Neural network based daily precipitation generator (NNGEN-P)

Daily weather generators are used in many applications and risk analyses. The present paper explores the potential of neural network architectures to design daily weather generator models. Focusing this first paper on precipitation, we design a collection of neural networks (multi-layer perceptrons...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Martínez, Fernando Luis, Penalba, Olga C., Segura, Enrique Carlos
Publicado: 2007
Materias:
artificial neural network
data set
empirical analysis
Markov chain
precipitation (climatology)
rainfall
simulation
Acceso en línea:https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_09307575_v28_n2-3_p307_Boulanger
http://hdl.handle.net/20.500.12110/paper_09307575_v28_n2-3_p307_Boulanger
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id paper:paper_09307575_v28_n2-3_p307_Boulanger
record_format dspace
spelling paper:paper_09307575_v28_n2-3_p307_Boulanger2023-06-08T15:52:37Z Neural network based daily precipitation generator (NNGEN-P) Martínez, Fernando Luis Penalba, Olga C. Segura, Enrique Carlos artificial neural network data set empirical analysis Markov chain precipitation (climatology) rainfall simulation Daily weather generators are used in many applications and risk analyses. The present paper explores the potential of neural network architectures to design daily weather generator models. Focusing this first paper on precipitation, we design a collection of neural networks (multi-layer perceptrons in the present case), which are trained so as to approximate the empirical cumulative distribution (CDF) function for the occurrence of wet and dry spells and for the precipitation amounts. This approach contributes to correct some of the biases of the usual two-step weather generator models. As compared to a rainfall occurrence Markov model, NNGEN-P represents fairly well the mean and standard deviation of the number of wet days per month, and it significantly improves the simulation of the longest dry and wet periods. Then, we compared NNGEN-P to three parametric distribution functions usually applied to fit rainfall cumulative distribution functions (Gamma, Weibull and double-exponential). A data set of 19 Argentine stations was used. Also, data corresponding to stations in the United States, in Europe and in the Tropics were included to confirm the results. One of the advantages of NNGEN-P is that it is non-parametric. Unlike other parametric function, which adapt to certain types of climate regimes, NNGEN-P is fully adaptive to the observed cumulative distribution functions, which, on some occasions, may present complex shapes. On-going works will soon produce an extended version of NNGEN to temperature and radiation. © Springer-Verlag 2006. Fil:Martinez, F. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. Fil:Penalba, O. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. Fil:Segura, E.C. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. 2007 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_09307575_v28_n2-3_p307_Boulanger http://hdl.handle.net/20.500.12110/paper_09307575_v28_n2-3_p307_Boulanger
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic artificial neural network
data set
empirical analysis
Markov chain
precipitation (climatology)
rainfall
simulation
spellingShingle artificial neural network
data set
empirical analysis
Markov chain
precipitation (climatology)
rainfall
simulation
Martínez, Fernando Luis
Penalba, Olga C.
Segura, Enrique Carlos
Neural network based daily precipitation generator (NNGEN-P)
topic_facet artificial neural network
data set
empirical analysis
Markov chain
precipitation (climatology)
rainfall
simulation
description Daily weather generators are used in many applications and risk analyses. The present paper explores the potential of neural network architectures to design daily weather generator models. Focusing this first paper on precipitation, we design a collection of neural networks (multi-layer perceptrons in the present case), which are trained so as to approximate the empirical cumulative distribution (CDF) function for the occurrence of wet and dry spells and for the precipitation amounts. This approach contributes to correct some of the biases of the usual two-step weather generator models. As compared to a rainfall occurrence Markov model, NNGEN-P represents fairly well the mean and standard deviation of the number of wet days per month, and it significantly improves the simulation of the longest dry and wet periods. Then, we compared NNGEN-P to three parametric distribution functions usually applied to fit rainfall cumulative distribution functions (Gamma, Weibull and double-exponential). A data set of 19 Argentine stations was used. Also, data corresponding to stations in the United States, in Europe and in the Tropics were included to confirm the results. One of the advantages of NNGEN-P is that it is non-parametric. Unlike other parametric function, which adapt to certain types of climate regimes, NNGEN-P is fully adaptive to the observed cumulative distribution functions, which, on some occasions, may present complex shapes. On-going works will soon produce an extended version of NNGEN to temperature and radiation. © Springer-Verlag 2006.
author Martínez, Fernando Luis
Penalba, Olga C.
Segura, Enrique Carlos
author_facet Martínez, Fernando Luis
Penalba, Olga C.
Segura, Enrique Carlos
author_sort Martínez, Fernando Luis
title Neural network based daily precipitation generator (NNGEN-P)
title_short Neural network based daily precipitation generator (NNGEN-P)
title_full Neural network based daily precipitation generator (NNGEN-P)
title_fullStr Neural network based daily precipitation generator (NNGEN-P)
title_full_unstemmed Neural network based daily precipitation generator (NNGEN-P)
title_sort neural network based daily precipitation generator (nngen-p)
publishDate 2007
url https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_09307575_v28_n2-3_p307_Boulanger
http://hdl.handle.net/20.500.12110/paper_09307575_v28_n2-3_p307_Boulanger
work_keys_str_mv AT martinezfernandoluis neuralnetworkbaseddailyprecipitationgeneratornngenp
AT penalbaolgac neuralnetworkbaseddailyprecipitationgeneratornngenp
AT seguraenriquecarlos neuralnetworkbaseddailyprecipitationgeneratornngenp
_version_ 1768544739172286464

Ejemplares similares