Modelling the Potential Impacts of Climate Change and Variability on Electricity Demand in Republic of Niger

Abstract

This study investigates the potential impact of climate change and variability on electricity demand under different Global Warming Levels (GWL1.5, GWL2.0, GWL2.5, and GWL3.0). First, to assess the sensitivity of electricity demand to climate variables, the Wavelet Transform Coherence (WTC) as well as Principal Component Analysis (PCA) were used. Secondly, to establish the relationship between electricity demand and climate variables, Multiple Linear Regression (MLR) and Artificial Neural Network (ANN) models have been used. Prior to the model development, the electricity demand data was de-trended to isolate only the influence of climate variables. Thirdly, to project the impact of climate change at specific GWL, the climate data from the reference period (1971-2000) was subtracted from that of GWL period. Results show that the electricity demand (DED) in Niger is positively correlated to Temperatures (Tmean, Tmax, Tmin), Cooling Degree-Days (CDD), and Heat Index (HI) and negatively correlated with Wind Speed (WSP) and Solar Radiation (SR). However, the electricity demand is more sensitive to temperatures, CDD, HI than SR and WSP. The regression models are able to adequately predict the electricity demand with a high coefficient of determination R2 (>0.8) and a relatively low Root Mean Square Error (RMSE<150MWh/day). In addition, the residual analysis reveals that the models comply with the basics assumptions of regression models. Furthermore, the results also show that the CORDEX simulations give a realistic representation of all the necessary climate variables used to model the electricity demand in Niger. The simulations project a robust increase in electricity demand at all the GWLs over Niger and indicate that the magnitude of the projection grows with increasing GWLs. Indeed, an increase of 4-16% of DED is projected depending on the magnitude of the warming. It is also worth noting that the magnitude of changes also differs with season, with the highest increase observed in March-May (MAM) and June-August (JJA) while December-February (DJF) displayed the lowest increase. For instance, the Regional Climate Models (RCMs) ensemble median project an increase of about 18% increase in DED for MAM and JJA while for DJF season, it only projects about 5% increase at GWL3.0. In addition to the increase in mean DED, simulations also project an increase in extreme electricity demand due to the increase of extreme temperatures and heatwaves over the country at all the GWLs. The study showed that climate change will affect both mean and peak DED at all the GWLs, with the magnitude of change growing with increasing GWLs. However, the study suggests the investigation of the roles of other factors to further the research, such as population change, future energy policy, urbanization, and economic growth that may also determine the future electricity demand for more robust projections.