Extended Abstract (548K)9.6
Tests of a satellite-based cloud initialization scheme for high latitude application in MM5
Xingang Fan, University of Alaska, Fairbanks, AK; and J. S. Tilley
A continuing problem with respect to regional weather and climate model simulations in high latitudes relates to the fact that the standard versions of many mesoscale models are not set up to provide for a proper initialization of cloud hydrometeor fields (e.g., cloud water, ice and snow mixing ratios, and rain water) Thus, the model has to spend time to spin-up these variables, and their associated cloud systems and latent heating, from other initial fields during the first few (0-6) hours of simulation. Adjustments that occur to the other model fields through this spin-up process then can result in a gradual drift of the model simulation away from the actual conditions and lead to a degraded forecast. If the model initial state were to contain clouds (via the associated hydrometeor mixing ratio fields), the model initial state would be closer to that of the real atmosphere and the spin-up process could be largely avoided, leading to improved simulations over both the short and long terms.
In this paper we describe a cloud diabatic initialization procedure (sometimes referred to as a 'hot start') developed for high latitude application with the Penn State/NCAR MM5 modeling system. In this procedure, the cloud related parameters are derived from MODIS satellite data, depending on availability. In particular, we derive the vertical profiles of temperature, atmospheric water vapor and approximate cloud water content from MODIS products and utilize them in this scheme. The scheme also takes advantage of the MM5 model's restart capability, which has been modified to accept new information and output 3-dimensional adjusted state fields and initialized cloud hydrometeor fields.
The cloud initialization scheme has been tested for a period during mid-August 2001 characterized by substantial clouds and precipitation over western Alaska and the southern slopes of the Brooks Range stemming from a series of short wave disturbances within a rapid westerly mid-tropospheric flow pattern. Results from these initial tests and a preliminary evaluation of the scheme performance will be discussed at the conference.
Session 9, NEW POLAR OBSERVATIONS AND APPLICATIONS: CLOUDS (Continued)
Thursday, 15 May 2003, 8:30 AM-10:15 AM
Previous paper Next paper