The Mountain Terrain Atmospheric Modeling and Observations (MATERHORN) Program: A Progress Report

Emerging important applications such as severe air pollution in cities and operation of aircraft and unmanned aerial vehicles in rugged terrain have resuscitated scientific and societal interests in mountain-terrain weather. Funded by a Multidisciplinary University Research Initiative (MURI) grant from the Office of Naval Research in 2011, the MATERHORN Program was designed to improve the predictability of weather in mountainous terrain through improvements in four critical components of research: Modeling, Experiments, Technology and Parameterizations. The program has achieved important scientific milestones with regard to identifying model deficiencies and knowledge gaps, conducting process studies, and developing knowledge and tools for model improvements.

Two comprehensive field studies were conducted at the Granite Mountain Atmospheric Science Test Bed (GMAST) of the US Army Dugway Proving Grounds. During September 25 to October 31, 2012, the field campaign focused on quiescent fair weather with wind speeds < 4 m/s, and during May 1 to May 30, 2013, it dealt with synoptic influence, moister surface conditions, and moderate (5 to 10 m/s) and strong (> 10m/s) wind periods. An unprecedented suite of high-end instrumentation was used, allowing investigations from tens of km to millimeters and hours to second scales. The technology development component includes the deployment of an unmanned aerial vehicle (UAV) with temperature, humidity and wind velocity sensors, an on-board data acquisition system, a navigation system and an in-situ calibrated turbulence measuring system with a combination of hot-film anemometers and a sonic anemometer. A newly developed fog aerosol sampling system (FASS) for droplet size distribution measurements is being incorporated to the UAV. A three-frequency sensor system for the collection Radio Frequency polarimetric data to infer soil moisture was designed, constructed and deployed in both field campaigns.

A host of phenomena that signify flow interactions across a range of scales, local forcing mechanisms (e.g., shadow propagation, topographic and land-use inhomogeneities) and energy cascade near the mountains through interacting physical phenomena were identified by the measurements. Numerous algorithms are being developed for data analysis (e.g., airborne Doppler lidar and microwave profiler retrieval algorithms) and field observations are interpreted using existing and new theoretical analyses and conceptual frameworks. Some of the observations are further investigated using controlled laboratory experiments and research-grade high resolution Large-Eddy Simulations (LES).

In the Modeling component, state-of-the-science mesoscale models (WRF and COAMPS) are employed, which are found to be more prone to forecast error when predicting in complex terrain than over flat terrain. Near-surface temperature forecasts over dryland mountainous regions are very sensitive to the soil moisture, formulation of soil characteristics, and choices of model parameters (coefficients) in land-surface models. Significant improvements could be achieved through improved soil moisture analysis and the use of soil thermal conductivity parameterizations appropriate for desert soil textures and very low soil moistures commonly found in arid regions. Different boundary-layer turbulent mixing schemes led to widely different predictions, calling for more terrain-appropriate parameterizations. Assimilation of near-surface observations could improve short-range forecasts over complex terrain, and ensemble methods with flow-dependent error estimates appeared superior to 3D variational methods that rely on static error estimates. Probabilistic methods for predicting observation impact, and determining optimal observation siting, in weakly-forced flows appeared tractable for short-range, near-surface forecasts. The immersed boundary method (IBM) implemented in WRF model could be used to obtain 50m resolution in complex terrain, thus improving the efficacy of nested simulations of complex terrain weather.

Further information can be found in www.nd.edu/~dynamics/materhorn