High-Resolution WRF Simulations of the METCRAX 2 Mesoscale Environment

High-resolution Weather Research and Forecasting (WRF) model simulations are performed for Intensive Observation Periods (IOPs) of the METCRAX 2 field campaign. The simulations test both the ability of the model to recreate the mesoscale environment of the crater and to help identify the source of some features of the background flow that were observed during the project.

A 1-km resolution nested domain covering Arizona is modeled for all 7 IOPs. Beginning with a standard model setup and 6 hours of spin-up time before sunset, all IOPs are simulated and verified against observations of temperature and wind taken at the Meteor Crater. These tests show a successful forecast of IOP 4, both in terms of the strength of the mesoscale katabatic drainage flow and its depth and vertical structure. Results from IOP 2 are also encouraging, although the timing of the modeled katabatic flow differs from observations. For the remaining five IOPs, the simulations do not verify well against observations - generally failing to generate a realistic drainage flow structure.

We attempt to improve simulations of all IOPs by adjusting the model parameterizations and changing the model domain and spin-up time. Simulations are shown to be predominantly sensitive to changes in the land-surface model and radiation schemes, with significant reduction in model error in runs using increased parameterization complexity. Changes in radiation scheme do not add significantly to the total model run time. In contrast, improved results from LSM changes lead to significant increases in computational expense. Increased vertical resolution in the lowest 1 km of the atmosphere also leads to improved modeling of the katabatic flow in some of the IOPs.

Our WRF simulations are used to investigate the mechanisms leading to different katabatic flow depths during different IOPs. Of particular interest is the deep (~500m) southwesterly flow during IOP 4 and whether this could be characterized as a katabatic flow at all heights, or if some other mesoscale circulation was present. Simulations show that the deep flow during IOP 4 is related to the upstream terrain and displays the properties indicative of a katabatic flow. The strong, deep flow was highly localized in the simulations and closely related to the locations of valleys emanating from the upstream plateau. Simulations with the upstream ridge smoothed and reduced in height show a corresponding decrease in katabatic flow depth during IOP 4, with less impact in IOPs where WRF is able to generate a drainage flow. Other mesoscale aspects of the flow relating to METCRAX 2 observations are also investigated.