Development of the Flux-Adjusting Surface Data Assimilation System for Mesoscale Models

Abstract

The flux-adjusting surface data assimilation system (FASDAS) is developed to provide continuous adjustments for initial soil moisture and temperature and for surface air temperature and water vapor mixing ratio for mesoscale models. In the FASDAS approach, surface air temperature and water vapor mixing ratio are directly assimilated by using the analyzed surface observations. Then, the difference between the analyzed surface observations and model predictions of surface layer temperature and water vapor mixing ratio are converted into respective heat fluxes, referred to as adjustment heat fluxes of sensible and latent heat. These adjustment heat fluxes are then used in the prognostic equations for soil temperature and moisture via indirect assimilation in the form of several new adjustment evaporative fluxes. Thus, simulated surface fluxes for the subsequent model time step are affected such that the predicted surface air temperature and water vapor mixing ratio conform more closely to observations. The simultaneous application of indirect and direct data assimilation maintains greater consistency between the soil temperature-moisture and the surface layer mass-field variables. The FASDAS is coupled to a land surface submodel in a three-dimensional mesoscale model and tests are performed for a 10-day period with three one-way nested domains. The FASDAS is applied in the analysis nudging mode for two coarse-resolution nested domains and in the observational nudging mode for a fine-resolution nested domain. Further, the effects of FASDAS on two different initial specifications of a three-dimensional soil moisture field are also studied. Results indicate that the FASDAS consistently improved the accuracy of the model simulations.

Impact of Land Surface Heterogeneity on Mesoscale Atmospheric Dispersion

“…While it is possible to determine the important vegetation characteristics at adequate spatial resolutions using satellite remote sensing, soil moisture measurements at the needed spatial and temporal resolution are scarce. However, techniques that constrain surface energy flux and moisture partitioning in the SVAT sub-model using either satellite observations of skin surface temperature (Jones et al 1998a,b;McNider et al 1994McNider et al , 2005 or remotely-sensed surface soil moisture estimates (Reichle et al 2002(Reichle et al , 2007(Reichle et al , 2008, or in situ soil moisture measurements (De Lannoy et al 2007), surface meteorological observations (Alapaty et al 2008), or other satellite data (Matsui et al 2007(Matsui et al , 2008 can be used indirectly to account for the spatial heterogeneity of soil moisture. Numerical model experiments conducted in this study show that atmospheric dispersion in both the convective and nocturnal boundary layers are sensitive to the nature of land-surface heterogeneity.…”

“…Thus the realistic specification of vegetation heterogeneity alone is not sufficient but needs to be coupled with a realistic distribution of soil moisture. Data assimilation techniques that utilise satellite skin temperature (McNider et al 2005), surface meteorological observations (Alapaty et al 2008), or an integrated land-surface assimilation analysis (Matsui et al 2007) are capable of indirectly accounting for soil moisture heterogeneity. Since the data assimilation techniques are aimed at constraining the SVAT model to generate a realistic spatial distribution of sensible and latent heat fluxes, it may appear that it is unnecessary to introduce satellite-derived spatial distributions of vegetation characteristics when such techniques are being utilised.…”

“…The sensitivity of atmospheric dispersion to the use of the assimilation techniques of Alapaty et al (2008), Matsui et al (2007) and McNider et al (2005) needs to be considered in future studies. Improvements in the performance of the dispersion modelling system utilising satellite vegetation data and data assimilation techniques also needs to be quantified by comparing against observations.…”

Impact of ASAR soil moisture data on the MM5 precipitation forecast for the Tanaro flood event of April 2009

“…Operational systems often use a Land Data Assimilation System (LDAS), which is essentially an offline version of the LSM forced with observed precipitation, radiation, and analyzed meteorology, so that the forecast starts with optimal soil moisture fields (e.g. Mitchell et al, 2004;Chen et al, 2007;Alapaty et al, 2008). Another way to initialize soil moisture is through dynamic adjustment within the mesoscale model simulation where soil moisture is nudged according to differences between modelled and analyzed observations of 2-m temperature and relative humidity, as described by Pleim and Xiu (2003).…”

Screen-level non-GTS data assimilation in a limited-area mesoscale model

“…They reported a general improvement of the meteorological parameters with surface and satellite data assimilation even in these strongly largescale-induced conditions. It has to be noted that they modified the surface fluxes using a so-called Flux-Adjusting Surface Data Assimilation System (FASDAS) developed by Alapaty et al (2008) which is probably able to act as a more effective perturbation in the model dynamics. Also Schraff (1997) showed that it is possible to improve the forecast of low stratus over the Alps using both surface and upper-air data with an observation-nudging method.…”