Abstract
Recent and future climate diagrams (surface air temperature, surface relative humidity, surface wind, and mean sea level pressure) for the Saudi Arabian Red Sea Coast are analysed based on hourly observations (2016–2020) and hourly ERA5 data (1979–2020) with daily GFDL mini-ensemble means (2006–2100). Moreover, GFDL mini-ensemble means are calculated based on the results of three GFDL simulations (GFDL-CM3, GFDL-ESM2M, and GFDL-ESM2G). Observation data are employed to describe the short-term current weather variability.
However, ERA5 data are considered to study the long-term current weather variability after bias removal via a comparison to observations. Finally, a bias correction statistical model was developed by matching the cumulative distribution functions (CDFs) of corrected ERA5 and mini-ensemble mean data over 15 years (2006–2020). The obtained local statistic were used to statically downscale GFDL mini-ensemble means to study the future uncertainty in the atmospheric parameters studied. There occurred significant spatial variability across the study area, especially regarding the surface air temperature and relative humidity, based on monthly analysis of both observation and ERA5 data. Moreover, the results indicated that the ERA5 data suitably describe Tabuk, Jeddah and Jizan weather conditions with a marked spatial variability. The best performance of ERA5 surface air temperature and relative humidity (surface wind speed and sea level pressure) data was detected in Tabuk (Jeddah). These data for the Saudi Arabian Red Sea coast, 1979–2020, exhibit significant positive trends of the surface air temperature and surface wind speed and significant negative trends of the relative humidity and sea level pressure. The GFDL mini-ensemble mean projection result, up to 2100, contains a significant bias in the studied weather parameters. This is partly attributed to the coarse GFDL resolution (2° × 2°). After bias removal, the statistically downscaled simulations based on the GFDL mini-ensemble mean indicate that the climate in the study area will experience significant changes with a large range of uncertainty according to the considered scenario and regional variations.