Vukovic A.,University of Belgrade |
Rajkovic B.,University of Belgrade |
Modelling for Environment's Sake: Proceedings of the 5th Biennial Conference of the International Environmental Modelling and Software Society, iEMSs 2010 | Year: 2010
Land Ice Sea Surface model (LISS) is a new model for prediction of soil temperature and soil moisture. It is a part of the Non-hydrostatic Multi-scale Model on B-grid (NMM-B). The skin temperature, that represents the temperature of the interface between ground and air, is calculated from surface energy balance. It includes total influence of the soil processes and vegetation cover. Evapotranspiration is parameterized with β parameter that takes into account evaporation from the bare soil, evaporation from interception reservoir and transpiration of the plants. Model has four layers and one or more layers for snow, depending on its amount. Soil temperatures are calculated using Fourier diffusion law and water content using Darcy law. LISS has been tested using two different data sets (Caumont, France 1986; Bondville, USA 1998) as well as against NOAH-LSM simulations. Annual balance of energy and water showed numerical stability. The annual diurnal variation of surface temperature is close to the observed value. RMSE for the surface temperature is 1.9°C for Bonville site. Surface fluxes in 36-hour period of snow growth simulations for Bondville are close to the observed values.
Hanna S.R.,Hanna Consultants |
Reen B.,Pennsylvania State University |
Hendrick E.,Epsilon Associates Inc. |
Santos L.,Air Quality Associates |
And 7 more authors.
Boundary-Layer Meteorology | Year: 2010
The objective of the study is to evaluate operational mesoscale meteorological model atmospheric boundary-layer (ABL) outputs for use in the Hazard Prediction Assessment Capability (HPAC)/Second-Order Closure Integrated Puff (SCIPUFF) transport and dispersion model. HPAC uses the meteorological models' routine simulations of surface buoyancy flux, winds, and mixing depth to derive the profiles of ABL turbulence. The Fifth-Generation Pennsylvania State University/National Center for Atmospheric Research Mesoscale Model (MM5) and the Weather Research and Forecast-Nonhydrostatic Mesoscale Model (WRF-NMM) ABL outputs and the HPAC ABL parameterisations are compared with observations during the International H2O Project (IHOP). The meteorological models' configurations are not specially designed research versions for this study but rather are intended to be representative of what may be used operationally and thus have relatively coarse lowest vertical layer thicknesses of 59 and 36 m, respectively. The meteorological models' simulations of mixing depth are in good agreement (±20%) with observations on most afternoons. Wind speed errors of 1 or 2 ms-1 are found, typical of those found in other studies, with larger errors occurring when the simulated centre of a low-pressure system is misplaced in time or space. The hourly variation of turbulent kinetic energy (TKE) is well-simulated during the daytime, although there is a meteorological model underprediction bias of about 20-40%. At night, WRF-NMM shows fair agreement with observations, and MM5 sometimes produces a very small default TKE value because of the stable boundary-layer parameterisation that is used. The HPAC TKE parameterisation is usually a factor of 5-10 high at night, primarily due to the fact that the meteorological model wind-speed output is at a height of 30 m for MM5 and 18 m for WRF-NMM, which is often well above the stable mixing depth. It is concluded that, before meteorological model TKE fields can be confidently used by HPAC, it would help to improve vertical resolution near the surface, say to 10 m or less, and it would be good to improve the ABL parameterisations for shallow stable conditions. © Springer Science+Business Media B.V. 2009.