PDF(Sci-hub)
Abstract：
Global Land Data Assimilation System (GLDAS) data are widely used for land-surface flux simulations. Therefore, the simulation accuracy using GLDAS dataset is largely contingent upon the accuracy of the GLDAS dataset. It is found that GLDAS land-surface model simulated runoff exhibits strong anomalies for 1996. These anomalies are investigated by evaluating four GLDAS meteorological forcing data (precipitation, air temperature, downward shortwave radiation and downward longwave radiation) in six large basins across the world (Danube, Mississippi, Yangtze, Congo, Amazon and Murray-Darling basins). Precipitation data from the Global Precipitation Climatology Centre (GPCC) are also compared with GLDAS forcing precipitation data. Large errors and lack of monthly variability in GLDAS-1996 precipitation data are the main sources for the anomalies in the simulated runoff. The impact of the precipitation data on simulated runoff for 1996 is investigated with the Community Atmosphere Biosphere Land Exchange (CABLE) land-surface model in the Yangtze basin, for which area high-quality local precipitation data are obtained from the China Meteorological Administration (CMA). The CABLE model is driven by GLDAS daily precipitation data and CMA daily precipitation, respectively. The simulated daily and monthly runoffs obtained from CMA data are noticeably better than those obtained from GLDAS data, suggesting that GLDAS-1996 precipitation data are not so reliable for land-surface flux simulations.
摘要：
全球土地数据同化系统（GLDAS）数据被广泛用于地表通量模拟。因此,使用GLDAS数据集的模拟准确性在很大程度上取决于GLDAS数据集的准确性。发现GLDAS地表模型模拟径流在1996年表现出强烈异常。通过评估四个GLDAS气象强迫数据(降水，空气温度,向下的短波辐射和向下的长波辐射)在世界六个大盆地(多瑙河，密西西比州,长江,刚果,亚马逊和默里达令盆地）。还将全球降水气候学中心（GPCC）的降水数据与GLDAS强迫降水数据进行了比较。GLDAS-1996降水数据的较大误差和缺乏月变异性是模拟径流异常的主要来源。使用长江流域的社区大气生物圈土地交换（CABLE）地表模型研究了1996年降水数据对模拟径流的影响，从中国气象局（CMA）获得了该地区的高质量局部降水数据。CABLE模型由GLDAS日降水数据和CMA日降水驱动，分别。从CMA数据获得的模拟每日和每月径流明显优于从GLDAS数据获得的径流，表明GLDAS-1996降水数据对于地表通量模拟并不那么可靠。
