Predicting methane emissions from livestock in Malaysia using the ARIMA model

Abstract

Purpose – The purpose of this paper is to compute the amount of methane generated from the waste of livestock from 1980 to 2008; then use the information in forecasting subsequent methane emissions by the sector in Malaysia from 2009 to 2020. Design/methodology/approach – The research study employed two approaches; computing methane emissions from 1980 to 2008 using the IPCC guidelines, and forecasting methane emissions for the animals from 2009 to 2020 using the autoregressive integrated moving average (ARIMA) model from the predictive analysis software (PASW-SPSS 18.0). Findings – Methane emissions from cattle, buffaloes and pigs accounted for over 95 per cent of total emissions and emissions from cattle are predicted to increase from 67.0 Gg in 2009 to 77.0 Gg by 2020. Emissions from the others will not be appreciable although poultry emissions will rise to 11.0 Gg by 2020. Attempt by the Malaysian Government to increase cattle production is not necessary at the moment as protein requirement has been met. Research limitations/implications – ARIMA model suffers from linear and data limitation: the future value of a variable assumed to be a linear function of several past observations in ARIMA is sometimes unrealistic. Large amounts of historical data are needed in ARIMA models in order to get desired results. The inventory of the animals was taken from 1980 to 2008.This needs to be improved upon by updating it to cover up to 2011 so that the forecast will start from 2012. Practical implications – The chosen ARIMA method has demonstrated its correctness in being adequate as a predicting tool for animal methane emissions. Policy makers can apply it so as to take practical steps to avoid these emissions. Originality/value – This is a novice idea as animal methane emission forecasting tool. This model will be of immense use and help in predicting methane emissions from livestock.