From August 18-23, 1994 an Intensive Observation Period (IOP)
of the NASA/USDA field experiment Washita '94 was conducted in
the nearly 600 square km. Little Washita Watershed in southwest Oklahoma to investigate land-atmosphere interactions and remote sensing of soil moisture and temperature. During this experiment, 3-hourly atmospheric radiosounding data was collected from 6 am local time until 6 pm local time. In addition, half-hourly surface measurements of latent and sensible heat flux, as well as standard meteorological variables were collected during these times.
We have coupled a sophisticated hydrological model (TOPLATS)
with a 1-d version of MM5 and applied this model to simulate atmospheric budgets of momentum, heat and moisture during this period. The hydrological model is applied at a 30 meter spatial resolution and an hourly temporal resolution, while the meteorological model is applied at a 12km resolution, with an "implied" advection step of 36 seconds.
In order to demonstrate the model response to assimilation of remotely sensed data, we will compare and contrast model budgets with and without the incorporation of 4 km spatial resolution precipitation data from NEXRAD and downward solar radiation from surface meteorological stations, which will serve as a surrogate for GOES-derived solar radiation. Special attention will be given to budgets of TKE and the development of shallow and deep convective clouds in the model versus what occured during the IOP.
The 1-D version of the MM5 model contains a full complement of
physics options including a 1 1/2 order TKE boundary layer scheme (the
Gayno-Seaman scheme), a deep convection parameterization (the Kain-Fritsch scheme), a shallow convection parameterization (the McHenry scheme), long-wave and short-wave atmospheric radiation schemes (Dudhia), and MM5's explicit microphysics package. TOPLATS is a TOPMODEL-based land-atmosphere transfer scheme orginally developed by Famiglietti and Woods and subsequently improved by Peters-Lidard to support boundary layer modeling and incorporation of remotely-sensed data.
We accomplish the coupling and transfer of forcing data between the two models, and between the two models and remotely-sensed data, through an I/O API extended to communicate with PVM mailboxes. This allows the two models to coordinate with each other and exchange data, while retaining their own fundamental spatio-temporal physical and computational characteristics. We will discuss aggregation and disaggregation of modeled fluxes at the coupled model interface in this context.
Preliminary simulations are ongoing as of 1 July 1998. We are
expecting that the assimilation of remotely-sensed data into the coupled model system will demonstrate desirable improvements in the response of the atmospheric column, because the surface fluxes computed by TOPLATS will be driven primarily by observed data. This will permit the modeled PBL to more accurately reflect the evolution of the natural PBL as it occured during selected, weakly-forced episodes of the Little Washita IOP. Following presentation of these results, we will enumerate our progress toward incorporation of the technique into a coupled 3-D version of the modeling system, and discuss its utility for both historical and predictive applications