The study period began June 30 and ended October 

 20, 1982. This time period was blocked into two periods 

 using August 15 as the midpoint so that each of the 

 combinations would be replicated. Within a block, the 

 four combinations were randomly assigned to four time 

 intervals, ranging from 10 to 14 days. The combinations 

 did not occur concurrently on any of the trails and were 

 never placed back to back, which v/ould have resulted in 

 one combination running for 20 to 28 days. Each time 

 interval included weekends or holidays when peaks in 

 visitor use were expected. Because visitor awareness of 

 the project could affect registration behavior, days on 

 which the signs were to be changed or relocated were 

 scheduled for the middle of the week when there were 

 fewer visitors. 



Data Collection 



Private party registration was determined by compar- 

 ing registration cards and outfitter schedules against a 

 photographic record that recorded actual use and regis- 

 tration behavior. The registration cards and outfitter 

 schedules were tabulated simultaneously with the film to 

 minimize confusion resulting from inaccurate informa- 

 tion. This worked better than tabulating each separately. 

 Registration cards were used to verify parties registered. 

 Outfitter schedules were used to help identify and elimi- 

 nate nonregistering commercial parties from the data. 



The film record was obtained from a camera system 

 developed by the USDA Forest Service Equipment De- 

 velopment Center in Missoula, MT (Gasvoda 1978). An 

 electronic trail traffic counter and camera were affixed to 

 trees 25 to 40 yards (23 to 36 meters) back from the 

 registration station. When the traffic counter infrared 

 beam was broken by a visitor, the camera was triggered 

 to expose two frames. At the 2-mile (3-km) location on 

 Pyramid Pass, placement and camouflage problems 

 prevented focusing on the registration station. At this 

 location, the camera was moved about 100 yards 

 (91 meters) down the trail. Visitors could be counted on 

 their way up to the station, but registration behavior 

 could not be observed and was inferred. 



RESULTS AND DISCUSSION 



Registration rates were influenced by location of the 

 registration station when all trails and user groups were 

 considered together (table 2). Moving the control sign 

 from the trailhead to a new location up the trail almost 

 doubled the registration rates. The experimental sign, 

 when located up the trail, resulted in a registration rate 

 of 69 percent, compared to 50 percent at the trailhead. 

 For signs at the farthest location, compliance rates were 

 60 percent for the control sign and 69 percent for the ex- 

 perimental sign. This difference is small, and is the only 

 comparison that is not highly significant statistically 

 (table 2). Registration rates must be examined separately 

 by user groups, however, before a decision to develop a 

 new sign for up-trail locations can be made. 



When data from all three trails were combined, and 

 the different types of user groups separated out, the new 

 sign farther up the trail ranked first for all users except 

 day horse riders (table 3). The day horse-user category 

 had so few observations that it was difficult to judge 

 which sign and location combination worked best with 

 this type of visitor. 



Day Hikers 



Day hikers responded very well to both signs located 

 up the trail. The new sign received 65 percent compU- 

 ance while the old sign received 64 percent. Location in- 

 fluenced the day hiker's registration rate more strongly 

 than either sign design. The new sign at the trailhead 

 did not elicit a high registration rate (44 percent), al- 

 though this was almost twice the rate for the old sign at 

 the trailhead. Day hikers may not consider themselves 

 as wilderness visitors and may not feel motivated to reg- 

 ister at the trailhead for this reason. For this group, see- 

 ing the registration station farther in the wilderness may 

 indeed symbolize wilderness entry (Lucas and Kovahcky 

 1981), and they respond favorably to registering. Of 

 course, some day hikers may never reach the registra- 

 tion station if it is located far up the trail, but other 

 monitoring techniques can be used to measure this use if 

 it is important. 



Table 2. — Combined registration data from Pyramid Pass, Owl Creek, and 

 Holland Lake trails^ for all visitor types 



Item 



Treatment A 

 0SNL2 



Treatment B 

 0S0L2 



Treatment 

 NSNL2 



Treatment D 

 NS0L2 



Number of parties 



110 



198 



135 



130 



Number registered 



66 



64 



93 



65 



Percent registration 



60 



32 



69 



50 



''statistical significance, tested by j^: 



1. Old sign by location (treatments A and B) {x^ = 22.2, df = 1, p<0.001) 



2. New sign by location (treatments C and D) (a-^ = 9.8, df = 1, p< 0.005) 



3. Old location by sign (treatments B and D) (^-2 = 10.3. df = 1, p<0.005) 



4. New location by sign (treatments A and 0) (jr^ = 2.1, df = 1, p<0.20). 

 ^OSNL = old sign, new location 



OSOL = old sign, old location 

 NSNL = new sign, new location 

 NSOL = new sign, old location. 



6 



