Weather analysis and short term forecasting in this nation currently relies on a couple of thousand automated weather stations reporting hourly. Imagine instead if there was the ability to collect millions of mobile observations at any time of the day or night from both urban and rural areas to support such analyses and forecasts. This exciting vision is part of a program by the U.S. Department of Transportation called the Vehicle Infrastructure Initiative (VII). The premise behind VII is that automobile manufacturers will soon be installing two-way radios in passenger vehicles that are connected to the computers that operate and monitor onboard systems. Roadside transceivers would be installed at many of the nation's intersections and along interstate and primary routes. Vehicles would be able to send “snapshots” of the status of internal systems back to the roadside infrastructure. These snapshots could include direct observations of atmospheric phenomena (such as ambient air temperature or windshield wiper state) or indirect observations of weather or pavement conditions through vehicle traction control or antilock breaking systems.

The Federal Highway Administration worked with Mitretek Systems to explore the feasibility of using observations from vehicles as a new, rich dataset for the weather and transportation communities. During the winter of 2005-2006, a project was launched that included several specially equipped vehicles that collected data during a variety of weather and road conditions. An analysis was performed that provided some initial estimates of temperature biases from vehicles versus in situ platforms. The analysis also explored whether other factors such as vehicle speed, traffic volume or different weather phenomena (such as sun angle, cloud cover, wind, or precipitation) had an effect on these readings. This paper will describe the methodology that was used for the data collection and provide summaries of many of these findings.