Estimating urban trail traffic: Methods for existing and proposed trails

The paper presents findings from the most comprehensive, continuous trail monitoring system in the United States, a network of 30 infrared monitors on five multiuse trails in Indianapolis, Indiana operating for periods of one to more than four years. We describe variation in traffic across different trails and segments of individual trails and present traffic ratios that describe variations in traffic by month, day of week, and time of day. We also present regression models for estimating traffic from temporal, weather, socio-demographic, and urban form variables. We propose and validate procedures for use of traffic ratios to extrapolate hourly counts to annual estimates and use of regression models to estimate traffic. Our procedures provide estimates within 20-30% of actual counts. Limitations of the models are noted. Potential applications range from forecasting traffic on proposed new trails to assessing the need for safety improvements such as stop lights at intersections.