| 65 | 0 | 150 |
| 下载次数 | 被引频次 | 阅读次数 |
【目的】降水作为水文循环的关键环节，受复杂地形条件及大气—海洋环流共同作用，对其进行准确表征具有较大挑战。为提高福建地区台风降水估计的精度与空间适应性，本文以福建省为研究区域，【方法】提出了Stacking-BTCH的融合框架，综合利用CMFD、ERA5-Land、GLDAS、IMERG四种降水产品及地面站点观测，在偏差校正基础上实现多源降水信息的加权融合，生成了一套高精度的融合降水数据集。【结果】基于16个独立站点的检验结果表明：在原始降水产品中CMFD综合表现最佳，相关系数为0.54，均方根误差为33.42mm/day；GLDAS对强降水刻画能力相对较弱，ERA5-Land与IMERG整体表现居中但存在区域性偏差。在综合各数据源优势后，融合产品的精度显著优于各原始及偏差校正后的单一降水产品，相关系数提升至0.72，均方根误差降至23.51mm/day。从各季节表现来看，融合降水产品在春、夏、秋的探测率均达1，误报率则呈现季节性变化，其中夏季表现最佳，春季相对偏高，约为0.2。【结论】结果表明，基于Stacking-BTCH框架构建的融合降水产品在不同气候背景下具有较为稳定和可靠的降水事件识别能力，整体性能优于单一降水产品。该融合方法在一定程度上缓解了站网稀疏及多源降水数据不确定性带来的影响，可为台风降水过程监测、致灾风险评估及业务预报提供数据支撑，有助于提升区域气候监测与灾害评估的精细化水平。
Abstract：[Objective] Precipitation， as a crucial component of the hydrological cycle， is jointly influenced by complex topographic conditions and atmospheric-ocean circulation， making its accurate characterization highly challenging. To improve the accuracy and spatial adaptability of typhoon precipitation estimation over Fujian， Fujian Province is selected as the study area. [Methods] A Stacking-BTCH merging framework was proposed， integrating four precipitation products—CMFD， ERA5-Land， GLDAS， and IMERG—together with rain gauge observations. On the basis of bias correction， the framework performed weighted merging of multi-source precipitation information to generate a high-accuracy merged precipitation dataset. [Results] Validation using 16 independent stations showed that among the original precipitation products， CMFD exhibited the best overall performance， with a correlation coefficient of 0.54 and an RMSE of 33.42 mm/day. GLDAS performed relatively poorly in capturing heavy rainfall， while ERA5-Land and IMERG showed moderate performance but with regional biases. After integrating the advantages of various data sources， the merged product demonstrated markedly higher accuracy than the original and bias-corrected individual precipitation products， with the correlation coefficient increasing to 0.72 and RMSE decreasing to 23.51 mm/day. From a seasonal perspective， the merged precipitation product achieved a probability of detection (POD) of 1.0 across spring， summer， and autumn. The false alarm ratio (FAR) exhibited seasonal variability， with the best performance observed in summer and relatively higher values in spring (approximately 0.2). [Conclusion] The results indicate that the merged precipitation product developed under the Stacking-BTCH framework exhibits stable and reliable precipitation event identification capability across different climatic conditions， with overall performance superior to that of individual precipitation products. This merging approach partially mitigates the impacts of sparse rain-gauge networks and uncertainties in multi-source precipitation datasets， and provides effective data support for typhoon-induced precipitation process monitoring， disaster-risk assessment， and operational forecasting， thereby contributing to improved refinement of regional climate monitoring and disaster assessment.
[1] GUNATHILAKE M B， AMARATUNGA Y V， PERERA A， et al. Statistical evaluation and hydrologic simulation capacity of different satellite-based precipitation products (SbPPs) in the Upper Nan River Basin， Northern Thailand[J]. Journal of Hydrology: Regional Studies， 2020， 32: 100743.
[2] FANG G S， ZHAO L， CAO S Y， et al. Estimation of tropical cyclone wind hazards in coastal regions of China[J]. Natural Hazards and Earth System Sciences， 2020， 20(6): 1617-1637.
[3] LIU W， LYU H S， ZHU Y H， et al. The performance of IMERG near-real-time estimations during the record-breaking Meiyu season in 2020[J]. Journal of Hydrology， 2024， 634: 131024.
[4] MA Z Q， XU J T， DONG B， et al. GMCP: A Fully Global Multisource Merging-and-Calibration Precipitation Dataset (1-Hourly， 0.1°， Global， 2000–the Present)[J]. Bulletin of the American Meteorological Society， 2025， 106(4): E596-E624.
[5] DOLLAN I J， MAINA F Z， KUMAR S V， et al. An assessment of gridded precipitation products over High Mountain Asia[J]. Journal of Hydrology: Regional Studies， 2024， 52: 101675.
[6] YANG L， SHI Z G， LIU R， et al. Evaluating the performance of global precipitation products for precipitation and extreme precipitation in arid and semiarid China[J]. International Journal of Applied Earth Observation and Geoinformation， 2024， 130: 103888.
[7] 杨鑫， 张建云， 周建中， 等. 基于ConvLSTM网络的多源降雨融合方法[J]. 华中科技大学学报(自然科学版)， 2022， 50(8): 33-39.
[8] 张良培， 何江， 杨倩倩， 等. 数据驱动的多源遥感信息融合研究进展[J]. 测绘学报， 2022， 51(7): 1317-1337.
[9] BHARTI V， SINGH C. Evaluation of error in TRMM 3B42V7 precipitation estimates over the Himalayan region[J]. Journal of Geophysical Research: Atmospheres， 2015， 120(24): 12458-12473.
[10] SINCLAIR S， PEGRAM G. Combining radar and rain gauge rainfall estimates using conditional merging[J]. Atmospheric Science Letters， 2005， 6(1): 19-22.
[11] LI M， SHAO Q X. An improved statistical approach to merge satellite rainfall estimates and raingauge data[J]. Journal of Hydrology， 2010， 385(1/2/3/4): 51-64.
[12] MANZ B， BUYTAERT W， ZULKAFLI Z， et al. High-resolution satellite-gauge merged precipitation climatologies of the Tropical Andes[J]. Journal of Geophysical Research: Atmospheres， 2016， 121(3): 1190-1207.
[13] CHAO L J， ZHANG K， LI Z J， et al. Geographically weighted regression based methods for merging satellite and gauge precipitation[J]. Journal of Hydrology， 2018， 558: 275-289.
[14] WEI L Y， JIANG S H， DONG J Z， et al. A combined extended triple collocation and cumulative distribution function merging framework for improved daily precipitation estimates over China’s mainland[J]. Journal of Hydrology， 2024， 641: 131757.
[15] 尹家波， 郭生练， 王俊， 等. 基于贝叶斯模式平均方法融合多源数据的水文模拟研究[J]. 水利学报， 2020， 51(11): 1335-1346.
[16] ZHAO N. A method for merging multi-source daily satellite precipitation datasets and gauge observations over Poyang Lake Basin， China[J]. Remote Sensing， 2023， 15(9): 2407.
[17] 王筱译， 吕海深， 朱永华， 等. 多源数据融合对IMERG降水产品的改进[J]. 中国农村水利水电， 2018(9): 25-29.
[18] BAEZ-VILLANUEVA O M， ZAMBRANO-BIGIARINI M， BECK H E， et al. RF-MEP: A novel Random Forest method for merging gridded precipitation products and ground-based measurements[J]. Remote Sensing of Environment， 2020， 239: 111606.
[19] ZAMAN B， MCKEE M， NEALE C M U. Fusion of remotely sensed data for soil moisture estimation using relevance vector and support vector machines[J]. International Journal of Remote Sensing， 2012， 33(20): 6516-6552.
[20] TIAN B R， CHEN H， YAN X， et al. A downscaling–merging scheme for monthly precipitation estimation with high resolution based on CBAM-ConvLSTM[J]. Remote Sensing， 2023， 15(18): 4601.
[21] ZHANG T T， LIANG Z M， WANG H J， et al. Merging multisatellite precipitation products using stacking method and the censored-shifted gamma ensemble model output statistics in China’s Beimiaoji basin[J]. Journal of Hydrology， 2023， 618: 129263.
[22] MENG C C， MO X G， LIU S X， et al. Improving near-real-time satellite precipitation products through multistage modified schemes[J]. Atmospheric Research， 2023， 292: 106875.
[23] 孟锐， 陈云瑶， 李彬权， 等. 基于机器学习和贝叶斯模型平均的多源降水融合方法[J]. 水利水电技术(中英文)， 2026， 57(3): 167-178.
[24] 黄思淇， 林木生. 闽江流域雨量站网优化研究[J]. 水文， 2025， 45(3): 72-77.
[25] 何杰， 阳坤， 李新， 等. 中国区域地面气象要素驱动数据集v2.0（1951-2020）[DS]. 2024.
[26] MU?OZ-SABATER J， DUTRA E， AGUSTí-PANAREDA A， et al. ERA5-Land: A state-of-the-art global reanalysis dataset for land applications[J]. Earth System Science Data， 2021， 13(9): 4349-4383.
[27] CHEN F， MITCHELL K， SCHAAKE J， et al. Modeling of land surface evaporation by four schemes and comparison with FIFE observations[J]. Journal of Geophysical Research: Atmospheres， 1996， 101(D3): 7251-7268.
[28] HOU A Y， KAKAR R K， NEECK S， et al. The global precipitation measurement mission[J]. Bulletin of the American Meteorological Society， 2014， 95(5): 701-722.
[29] YUAN F， ZHANG L M， WIN K， et al. Assessment of GPM and TRMM multi-satellite precipitation products in streamflow simulations in a data-sparse mountainous watershed in Myanmar[J]. Remote Sensing， 2017， 9(3): 302.
[30] 李漠雨， 董少睿， 郭英香. 青藏高原东部三种降水数据产品的适用性评估[J]. 干旱区研究， 2025， 42(7): 1173-1183.
[31] LIU W， LYU H S， ZHU Y H， et al. The explanation of small- and medium-watershed-scale bias variability in IMERG in Chinese humid regions [J]. Atmospheric Research， 2025， 320: 108075.
[32] SU J B， LYU H S， CROW W T， et al. The effect of spatiotemporal resolution degradation on the accuracy of IMERG products over the Huai River Basin[J]. Journal of Hydrometeorology， 2020， 21(5): 1073-1088.
[33] BREIMAN L. Stacked regressions[J]. Machine Learning， 1996， 24(1): 49-64.
[34] HE X L， XU T R， XIA Y L， et al. A Bayesian three-cornered hat (BTCH) method: Improving the terrestrial evapotranspiration estimation[J]. Remote Sensing， 2020， 12(5): 878.
[35] 赵君， 刘雨， 徐进超， 等. 基于贝叶斯三角帽法的多源降水数据融合分析及应用[J]. 水科学进展， 2023， 34(5): 685-696.
[36] 吴静雯， 颜悠逸， 殷方旭， 等. 台风降水的WRF模拟: 以福建省杜苏芮台风为例[J]. 水利水电技术(中英文)， 2025， 56(11): 1-20.
基本信息:
中图分类号:TP181;P339;P426.6
引用信息:
[1]张晗,吴伟浩,黎国杭,等.基于机器学习与贝叶斯三角帽法的多源台风降水融合与监测能力评估[J].水利水电技术(中英文)().
基金信息:
国家自然科学基金(42471019); 福建省水利科技项目(FISOZB-20250025)
2026-07-15
2026-07-15
2026-07-15