P1.6 USE OF OKLAHOMA MESONET AND NATIONAL WEATHER SERVICE DATA IN THE DEVELOPMENT OF CONTROL STRATEGIES FOR WINTER-TIME BRIDGE HEATING Kala N. Pandit * Oklahoma State University, Stillwater, Oklahoma Brian K. Callihan The Dow Chemical Company, Freeport, Texas Derek S. Arndt Oklahoma Climatological Survey, Norman, Oklahoma Steve C. Jenks Oklahoma State University, Stillwater, Oklahoma James R. Whiteley Oklahoma State University, Stillwater, Oklahoma Ronald L. Elliott Oklahoma State University, Stillwater, Oklahoma * Corresponding author address: Kala N Pandit, 111 Agricultural Hall, Department of Biosystems and Agricultural Engineering, Oklahoma State University, Stillwater, OK 74078; E-mail: pkala@okstate.edu 1. INTRODUCTION Wintertime maintenance of highways has been a crucial task for transportation authorities in North America and Europe. Development of new technologies and improved weather forecasting procedures to reduce icy conditions on roadways and bridges has been a subject of research since the early sixties (Highway Research Board 1964; Henderson 1963). Technology related to the heating of bridge decks and overpasses to combat preferential icing during freezing and sub-freezing conditions has drawn the attention of transportation authorities in the US since the seventies (Ferrara and Yenetchi 1976; Lee et al. 1986; Anonymous 1998). The Geothermal Smart Bridge (GSB) project, which is currently under development at Oklahoma State University (OSU) is aimed at developing heating systems for bridge decks using ground source heat pump technology. The overall mission of the GSB project is to "research, design, and demonstrate technically feasible, economically acceptable, and environmentally compatible Smart Bridge Systems to enhance the nation's highway system safety and to reduce its life cycle cost". The four major components of the bridge heating system are the ground loop heat exchanger, heat pump, bridge deck, and control system. This paper focuses on the first-generation control system that has been designed using computer modeling with weather data as the input. The basic premise of the first-generation control system module (CSM) relies on a rule-based system guided by the weather data available at nearby stations. However, on-site pavement/bridge-deck sensors can also be used to supplement the weather information. The control strategies for the ground source heat pumping, which are developed with input from real-time and near real- time weather information, are discussed in more detail in section 3 of this paper. 2. WEATHER DATA AND FORECAST PRODUCTS 2.1 Oklahoma Mesonet Data The Oklahoma Mesonetwork (Mesonet), which was developed through a partnership between the University of Oklahoma (OU) and OSU, is a network of 114 automated weather stations distributed around the State of Oklahoma at an average distance of 32 km between stations (Elliott et al. 1994; Brock et al. 1995; Shafer et al. 2000). At each station, thirteen atmospheric and subsurface variables are measured and 5-min data summaries are relayed every 15-min to a central processing site in Norman, Oklahoma. The data from each remote station are broadcast over radio waves to a nearby sheriff, police or highway patrol station. The data then enter the Oklahoma Law Enforcement Telecommunications System (OLETS) and are sent through the main OLETS office in Oklahoma City to the Oklahoma Climatological Survey (OCS) housed at the OU campus in Norman. Mesonet data have been found to be of high quality and useful for a variety of research applications (e.g., meteorology, hydrology, agriculture, etc.). Road Weather Information Systems (RWIS) are receiving considerable attention nationally, but a comprehensive system of this type has not yet been implemented in Oklahoma. The high space and time resolution of