CHIRPS和GLEAM卫星产品在中国的干旱监测效用评估

为评估具有分辨率高、覆盖范围广、数据序列动态连续优点的卫星降水和潜在蒸散发产品在中国干旱监测中的应用能力,以CGDPA(China gauge-based precipitation daily analysis dataset)和PM-Ep(potential evapotranspiration calculated by penman-monteith)为基准数据,以标准化降水蒸散指数(standardized precipitation evapotranspiration index, SPEI)为干旱监测指标,利用6个精度统计指标和分类度量指标评估CHIRPS(climate hazards group infrared precipitation with station data)、GLEAM-Ep(global land-surface evaporation:the Amsterdam methodology, potential evapotranspiration)在中国的干旱监测效用。结果表明:1)CHIRPS和GLEAM-Ep在中国东部和西南地区表现良好。由于气象站格网的稀疏,它们在中国西部大部分地区的表现存在较大不确定性。2)SPEIs(基于CHIRPS和GLAEM-Ep计算的SPEI)在中国东部和西南地区对干旱模拟效果较好,其与SPEIm(基于CGDPA和PM-Ep计算的SPEI)的相关系数(CC)在东南、西南、长江中下游地区、华北均大于0.6,尤其是相对湿润地区的CC值在0.70以上且均方根误差低于0.8。3)SPEIs3(3个月的SPEIs)在东南、西南、长江中下游地区、华北对干旱面积比的演变过程模拟效果较好,持续干旱模拟成功率分别为0.93、0.57、0.71、0.81及干旱空间分布模拟精度较高。总体而言,基于长序列CHIRPS和GLEAM-Ep计算的SPEI可适用于中国东部和西南地区的干旱监测与评价,可为该地区的干旱事件监测与预警、水资源管理提供可靠的参考依据。

Drought monitoring utility assessment of CHIRPS and GLEAM satellite products in China

Quantitatively evaluation of satellite precipitation and potential evapotranspiration products with the high resolution, wide coverage and dynamic continuity is an important foundation for drought monitoring over mainland China. This paper examined the accuracy of products and the utility of drought monitoring using the newly released long-term (more than 30 years) satellite precipitation products (climate hazards group infrared precipitation with station, CHIRPS) data version 2.0 and potential evapotranspiration products (global land-surface evaporation: the Amsterdam methodology, GLEAM) dataset. We selected standardized precipitation evapotranspiration index (SPEI) and six statistical and classification measures indexes to evaluate accuracy and the utility of CHIRPS and GLEAM-Ep (potential evapotranspiration derived from GLEAM), with China Gauge-based Precipitation Daily Analysis dataset (CGDPA) and potential evapotranspiration calculated by Penman-Monteith (PM-Ep) as evaluation reference data sets. The results showed that CHIRPS and GLEAM-Ep performed well in eastern and southwest of China. However, their performance in the most areas of western China was uncertain due to the sparsity of meteorological stations. SPEIs (SPEI value calculated by the CHIRPS and GLEAM-Ep) had a good performance on drought monitoring in eastern and southwestern China. Values of correlation coefficient (CC) between SPEIs and SPEIm (SPEI value calculated by the CGDPA and PM-Ep) were more than 0.6 in southeastern China, southwestern China, the middle and lower reaches of the Yangtze River region and northern China, especially in relatively humid areas, which the correlation coefficient values were more than 0.7 and the root mean-squared error (RMSE) values were less than 0.8. In order to further evaluate the performance of SPEIs in drought monitoring at spatial and temporal scales, the SPEIs3 (SPEIs at three-month scale) and SPEIm3 (SPEIm at three-month scale) in southeastern China, southwestern China, the middle and lower reaches of the Yangtze River region and north China were compared. SPEIs3 had a better simulation effect on the evolution process of the ratio of drought area in those regions during1981-2016. And the success ratio of simulating drought duration was about 0.93, 0.57, 0.71 and 0.81, respectively. Moreover, SPEIs3 displayed a high accuracy over the four regions in capturing drought, which was the spatial distribution of typical drought events on specific month. In conclusion, SPEI based on long-term CHIRPS and GLEAM-Ep could be applied to monitor and evaluate drought in eastern and southwestern China. This paper can provide reliable reference for monitoring and warning of drought events and water resources management in those regions.