Marginal Impact of climate variability on crop yields in Ghana

Abstract

The impacts of climate variability indicators on crop yields have attracted global attention over the years. This is particularly so because of the growing concern about the effect of climate change on sustainable food production, especially in extreme climate-risk-prone regions in sub-Saharan Africa. The aim of the present paper is to study the marginal effect of climate variables on maize yields in northern Ghana. In this paper, we apply the random forest machine learning approach, Mann-Kendall test, Sen's slope, using data from four study locations in the Upper East region of Ghana to analyze the phenomena. It was evident that maximum and minimum temperatures are the most important climate variability indicators, which have the greatest impact on maize yields. Maize yield in the Upper East region tends to decrease when temperatures exceed 32°C and increase when temperatures are around 24°C. Our results generally show that in all the studied districts, maize yield tends to be at its maximum when the temperature is around 24°C. These findings are crucial to the efforts being made in addressing the emerging challenges confronting smallholder farmers in sustainable food production due to persistent climate change as well as contributing to the achievement of Sustainable Development Goals 1, 2 and 13.

Keywords

• Sub-Saharan Africa
• Climate change
• Machine learning
• Maize yields
• Sustainable food production
• Precision agriculture