The invasive winter–annual cheatgrass (Bromus tectorum) has degraded public and private lands throughout the Western United States, particularly in the Great Basin region of Utah (Knapp 1996). Cheatgrass invasion supports a loss of native vegetation, heightened fire risk and decline in biodiversity has potentially severe implications for land managers, wildlife and the general public (see reviews in Knapp 1996, Zouhar 2003). In likely climate change and land use change scenarios, cheatgrass cover may spread even more than the current extent due to the tolerance of cheatgrass to disturbance and drought (Chambers and Pellant 2008).

In the United States, it is estimated that losses due to invasive species cost upwards of $120 billion dollars per year (Pimental et al 2005). In the Western United States, cheatgrass has become a major invasive plant. In the Great Basin region alone, the Bureau of Land Management (BLM) estimates that cheatgrass has spread to over 1/3 of the total land area, with more and more acres being invaded yearly. While the total cost of cheatgrass invasion is difficult to determine, its contribution to ecosystem degradation, loss of habitat and forage for wild and domestic animals, losses due to cheatgrass-fueled fires coupled with resources spent on habitat restoration and fire suppression has made it clear that it has made a significant impact on the regional as well as national economy and ecology.

Following the ten–year monitoring report for nesting raptors in Northwest Utah by HawkWatch International, Inc. (HWI) (available here [see “for further reading’ for full URL]) suggested cheatgrass dominated habitats may act as an ecological “trap” for Burrowing Owls (Athene cunicularia) and Ferruginous Hawks (Buteo regalis). An ecological trap is an in appropriate attraction to poor quality habitat (Gates and Gysel 1978). Research from southern Idaho and northern Utah suggests that cheatgrass vegetation my support lower abundances of prey species, such as ground squirrels, black-tailed jackrabbits and other small mammals. (U.S. Department of Interior 1996, Hanser and Huntly 2006, Steenhof et al. 2006) Because of this potential depression in prey availability, areas that are invaded by cheatgrass could have potential negative implications for a variety of raptor species that rely on these lands for resources.

The study region is comprised mostly of public land (BLM) and inholdings of DoD installations, defined loosely by the boundaries of the Military Operating Area (MOA) in the West Desert of Utah. This area is also home to a diverse raptor population, which has been monitored over the past few decades by non–profit organizations (HWI and the Raptor Inventory Nest Survey [RINS]), volunteers, the DoD and BLM. The Utah Legacy Raptor Project (ULRP) partnership, consisting of military (Natural Resources staff from Dugway Proving Ground, Hill Air Force Base), non-profit (HWI and Raptor Inventory Nest Survey [RINS]), and agency personnel (i.e., BLM, US Fish and Wildlife Service [USFWS], Utah Division of Wildlife Resources [UDWR]) was formed in 2009 to further research this important issue for three focal species: the Golden Eagle (Aquila chrysaetos), Ferruginous Hawk, and Burrowing Owl. The Ferruginous Hawk and Burrowing Owl are open-country raptors adapted to native grasslands, shrubsteppes and deserts of the Western US and currently are Utah State Species of Concern (Haug et al. 1993, Bechard and Schmutz 1995, UDWR 2004). The third study species, the Golden Eagle, is not a species of concern in Utah; but research in Southern Idaho, Northern Nevada, and North Central Utah has suggested populations are declining in the West. Golden Eagles have received increasing regulatory attention recently as a result of issues with wind development siting.

HawkWatch International developed methods and directed the analysis to investigate the long-term effects of cheatgrass invasionof landscapes on raptors. In order to achieve our results, we needed to answer the following questions:

• Where do raptors nest in the study area?
• Where does cheatgrass occur in the study area?
• What geographical and environmental features are associated with cheatgrass coverage and invasion?
• What geographical and environmental factors are associated with raptor habitat?
• What spatial and prey resources support raptor populations in the study area?
• Is there a difference in nesting success and productivity between nests that occur in uninvaded areas and areas that are currently invaded by cheatgrass?
• What areas do we know little about in terms of raptor use and nesting?
• Are there potential areas to target for conservation (e.g. areas that are currently uninvaded by cheatgrass that might be good habitat for raptors or areas that seem to be unimportant to raptors that might be targeted for recreation/military use?)

HWI led the effort to compile nesting data from the project partners, which involved sorting through over 10 years of data for all agencies and non profits involved. All in all, there were over 14,000 nesting records compiled, from which we created territories by grouping nests according to nesting histories and local field knowledge. From these data, we were able to determine 167 Golden Eagle, 124 Ferruginous Hawk, and 282 Burrowing Owl territories. We calculated territory metrics for each species, including inter-nest distance and inter-territory distance. From our data, we can extrapolate estimates for breeding population size for each species. We multiplied the average activity per surveyed territory by total concurrently known territories to estimate that there are an average of 90 Golden Eagle pairs, 44 Ferruginous Hawk pairs and 56 Burrowing owl pairs during the study period.

While investigating trends within the study area, we discovered that Golden Eagle nesting activity declined from around 50% to around 25% after a particularly high number of large wildfires within the study area (below). This trend represents a potentially alarming decline for this species, given the regulatory interest and potential decline in eagles today.

The nesting sites mentioned above were then used to create predictive habitat maps for each species, modeling both nesting habitat and surrounding habitat that is most similar to areas known to be important for foraging. To achieve this, HWI scientists used MaxEnt, a habitat modeling program that excels at using small sample sizes when true absences area unknown (Elith et al 2011). These habitat models had misclassification rates of 9%, 13% and 17% for Golden Eagles, Ferruginous Hawks and Burrowing Owls, respectively. The next step was to create a Discriminant Function Analysis (DFA; McGarigal et al. 2000) that indicated the importance of habitat types to areas surrounding known nests. From this analysis, we were able to rank areas within the original MaxEnt model that are surrounded by the highest-quality foraging habitat. These habitat maps have helped project partners to direct survey effort, and also give agency land managers potential areas within their jurisdiction for conservation management.

Left to right: Golden Eagle, Ferruginous Hawk and Burrowing Owl predicted habitat. Yellow corresponds with results from MaxEnt modeling, depicting nesting habitat. Red corresponds with areas within the MaxEnt model that are also surrounded by landscapes most similar to known nesting territories.

The next step of our analysis involved determining where in the West Desert we can expect to find cheatgrass. Visually, in some areas, it appears that cheatgrass is virtually ubiquitous, but a strong scientific approach was needed in order to assure robust consideration of all environmental factors. The remote sensing and GIS lab at Utah State University performed the remote sensing to determine cheatgrass coverage within the study area, using MODIS data.

USU selected images that corresponded with times of green up, peak greenness and senescence of cheatgrass, an effective way of analyzing cheatgrass coverage because cheatgrass has much earlier phenology than native flora within the region. These images were then used to model cheatgrass distribution within the study area, creating continuous maps of potential cheatgrass coverage for each year of the study (1998-2012) based on NDVI (Normalized Difference Vegetation Index), an index that differentiates vegetation from surrounding landscapes. Hawkwatch scientists used these maps to create binary maps of cheatgrass coverage, with “1” indicating areas that are covered by cheatgrass, and “0” indicating areas that are free of cheatgrass invasion. We then mosaicked these maps into four time periods, due to the considerable variability between years. These collated images where then used to determine which environmental factors were associated with cheatgrass invasion, which made it possible to determine factors associated with cheatgrass invasion. We followed methods set forth by Bradley and Mustard (2006), creating “bins” for each environmental variable and calculating probability of cheatgrass occurrence in each, thus assessing coverage relative to elevation, aspect, slope, average precipitation, soil water capacity, distance to human landscapes, and distance to linear features. Therefore, we could map areas that were currently uninvaded, but at high risk of invasion based on these factors.

An example of cheatgrass coverage in relation to one environmental factor, distance to human dominated landscapes. From this graph, we can determine that as one travels farther from human landscapes, the probability of cheatgrass coverage decreases.

The intersection of cheatgrass invasion maps and habitat maps helped to highlight areas of relative importance for land managers (see below). In the central to southeast corner of the study area, in the Fillmore and Cedar City BLM Field Office jurisdictions, there are large sections of the landscape that are considered high risk for cheatgrass invasion. These areas also correspond to regions that we know relatively little about in terms of raptor nesting and resource use. HawkWatch International hopes to use this information to leverage conservation efforts at a larger scale within the Great Basin region.

This map indicates the intersection between Golden Eagle predicted territories, currently known nesting territories and cheatgrass invasion potential. Note that in the central and south study area, there is little known about nesting territories, and there is a high percentage of the landscape that is at high risk of invasion with no current cheatgrass coverage.

The final piece of our analysis was to determine if there was a difference in productivity and nest success between areas in low, medium and high cheatgrass. We found that there was an association between higher cheatgrass coverage and lower activity for each of the focal species, but we determined that Golden Eagles are most affected by this. Because Golden Eagles rely on a prey base that is comprised of at least two reliable mammal species populations (typically this is black-tailed jack rabbits and one other population), loss of shrub cover due to fires and transition to cheatgrass monoculture could potentially prevent the establishment of a proper prey structure for these species, and therefore contribute to overall decline in nesting activity and success.

This presents a more nuanced story of cheatgrass invasion and its effects on raptor productivity, in that the depression we see in nesting starts and success is likely due to shrub loss from intense fires and decrease in prey base, and not necessarily due to our original ecological sink hypothesis; that cheatgrass monoculture falsely presents a habitat that appears to be suitable habitat for nesting raptors. The spread of cheatgrass is also a result of increased fire, which in turn, prevents the development of shrubs necessary to support resources for Golden Eagles.

Similar to the previous map, this map shows predicted areas of cheatgrass presence and areas at risk of invasion. Overlaid on top are fire polygons for each year between 1998-2010, provided from the BLM. Please note that this does not include fires that occurred on DoD property.

This project would not have been possible without the support of the Department of Defense Legacy Natural Resources Management Program, who provided funding for all three years of the study. Additionally, we would like to thank our project partners; the natural resources staff at Dugway Proving Ground and Hill Air Force Base, the volunteers and staff at Raptor Inventory Nest Survey, Kent Keller, the Utah Department of Wildlife Resources, the Salt Lake City Field Office of the Bureau of Land Management and US Fish and Wildlife Service.

Through our cooperative efforts, we were able to achieve the following objectives, many of which comprise firsts for large-scale conservation efforts in the Great Basin region:

• Creation of a partnership comprised of 2 DoD installations (i.e., DPG and HAFB), two raptor-focused not for profit organizations profits (i.e., HWI and RINS), BLM, USFWS and UDWR.
• Compilation of multiple years of nest activity data for each of the focal species into one database, with some records dating back over 30 years and encompassing almost 40,000 square kilometers, the largest geographical long–term dataset compiled for these species.
• Compilation of over 14,000 previously unassembled nesting records (1998–2011), representing over 1,000 nests and 573 “territories” or nest clusters for Golden Eagles, Ferruginous Hawks, and Burrowing Owls for the entire study area. This involved combining historical records, previous reports, and field-based observations to place nests into associated clusters.
• Mapping of the previously mentioned territories, providing a visual representation of where efforts have been undertaken and where birds have been found. Additionally, calculation of landscape metrics for each of the focal species, depicting territory and nest spacing.
• Creation of predictive habitat maps for all three study species for the study area.
• Creation of cheatgrass coverage maps from MODIS images for all years of the study period (1998-2010). Completion of statistical and geospatial analysis of cheatgrass invasion risk, including landscape risk factors and predictive maps of potential risk factors associated with invasion potential.
• Analysis of cheatgrass invasion risk potential associated with currently occupied and predicted habitat. This analysis and resulting maps offers significant management implications for land managers within the study area, highlighting areas of invaded and high future invasion risk relative to raptor habitat.
• Development of standardized protocols that can be used in other regions for organizations and agencies to use to monitor raptor populations. This document includes model data sheets, nest chronology information for our region by species (includes other species besides the ULRP focal species), suggestions for safe field practices, and information on how to complete surveys without disturbing nesting birds.
• Development of protocols for use in engaging citizen scientists in large-scale and long-term monitoring projects for the collection of data valuable to decision makers.
• Development of a list of management recommendations for use by federal and state managers for each of the focal species.

________

This publication is a product of the ULRP’s final report which describes in detail the results of two years of our research. That publication is:
Slater, S. J., K. W. Frye Christensen, R. N. Knight, R. MacDuff, and K. Keller. 2012. Great Basin Avian Species-at-risk and Invasive Species Management through Multi-Agency Monitoring and Coordination Final Report. Department of Defense, Legacy Resources Management Program (Project #10–102).

For further reading:

• HawkWatch International: www.hawkwatch.org
• HawkWatch International’s Ten-Year Report on Nesting Raptors in Northwest Utah: http://hawkwatch.org/images/stories/Conservation_Science/Publications_and_Reports/Technical_Reports_-_Current_Projects/GBRNS_UT_10yr_report_2010.pdf
• The Legacy Final Report, prepared by HawkWatch International. Available soon at www.hawkwatch.org
• Legacy ULRP Nest Monitoring Manual: Available soon at www.hawkwatch.org
• Legacy Resources Program: https://www.dodlegacy.org/legacy/intro/about.aspx

Literature Cited:

• Bechard, M. J., and J. K. Schmutz. 1995. Ferruginous Hawk (Buteo regalis). In The Birds of North America, No. 172 (A. Poole and F. Gill, eds.). The Academy of Natural Sciences, Philadelphia, and the American Ornithologists’ Union, Washington D.C.
• Bradley, B.A. and J.F. Mustard. 2006. Characterizing the landscape dynamics of an invasive plant and risk of invasion using remote sensing. Ecological Applications 16:1132–1147.
• Chambers, J. C., N. Devoe, and A. Evenden, Angela (eds.). 2008. Collaborative management and research in the Great Basin — examining the issues and developing a framework for action. Gen. Tech. Rep. RMRS-GTR-204. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 66 pp
• Elith, J., S. Phillips, T. Hastie, M. Dudik, Y. En Chee, and C. Yates. 2011. A statistical explanation of MaxEnt for ecologists. Diversity and Distributions 17:43–57
• Gates, J. E., and L. W. Gysel. 1978. Avian nest dispersion and fledging success in field-forest ecotones. Ecology 59:871–883
• Hanser, S. E., and N. J. Huntly. 2006. The biogeography of small mammals of fragmented sagebrush-steppe landscapes. Journal of Mammalogy 87:1165–1174.
• Haug, E. A., B. A. Millsap, and M. S. Martell. 1993. Burrowing Owl (Speotyto cunicularia). In the Birds of North America, No. 61 (A. Poole and F. Gill, eds.). The Academy of Natural Sciences, Philadelphia, PA, and the American Ornithologists’ Union, Washington D.C.
• Knapp, P. A. 1996. Cheatgrass (Bromus tectorum L.) dominance in the Great Basin desert. Global Environmental Change 6:37–52.
• McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate statistics for wildlife and ecology research. Springer-Verlag, New York, NY.
• Pimental, D., R. Zuniga, and D. Morrison. 2005. Update on the environmental and economic costs associated with invasive species in the United States. Ecological Economics 52:273–278.
• Steenhof, K., E. Yensen, M. N. Kochert, and K. L. Gage. 2006. Populations and habitat relationships of Piute ground squirrels in southwestern Idaho. Western North American Naturalist 66:482–491.
• UDWR (Utah Division of Wildlife Resources). 2004. Utah sensitive species list. Utah Department of Natural Resources, Utah Division of Wildlife Resources, Salt Lake City, Utah. 7 pp + appendices.
• U.S. Department of the Interior. 1996. Effects of military training and fire in the Snake River Birds of Prey National Conservation Area. BLM/IDARNG Research Project Final Report. U.S. Geological Survey, Biological Resources Division, Snake River Field Station, Boise, ID. 130pp.