Assessment of the impacts of Climate change, migration, food security and Vulnerability in the rural Gambia

Abstract

This thesis concentrated on three essential objectives. The first addressed the time series econometrics part on climate change, food availability, and other variables in The Gambia. The second objective analyses the impacts of food security on migration response by looking at the climate change interaction. Objective three examines farmers' vulnerability to change in climate at households and regions. Four main approaches are adopted to address these central themes. The first tactic is time series analysis using VAR, FEVD, Impulse response function, granger causality tests, ARDL, and ECM methodologies. It is based on estimating the unit root test using the ADF, PP, and KPSS tests. After unit root tests, we can select which time series method is more appropriate and why.The forest errors variance decomposition (FEVD) and impulse response function (IRF) follow after the VAR models. So after, we also determine the F- bounds statistics values to see whether co-integration exists among the variables and whether we should select ARDL in the short run or ARDL in the long run. Therefore, appropriate model is ADRL long run, which contains the error correction parts. In the second approach, we used the multilevel version of the conventional logistic regression to predict the probability of odds of moving to an international destination from the rural areas of The Gambia, where households i is located. In estimating this model, we started with the based models, climate change only model, food security only model, and climate change and food security interaction model to find the probability of the international move from the rural Gambia. The third approach is divided into two approaches. In the first approach of the third, we used one of the econometrics approaches to measure vulnerability to poverty called vulnerability to expected poverty approaches. This tactic is based on valuing the probability of log consumption below or equal to the poverty line, given that climate shocks or other household characteristic shocks affect household consumption levels (See details on the methodology of objective 3).The method is used to assess the household level's vulnerability using the single cross-sectional data we collected. We used dependent variables as a change in total consumption expenditure. In the second approach of the third, we used the integrated approaches to understand vulnerability, and the Principal Components Analysis was the best approach for this kind of study. We employed the PCA to create vulnerability indices for each vulnerability component and conduct the analysis across the rural regions in The Gambia. We attached the positive values to adaptive capacity and negative values for exposure and sensitivity indicators of vulnerability as adopted by IPCC, 2001, 2007 and after we validated the vulnerability index by using the different complete variables. The analysis results for objective one stated that: Climate change affects agriculture negatively and, in turn, harms food availability. There is a positive correlation between food security and agriculture value added. The results also found that growth increases with the growth rate of agriculture and food availability. Finally, in theme one, as population growth increases annually, food availability decreases in The Gambia. For migration responses, we found that the food security status of the households slightly increases migration to an international move. Food insecurity households do not migrate. The results further reveal that floods and temperature changes do not cause migration, but heavier rainfall, changes in rainfall, and drought lead to migration. We also found that remittance received households are more likely to migrate than non-remittance received households. Income, employment as farmers, stop at secondary education, and illiterate, the probability of migration in these categories is much higher according to our findings. Finally, with food security consumption level, the likelihoods of migration become weaker. Lastly, the final chapter results are summarized: secondary education increases vulnerability to poverty and employment as farmers increase vulnerability to poverty. The empirical results finding stated that floods, changes in rainfall, and drought increase the probability of future vulnerability, looking at the changes in total consumption expenditure as a proxy variable for vulnerability. In addition, indicators-based approaches revealed that rural regions have different vulnerability levels to climate change. All the rural regions were more vulnerable to climate change due to high exposure and sensitivity and low adaptive capacity. Moreover, the thesis finally reveals that according to PCA's vulnerability components and vulnerability indices, farmers in these regions are highly vulnerable from socioeconomic and biophysical attributes to climate change. Furthermore, the outcomes of validation indicated that NGO support reduces farmer’s vulnerability to change in climate in the rural Gambia by 82 percent, whereas government support increases farmer’s vulnerability to climate change by 79 percent.