Statistical methods for downscaling and bias correcting CMIP6 data across Central and Southern Florida
Abstract
Climate change is expected to intensify the frequency and intensity of meteorologic variables that drive hydroclimatic extremes (floods and droughts), which are linked with the safety of infrastructure and well-being of communities. Global climate models (GCMs) can be used to assess changes in meteorologic variables under climate change. These models must be downscaled and bias corrected against field observations for local and regional applications. For doing so, statistical downscaling techniques can be used to create correlations between the field observations and GCMs. Here, we applied two downscaling techniques—Quantile Mapping (QM) and Delta Method (DM)—to downscale daily precipitation and air temperature produced by eight GCMs for the Coupled Model Intercomparison Project Phase 6 (CMIP6) climate scenarios across the Central and Southern Florida, an area with valuable ecosystem (e.g., Everglades National Park) and large population centers (e.g., Miami) that has experienced severe hydroclimatic extremes like floods in the past. We collected observed daily precipitation and air temperature from weather stations for period 1985-2014. Error metrics were calculated and statistical tests were performed to evaluate the efficiency of the downscaling. The downscaled GCMs were then applied to project daily precipitation and air temperature up to 2100 under three future scenarios, shared socioeconomic pathways (SSPs), SSP1-2.6, SSP2-4.5, and SSP5-8.5. Our preliminary results showed that BCC-CSM2-MR model outperformed the other seven GCMs for generating daily precipitation and air temperature data. The results can be utilized in understating changes in hydrology and characteristics of hydroclimatic extremes under a changing climate.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.A45A..68H