|Abstract:||Domestic livestock grazing occurs in virtually all sagebrush habitats and is a prominent disturbance factor. By affecting habitat condition and trend, grazing influences the resources required by, and thus, the distribution and abundance of sagebrush-obligate wildlife species (for example, sage-grouse Centrocercus spp.). Yet, the risks that livestock grazing may pose to these species and their habitats are not always clear. Although livestock grazing intensity and associated habitat condition may be known in many places at the local level, we have not yet been able to answer questions about use, condition, and trend at the landscape scale or at the range-wide scale for wildlife species. A great deal of information about grazing use, management regimes, and ecological condition exists at the local level (for individual livestock management units) under the oversight of organizations such as the Bureau of Land Management (BLM). However, the extent, quality, and types of existing data are unknown, which hinders the compilation, mapping, or analysis of these data. Once compiled, these data may be helpful for drawing conclusions about rangeland status, and we may be able to identify relationships between those data and wildlife habitat at the landscape scale. The overall objective of our study was to perform a range-wide assessment of livestock grazing effects (and the relevant supporting data) in sagebrush ecosystems managed by the BLM. Our assessments and analyses focused primarily on local-level management and data collected at the scale of BLM grazing allotments (that is, individual livestock management units). Specific objectives included the following:
1. Identify and refine existing range-wide datasets to be used for analyses of livestock grazing effects on sagebrush ecosystems.
2. Assess the extent, quality, and types of livestock grazing-related natural resource data collected by BLM range-wide (i.e., across allotments, districts and regions).
3. Compile and synthesize recommendations from federal and university rangeland science experts about how BLM might prioritize collection of different types of livestock grazing-related natural resource data.
4. Investigate whether range-wide datasets (Objective 1) could be used in conjunction with remotely sensed imagery to identify across broad scales (a) allotments potentially not meeting BLM Land Health Standards (LHS) and (b) allotments in which unmet standards might be attributable to livestock grazing. Objective 1: We identified four datasets that potentially could be used for analyses of livestock grazing effects on sagebrush ecosystems. First, we obtained the most current spatial data (typically up to 2007, 2008, or 2009) for all BLM allotments and compiled data into a coarse, topologically enforced dataset that delineated grazing allotment boundaries. Second, we obtained LHS evaluation data (as of 2007) for all allotments across all districts and regions; these data included date of most recent evaluation, BLM determinations of whether region-specific standards were met, and whether BLM deemed livestock to have contributed to any unmet standards. Third, we examined grazing records of three types: Actual Use (permittee-reported), Billed Use (BLM-reported), and Permitted Use (legally authorized). Finally, we explored the possibility of using existing Natural Resources Conservation Service (NRCS) Ecological Site Description (ESD) data to make up-to-date estimates of production and forage availability on BLM allotments. Objective 2: We investigated the availability of BLM livestock grazing-related monitoring data and the status of LHS across 310 randomly selected allotments in 13 BLM field offices. We found that, relative to other data types, the most commonly available monitoring data were Actual Use numbers (permittee-reported livestock numbers and season-of-use), followed by Photo Point, forage Utilization, and finally, Vegetation Trend measurement data. Data availability and frequency of data collection varied across allotments and field offices. Analysis of the BLM‘s LHS data indicated 67 percent of allotments analyzed were meeting standards. For those not meeting standards, livestock were considered the causal factor in 45 percent of cases (about 15 percent of all allotments). Objective 3: We sought input from 42 university and federal rangeland science experts about how best to prioritize rangeland monitoring activities associated with ascertaining livestock impacts on vegetation resources. When we presented a hypothetical scenario to these scientists and asked them to prioritize monitoring activities, the most common response was to measure ground and vegetation cover, a variable that in many cases (10 of 13 field offices sampled) BLM had already identified as a monitoring priority. Experts identified several other traditional (for example, photo points) and emerging approaches (for example, high-resolution aerial photography) to monitoring. Objective 4: We used spatial allotment data (described in Objective 1) and remotely sensed vegetation data (sagebrush cover, herbaceous vegetation cover, litter and bare soil) to assess differences in allotment LHS status ("Not met" vs. "Met"; if "Not met" - livestock-caused vs. not). We then developed logistic regression models, using vegetation variables to predict LHS status of BLM allotments in sagebrush steppe habitats in Wyoming and portions of Montana and Colorado. In general, we found that more consistent data collection at the local level might improve suitability of data for broad-scale analyses of livestock impacts. As is, data collection methodologies varied across field offices and States, and we did not find any local-level monitoring data (Actual Use, Utilization, Vegetation Trend) that had been collected consistently enough over time or space for range-wide analyses. Moreover, continued and improved emphasis on monitoring also may aid local management decisions, particularly with respect to effects of livestock grazing. Rangeland science experts identified ground cover as a high monitoring priority for assessing range condition and emphasized the importance of tracking livestock numbers and grazing dates. Ultimately, the most effective monitoring program may entail both increased data collection effort and the integration of alternative monitoring approaches (for example, remote sensing or monitoring teams). In the course of our study, we identified three additional datasets that could potentially be used for range-wide analyses: spatial allotment boundary data for all BLM allotments range-wide, LHS evaluations of BLM allotments, and livestock use data (livestock numbers and grazing dates). It may be possible to use these spatial datasets to help prioritize monitoring activities over the extensive land areas managed by BLM. We present an example of how we used spatial allotment boundary data and LHS data to test whether remotely sensed vegetation characteristics could be used to predict which allotments met or did not meet LHS. This approach may be further improved by the results of current efforts by BLM to test whether more intensive (higher resolution) LHS assessments more accurately describe land health status. Standardized data collection in more ecologically meaningful land units may improve our ability to use local-level data for broad-scale analyses.