Summary

Under our previous ARM support, we developed an lntegrated Data Assimilation and Sounding System (IDASS). The IDASS system couples in situ and remote sensing instruments with a state-of-the-art mesoscale data assimilation system to provide a time-continuous description of three-dimensional atmospheric structure with a horizontal resolution of approximately 5 km' over a domain of 200x200 km2. The observing system associated with IDASS is the lntegrated Sounding System (ISS), which combines numerous remote and in situ sensors into a single transportable unit. Combining sensor platforms into a single unit allows us to take advantage of their positive attributes, while overcoming their singular limitations.

Four-dimensional assimilation of ISS and other data sets into a non-hydrostatic version of the NCAR/Pennsylvania State University mesoscale model is being pursued by David Parsons and Walt Dabberdt, together with MMM scientists Bill Kuo, Jim Dudhia, and YongRun Guo. The project is funded by DOE's ARM program. Results obtained using observing simulation system experiments (OSSE) suggest that data assimilation can improve the accuracy of model simulations using only relatively modest arrays of sensing systems. As expected, the OSSEs show that for mesoscale domains an accurate determination of the lateral boundary conditions is a strong constraint on quality of the simulation. The potential improvement resulting from data assimilation is being quantified using data taken during a ten-day field experiment at the ARM/CART site in June 1993. For this ten-day period, simulations will be conducted with and without data assimilation. The data from three CLASS sounding sites (approximately 90 soundings) will be withheld from the assimilation and used to quantitatively measure the potential improvement obtained from employing data assimilation.