基于广义加性模型估算1979-2014年三峡库区降水及其特征分析

doi:10.11821/dlxb201707007

摘要/Abstract

摘要：

高精度、长时间序列、空间连续的降水产品是气候分析、水文模拟等众多研究领域中的重要输入。近期欧洲研究人员融合了3种不同类型的降水数据：站点观测、遥感产品、再分析资料,制作了全球1979-2014年3小时一次0.25°空间分辨率的降水产品（Multi-Source Weighted-Ensemble Precipitation, MSWEP）,凭借高时空分辨率及其对多源信息的挖掘和融合,该产品一经发布即受到广泛关注。本文利用三峡库区及附近地区气象站点的降水资料对MSWEP月降水数据进行评估,采用广义加性模型算法（General Additive Model,GAM）融合站点降水空间插值结果和MSWEP产品,对三峡库区融合后降水进行分析。主要结论为：① 降水估算精度呈现冬春季偏高、夏秋季偏低的特征,MSWEP产品与站点插值方法具有互补性,前者对夏秋季降水估算精度更高,后者对冬春季降水估算精度更高;② GAM算法可以充分发挥站点插值和MSWEP数据各自的优势,提高区域降水估算精度,与融合前相比,均方根误差减少了17%~50%,相关系数r提高了10%~30%;③ 2003年库区蓄水前后降水变化的主要特征有：库区中部长江以南地区汛期降水（5-10月）下降,库区西部干季（1-4月,11-12月）降水增加,库区外围西北部（大巴山地区）汛期降水增加,降水空间格局异质性增加,干季降水占全年降水比例升高。

关键词: 降水, 三峡库区, MSWEP, GAM

Abstract:

Long time-series, spatially-contiguous and accurate precipitation products are one of the most important inputs for various studies, including climate change detection, hydrological modelling, drought monitoring, etc. However, due to its high spatio-temporal variability, precipitation is one of the meteorological elements that are most difficult to estimate. Recently, a new global gridded precipitation dataset that merges gauge measurements, satellite products, and reanalysis data, has been produced. Owing to its high spatial resolution, long time span, and comprehensive combination of different precipitation data sources, MSWEP data have received wide attention since its release. In this study, we first evaluated accuracy of MSWEP monthly precipitation using local gauge measurements in the Three Gorges Dam region, then produced more accurate precipitation data by combining MSWEP and gauge measurements with the GAM (Generalized Additive Model) method, and finally analysed precipitation changes before and after the dam water level rose to 135 m in June 2003. Main conclusions are drawn as follows. (1) No matter what methods are used, estimation accuracy of precipitation shows strong seasonality: more accurate in cold-dry season (spring and winter), while less accurate in hot-wet season (summer and autumn). Gauge measurement and MSWEP are complementary, with the former being more accurate in cold-dry season and the latter being more accurate in hot-wet season. (2) GAM can take advantages of both gauge measurements and MSWEP with flexibility and achieve more accurate precipitation estimation (rmse decreases by 17%-50%, and r increases by 10%-30%). There are still great seasonal variations in accuracy, with rmse being 8-20 mm in spring and winter and 20-50 mm in summer and autumn. (3) Based on the precipitation estimation results obtained in step 2, we found the following phenomena after the water level rose to 135 m in 2003: 1) dam regions south of the Yangtze River show a precipitation reduction over May-October; 2) dam regions in the western part show a precipitation increase over November-April; 3) The northwestern part outside core dam regions shows a precipitation increase over May-October, which is consistent with the results of other studies; 4) precipitation shows an increase in spatial heterogeneity, but a slight decrease in seasonality conveyed by an increase in proportion of precipitation in dry season.

Key words: precipitation, Three Gorges Dam region, MSWEP, GAM

王圆圆, 郭徵, 李贵才, 郭兆迪. 基于广义加性模型估算1979-2014年三峡库区降水及其特征分析[J]. 地理学报, 2017, 72(7): 1207-1220.

Yuanyaun WANG, Zheng GUO, Guicai LI, Zhaodi GUO. Precipitation estimation and analysis of the Three Gorges Dam region (1979-2014) by combining gauge measurements and MSWEP with generalized additive model[J]. Acta Geographica Sinica, 2017, 72(7): 1207-1220.

图/表 10

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

参考文献 24

[1]	Dalya C, Neilson R P, Phillips D L.A statistical-topographic model for mapping climatological precipitation over mountainous terrain. Journal of Applied Meteorology, 1994, 33: 140-158.
[2]	Herold N, Alexander L V, Donat M G, et al.How much does it rain over land? Geophysical Research Letter, 2015, 43: 341-348. doi: 10.1002/2015GL066615
[3]	Prein A F, Gobiet A.Impacts of uncertainties in European gridded precipitation observations on regional climate analysis. International Journal of Climatology, 2016. doi: 10.1002/joc.4706. doi: 10.1002/joc.4706 pmid: 5214405
[4]	Beck H E, van Dijk A I J M, Levizzani V, et al. MSWEP: 3-hourly 0.25° global gridded precipitation (1979-2015) by merging gauge, satellite, and reanalysis data. Hydrol. Earth Syst. Sci. Discuss., 2016. doi:10.5194/hess-2016-236.
[5]	Huffman G J, Bolvin D T, Nelkin, E J.et al.The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales, Journal of Hydrometeorology, 2007, 8: 38-55.
[6]	Funk C, Verdin A, Michaelsen J, et al.A global satellite-assisted precipitation climatology. Earth System Science Data, 2015, 7: 275-287. doi: 10.5194/essdd-8-401-2015
[7]	Kidd C, Bauer P, Turk J, et al.Inter comparison of high-resolution precipitation products over Northwest Europe. Journal of Hydrometeorology, 2012, 13: 67-83. doi: 10.1175/JHM-D-11-042.1
[8]	Wu J, Gao X J, Giorgi F, et al.Climate effects of the Three Gorges Reservoir as simulated by a high resolution double nested regional climate model. Quaternary International, 2012, 282: 27-36. doi: 10.1016/j.quaint.2012.04.028
[9]	Miller N, Jin J M, Tsang C F.Local climate sensitivity of the Three Gorges Dam. Geophysical Research Letter, 2005, 32, L16704. doi: 10.1029/2005GL022821
[10]	Wu L G, Zhang Q, Jiang Z H.The Three Gorges Dam affects regional precipitation. Geophysical Research Letter, 2006, 33: L13806. doi: 10.1029/2006GL026780. doi: 10.1029/2006GL026780
[11]	Li Bo, Tang Shihao.Local Precipitation changes induced by the Three Gorges Reservoir based on TRMM observations. Resources and Environment in the Yangtze Basin, 2014, 23(5): 617-625. doi: 10.11870/cjlyzyyhj201405004
	[李博, 唐世浩. 基于TRMM卫星资料分析三峡蓄水前后的局地降水变化. 长江流域资源与环境, 2014, 23(5): 617-625.] doi: 10.11870/cjlyzyyhj201405004
[12]	Xiao C, Yu R C, Fu Y F.Precipitation characteristics in the Three Gorges Dam vicinity. International Journal of Climatology, 2010, 30: 2021-2024. doi: 10.1002/joc.1963
[13]	Ma Zhanshan, Zhang Qiang, Qin Yanyan.Numerical simulation and analysis of the effect of Three Gorges Reservoir Project on the regional climate change. Resources and Environment in the Yangtze Basin, 2010, 19(9): 1044-1052.
	[马占山, 张强, 秦琰琰. 三峡水库对区域气候影响的数值模拟分析. 长江流域资源与环境, 2010, 19(9): 1044-1052.]
[14]	Zhao F, Sheperd M.Precipitation changes near Three Gorges Dam, China (Part I): A spatiotemporal validation analysis. Journal of Hydrometeorology, 2012, 13: 735-745. doi: 10.1175/JHM-D-11-061.1
[15]	Tian, Y D, Peters-Lidard C D. A global map of uncertainties in satellite-based precipitation measurements. Geophysical Research Letters, 2010, 37: L24407. doi: 10.1029/2010GL046008
[16]	Wood S N.Thin plate regression splines. Journal of the Royal Statistical Society Series B (Statistical Methodology), 2003, 65(1): 95-114.
[17]	Parmentier B, McGill B, Wilson A M, et al. An assessment of methods and remote-sensing derived covariates for regional predictions of 1 km daily maximum air temperature. Remote Sensing, 2014, 6: 8639-8670. doi: 10.3390/rs6098639
[18]	Ouarda T B M J, Charron C, Marpu P R. The generalized additive model for the assessment of the direct, diffuse, and global solar irradiances using SEVIRI images, with application to the UAE. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016, 9(4): 1553-1566. doi: 10.1109/JSTARS.2016.2522764
[19]	Zhang X, Liang S, Song Z, et al.Local adaptive calibration of the satellite derived surface incident shortwave radiation product using smoothing spline. IEEE Transactions on Geoscience and Remote Sensing, 2016. doi: 10.1109/TGRS. 2015.2475615. doi: 10.1109/TGRS.2015.2475615
[20]	Park N W, Chi K H.Quantitative assessment of landslide susceptibility using high resolution remote sensing data and a generalized additive model. International Journal of Remote Sensing, 2008, 29(1): 247-264. doi: 10.1080/01431160701227661
[21]	Package 'mgcv'. https://cran.r-project.org/web/packages/mgcv/mgcv.pdf.
[22]	Hamed K H.Trend detection in hydrologic data: The Mann-Kendall trend test under the scaling hypothesis. Journal of Hydrology, 2008, 349(3/4): 350-363. doi: 10.1016/j.jhydrol.2007.11.009
[23]	Wang Yuanyuan, Li Guicai, Zhang Yan.Regional representativeness analysis of national reference climatological stations based on MODIS/LST product. Journal of Applied Meteorological Science, 2011, 22(2): 214-220. doi: 10.3969/j.issn.1001-7313.2011.02.010
	[王圆圆, 李贵才, 张艳. 利用MODIS/LST产品分析基准气候站环境代表性. 应用气象学报, 2011, 22(2): 214-220.] doi: 10.3969/j.issn.1001-7313.2011.02.010
[24]	Hu Ting, Zhou Jiangxing, Dai Kan.Application of USCRN Station Density Strategy to China Climate Reference Network. Journal of Applied Meteorological Science, 2012, 23(1): 40-46.
	[胡婷, 周江兴, 代刊. USCRN气候基准站网布局理论在我国的应用. 应用气象学报, 2012, 23(1): 40-46.]