Utilizing Mesoscale Model Output within the SLAM -P model Framework

摘要

With the advent of fast desktop computers, mesoscale models are being run to supply data atJr types of air pollution problems. However, key algorithms in many of these transport and diffusion models have not been designed to handle the various types of data that are available from these forecast models. An example of this is found in the SL/EM (Short-range Layered Atmospheric Model) for particulates model which handles both gaseous and particulate dispersion. Up to recently, SLAW has made use of mixing depths as a twkdBily function that takes the mixing depth at a trajectory location at sunrise and sunset and uses these values to control vertical puff splitting and maximum plume growth. This procedure, while easy to implement, does have a couple of major drawbacks. First, the afternoon maximum mixing height is applied throughout daytime. However, we know that the true mixing depth increases slowly after sunrise and then grows rapidly later in the morning reaching a maximum inthe afternoon. A second drawback is the assumption that the mixing depth within a puff remains constant relative to puff movement. However, we know that the true mixing depth changes in time through advection over varying surface conditions and throughdifferent synoptic weather patterns. Through the use of hourly mixing depths from a mesoscale model such as RAMS (Regional Atmospheric Modeling System) it is hoped that these drawbacks will be eliminated and will produce more realistic transport layer depth calculations and splitting conditions. This paper will describe the implementation of these hourly mixing depths within the SLAMP modeling framework. In addition, results from the tracer data set ANATEX (Across North America Tracer Experiment) will also be presented.