for the different methods, confirm that the warming for the Tibetan Plateau is not only elevation‐dependent, but also latitude‐dependent, and the magnitude of altitudinal amplification trend is two orders of magnitude larger than that of latitudinal amplification trend on the annual and seasonal scales over the period 1961–2015. The annual warming is amplified by 0.066°C 10 year−1·km−1 of altitude and 0.016 × 10−2°C 10 year−1·km−1 of latitude (or 0.018°C 10 year−1·degree−1 of latitude), with the highest altitudinal (latitudinal) amplification in winter (autumn), followed by autumn (winter) and the lowest in spring and summer. Meanwhile, the estimates of altitude and latitude components of the regional warming shows that the contribution of latitude to the regional warming is larger than that of altitude on the annual scale and for all seasons excepting summer, due primarily to the much larger range of latitude than the range of altitude covered by all the stations used. Moreover, analysis of the number of snow cover days demonstrates that the snow albedo feedback has played a significant role in enhancing the warming in the Tibetan Plateau during the recent sub‐period 1996–2015, due to the decrease of snow cover days in the spring, autumn and winter seasons. These results highlight an integrated picture of Tibetan Plateau warming amplification, and a clear snow albedo feedback effect on the regional amplification, allowing better investigation of impacts and mechanisms.