Table 1 — A classification of the 16 vegetation types 

 based on classification schemes developed 

 by Mueller-Dombois and Ellenberg (1974) 



and Vankat (1990) 



Closed Forests 



Evergreen coniferous forests 



Picea rubens-Abies subalpine forest 



Picea- Abies/ Lycopodium (White Mountains) 

 Picea- Abies/ Dryopteris (Smoky Mountains) 

 Picea engelmannii- Abies lasiocarpa subalpine forest 



Picea- Abies/ Vaccinium (Rocky Mountains) 

 Pseudotsuga menziesii-mlxed conifer upper montane 

 forest 



PseudotsugaJPachistima (Cascade Mountains) 



Cold-deciduous forest with evergreen trees 



Acer saccliarum-Fagus grandifolia-Betula alleghaniensis 

 forest 



Northern hardwood//.eers/a (White Mountains) 

 Northern hardvjood/ Maianthemum (White Mountains) 

 Gray beech/Carex (Smoky Mountains) 

 Populus tremuloides forest 



Popuius/ Geranium (Rocky Mountains) 



Cold-deciduous forest without evergreen trees 



Cove hardvjood/ Amphicarpa (Smoky Mountains) 



Dwarf-Scrub Communities 



Evergreen dwarf scrub 

 Subalpine heath 



Phyllodoce (Cascade Mountains)^ 



Terrestrial Herbaceous Communities 



Meadow and grasslands 



Below timberline (anthropogenic) 



Potentilla old-field (Smoky Mountains)^ 

 Above timberline 



Carex bigelowii (White Mountains)^ 

 Carex nigricans (Cascade Mountains) 

 Kobresia (Rocky Mountains) 

 Trifolium (Rocky Mountains) 



Perennial forb communities 



Valeriana (Cascade Mountains)^ 



' These vegetation types have similarities to other vegetation types 

 as well. Some of the plots in the Phyllodoce and Valeriana types 

 have an open overstory of Picea engelmannii- Abies lasiocarpa. The 

 Potentilla old-field is reverting to a deciduous forest. The Carex 

 bigelowii meadow has a minor dwarf-scrub component. 



ground cover in the Smoky Mountains gray beech for- 

 est. Most of the forb ground covers consisted of species 

 that were erect and caulescent (with a definite leafy 

 stem); however, the forbs that provided most of the 

 cover in the Trifolium meadow in the Rockies were 

 prostrate or scapose (without definite leafy stems). 



EXPERIMENTAL METHODS 



The experimental design follows the standard pro- 

 tocol suggested by Cole and Bayfield (1993). Four 



replicate sets of experimental trampling lanes were 

 established in each vegetation type. Each set consisted 

 of five lanes, each 0.5 m wide and 1.5 m long. Where 

 the ground was sloped, lanes were oriented parallel 

 to contours. Slopes were never more than 10 percent. 

 Treatments were randomly assigned to lanes. One 

 lane was a control and received no trampling. The 

 other lanes usually received either 25, 75, 200, or 500 

 passes. Treatments in the highly resistant C. nigricans 

 (Cascades) and Kobresia (Rockies) types were 75, 200, 

 500, and 700 passes. A pass was a one-way walk, at 

 a natural gait, along the lane. Tramplers weighed 

 about 70 kg. They wore lug-soled boots. 



Measvirements were taken on each lane in two adja- 

 cent 30- by 50-cm subplots. The following parameters 

 were measured: 



1. The cover of each vascular plant species and of 

 lichens and mosses. Visual estimates were recorded 

 as the closest of the following values: 0, 1, 5, 10, 15, 

 20, 30, 40, 50, 60, 70, 80, 90, or 100 percent. 



2. The cover of bare ground (ground not covered by 

 five vegetation). Visual estimates used the same val- 

 ues as for individual species. 



3. Mean vegetation height. We used a point quad- 

 rat frame v^dth five pins, each 5 cm from the next. 

 The fi-ame was placed 10 times, systematically, along 

 the length of the subplot. The pins were dropped to 

 the ground. When the pin hit bare ground, a was 

 recorded. When it hit live vegetation, the height of 

 the pin strike was recorded to the nearest 1 cm. 



Trampling treatments were administered in early 

 summer of 1988. Initial measurements were taken 

 before trampling. FoUowup measurements were taken 

 shortly after trampling and 1 year after trampling. 

 Height measurements were taken immediately after 

 trampling. Cover estimates were taken 2 weeks after 

 trampling. This lag made it easier to distinguish dam- 

 aged but living vegetation from dead vegetation. 



DATA ANALYSIS 



The types of vegetation change described are (1) the 

 amount of bare ground, (2) vegetation cover, (3) veg- 

 etation height, (4) species richness (the number of 

 species), and (5) species composition. Indices of dura- 

 bility were determined for the vegetation types. The 

 responses of individual species were also described. 



Bare Ground 



Mean bare ground (the proportion of the measure- 

 ment plot not covered by Hve vegetation) is presented, 

 before and after trampling and after 1 year, for each 

 trampling intensity between and 500 or 700 passes. 

 This provides a straightforward measure of changes 

 in vegetation cover after trampling. Bare ground 

 should not be confused with exposed mineral soil. 



2 



