Uncertainty Quantification of Modeled Wind Speed Using Different WRF PBL Schemes for Puerto Rico

The planetary boundary layer (PBL) parameterization is the primary influencing element when modeling wind speed, from the point of view of physics schemes in the Weather Research and Forecasting (WRF) model. As part of the Puerto Rico Grid Resilience and Transitions to 100% Renewable Energy Study (PR100), a 20-years of wind resource data set has been developed using the WRF model to support wind energy development considerations for the PR100 project. In this study, we investigated the modeling uncertainty of wind speed stemming from various PBL parameterizations that use different algorithms within the WRF model. Numerical experiments covering 20 years (2001-2020) at 3-km using 11 different PBL schemes were implemented to explore the uncertainty in modeling wind speed for onshore and offshore locations in Puerto Rico. The spread index (SI) was used to quantify the wind modeling uncertainty in the 20-year data. We analyzed the SI calculated over various scales (e.g., diurnal, vertical, and spatial scales, etc.) for four land-based wind (LBW) and four offshore wind (OSW) sites. The PBL’s atmospheric conditions can be characterized into two categories: unstable PBL (mainly during the day) and stable PBL (primarily at night). We therefore focused on analyzing the SI for daytime and nighttime, respectively. For the wind shear (10m–200m) for the selected OSW and LBW sites, the WRF model showed SI ranging from 39%–94% (daytime OSW), 50%–75% (daytime LBW), 37%–60% (nighttime OSW), and 57%–143% (nighttime LBW). This presentation will describe the uncertainty quantification of WRF-based wind modeling in detail as well as summarize comprehensive analysis performed for the PR100 study during the past 2 years.