where that use occurs. These relationships are compli- 

 cated by the fact that the importance of both differences 

 in amount of use and differences between habitat types 

 varies with use intensity. In the forested habitat types, 

 differences in amount of use only affect vegetation loss 

 profoundly if use is less than about 300 passes per sea- 

 son. At higher use levels, differences in amount of use 

 have little effect on cover. Differences in the susceptibil- 

 ity of habitat types to vegetation loss, after one season 

 of use, are most pronounced at use levels of 300 to 900 

 passes. 



Managers can influence the spatial distribution of use 

 in two ways that can reduce impact. First, they can in- 

 fluence the vegetation types on which people camp, and 

 second, they can influence whether camping occurs on 

 sites that have been used frequently, infrequently, or 

 never before. 



The results of this study show that, in the study area, 

 camping will cause less vegetation loss in a grassland 

 than in a forest. This corroborates the results of a study 

 of existing sites in the Bob Marshall Wilderness (Cole 

 1983). If camping occurs in forest, it is best to camp in a 

 type with lots of Xerophyllum tenax. If frequent use of 

 sites is likely, sites might best be located in a forested 

 type like ABLA/CLUN. The fragiHty of vegetation 

 becomes irrelevant with heavy trampling because most 

 vegetation is eliminated from any site. Nevertheless, the 

 thick organic horizons of the ABLA/CLUN type make it 

 less prone to mineral soil exposure and severe soil 

 compaction. 



The apparent effect of local differences in species com- 

 position on amount of vegetation loss suggests that care 

 about where visitors walk, build a fire, and set up a tent 

 can have a great influence on amount of impact. If visi- 

 tors can recognize particularly fragile species (see 

 table 8) and avoid trampling them, vegetation loss could 

 be minimized. Again, this is only likely to be effective on 

 infrequently used campsites. 



For vegetation loss and soil compaction, aggregate im- 

 pact in any local area— a concern for both the severity 

 and areal extent of impact— will generally be minimized 

 when use is directed to whichever site has already been 

 most frequently used. The following example illustrates 

 this point. After treatment, the 100-pass lane of the 

 ABLA/CLUN type (a IG-ft'^ [l.S-m^] area) loses 61 per- 

 cent cover (table 4). The next increment of 100 passes 

 will be less damaging if applied to the 100-pass lane 

 rather than an unused lane. Two hundred passes on one 

 lane would remove 74 percent cover over a 1.5-m^' area— 

 12 ft^ (1.11 m-) of vegetation would be lost; 100 passes 

 each along two lanes would remove 61 percent cover 

 over a 32-ft'^ (3-m-) area— a loss of 20 ft-' (1.83 m') of 

 vegetation. In other words, less vegetation will be lost 

 when use is concentrated on one site as opposed to being 

 distributed over more sites. 



The exception to this is the FESC-FEID grassland 

 where it would be best to keep total use on any site 

 below 1,200 passes. At this level of use, no cover loss 

 occurs. Of course, trampling at this level year after year 

 might cause pronounced changes. We will be in a posi- 

 tion to evaluate this possibility after further seasons of 

 trampling. 



The curve relating mineral soil exposure to amount of 

 use appears to be a straight hne. That is, any incremen- 

 tal increase in use has about the same effect, regardless 

 of how much use the site has already received. This sug- 

 gests that aggregate impacts for mineral soil exposure 

 will be about the same whether use is concentrated or 

 spread among a number of sites. 



Although impacts appear to usually be minimized by 

 concentrating use on a few sites, minimizing amount of 

 ecological change is only one goal of management. 

 Esthetic concerns and the provision of opportunities for 

 solitude and primitive and unconfined recreation are also 

 important. It might be logical, for example, for manage- 

 ment to decide that a campsite is acceptable if vegeta- 

 tion cover exceeds 50 percent and mineral soil exposure 

 is less than 5 percent. We can see from tables 4 and 9 

 that to achieve this, use could be no more than 40 

 passes in ABLA/CLUN, 80 passes in PSME/SYAL, 100 

 passes in ABLA/CLUN-VACA, 150 passes in 

 ABLA/VACA, 700 passes in ABLA/XETE, and 1,200 

 passes in FESC-FEID sites. 



Obviously, to make such predictions useful it is neces- 

 sary to convert number of passes into a unit of use 

 managers can apply. This cannot be done exactly, but 

 by using a number of assumptions and some observa- 

 tions of camping activities it is possible to develop a 

 reasonable conversion. 



I counted numbers of tramples associated with build- 

 ing a fire and cooking a simple meal. In one-half hour of 

 meal preparation, one person took about 500 steps 

 within the immediate cooking area— an area of about 

 100 ft' (10 m-). From this, I estimated that in an over- 

 night stay, one person would take a minimum of 1,500 

 steps in this area; 500 at breakfast, 500 at dinner, and 

 500 during the other hours in camp. A number of studies 

 undertaken in diverse wildernesses (for example, 

 Heberlein and Dunwiddie 1979; Lucas 1980; Leonard and 

 others 1978) have shown that the most common party 

 size is two and median and mean party size is usually 

 three or four. If we take three as an average number, I 

 would estimate a minimum of about 4,000 steps in the 

 central 100 ft'^ (10 m') of the campsite. This is slightly 

 less than 3 times 1,500 because of work-sharing in any 

 cooking effort. Pressures exerted will be higher than this 

 because impacts caused by standing, sitting, and lying 

 are not included. Leney (1974) has shown that these ac- 

 tivities do have a significant impact, although they are 

 less destructive than moving about. Liddle (1975b) 

 states, for example, that the vertical force applied by a 

 standing man is about 2.8 Vol'w? (200 g/cm^) compared 

 with a force of 800 Ib/in^ (57,000 g/cm'^) applied when in 

 motion. Consequently, 4,000 steps per overnight stay by 

 a typical party of three should be considered a conserva- 

 tive estimate. 



The next problem is to relate this number of steps to 

 number of passes in the lanes during the experiment. 

 The treatments were administered in lanes that are 1 ft 

 (0.3 m) wide and 16.4 ft (5 m) long. Most people took an 

 average of seven steps in each lane, so each step occurs 

 in a 1- by 2.3-ft (0.3- by 0.714-m) area (2.3 ft^; 0.214 m^). 

 Therefore, during each pass, each 0.214-m^ area is 

 stepped on once. If we assume that 4,000 steps 



32 



