(1 night's use) are equally distributed throughout the 

 central 100 ft- (10 m-) of the campsite, then each 0.214 m- 

 area of the central campsite is trampled 87 times— the 

 equivalent of 87 passes. This should be a conservative 

 estimate, but let us assume that 75 to 150 passes will 

 simulate the amount of trampling that occurs in 1 night 

 of use by a typical party of three in the central 100 ft- 

 (10 m^) of a campsite. Peripheral parts of the campsite 

 would receive less use. 



We tested these estimates by measuring vegetation 

 response to actual camping on two of these vegetation 

 types. Two people spent 1 night each on previously un- 

 disturbed sites near the trampling plots in the ABLA/ 

 CLUN and ABLA/XETE types. Cover measurements, 

 similar to those taken on trampling plots, were taken on 

 nine quadrats systematically dispersed in a 100-ft- 

 (lO-m'^) area around a fire ring and nine around a tent 

 site in each type. Measurements were taken before and 

 after the 8-week trampling period; the night of camping 

 occurred in the middle of this period. 



Relative cover around the fire ring after 1 night of 

 camping was 45 percent in ABLA/CLUN and 78 percent 

 in ABLA/XETE. From table 4 and figure 10, this level 

 of impact corresponds most closely with 100 passes in 

 each of these types. The relative cover of the most abun- 

 dant species around the fire ring was comparable to that 

 found after 75 to 200 passes, depending on the species. 

 Thus, the estimate of 75 to 150 passes as simulating the 

 effect of a typical party in the central part of the camp- 

 site appears realistic. 



Relative cover around and under the tent was 92 per- 

 cent in ABLA/CLUN and 95 percent in ABLA/XETE. 

 In both cases this corresponds most closely with the 

 effect of 15 passes. For the most abundant species, rela- 

 tive cover was comparable to that found after 5 to 25 

 passes, depending on the species. Thus, the effect of 5 to 

 25 passes appears to simulate the effect of a typical 

 party around the tent. 



A large party of, say, 12, although four times as large, 

 would probably not trample any given area four times as 

 much. Per capita impacts on any square meter will be 

 less than with a small party because tasks are shared. I 

 estimate that the central area would be trampled three 

 times as much as with a party of three, an intensity 

 equivalent to about 250 to 400 passes. Because people in 

 a large party must spread out more, however, this high 

 level of trampling on the campsite occurs over a much 

 larger area (i.e., the effects of the 250 to 400 passes will 

 apply to an area perhaps two to three times as large as 

 100 ft- [10 m-]). 



Using these estimates, acceptable use frequencies for 

 the average party are: no use for ABLA/CLUN; 1 night 

 of use for PSME/SYAL, ABLA/CLUN-VACA, and 

 ABLA/VACA; 5 to 8 nights for ABLA/XETE; and 10 to 

 15 nights for the FESC-FEID grassland. Campsites used 

 by large parties would only be acceptable in 

 ABLA/XETE (2 to 3 nights of use) and FESC-FEID 

 (3 to 4 nights of use). 



Clearly, such predictions should be treated cautiously 

 and conservatively. Cumulative changes from prolonged 

 trampling will also probably be more severe, making 

 desirable use frequencies even lower. Nevertheless, the 

 estimates do compare favorably with measures of vege- 

 tation loss on campsites in similar vegetation types 

 receiving estimated amounts of use (Cole and Fichtler 

 1983). 



In addition to corroborating the appropriateness of 

 management suggestions derived from earlier studies of 

 existing campsites (Cole 1982c, 1983; Cole and Fichtler 

 1983), this study provides more precise quantification of 

 use levels, for individual habitat types, at which disper- 

 sal or concentration is likely to be appropriate. Except 

 in the grassland, concentration is always better at 

 minimizing ecological change, particularly vegetation 

 loss. If objectives such as allowing no more than 50 per- 

 cent vegetation loss are developed, however, dispersal 

 can be desirable, particularly where use levels are not 

 high and vegetation is resistant (for example, in the 

 FESC-FEID and ABLA/XETE habitat types). 



Exactly what these use thresholds are will depend 

 upon stated objectives. The previous example, using an 

 objective of no more than 50 percent vegetation loss and 

 5 percent soil exposure, shows how this can be done. 

 But there is no magic in these objectives. Managers may 

 want to maintain more or less cover than this. The point 

 is that we are getting closer to a point where we can 

 predict quantitatively the effects of alternative use con- 

 figurations on campsite conditions. 



In assessing the applicability of these results else- 

 where, it is worth reiterating that the FESC-FEID 

 grassland is the most resistant vegetation type and 

 ABLA/XETE is the most resistant forested type studied 

 experimentally. The graphs in figures 1 and 10 can be 

 used in areas with similar vegetation types to assess the 

 likely effects of initial use of undisturbed sites. For some 

 unstudied ecosystem types, such as deserts, however, 

 there are no available estimates of the effects of given 

 amounts of use. 



33 



