Assessing the Global Wind Atlas and local measurements for bias correction of wind power generation simulated from MERRA-2 in Brazil

TL;DR

This study evaluates the effectiveness of the Global Wind Atlas and local measurements for bias correction of MERRA-2 wind data in Brazil, revealing that GWA improves large-scale results but not local wind park predictions.

Contribution

It provides a comprehensive analysis of bias correction methods across multiple spatial levels in Brazil, comparing global and local data sources for wind power simulation.

Findings

GWA improves large-scale wind power simulation results.

Local measurements perform similarly to GWA for bias correction.

Complex interpolation methods do not significantly enhance simulation accuracy.

Abstract

NASAs MERRA-2 reanalysis is a widely used dataset in renewable energy resource modelling. The Global Wind Atlas (GWA) has been used to bias-correct MERRA-2 data before. There is, however, a lack of an analysis of the performance of MERRA-2 with bias correction from GWA on different spatial levels - and for regions outside of Europe, China or the United States. This study therefore evaluates different methods for wind power simulation on four spatial resolution levels from wind park to national level in Brazil. In particular, spatial interpolation methods and spatial as well as spatiotemporal wind speed bias correction using local wind speed measurements and mean wind speeds from the GWA are assessed. By validating the resulting timeseries against observed generation it is assessed at which spatial levels the different methods improve results - and whether global information derived from the GWA can compete with locally measured wind speed data as a source of bias correction. Results show that (i) bias correction with the GWA improves results on state, sub-system, and national-level, but not on wind park level, that (ii) the GWA improves results comparably to local measurements, and that (iii) complex spatial interpolation methods do not contribute in improving quality of the simulation.