冰川区不同气温估算方法评估——以藏东南帕隆4号冰川为例

冰川作为地表特殊的下垫面，冰川区内气温明显低于同高度非冰川区大气温度。如何利用低海拔非冰川区观测资料精确估算高海拔冰川区气温，直接关系着青藏高原冰川消融估算及其水文效应的评估。利用架设在藏东南帕隆藏布4号冰川不同高度带的四台自动气象站资料，分析了冰川区与非冰川区气温的波动特征，评估了迄今为止通用的线性递推模型（DT模型）、分段拟合模型（SM模型）和简化热力学模型（GB模型）三种方法在藏东南冰川区气温估算方面的应用效果。对比研究发现：SM模型在帕隆4号冰川上的模拟效果最为理想且操作相对简单；传统DT模型在消融区存在严重的高估，帕隆4号冰川表面夏季（6-8月）正积温的高估比例接近39%；GB模型由于受到诸如冰川风边界层厚度等不确定性的影响，降低了大范围温度估算的可操作性。

Evaluation on three methods of temperature estimation at glacierized reigons: taking the Parlung No.4 Glacier in the southeastern Tibetan Plateau as example

As a special underlying surface of the earth's surface, the temperature within katabatic wind boundary layer is significantly lower than the temperature of free atmosphere at the same altitude because of the influence of katabatic wind. How to accurately estimate the temperature of glacierized regions by using the data of non-glacierized regions is important for the accurate assessment of glacier melting and their hydrological influence on the Tibetan Plateau. Based on four automatic weather stations (AWSs) at different elevations on Parlung No.4 Glacier in the southeastern Tibetan Plateau, we analyzed the characteristics of temperature fluctuation between the glacierized and the non-glacierized area. Meanwhile we evaluated the performance of three popular estimation methods on air temperature:the traditional linear model (DT model), piecewise linear model (SM model) and simplified thermodynamic model (GB model) respectively. Our results evidenced that the SM model has the strong ability to estimate the air temperature at three AWSs; the application of the traditional DT model will lead to significant overestimation in ablation region, with 39% overestimation of the positive accumulated temperature in summer (June to August). The GB model increases the uncertainty of estimated temperature due to the uncertainty of many parameters including the thickness of katabatic boundary layer, which reduces its operability on the large-scale temperature estimation.