Please use this identifier to cite or link to this item:
Authors: Jibril, Ibrahim
Mohammed, Jiya mamman
John, Musa Jiya
ADEOYE, Peter Aderemi
Keywords: ARIMA-model,
MLR-model and Trend
Climatic Elements,
Issue Date: 25-Feb-2016
Publisher: Proceedings of the Fourth International Conference on Engineering and Technology Research February 23 - 25, 2016 ISBN: 978-2902-58-6 Volume 4
Abstract: Extreme events of atmospheric phenomena are often non-deterministic in nature, and this has been a major constraint in achieving agricultural sustainability, which directly affects economic advancement, most especially of developing countries. This call for an urgent look at key climatic phenomena and finding simpler, but most reliable ways of predicting them in order to make proper plan against reoccurrence. To facilitate this research work, 29 years information of the observed relative humidity of Ogun State was obtained from the Federal Ministry of Water Resources, Abeokuta, Nigeria. The data collected covers the periods between 1982 and 2009 and were pre-whitened and aggregated into monthly and annual time series to clear the doubt of outliers. The Mann-Kendal non-parametric test, Long-range dependency test and test for serial dependence were carried out. The Mann-Kendal Z-value obtained was -1.37, which gives no reason to expect the presence of trend in the time series, but the Sen.’s slope trend line indicated slight decreasing trend. The spectral density analysis showed high variance to lower frequency, signifying a positive correlation which was in line with the Durbin-Watson test that gives a d-value of 1.28. No evidence of seasonal effect in the series as clearly depicted by the monthly Periodogram, and the data was therefore treated basically as stochastic. The data was divided into two and the first 20 years was used for model development, while the remaining 9 years was used for validation. Multiple Linear Regression (MLR) and Autoregressive Moving Average (ARMA) models were considered. The results indicated that there may be continues decrease in the amount of air moisture for a while. However, the best predictive model was found to be ARMA, though MLR give better validation. It is therefore recommended that other climatic parameters be looked into for proper planning.
Appears in Collections:Agric. and Bioresources Engineering

Files in This Item:
File Description SizeFormat 
LAUTECH 2016.pdf705.77 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.