边界层参数化方案对文昌近地层湿度模拟的影响

Influence of Boundary Layer Parameterization Scheme on Near-Surface Humidity Simulation in Wenchang

摘要: 本文基于WRF中尺度预报模式，采用10种边界层参数化方案(YSU、MYJ、QNSE、MYNN2、MYNN3、GBM、BouLac、UW、TEMF、Shin-Hong)，对2015年12个个例的近地层湿度开展了数值模拟。通过浅层风塔观测数据进行对比，研究了不同边界层参数化方案对文昌近地层湿度模拟差异。结果表明：随着近地层高度的增加，三月到七月的观测相对湿度增加，模拟相对湿度变化不大；其余月份观测和模拟的相对湿度均会降低；模拟和观测的均方根误差随高度增加而增加，相关系数随高度增加则会减小。通过对均方根误差和相关系数的分析，MYJ方案模拟数据接近观测数据，并且可以模拟出每个月份的日变化趋势，是模拟结果最好的方案。TEMF方案的均方根误差最大且相关系数表现最差，是模拟效果最差的方案。

Abstract: Based on the WRF mesoscale prediction model, 10 boundary layer parameterization schemes (YSU, MYJ, QNSE, MYNN2, MYNN3, GBM, BouLac, UW, TEMF and Shin-Hong) are used to simulate the near surface humidity of 12 cases in 2015. By comparing the observation data of shallow wind tower, the simulation differences of near stratum humidity in Wenchang under different boundary layer parameterization schemes are studied. The results show that with the increase of near ground height, the observed relative humidity increases from March to July, and the simulated relative humidity changes little. The relative humidity observed and simulated in other months will decrease; the root mean square error of simulation and observation increases with the increase of height, and the correlation coefficient decreases with the increase of height. Through the analysis of root mean square error and correlation coefficient, the simulation data of MYJ scheme is close to the observation data, and the daily variation trend of each month can be simulated. It is the best scheme with simulation results. TEMF scheme has the largest root mean square error and the worst correlation coefficient, which is the scheme with the worst simulation effect.