| Popis: |
Effective representation of soil heterogeneity in land surface models is crucial for accurate weather and climate simulations. The NOAH-MP land surface model uses dominant soil texture from State Soil Geographic (STATSGO)/Food and Agriculture Organization (FAO) datasets, considerably introducing uncertainty in the simulation of soil hydrothermal changes and terrestrial water and energy fluxes, at a fine scale. This study investigates the likely added value of incorporating an alternative high resolution soil grid data at different depths, for a better representation of soil hydrothermal dynamics in NOAH-MP v4.3. The model is set up at 1 km grid space over all Ireland domain and soil layer thicknesses of 0.07, 0.21, 0.72 and 1.55 m, with a cummulative soil depth of 2.55 m. The thicknesses are selected to match the layers of initial soil input fields. Model experiments are carried out based on two soil data options namely, (1) the STATSGO/FAO dominant soil texture and (2) the 250 m global soil grid textural compositions from the International Soil Reference and Information Centre (ISRIC), in combination with PedoTransfer Functions (PTFs). The current model integration is applied within the high resolution land data assimilation (HRLDAS) framework to simulate soil temperature and soil liquid water, and evaluated for wet and dry periods using observations from the newly established Terrain-AI data platforms (terrainai.com). Ultimately, the study highlights the importance of using realistic dynamic soil information, which could provide insightful scientific contributions to better monitor surface climate and the influences on land use and land management under climate change. |
