Datasets and ensemble scripts for: Multi-Model Ensemble Projections of Wind Power Density in Northeast Brazil

This repository contains the processed datasets and scripts used in the study“Multi-Model Ensemble Projections of Wind Power Density in Areas of Northeast Brazil for the Recent Past and Future”. The dataset includes:(i) regional wind speed time series derived from CORDEX-CORE Regional Climate Models and ECMWF-ERA5 reanalysis;(ii) scripts used to generate multi-model ensembles using Arithmetic Mean (AM), Convex Combination (CC), Principal Component Regression (PCR), and Machine Learning techniques (kNN, Decision Tree, Random Forest, Artificial Neural Networks, and Support Vector Machine);(iii) wind power density (WPD) results for the historical period (1986–2005), near future (2041–2060), and distant future (2080–2099) under RCP2.6 and RCP8.5 scenarios. All scripts are provided in R and Python and allow reproducibility of the ensemble generation and WPD calculations.