Comparison of annual rainfall forecasting in Kerman province using artificial neural network and multiple linear regression models

Rezaei, M. (2013) Comparison of annual rainfall forecasting in Kerman province using artificial neural network and multiple linear regression models. Masters thesis, University of Zabol.

[img]
Preview
Text
Comparison of annual rainfall forecasting in Kerman province using artificial neural network and multiple linear regression models.pdf

Download (19kB) | Preview

Abstract

Today the world is being faced by the Climate-change Phenomenon as a serious problem. It seems that the ecosystems are vulnerable to these changes specifically.Hence, studying of long-term climate change trends are important in climatic investigating and future forecasting. Precipitation phenomenon is a function of many factors that it's forecast with usual statistical methods has low accuracy. There is several techniques to simulation of future climate periods that usage of General Circulation model data (GCM) is the most reliable of them.GCM Models are able to simulate the detain large surfaces,alone. Therefore, Down scaling GCM data in stationary level by means of different techniques is necessary for using of this data. In this study, by Using of HadCM3 model output sunder A2 scenario,SDSM down scaling models and artificial neural network, the precipitation has been predicted for three periods: (2010-2039), (2040-2069) and (2070-2099).Then, the results of the two models were evaluated and compared according to statistical criteria. Results indicate that the artificial neural network model has higher performance in most stations and annual rainfall of the Kerman province is being declined until 2100

Item Type: Thesis (Masters)
Uncontrolled Keywords: Annual rainfall forecasting,Artificial neural network, Multiple linear regression, Kerman province
Subjects: Q Science > QE Geology
T Technology > T Technology (General)
Depositing User: admin admin1 admin2
Date Deposited: 27 Jan 2016 07:54
Last Modified: 27 Jan 2016 07:54
URI: http://eprints.uoz.ac.ir/id/eprint/32

Actions (login required)

View Item View Item