Charles Explorer logo
🇨🇿

Magnitude of Internal Climate Variability Based on Pre-industrial CMIP6 Models over Central Europe

Publikace na Matematicko-fyzikální fakulta |
2023

Tento text není v aktuálním jazyce dostupný. Zobrazuje se verze "en".Abstrakt

Understanding the nature of the Internal Climate Variability (ICV) as in- herent to the Earth's climate system is crucial for global climate models (GCMs) and regional climate models (RCMs) analyses, since its existence is inevitable in all-time series of any meteorological variable. This obviously makes the ICV a source of un- certainties in climate models' outputs. Preindustrial control runs, i.e., GCM simula- tions with constant external forcings, represent useful source of data for ICV analysis.

The present study aims to assess the ICV magnitude from these simulations in order to provide a useful benchmark for further performance analysis of GCMs or RCMs over central Europe. The approach consists of computing the standard deviation (SD) and

Inter-quartile-range (IQR) of near-surface air temperature for all the available pre-in- dustrial control simulations within a selected time-period of 500 years. It encompasses the comparison of estimated ICV magnitudes between the GCMs. The whole analysis is conducted on monthly, seasonal, and annual time-scales over central Europe. All the models depict the annual cycle, seasonal and yearly time-series of near-surface air temperature over the study area despite of the values and data spread differences. The

SD and IQR show that MRI-ESM2-0 (EC-Earth3-CC and EC-Earth3-Veg) display smallest (largest) magnitude(s) compared to rest of the models throughout almost the whole analyses. This has been confirmed from their data spread characteristics, in which MRI-ESM2-0 displays smaller variability, while EC-Earth3-CC and EC-

Earth3-Veg exhibit larger variability. The inter-quartile-range analysis displays almost the same pattern as the standard deviation. However, by considering their differences, the IQR is likely preferable since it is a robust metric that cannot be significantly af- fected by the existence of outliers in the datasets.