Assessing the Quality of Wind Observations Using a Regional Scale Model

The Remote Automated Weather Station (RAWS) network began operating in the U.S. in the early 1980s, and has become a large multi-agency partnership to collect hourly fire weather data for over 2000 stations across the U.S. Primary weather measurements from RAWS include temperature, relative humidity, wind speed and direction, precipitation, and solar radiation, which are operationally used in calculating fire danger indices. Historical RAWS are used among other things to calculate fire danger in relation to fire activity. RAWS observation standards provide station measurement, maintenance and siting guidelines. A unique aspect of RAWS is that most of the stations at least in the western U.S. are located at mid- and high-elevations, generating complex terrain measurements that no other large observing network represents. A recent reanalysis project for California and Nevada produced a homogeneous 10-year (2004-2013) 2-km hourly climatology using the Weather and Research Forecast (WRF) model. In assessing wind speed as part of the validation process, we saw that a number of RAWS locations appeared to be underrepresenting wind speed. Examination of metadata photos revealed that the wind sensors of these RAWS are not sufficiently exposed, due likely to vegetation growth and maturation over the years. In this presentation, we highlight the RAWS network and its value, and describe how site changes are influencing wind measurements. We will also discuss how a regional model, such as WRF, can be used to quickly assess the quality of wind measurements for a large region.