Comprehensive evaluation of Ensemble Multi-Satellite Precipitation Dataset using the Dynamic Bayesian Model Averaging scheme over the Tibetan plateau
Comprehensive evaluation of Ensemble Multi-Satellite Precipitation Dataset using the Dynamic Bayesian Model Averaging scheme over the Tibetan plateau
复制标题
青藏高原集合多星降水数据集动态贝叶斯模型平均方案综合评价
DOI：
10.1016/j.jhydrol.2017.11.050
复制
发表时间：
2018
影响因子：
6.4
通讯作者：
Hong Yang
中科院分区：
文献类型：
--
作者：
Ma Yingzhao;Yang Yuan;Han Zhongying;Tang Guoqiang;Chu Zhigang;Maguire Lane;Hong Yang
本研究的目的是使用动态贝叶斯模型平均方案 (EMSPD-DBMA) 综合评估 2001 年至 2015 年青藏高原 (TP) 日尺度和 0.25° 尺度的新集合多卫星降水数据集。针对仪表观测值的误差分析表明，EMSPD-DBMA 捕获了每日降水量的时空模式，可接受的相关系数 (CC) 为 0.53，相对偏差 (RB) 为 -8.28%。此外，2014年和2015年夏季，EMSPD-DBMA在几乎所有指标上都优于IMERG和GSMaP-MVK，最低的RB和均方根误差（RMSE）值分别为-2.88%和8.01mm/d。它还比 IMERG 和 GMap-MVK 更好地再现了日降雨量方面的概率密度函数 (PDF)，并更好地估计了中雨和强降雨。此外，长江上游地区耦合路由和超额存储（CREST）模型的水文评估表明，EMSPD-DBMA强制模拟在径流预测方面表现出令人满意的水文性能，与校准和验证期间的仪表强制模拟（0.88和0.60）相比，纳什-萨特克利夫效率系数（NSE）值为0.82和0.58。 分别。 EMSPD-DBMA 还比新的多源加权集合降水版本 2 (MSWEP V2) 产品更适合峰值流量模拟，这表明集合卫星降水数据的水文应用前景广阔。本研究属于青藏高原多卫星混合降水资料的早期综合评价，对于完善复杂地形地区的DBMA算法具有重要意义。
The objective of this study is to comprehensively evaluate the new Ensemble Multi-Satellite Precipitation Dataset using the Dynamic Bayesian Model Averaging scheme (EMSPD-DBMA) at daily and 0.25° scales from 2001 to 2015 over the Tibetan Plateau (TP). Error analysis against gauge observations revealed that EMSPD-DBMA captured the spatiotemporal pattern of daily precipitation with an acceptable Correlation Coefficient (CC) of 0.53 and a Relative Bias (RB) of −8.28%. Moreover, EMSPD-DBMA outperformed IMERG and GSMaP-MVK in almost all metrics in the summers of 2014 and 2015, with the lowest RB and Root Mean Square Error (RMSE) values of −2.88% and 8.01 mm/d, respectively. It also better reproduced the Probability Density Function (PDF) in terms of daily rainfall amount and estimated moderate and heavy rainfall better than both IMERG and GSMaP-MVK. Further, hydrological evaluation with the Coupled Routing and Excess STorage (CREST) model in the Upper Yangtze River region indicated that the EMSPD-DBMA forced simulation showed satisfying hydrological performance in terms of streamflow prediction, with Nash-Sutcliffe coefficient of Efficiency (NSE) values of 0.82 and 0.58, compared to gauge forced simulation (0.88 and 0.60) at the calibration and validation periods, respectively. EMSPD-DBMA also performed a greater fitness for peak flow simulation than a new Multi-Source Weighted-Ensemble Precipitation Version 2 (MSWEP V2) product, indicating a promising prospect of hydrological utility for the ensemble satellite precipitation data. This study belongs to early comprehensive evaluation of the blended multi-satellite precipitation data across the TP, which would be significant for improving the DBMA algorithm in regions with complex terrain.
登录
查看更多内容
DOI：
10.1175/jam2173.1
发表时间：
2004-12-01
期刊：
JOURNAL OF APPLIED METEOROLOGY
影响因子：
--
作者：
Hong, Y;Hsu, KL;Gao, XG
通讯作者：
Gao, XG
实时 Trmm 多卫星降水分析的全球视图：对其后续全球降水测量任务的影响
影响因子：
8
作者：
Bin Yong;Die Liu;Jonathan J. Gourley;Yudong Tian;George J. Huffman;Liliang Ren;Yang Hong
通讯作者：
Yang Hong
影响因子：
3.8
作者：
Daqing Yang;B. Ye;A. Shiklomanov
通讯作者：
Daqing Yang;B. Ye;A. Shiklomanov
影响因子：
3.8
作者：
Y. Mei;E. Nikolopoulos;E. Anagnostou;M. Borga
通讯作者：
Y. Mei;E. Nikolopoulos;E. Anagnostou;M. Borga
DOI：
10.1002/joc.4045
发表时间：
2015-06
期刊：
International Journal of Climatology
影响因子：
--
作者：
Yingzhao Ma;Yinsheng Zhang;Daqing Yang;S. B. Farhan
通讯作者：
Yingzhao Ma;Yinsheng Zhang;Daqing Yang;S. B. Farhan