Understanding Differences in California Climate Projections Produced by Dynamical and Statistical Downscaling. Issue 19 (5th October 2020)

• Record Type: Journal Article
• Authors: Walton, Daniel; Berg, Neil; Pierce, David; Maurer, Ed; Hall, Alex; Lin, Yen‐Heng; Rahimi, Stefan; Cayan, Dan
• Is Part Of: Journal of geophysical research. Volume 125:Issue 19(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-10-05
• Subjects: downscaling -- California -- regional climate -- snow albedo feedback; Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1029/2020JD032812
• Languages: English
• ISSNs: 2169-897X

Abstract

Abstract: We compare historical and end‐of‐century temperature and precipitation patterns over California from one dynamically downscaled simulation using the Weather Research and Forecast (WRF) model and two simulations statistically downscaled using Localized Constructed Analogs (LOCA). We uniquely separate causes of differences between dynamically and statistically based future climate projections into differences in historical climate (gridded observations versus regional climate model output) and differences in how these downscaling techniques explicitly handle future climate changes (numerical modeling versus analogs). In these methods, solutions between different downscaling techniques differ more in the future compared to the historical period. Changes projected by LOCA are insensitive to the choice of driving data. Only through dynamical downscaling can we simulate physically consistent regional springtime warming patterns across the Sierra Nevada, while the statistical simulations inherit an unphysical signal from their parent Global Climate Model (GCM) or gridded data. The results of our study clarify why these different techniques produce different outcomes and may also provide guidance on which downscaled products to use for certain impact analyses in California and perhaps other Mediterranean regimes. Plain Language Summary: Climate scientists commonly use global climate models (GCMs) to make end‐of‐century climate projections based on the primitive governing equations. Because of the limitations of current computing resources, GCMs cannot simulate and provide sufficiently high enough spatial resolution to capture the large geographic landscape and potentially diverse climate change patterns across California. As such, scientists typically apply two different approaches to generate high‐resolution regional climate projections: (1) feed coarse‐resolution GCM data into to higher‐resolution regional climate models (RCMs) to generate high‐resolution physically based climate change signals and (2) feed the GCM data into statistical models "trained" on historical observation data. In our study, we compare simulations using the former method with those using the latter method, examining April surface maximum temperature and annual precipitation projections. We find that the methods used here tend to yield similar projections if applied to a historical period (1990s), but their projections can be quite different for a future period (2090s), especially for precipitation. For our statistical methodology, the nature of projections is relatively insensitive to our choice of training data. Results here may help to guide others on methodology choice when projecting climate change across California or indeed other Mediterranean climate locales. Key Points: The climate change signal projected from the WRRF dynamical versus the LOCA statistical methods differs Differences between the two downscaling methods are larger for future projections than for the historical period LOCA statistical projections are relatively insensitive to training data