Loading...
Building a high-resolution solar radiation database for solar resource planning under climate change scenarios
Bailey, Maggie D.
Bailey, Maggie D.
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2024
Date Submitted
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
2025-10-04
Abstract
As the photovoltaic (PV) industry moves to extend plant lifetimes to more than 50 years, assumptions that current and historical weather represent future solar radiation trends may not be appropriate, especially as our climate changes rapidly. This thesis is a framework for constructing a state-of-the-art data product to project common solar radiation variables for future years at a high resolution in space and time. It can serve as a primary tool for PV facility planning and power grid modeling. The data product relies on regional climate models (RCMs), downscaled from general circulation models (GCMs) that operate on a global scale. RCMs are often not at a spatial or temporal resolution that is useful for planning. Therefore, generating higher resolution data from available RCMs for future climate path scenarios is required.
This thesis addresses three research problems that result from building towards this final data product. First, we examine the uncertainty of regridding multiple RCMs from their native grid resolutions to a common grid. While this is a frequent step in any multi-modal climate model analysis, the uncertainty and downstream effects associated with it have not been widely studied. Second, we adapt a common bias correction method, quantile mapping, for solar radiation data to correct for systematic biases present in RCMs and assess the effectiveness of this method as well as sources of persistent biases. Finally, this thesis proposes a novel temporal downscaling method to generate hourly solar radiation data given daily averages that is adaptable to any location and time of year. We also implement spatial downscaling for solar radiation data from the 20 km to 8 km grid resolution. The final data product will be publicly available in October 2024.
Associated Publications
Rights
Copyright of the original work is retained by the author.