利用空间随机森林方法提升GPM卫星遥感降水质量

胡保健; 李伟; 陈传法; 胡占占

doi:10.11834/jrs.20221222

大气与海洋 | 浏览量 : 0 下载量: 149 CSCD: 0 更多指标

R-PDF
PDF
导出
分享
收藏
专辑

利用空间随机森林方法提升GPM卫星遥感降水质量
Improving the quality of remotely sensed precipitation product from GPM satellites by using a spatial random forest
2024年28卷第2期页码：414-425
纸质出版日期： 2024-02-07 ，
DOI： 10.11834/jrs.20221222

扫描看全文

胡保健，李伟，陈传法，胡占占.2024.利用空间随机森林方法提升GPM卫星遥感降水质量.遥感学报，28（2）： 414-425

Hu B J，Li W，Chen C F and Hu Z Z. 2024. Improving the quality of remotely sensed precipitation product from GPM satellites by using a spatial random forest. National Remote Sensing Bulletin， 28（2）：414-425
胡保健，李伟，陈传法，胡占占.2024.利用空间随机森林方法提升GPM卫星遥感降水质量.遥感学报，28（2）： 414-425 DOI： 10.11834/jrs.20221222.

Hu B J，Li W，Chen C F and Hu Z Z. 2024. Improving the quality of remotely sensed precipitation product from GPM satellites by using a spatial random forest. National Remote Sensing Bulletin， 28（2）：414-425 DOI： 10.11834/jrs.20221222.

摘要

卫星遥感降水产品是当前获取大范围、连续性降水观测的主要来源，但目前已有的卫星遥感降水产品空间分辨率粗糙，且存在一定的系统偏差。为此，本文充分考虑高分辨率环境变量（包括地形、NDVI、地表温度、经纬度）对降水影响以及邻近遥感降水（站点）空间相关性，构建了一种双阶段空间随机森林SRF（Spatial Random Forest）方法（SRF-SRF）。以四川省2015年—2019年GPM（Global Precipitation Measurement Mission）月降水数据为例，借助SRF-SRF对其质量提升，并将计算结果与现有7种方法比较，包括地理加权回归（GWR）、反向传播神经网络（BPNN）、随机森林（RF）、站点降水Kriging插值（Kriging）、经SRF降尺度后的地理差异分析校正（SRF-GDA）、经双线性插值降尺度后的SRF校正（Bi-SRF）以及年降水经SRF降尺度后按月比例分解并利用SRF校正（SRFdis）等。实验分析表明：（1）在月尺度上，与原始GPM相比，SRF-SRF的平均绝对误差（MAE）降低了19.51%，中误差（RMSE）降低了16.35%，而且精度优于其他方法；在季尺度上，SRF-SRF在冬季误差最小，在夏季误差最大，但其计算精度均优于其他方法；在年尺度上，基于SRF的4种方法（包括SRF-SRF、SRF-GDA、Bi-SRF和SRFdis）优于GWR、BPNN、RF，并且SRF-SRF计算精度优于单阶段的Bi-SRF和SRF-GDA。（2）SRF-SRF降水产品空间分布连续性较好，且局部降水细节得到明显提升。（3）借助RF对各自变量重要性分析得出，降水空间相关性对卫星遥感降水质量提升具有重要作用。（4）基于月尺度的SRF-SRF融合校正效果优于基于年尺度的SRFdis，表明NDVI可用于该区域月尺度降水质量提升。

Abstract

Satellite remote-sensing precipitation products are currently the main source for obtaining large-scale and continuous precipitation observations. However

currently available satellite remote-sensing precipitation products have coarse spatial resolution and suffer from certain systematic biases. Thus

this paper aims to downscale the precipitation data and remove its inherent systematic biases.

This paper proposes a two-stage Spatial Random Forest (SRF) method (SRF-SRF) by fully considering the influence of high-resolution environmental variables (including topography

NDVI

surface temperature

latitude

and longitude) on the precipitation and the spatial correlation of neighboring remotely sensed precipitation (stations). Taking the Global Precipitation Measurement Mission (GPM) monthly precipitation data of Sichuan Province from 2015—2019 as an example

its quality is enhanced with the help of SRF-SRF. The calculation results are compared with those of seven existing methods

including Geo-Weighted Regression (GWR)

Back-Propagation Neural Network (BPNN)

Random Forest (RF)

Kriging interpolation of station precipitation (Kriging)

Geographic Difference Analysis correction after downscaling by SRF (SRF-GDA)

SRF correction after downscaling by bilinear interpolation (Bi-SRF)

and annual precipitation downscaled by SRF. Subsequently

the results are scaled by month and corrected using SRF (SRFDis).

This paper proposes a two-stage satellite precipitation product-quality enhancement method that considers spatial correlation. The method takes into account the spatial autocorrelation between precipitation and combines downscaling and calibration while integrating environmental factors. Accordingly

the spatial resolution and accuracy of precipitation products improve. Experimental results show that the new method outperforms the other seven classical methods and is more applicable to the quality improvement of precipitation products in complex terrain.

Experimental analysis shows the following

(1) At the monthly scale

compared with the original GPM

the mean absolute error (MAE) of SRF-SRF is reduced by 19.51%

and the medium error (RMSE) is reduced by 16.35%. The accuracy is better than those of other methods. At the seasonal scale

SRF-SRF has the smallest error in winter and the largest error in summer

but its calculation accuracy is better than those of other methods. At the annual scale

the four SRF-based methods (including SRF-SRF

SRF-GDA

Bi-SRF

and SRFdis) outperform GWR

BPNN

and RF. The accuracy of SRF-SRF is higher than that of Bi-SRF and SRF-GDA. (2) The spatial-distribution continuity of SRF-SRF precipitation products is better

and the local precipitation details are significantly improved. (3) The spatial correlation of precipitation plays an important role in the improvement in GPM precipitation quality. (4) SRF-SRF based on the monthly scale is better than SRFdis based on the annual scale. This finding indicates that NDVI can be used for precipitation-quality enhancement at the monthly scale in Sichuan province.

关键词

遥感降水降尺度点面融合随机森林GPM机器学习

Keywords

remote sensingprecipitationdownscalingpoint and surface fusionrandom forestGPMmachine learning

references

Baez-Villanueva O M, Zambrano-bigiarini M, Beck H E, Mcnamara I, Ribbe L, Nauditt A, Birkel C, Verbist K, Giraldo-Osorio J D and Xuan Thinh N. 2020. RF-MEP: a novel random forest method for merging gridded precipitation products and ground-based measurements. Remote Sensing of Environment, 239: 111606 [DOI: 10.1016/j.rse.2019.111606http://dx.doi.org/10.1016/j.rse.2019.111606]

Brodeur Z P and Steinschneider S. 2020. Spatial bias in medium-range forecasts of heavy precipitation in the Sacramento River Basin: implications for water management. Journal of Hydrometeorology, 21(7): 1405-1423 [DOI: 10. 1175/JHM-D-19-0226.1http://dx.doi.org/10.1175/JHM-D-19-0226.1]

Chen C F and Li Y Y. 2019. A fast global interpolation method for digital terrain model generation from large LiDAR-derived data. Remote Sensing, 11(11): 1324 [DOI: 10.3390/rs11111324http://dx.doi.org/10.3390/rs11111324]

Fu Y, Xia J Z, Yuan W P, Xu B, Wu X X, Chen Y and Zhang H C. 2016. Assessment of multiple precipitation products over major river basins of China. Theoretical and Applied Climatology, 123(1/2): 11-22 [DOI: 10.1007/s00704-014-1339-0http://dx.doi.org/10.1007/s00704-014-1339-0]

Hu Q F, Yang D W, Li Z, Mishra A K, Wang Y T and Yang H B. 2014. Multi-scale evaluation of six high-resolution satellite monthly rainfall estimates over a humid region in China with dense rain gauges. International Journal of Remote Sensing, 35(4): 1272-1294 [DOI: 10.1080/01431161.2013.876118http://dx.doi.org/10.1080/01431161.2013.876118]

Hu S, Han J, Zhan C S and Liu L M Z. 2020. Spatial downscaling of remotely sensed precipitation in Taihang Mountains. Geographical Research, 39(7): 1680-1690

胡实, 韩建, 占车生, 刘梁美子. 2020. 太行山区遥感卫星反演降雨产品降尺度研究. 地理研究, 39(7): 1680-1690 [DOI: 10.11821/dlyj020190545http://dx.doi.org/10.11821/dlyj020190545]

Immerzeel W W, Rutten M M and Droogers P. 2009. Spatial downscaling of TRMM precipitation using vegetative response on the Iberian Peninsula. Remote Sensing of Environment, 113(2): 362-370 [DOI: 10.1016/j.rse.2008.10.004http://dx.doi.org/10.1016/j.rse.2008.10.004]

Islam M A, Yu B F and Cartwright N. 2020. Assessment and comparison of five satellite precipitation products in Australia. Journal of Hydrology, 590: 125474 [DOI: 10.1016/j.jhydrol.2020.125474http://dx.doi.org/10.1016/j.jhydrol.2020.125474]

Ji T, Liu R, Yang H, He T R and Wu J F. 2015. Spatial downscaling of precipitation using multi-source remote sensing data: a case study of sichuan-chongqing region. Journal of Geo-Information Science, 17(1): 108-117

嵇涛, 刘睿, 杨华, 何太蓉, 吴建峰. 2015. 多源遥感数据的降水空间降尺度研究——以川渝地区为例. 地球信息科学学报, 17(1): 108-117 [DOI: 10.3724/SP.J.1047.2015.00108http://dx.doi.org/10.3724/SP.J.1047.2015.00108]

Jia S F, Zhu W B, Lű A F and Yan T T. 2011. A statistical spatial downscaling algorithm of TRMM precipitation based on NDVI and DEM in the Qaidam Basin of China. Remote Sensing of Environment, 115(12): 3069-3079 [DOI: 10.1016/j.rse.2011.06.009http://dx.doi.org/10.1016/j.rse.2011.06.009]

Jiang S H, Wei L Y, Ren L L, Xu C Y, Zhong F, Wang M H, Zhang L Q, Yuan F and Liu Y. 2021. Utility of integrated IMERG precipitation and GLEAM potential evapotranspiration products for drought monitoring over mainland China. Atmospheric Research, 247: 105141 [DOI: 10.1016/j.atmosres.2020.105141http://dx.doi.org/10.1016/j.atmosres.2020.105141]

Jing W L, Yang Y P, Yue X F and Zhao X D. 2016. A spatial downscaling algorithm for satellite-based precipitation over the tibetan plateau based on NDVI, DEM, and land surface temperature. Remote Sensing, 8(8): 655 [DOI: 10.3390/rs8080655http://dx.doi.org/10.3390/rs8080655]

Jongjin B, Jongmin P, Dongryeol R and Minha C. 2016. Geospatial blending to improve spatial mapping of precipitation with high spatial resolution by merging satellite-based and ground-based data. Hydrological Processes, 30(16): 2789-2803 [DOI: 10.1002/hyp.10786http://dx.doi.org/10.1002/hyp.10786]

Karbalaye Ghorbanpour A, Hessels T, Moghim S and Afshar A. 2021. Comparison and assessment of spatial downscaling methods for enhancing the accuracy of satellite-based precipitation over Lake Urmia Basin. Journal of Hydrology, 596: 126055 [DOI: 10.1016/J.JHYDROL.2021.126055http://dx.doi.org/10.1016/J.JHYDROL.2021.126055]

Kim K H, Kim M S, Lee G W, Kang D H and Kwon B H. 2013. The adjustment of radar precipitation estimation based on the kriging method. Journal of the Korean Earth Science Society, 34(1): 13-27 [DOI: 10.5467/JKESS.2013.34.1.13http://dx.doi.org/10.5467/JKESS.2013.34.1.13]

Lai X and Gong Y F. 2017. Relationship between atmospheric heat source over the Tibetan Plateau and precipitation in the Sichuan-Chongqing region during summer. Journal of Meteorological Research, 31(3): 555-566 [DOI: 10.1007/s13351-017-6045-2http://dx.doi.org/10.1007/s13351-017-6045-2]

Li J and Zhang X. 2015. Downscaling method of TRMM satellite precipitation data. Scientia Geographica Sinica, 35(9): 1164-1169

李净, 张晓. 2015. TRMM降水数据的空间降尺度方法研究. 地理科学, 35(9): 1164-1169 [DOI: 10.13249/j.cnki.sgs.2015.09.1164http://dx.doi.org/10.13249/j.cnki.sgs.2015.09.1164]

Lu X Y, Tang G Q, Wang X Q, Liu Y, Jia L H, Xie G H, Li S and Zhang Y X. 2019. Correcting GPM IMERG precipitation data over the Tianshan Mountains in China. Journal of Hydrology, 575: 1239-1252 [DOI: 10.1016/j.jhydrol.2019.06.019http://dx.doi.org/10.1016/j.jhydrol.2019.06.019]

Lu X Y, Tang G Q, Wang X Q, Liu Y, Wei M and Zhang Y X. 2020. The development of a two-step merging and downscaling method for satellite precipitation products. Remote Sensing, 12(3): 398 [DOI: 10.3390/rs12030398http://dx.doi.org/10.3390/rs12030398]

Lu X Y, Wei M and Wang X Q. 2017. Correction of TRMM monthly precipitation data from 1998 to 2013 in Xinjiang. Journal of Applied Meteorological Science, 28(3): 379-384

卢新玉, 魏鸣, 王秀琴. 2017. TRMM月降水量产品在新疆地区的订. 应用气象学报, 28(3): 379-384 [DOI: 10.11898/1001-7313.20170311http://dx.doi.org/10.11898/1001-7313.20170311]

Ma J H, Qu C, Zhang H X and Xia Y Q. 2013. Spatial downscaling of TRMM precipitation data based on DEM in the upstream of Shiyang River Basin during 2001-2010. Progress in Geography, 32(9): 1423-1432

马金辉, 屈创, 张海筱, 夏燕秋. 2013. 2001-2010年石羊河流域上游TRMM降水资料的降尺度研究. 地理科学进展, 32(9): 1423-1432 [DOI: 10.11820/dlkxjz.2013.09.012http://dx.doi.org/10.11820/dlkxjz.2013.09.012]

Ma Z Q, He K, Tan X, Xu J T, Fang W Z, He Y and Hong Y. 2018. Comparisons of spatially downscaling TMPA and IMERG over the Tibetan Plateau. Remote Sensing, 10(12): 1883 [DOI: 10.3390/rs10121883http://dx.doi.org/10.3390/rs10121883]

Markonis Y, Papalexiou S M, Martinkova M and Hanel M. 2019. Assessment of water cycle intensification over land using a multisource global gridded precipitation DataSet. Journal of Geophysical Research: Atmospheres, 124(21): 11175-11187 [DOI: 10.1029/2019JD030855http://dx.doi.org/10.1029/2019JD030855]

Njuki S M, Mannaerts C M and Su Z B. 2020. An improved approach for downscaling coarse-resolution thermal data by minimizing the spatial averaging biases in random forest. Remote Sensing, 12(21): 3507 [DOI: 10.3390/rs12213507http://dx.doi.org/10.3390/rs12213507]

Pan Y, Shen Y, Yu J J and Zhao P. 2012. Analysis of the combined gauge-satellite hourly precipitation over China based on the OI technique. Acta Meteorologica Sinica, 70(6): 1381-1389

潘旸, 沈艳, 宇婧婧, 赵平. 2012. 基于最优插值方法分析的中国区域地面观测与卫星反演逐时降水融合试验. 气象学报, 70: 1381-1389 [DOI: 10.11676/qxxb2012.116http://dx.doi.org/10.11676/qxxb2012.116]

Renard B, Kavetski D, Leblois E, Thyer M, Kuczera G and Franks S W. 2011. Toward a reliable decomposition of predictive uncertainty in hydrological modeling: characterizing rainfall errors using conditional simulation. Water Resources Research, 47(11): W11516 [DOI: 10.1029/2011WR010643http://dx.doi.org/10.1029/2011WR010643]

Sekulić A, Kilibarda M, Heuvelink G B M, Nikolić M and Bajat B. 2020. Random forest spatial interpolation. Remote Sensing, 12(10): 1687 [DOI: 10.3390/rs12101687http://dx.doi.org/10.3390/rs12101687]

Sharifi E, Saghafian B and Steinacker R. 2019. Downscaling satellite precipitation estimates with multiple linear regression, artificial neural networks, and spline interpolation techniques. Journal of Geophysical Research: Atmospheres, 124(2): 789-805 [DOI: 10.1029/2018JD028795http://dx.doi.org/10.1029/2018JD028795]

Shen Y, Xiong A Y, Hong Y, Yu J J, Pan Y, Chen Z Q and Saharia M. 2014a. Uncertainty analysis of five satellite-based precipitation products and evaluation of three optimally merged multi-algorithm products over the Tibetan Plateau. International Journal of Remote Sensing, 35(19): 6843-6858 [DOI: 10.1080/01431161.2014.960612http://dx.doi.org/10.1080/01431161.2014.960612]

Shen Y, Zhao P, Pan Y and Yu J J. 2014b. A high spatiotemporal gauge-satellite merged precipitation analysis over China. Journal of Geophysical Research: Atmospheres, 119(6): 3063-3075 [DOI: 10.1002/2013jd020686http://dx.doi.org/10.1002/2013jd020686]

Shi W J, Yue T X, Shi X L and Song W. 2012. Research progress in soil property interpolators and their accuracy. Journal of Natural Resources, 27(1): 163-175

史文娇, 岳天祥, 石晓丽, 宋伟. 2012. 土壤连续属性空间插值方法及其精度的研究进展. 自然资源学报, 27(1): 163-175 [DOI: 10.11849/zrzyxb.2012.01.017http://dx.doi.org/10.11849/zrzyxb.2012.01.017]

Song F, Hu Q and Qian W H. 2004. Quality control of daily meteorological data in China, 1951-2000: a new dataset. International Journal of Climatology, 24(7): 853-870 [DOI: 10.1002/joc.1047http://dx.doi.org/10.1002/joc.1047]

Sun R C, Yuan H L, Liu X L and Jiang X M. 2016. Evaluation of the latest satellite-gauge precipitation products and their hydrologic applications over the Huaihe River Basin. Journal of Hydrology, 536: 302-319 [DOI: 10.1016/j.jhydrol.2016.02.054http://dx.doi.org/10.1016/j.jhydrol.2016.02.054]

Wang Y D, Nan Z T, Chen H and Wu X B. 2016. Correction of CMORPH daily precipitation data over the Qinghai-Tibetan Plateau with K-nearest neighbor model. Remote Sensing Technology and Application, 31(3): 607-616

王玉丹, 南卓铜, 陈浩, 吴小波. 2016. 基于K最近邻模型的青藏高原CMORPH日降水数据的订正研究. 遥感技术与应用, 31(3): 607-616 [DOI: 10.11873/j.issn.1004-0323.2016.3.0607http://dx.doi.org/10.11873/j.issn.1004-0323.2016.3.0607]

Wei L Y, Jiang S H, Ren L L, Zhang L Q, Wang M H and Duan Z. 2020. Preliminary utility of the retrospective IMERG precipitation product for large-scale drought monitoring over mainland China. Remote Sensing, 12(18): 2993 [DOI: 10.3390/rs12182993http://dx.doi.org/10.3390/rs12182993]

Xie P P and Xiong A Y. 2011. A conceptual model for constructing high-resolution gauge-satellite merged precipitation analyses. Journal of Geophysical Research: Atmospheres, 116(D21): D21106 [DOI: 10.1029/2011jd016118http://dx.doi.org/10.1029/2011jd016118]

Yang M X, Liu G D, Chen T, Chen Y and Xia C C. 2020. Evaluation of GPM IMERG precipitation products with the point rain gauge records over Sichuan, China. Atmospheric Research, 246: 105101 [DOI: 10.1016/j.atmosres.2020.105101http://dx.doi.org/10.1016/j.atmosres.2020.105101]

Zhang W, Liu D, Zheng S J, Liu S Y, Loáiciga H A and Li W K. 2020. Regional precipitation model based on geographically and temporally weighted regression kriging. Remote Sensing, 12(16): 2547 [DOI: 10.3390/rs12162547http://dx.doi.org/10.3390/rs12162547]

Zhou Z T, Guo B, Su Y Z, Chen Z S and Wang J. 2020a. Multidimensional evaluation of the TRMM 3B43V7 satellite-based precipitation product in mainland China from 1998-2016. PeerJ, 8: e8615 [DOI: 10.7717/peerj.8615http://dx.doi.org/10.7717/peerj.8615]

Zhou Z T, Guo B, Xing W X, Zhou J, Xu F L and Xu Y. 2020b. Comprehensive evaluation of latest GPM era IMERG and GSMaP precipitation products over mainland China. Atmospheric Research, 246: 105132 [DOI: 10.1016/j.atmosres.2020.105132http://dx.doi.org/10.1016/j.atmosres.2020.105132]

文章被引用时，请邮件提醒。

提交

融合特征优选与随机森林算法的GF-6影像东北一季稻遥感提取

人工智能时代的遥感变化检测技术：继承、发展与挑战

融合无人机和卫星影像的温带疏林草原木本和草本植物覆盖度遥感估算

融合Sentinel-2数据的高分五号高光谱数据降尺度