Estimating climate — Demand Nexus to support longterm adequacy planning in the energy sector

Projecting future energy demand is extremely complex, due to the highly uncertain future policy, technology, land-use and climate change. Existing energy-economy tools (e.g., NEMS, MARKAL) can project future demand accounting for factors such as uncertain technology, price, etc. However, these tools are incapable of incorporating uncertainties associated with shifts in demand due to climate change. In this paper, we propose a Bayesian non-parametric method to capture the energydemand — climate nexus. The probabilistic estimates from our proposed models can be integrated with existing models to project future demand. The Bayesian framework can also allow stakeholders to apply appropriate prior probabilities on relevant climate variables based on expert knowledge. To illustrate applicability of our proposed method, we used residential and commercial energy sectors in the state of Indiana as a case study. Our results indicate maximum-sustained-wind-speed, mean-dew-point-temperature and snowfall are the most important predictors of the residential and commercial energy-demands.