Plant Identification. Plant nomenclature and nativity followed the USDA Plants Database (USDA, NRC 2004). Where possible, all plants were identified to the specific level. When field identification was not reliable or hybridization was suspected, the plants were identified to the generic level. Notable examples of this are the broad-leaved Chenopodium spp., all Lotus spp., and all Solidago spp. In each of these instances, all taxa were only identified to the genus. It is highly likely that the broad-leaved members of the Chenopodium genus include the nonnative species Chenopodium album L. var. album. There are, however, three other species of broad-leaved Chenopodium present in this region which are difficult-to-impossible to reliably distinguish in the field. These species are C. fremontii S. Wats., C. berlandieri Moq., and C. incanum (S. Wats.) Xxxxxx which are all native to the United States. We have chosen a conservative identification and consider all of the Chenopodium spp. as native. It should be recognized that C. album var. album may be present on our plots and therefore the numbers we report for nonnative species are potentially low. Cattle Grazing Mt. Trumbull has been grazed by domestic livestock at various intensities since Euro- American settlement in the late 1800’s. Cattle grazing was excluded from the study area upon initiation of the restoration treatments. In July 2002, 64 head of cattle were reintroduced and grazed through October of the same year. The following year, 76 head of cattle were grazed during the same months as in 2002. Data Analysis Our study incorporates a Before-After/Control-Impact (BACI) design (Xxxxxxx-Xxxxx and Xxxxxxx 1986, Green 1993, Xxxxxxxxx 1994). While this design is not consistent with true replication and randomization (Xxxxxxxx 1984), it allowed us to examine ecosystem response to restoration treatments across a large landscape (Xxx Xxxxxxx et al. 2001). The treatment area was organized into separate land units based on topography and management objectives (Fig. 2). Restoration treatments were implemented based on the geographic boundaries of these management units. For the purposes of this study, each study plot was analyzed as an independent sample point. We have not attempted to account for spatial autocorrelation in our analyses. Percent frequency was calculated by dividing the number of plants recorded on both line transects by 332 (the number of possible hits per each plot). The data were not consistently normally distribut...
Plant Identification
