古尔班通古特沙漠大气边界层参数化方案的模拟评估[1]

为准确描述我国最大的固定/半固定沙漠-古尔班通古特沙漠区域的大气边界层结构，本文利用该沙漠腹地2017年的梯度铁塔和通量观测数据，基于中尺度气象模式WRF (Weather Research and Forecast v3.7.1)，分析了5种边界层参数化方案在古尔班通古特沙漠的适用性。结果表明：1）采用WRF模拟沙漠腹地近地层内的边界层特征时，2m气温的模拟存在冷偏差，5种边界层参数化方案均能较好地模拟出四个季节2m气温的日变化特征，其中非局地方案ACM2(Asymmetric Convective Model version 2)对2m气温效果最好，局地方案BL方案的模拟偏差最大；2）5种边界层参数化方案均能够模拟出10m风速的日变化特征，其中局地方案BL(Bougeault-Lacarrere)对10m风速效果最佳；3）采用WRF模拟沙漠近地层内的地表通量特征时，感热通量存在高估现象，潜热通量存在低估现象，5种边界层参数化方案均能较好地模拟出四个季节模拟时间段内地表净辐射通量的日变化特征，其中局地方案MYJ(Mellor-Yamada-Janjie)的模拟精度最高。

Simulation Evaluation of Parameterization Scheme for Atmospheric Boundary Layer in Gurbantunggut Desert

As the largest fixed/semi-fixed desert in China, the Gurbantünggüt Desert has a certain impact on the local regional climate. To accurately describe the atmospheric boundary layer process in the desert region, this paper analyzes the applicability of five Planetary Boundary Layer (PBL) parameterization schemes, based on the gradient tower and EC flux system observation data of the desert hinterland in 2017. The results show that: 1) there is a cold biases in the simulation of 2m air temperature. The five PBL parameterization schemes can simulate the diurnal variation characteristics of the 2m temperature well in the four seasons, and the ACM2 (Asymmetric Convective Model version 2) has the best simulation effect on 2m temperature, the BL (Bougeault-Lacarrere) scheme has the largest simulation deviation. 2) The five PBL parameterization schemes can roughly simulate the daily variation characteristics of 10m wind speed, and the BL scheme shows better effect on 10m wind speed. 3) When WRF is used to simulate the surface flux characteristics in the near-surface layer of desert, the sensible heat flux is overestimated, and the latent heat flux is underestimated. The five PBL parameterization schemes can simulate the daily variation characteristics of the net radiant flux well, in which the MYJ (Mellor-Yamada-Janjie) yields highest simulation accuracy.