Sampling characteristics of the bus route survey technique in the James River, Virginia

The bus route survey technique is a new on-site angler survey technique that was developed for small rivers, with remote access points and low angler use. The technique employs vehicle counts to collect angler effort information. Interviews are conducted at access points to collect user characteristic and catch data from arriving, mid-trip, or departing anglers. This technique was modified to sample total recreational use on 306 km of the James River from Glen Wilton to Richmond, Virginia. The river was spatially stratified into 6 areas, two urban areas (Areas 5 and 6), and four rural areas (Areas 1-4). The two-year study was conducted from March through November, in 1988 and 1989.

In order to compare the bus route technique in a large riverine system, I conducted four 4-day intensive sampling days, two weekend and two week days, in Areas 1 and 4 during each year. During these periods a complete as possible access-point survey was conducted simultaneously with the bus route survey. Surveyors for both techniques collected effort, user characteristic, and catch data from several user groups. Aerial flights were also conducted to estimate effort during the 4-day intensive sampling periods. Effort estimates of the bus route and complete access-point surveys were similar in both areas, however aerial surveys collected 205 (29%) more hours of effort per sample because it included users accessing the river through undefined or private access points. Data collected on most user characteristics were not significantly different among methods. In three of four sampling periods catch was not significantly different between the bus route and complete access-point techniques, but in 1988, Area 4, catch was statistically higher for the complete access-point surveys. To solve this problem of underestimating case with the bus route survey more afternoon samples must be conducted to intercept departing anglers.

In areas with more than 15 access points or driving time between access points is longer than half the survey day, the bus route can be extended over a two-day period, or a sub-sample of a number of access points can be surveyed. A 4-day intensive sampling period was conducted in 1989, Area 1 during the summer season to compare effort, user characteristics, and catch data between two-day, sub-sampling, and complete access-point surveys. Also, 26 sampling days were conducted in 1989, Area 1 during the summer season to compare effort, user characteristics, and catch between two-day and sub-sampling bus route techniques. No significant difference in effort, and certain (9) user characteristics, or angler catch data was detected between bus route modifications or the complete access-point survey. Also, no significant differences in effort, user characteristics (11), and angler catch variables (2) were detected between the two-day and sub-sampling methods that were conducted throughout the entire summer period.

The bus route survey was designed to have equal daily probabilities. When unequal daily probabilities are used, the daily effort formula becomes complex. Daily effort, user characteristics, and catch data were compared from interviews conducted in morning and afternoon samples in 1989, Area 4 during the summer season. Results showed no statistical difference in effort although a mean of 72 (52%) more hours of effort were collected per afternoon sampling period. Most user characteristics were not significantly different, but a larger proportion of departing users were interviewed in the afternoon period (54%) then the morning period (22%). Because only two departing interviews in the early sampling period were anglers, catch estimates could not be compared between the two periods. If collecting catch data is an important survey objective, then more afternoon bus route sampling periods must be conducted.