摘要:利用最新的高时空分辨率（1 km、1 h）的中国气象局高分辨率陆面数据同化系统（HRCLDAS-V1.0）大气近地面强迫资料，驱动由NCAR发展的通用陆面模式（CLM），对青藏高原地区2015年1月1日至9月30日的土壤湿度开展了模拟研究。结果表明模拟得到的高时空分辨率（1 km、1 h）土壤湿度能够体现出青藏高原地区从东南向西北逐渐变低的空间分布特征，较好地表现出各层土壤湿度的时间变化特征，6~9月土壤湿度波动较大，1~5月波动较平缓，上层土壤湿度变幅较大，深层变化较平缓。0~5 cm、0~10 cm和10~40 cm深度土壤湿度模拟结果与观测值的相关系数均在0.8以上，其中0~5 cm土层的相关系数达到0.92，各层土壤湿度观测值与模拟值的均方根误差变化则相反，3个土层土壤湿度模拟结果与观测值的偏差均小于0.04 mm3 mm-3，但模式对于研究时段土壤湿度变化的低值有高估现象，且模拟能力随着土层深度的加深而减弱。
关键词:青藏高原/
HRCLDAS/CLM系统/
陆面模式/
土壤湿度
Abstract:China Meteorological Administration High Resolution Land Data Assimilation System (HRCLDAS-V1.0) is used to drive the Community Land Model (CLM) for soil moisture simulation from 1 Jan to 30 Sep 2015 over the Tibetan Plateau region. Results show that the simulated soil moisture with high spatial-temporal resolution (1 km, 1 h) can reflect the spatial distribution characteristics of soil moisture in the Tibetan Plateau, where the soil moisture gradually decreases from the southeast to the northwest. Temporal variability of the soil moisture shows large changes from Jun to Sep, while the changes from Jan to May are more gentle at different depths of soil. Large changes are found in the upper soil layers, while changes are often small in deep soil layers. The correlation coefficient between the observations and simulated soil moisture is above 0.8 of 0-5-cm, 0-10-cm, and 10-40-cm depths, and reaches 0.92 for the layer of 0-5-cm depth. The changes of the root mean square errors for simulated and observed soil moisture are opposite, and the biases of simulated soil moisture are less than 0.04 mm3 mm-3 in all layers. During the study period, the low value of soil moisture is overestimated, and the simulation ability of the model for soil moisture decreases with the depth of the soil.
Key words:Tibetan Plateau/
HRCLDAS/CLM System/
Land surface model/
Soil moisture
PDF全文下载地址:
http://www.iapjournals.ac.cn/qhhj/article/exportPdf?id=20180405
