Evaluation of the regional climate model RegCM4.7 over the Carpathian region for very wet and average years
- 1Department of Meteorology, Eötvös Loránd University, Budapest, Hungary (kalmartimea@caesar.elte.hu)
- 2Faculty of Science, Excellence Center, Eötvös Loránd University, Martonvásár, Hungary
Precipitation is one of the most important climate variables in many aspects due to its key impact on agriculture, water management, etc. However, it remains a challenge for climate models to realistically simulate the regional patterns, temporal variations, and intensity of precipitation. The difficulty arises from the complexity of precipitation processes within the atmosphere stemming from cloud microphysics, cumulus convection, large-scale circulations, planetary boundary layer (PBL) processes, and many others. This is especially true for heterogeneous surfaces with complex orography such as the Carpathian region. Thus, the Carpathian Basin, with its surrounding mountains, requires higher model resolution, along with different parameterizations, compared to more homogenous regions. The aim of the study is to reproduce the historical precipitation pattern through testing the parameterization of surface processes. The appropriate representations of land surface component in climate models are essential for the simulation of surface and subsurface runoff, soil moisture, and evapotranspiration. Furthermore, PBL strongly influences temperature, moisture, and wind through the turbulent transfer of air mass. The current study focuses on the newest model version of RegCM (RegCM4.7), with which we carry out simulations using different parameterization schemes over the Carpathian region. We investigate the effects of land-surface schemes (i.e. BATS - Biosphere-Atmosphere Transfer Scheme and CLM4.5 - Community Land Model version 4.5) in the regional climate model. Studies over different regions have shown that CLM offers improvements in terms of land–atmosphere exchanges of moisture and energy and associated surface climate feedbacks compared with BATS. Our aim includes evaluating whether this is the case for the Carpathian region.
Four 1-year-long experiments both for 1981 and 2010 (excluding the spin-up time) are completed using the same domain, initial and lateral atmospheric boundary data conditions (i.e. ERA-Interim), with a 10 km spatial resolution. These years were chosen because 1981 was a normal year in terms of precipitation, while 2010 was the wettest year in Hungary from the beginning of the 20th century. We carry out a detailed analysis of RegCM outputs focusing not only on standard climatological variables (precipitation and temperature), but also on additional meteorological variables, which have important roles in the water cycle (e.g. soil moisture, evapotranspiration). The simulations are compared with the CARPATCLIM observed, homogenised, gridded dataset and other databases (ESA CCI Soil Moisture Product New Version Release (v04.5) and Surface Solar Radiation Data Set - Heliosat (SARAH)). It is found that the simulated near-surface temperature and precipitation are better represented in the CLM scheme than in the BATS when compared with observations, both over the lowland and mountainous area. The model simulations also show that the precipitation is overestimated more over mountainous area in 2010 than in 1981.
How to cite: Kalmár, T., Pieczka, I., and Pongrácz, R.: Evaluation of the regional climate model RegCM4.7 over the Carpathian region for very wet and average years, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-330, https://doi.org/10.5194/egusphere-egu2020-330, 2019
Comments on the display
AC: Author Comment | CC: Community Comment | Report abuse
Dear Tímea,
What do you think are the most important differences between in the BATS and CLM models for your work? Why do the results differ so much between them?
Similarly, what do you think are the differences between the Holstag and UW schemes? It seems that their effects aren't quite as large as the difference between land models, do you have thoughts on why that could be?
Dear Angeline,
Thank you for your question. Firstly, let me start with the short descriptions about the four parameterization schemes.
The behaviours of the two land models are quite different. BATS scheme has one vegetation layer, one snow layer, two soil temperature levels and three soil moisture levels. Heat fluxes, water vapour and momentum at the surface are calculated on the basis of the drag coefficients obtained from Monin-Obukhov similarity theory applied to the surface layer. CLM4.5 is a more sophisticated model than BATS; the former includes more elaborate surface characteristics, with more soil and snow layers, and uses explicit treatments for both liquid water and ice. The surface energy fluxes are calculated separately for snow-covered, water-covered, and snow/water-free portions of vegetated and cropland units, snow-covered and snow-free portions of glacier land units.
According to the planetary boundary schemes, Holtslag is a non-local parameterization scheme and the calculation of PBL height is based on the bulk Richardson number. The UW is a 1.5-order local, down-gradient diffusion parameterization in which the velocity scale is based on turbulent kinetic energy.
We are still working on exploring the interactions between the parameterization schemes, but the main findings so far can be summarized as follows:
The treatment of the soil moisture is different between the two land surface schemes that affects the whole hydrological cycle and is a key variable of the land-atmosphere system. For instance, BATS simulations contain more available water/moisture leading to more cloud (and rain) and less incoming shortwave radiation, which results in lower temperature. The effect of the PBL scheme is smaller than the land surface model, but we also can see differences (namely, lower temperature values occur with the UW scheme). This detected cooling is ascribed to a general reduction in lower tropospheric eddy heat diffusivity with the UW scheme.