Twelve months of automated hourly pedestrian counts in downtown Montpelier, Vermont (pop. 8,035) were analyzed along with weather data (temperature, relative humidity, precipitation, and wind) to determine the factors affecting count variability. The study is unique in that a large amount of data in a single location were collected in a locale with a very large rangeof weather conditions. Results indicate consistent patterns in relative volumes by hour of the day and week of year that support the fact that good adjustment factors can be developed to use with time-limited counts to estimate usage and pedestrian exposure to accidents. However, weak predictive relationships were found between weather variables, season and pedestrian volumes (no more than 30% of variation and 20% of volume are affected). Precipitation and season do impact pedestrian levels even when time of day and day of week are controlled, but other larger unmeasured factors are at play. Even in a relatively severe climate the impact of weather on walking levels in a business and commercial downtown is large enough to consider programs and counter measures that might increase walking in adverse weather.
Abstract