An improved three-layer machine learning searching algorithm for regional frequency analysis of extreme rainfall events
YANG Zhe1,2,3, YANG Kun2,3, LYU Juan2,3, ZUO Huiqiang4, YIN Jianming5
1. Post-doc workstation, China Reinsurance(Group) Co., Ltd., Beijing 100033; 2. China Institute of Water resources and Hydropower Research, Beijing 100038; 3. Research Center on Flood and Drought Disaster Reduction of the Ministry of Water Resources, Beijing 100038; 4. China Property & Casualty Reinsurance Company LTD, Beijing 100033; 5. China Re CRM, Chongqing 400020
Abstract:According to the physical mechanism of extreme rainfall formation, combined with machine learning techniques such as isometric feature mapping and influence region, an improved algorithm is proposed based on the regional frequency analysis method of three-layer searching method, which reduces the uncertainty of the quantile value of the rainfall intensity-duration-frequency (IDF) model. According to the climate and terrain characteristics of the target site, the existing three-layer searching method uses the feature selection method to select the most representative geographical or meteorological elements as the similarity factors to form a homogeneous population, and then obtain the rainfall intensity quantile value with low uncertainty. Considering the nonlinear correlation between feature elements, feature extraction method and supervised clustering method are added on the basis of the existing three-layer searching method to reduce the influence of the nonlinear correlation between feature elements on homogeneous group clustering and its dependence on the input site data. The method was tested at rainfall stations in British Columbia, Canada, and the test results show that the method can further reduce the uncertainty of IDF estimation results.
[1] Stott P A.Detection and attribution of climate change:A regional perspective[J].Wiley Interdisciplinary Reviews-Climate Change, 2010, 1(2):192-211. [2] Trenberth K E.Attribution of climate variations and trends to human influences and natural variability[J].Wiley Interdisciplinary Reviews-Climate Change, 2011, 2(6):925-930. [3] Mondal A,Mujumdar P P. Modeling non-stationarity in intensity, duration and frequency of extreme rainfall over India[J].Journal of Hydrology, 2015, 521:217-231. [4] Simonovic S P,Schardong A,Sandink D. Mapping extreme rainfall statistics for canada under climate change using updated intensity-duration-frequency curves[J].Journal of Water Resources Planning and Management, 2017, 143(3):1-12. [5] Fernandes L G,Rodrigues R R.Changes in the patterns of extreme rainfall events in southern Brazil[J].International Journal of Climatology, 2017, 1352(8):1337-1352. [6] Ma S,Zhou T,Dai A,etal.Observed changes in the distributions of daily precipitation frequency and amount over China from 1960 to 2013[J].Journal of Climatology, 2015, 28(17):6960-6978. [7] Hosking J R M, Wallis J R. Regional frequency analysis:an approach based on l-moments[M].New York:Cambridge University Press, 1997. [8] Easterling D R. Regionalization of thunderstorm rainfall in the contiguous United States[J].International Journal of Climatology, 1989, 9(6):567-579. [9] Gaál L,Kysel J. Comparison of region-of-influence methods for estimating high quantiles of precipitation in a dense dataset in the Czech Republic[J].Hydrology and Earth System Sciences, 2009, 13(11):2203-2219. [10] Yang T. Regional frequency analysis and spatio-temporal pattern characterization of rainfall extremes in the Pearl River Basin, China[J].Journal of Hydrology, 2010, 380(3/4):386-405. [11] Satyanarayana P, Srinivas V V. Regional frequency analysis of precipitation using large-scale atmospheric variables[J].Journal of Geophysical Research, 2008, 113(D24):1-16. [12] Ahmad N H,Othman I R,DeniS M. Hierarchical Cluster Approach for Regionalization of Peninsular Malaysia based on the Precipitation Amount[J]. Journal of physics.Conference series, 2013, 423:012018. [13] Gabriele S, Chiaravalloti F.Searching regional rainfall homogeneity using atmospheric fields[J].Advances in Water Resources, 2013, 53:163-174. [14] Burn D H. A framework for regional estimation of intensity-duration-frequency(IDF)curves[J].Hydrological Processes, 2014, 28(14):4209-4218. [15] Asong Z E,Khaliq M N,Wheater H S. Regionalization of precipitation characteristics in the Canadian Prairie Provinces using large-scale atmospheric covariates and geophysical attributes[J]. Stochastic Environmental Research and Risk Assessment, 2015, 29(3):875-892. [16] Haddad K,Johnson F,Rahman A, et al.Comparing three methods to form regions for design rainfall statistics:Two case studies in Australia[J].Journal of Hydrology, 2015, 527:62-76. [17] Yang Z,Burn D H. Automatic feature selection and weighting for the formation of homogeneous groups for regional IDF estimation[J]. Journal of Hydrology, 2019, 575(5):292-307. [18] Collier C G. The impact of urban areas on weather[J].Quarterly Journal of the Royal Meteorological Society, 2006, 132(614):1-25. [19] Shepherd J M. A review of current investigations of urban-induced rainfall and recommendations for the future[J].Earth Interactions, 2005, 9(12):1-27. [20] Houze R A. Observed structure of mesoscale convective systems and implications for large-scale heating[J].Quarterly Journal of the Royal Meteorological Society, 1989, 115(487):425-461. [21] Houze R A.Stratiform precipitation in the tropics:A meteorological paradox?[J].American Ceramic Society Bulletin, 1997, 78(10):2179-2196. [22] Li D,Bou-Zeid E,Baeck M L, etal.Modeling Land Surface Processes and Heavy Rainfall in Urban Environments:Sensitivity to Urban Surface Representations[J].Journal of Hydrometeorology,2013, 14(4):1098-1118. [23] Tenenbaum J B, De Silva V, Langford J C. A Global Geometric Framework for Nonlinear Dimensionality Reduction[J].Science, 2000, 290(5500):2319-2323. [24] Wu Y,Chan K L,Wang L.Face recognition based on discriminative manifold learning[J].Pattern Recognition, 2004, 4:171-174. [25] Burn D H. Evaluation of regional flood frequency analysis with a region of influence approach[J].Water Resources Research, 1990, 26(10):2257-2265. [26] Shu C,Burn D H. Homogeneous pooling group delineation for flood frequency analysis using a fuzzy expert system with genetic enhancement[J]. Journal of Hydrology, 2004, 291(1/2):132-149. [27] Castellarin A,Burn D H, Brath A. Homogeneity testing:How homogeneous do heterogeneous cross-correlated regions seem?[J]. Journal of Hydrology, 2008, 360(1-4):67-76. [28] Murugavel P,Pawar S D,Gopalakrishnan V. Trends of convective available potential energy over the Indian region and its effect on rainfall[J].International Journal of Climatology,2012, 32(9):1362-1372. [29] Meukaleuni C, Lenouo A, Monkam D. Climatology of convective available potential energy(CAPE)in ERA-Interim reanalysis over West Africa[J].Atmospheric Science Letters, 2016, 17(1):65-70. [30] Mostofi Zadeh S, Durocher M, Burn D H, etal. Pooled flood frequency analysis:a comparison based on peaks-over-threshold and annual maximum series[J].Hydrological Sciences Journal, 2019, 64(2):121-136.