Characterizing and forecasting UPLB rainfall through neural networks approach.

Date

10-2009

Degree

Bachelor of Science in Applied Physics

College

College of Arts and Sciences (CAS)

Adviser/Committee Chair

Hannah Rissah E. Forio

Co-adviser

Sharon P. Lubag

Abstract

This study aims to characterize and forecast UPLB rainfall using neural networks. Meteorological data from 1959 to 200/1 which included the daily rainfall, daily mean temperature, daily relative humidity and daily sunshine duration from the Agrometeorology and Farm Structure Division, College of Engineering and Agro-Industrial Technology were considered in this study. The nonlinear function approximation shows that in UPLB rainy season occurs in the months of June to November and dry season follows throughout the rest of the year. Also, the input variable predictors of rainfall were found to he the daily mean temperature, daily relative humidity, daily sunshine duration and previous day rainfall amount. In 1982-2008, a relative increase in the mean temperature was observed. This change has raised the mean amount of rainfall by —0.6 millimeters over the past 27 years. Forecasting in this study evaluated the forecast values of neural networks architecture by a comparison over the past records. Moreover, forecasting the UPLB rainfall was done in three basis; the 366 days, days in month and monthly basis. Based on the results, forecasting the amount of rainfall in UPLB is more applicable during monthly basis.

Language

English

Location

UPLB Main Library Special Collections Section (USCS)

Call Number

Thesis

Document Type

Thesis

This document is currently not available here.

Share

COinS