Scholarly record

ANALYSIS OF WORLD GREENHOUSE GAS EMISSIONS CAUSED BY AGRICULTURE

O. V. Mirolyubova, N. Didenko, D. Skripnuk, K. Kikkas, E. A. Samylovskaya

Abstract

Greenhouse gas emissions produced by agriculture, animal husbandry, fishery and forestry activities continue to increase, although not as fast as emissions produced by other human activities. The paper analyzes the impact of agricultural, livestock, fishery and forestry human activities on greenhouse gas emissions. The goal is to build a model of greenhouse gases being changed due to human activities’ factors. As a theoretical model the authors have chosen ADL-model (auto-regressive-distributed-lags-model), in which the current values of the series depend both on past values of the series and on current and past values of other time series. In the model, the endogenous variable is the emissions of greenhouse gases caused by agriculture (tons of CO2 equivalent) in a t time period (y?t). The exogenous variables are greenhouse gas emissions caused by agriculture in the previous periods (y?t-j ) and economic indicators, evaluating the sphere associated with agriculture, livestock, fishing and forestry activities. They are as follows: World: Machinery for Food, Beverage and Tobacco Processing: Production (turnover) MSP, USD million ( X?1t ); World: Machinery, Agricultural and Forestry Machinery: Production (turnover) MSP, USD million (X?2t ); Animal Husbandry, Industrial: Primary Materials, USD million (X?3t); Arable Land, sq km ( X?4t ); Animal waste - Production, Terajoules ( X?5t ); Waste Generated by Agriculture, Forestry and Fishing, tons ( X?6t ). Six exogenous variables (...) have been selected by means of qualitative and statistical methods of analysis. 38 economic indicators, evaluating primary, secondary and tertiary sectors of the global economy were analyzed. Indicator “greenhouse gas emissions produced by agriculture in previous time periods” ( X?t-j) shows the effect on the current evaluation of the process of greenhouse gases emission due to agricultural activities. Primary statistical information on the endogenous and exogenous variables has been collected from the countries over the period of 1998 - 2015. Endogenous and exogenous variables for the global economy on the basis of primary country statistical information were defined as the geometric average of the countries’ indices in year t. The paper presents the methodology of the empirical validation of the model. The method includes the following main stages: inspection of the time series of the variables for their stationarity using Dickey - Fuller test; the choice of the endogenous variable lags, which have a strong correlation with the variable value in the last period; check the tightness of the endogenous variable’s connection with the exogenous variables, exogenous variables’ test of multicollinearity; testing the significance of autocorrelation coefficients using the Ljung-Box Q-test; verification of the pair correlation coefficients for their significance using Student's t-test; determination of model coefficients using regression analysis; test of the regression equation and the regression equation coefficients’ significance. In conclusion, the article provides some ideas about the influence of human activities on the change of greenhouse gas emissions produced by agriculture, animal husbandry, fishery and forestry activities and defines the perspective of further research.

Publication details

Back to publication list