More robust offshore wind energy planning through model ensembling

Abstract

This research performs an ex-ante assessment of the 19 high potential areas for offshore wind energy (HPA-OWE) allocated in four maritime spatial planning subdivisions of Spain. A 39 geo-statistical criteria pool was developed and categorized into five planning tiers (coexistence, socio-ecological, spatial-efficiency, energy-equity, technical/technological). An ensemble of three multi-criteria decision analysis (MCDA) techniques coupled with a Monte Carlo method based on a large, uniform number of randomly distributed criteria weights is applied for more robust priority rankings of HPA-OWE. The co-existence tier indicates that HPA-OWE should be prioritized in the North Atlantic and in the Levantine-Balearic planning subdivision. The application of machine learning on the MCDA results identified criteria that most influence the rank of each HPA-OWE at planning subdivision. The outcomes highlight the need to include place-based data to better take into account spatial inequalities in coastal regions and re-balance them with socio-economic and energetically privileged coastal territories