Invasive annual grasses are proliferating across western rangelands at an alarming rate, leading to altered fire regimes, degraded wildlife habitat and loss of forage for wildlife and cattle.^1,2 Land managers cannot keep pace with the unrelenting waves of invasion by exotic annual species (such as cheat-grass [Bromus tectorum L.], medusahead [Taeniatherum caputmedusae (L.) Nevski], and ventenata [Ventenata dubia (Leer) Coss.]).³ The continued expansion of invasive annual grasses is a complex ecosystem management problem. The problem has multiple interactive social and ecological causes, strong self-perpetuating properties, ecosystem level effects, and is within a temporally and spatially variable environment. As such, no single “silver-bullet” solution exists. Instead, an effective multiyear management strategy varying from site-to-site is needed. In addition, the invasive annual grass issue is too complex and large to be solved by any single organization and significant knowledge gaps continue to persist.^4,5 This “wicked” problem will persist and require continued adaptive management even if land managers were granted unlimited resources and discretion to address it.⁶ This is not to say we do not need additional resources; knowledge-limited problems that vary markedly over time and space require sustained investment in adaptive management and research. Tackling this problem requires prolonged investment (see Smith et al. this issue)⁷ from a diversity of actors with complementary capacities and supportive policies to shift focus from an event-based “silver-bullet” approach to managing invasive annual grasses annually (see Johnson et al. this issue).⁸

A dominant focus of this Special Issue is on implications of scientific discoveries for improving the outcomes of invasive annual grass management and sagebrush ecosystem restoration in the northern Great Basin of the western United States (e.g., Davies et al.; Baughman et al. this issue).^9,10 This science-centric focus demonstrates significant knowledge gaps exist surrounding management of invasive annual grasses. However, science plays an important role in addressing invasive annual grass management. Scientific discovery often only informs a logical starting point for management because researched management practices do not always align with the fiscal, logistical, and societal realities of large-scale management of invasive annual grasses across mixed ownership landscapes.¹¹ Furthermore, researchers frequently conduct science within a small range of temporal and spatial variation compared with land managers who contend with large geographies and long planning horizons.¹² Experimental designs including treatment replication over space and time broaden inferences about management efficacy across wider climatic, environmental, and biotic conditions. However, they only account for a small fraction of the whole range of ecosystem variability.Such research is not conducted in vain and is important to informing viable alternatives. In practice, however, scaling up implementation to management scales often produces results not aligned with expectations. Furthermore, many studies of land treatments that failed are underreported in the peer-reviewed literature,¹³ which contribute to an inflation of expected efficacy from land treatments. Tackling the complex problem of invasive annual grass management requires a science-based approach, strong partnerships, and a platform for continuous learning and communication.^4,5,11,14

We suggest more case studies of real-world management efforts are needed to expand existing knowledge for informing the efficacy of invasive annual grass management strategies and treatments under a wider range of conditions. The value of these case studies is maximized when they are designed to address specific management/research questions from the start. Therefore, we suggest, as others have, closer collaboration between researchers and land managers (and other stakeholders), and better integration of research within the management process where science evaluates the effectiveness of multiple strategies when and where management opportunities arise.⁵ Additionally, we advocate for a dynamic management database to capture the variability and successes or failures of past treatments and provide ongoing learning between managers and researchers.¹⁴

A case for more case studies

Case studies offer an opportunity to evaluate the efficacy of current science at management scales, further inform scientific research by tightening the link between science and management,¹⁵ and cross-pollinate lessons learned among agencies.¹¹ We use four management case studies shared at the 2020 Invasive Annual Grass Workshop organized by the High Desert Partnership, the SageCon Partnership, and Oregon State University, to highlight lessons learned and opportunities to improve success with the complexities of managing invasive annual grasses. These case studies were conducted in Oregon and shared by the Burns District Bureau of Land Management (BLM), Vale District BLM, Burns Paiute Tribe, and Crooked River Weed Management Area. These case studies yielded valuable information about what did and did not work, conditions contributing to success or failure, as well as challenges and opportunities associated with scaling up management, that otherwise might not have been broadly shared. We present summaries of the four case studies where we identify the lessons learned and the opportunities to advance future management efforts to inform the direction for new science (Table 1). When applicable, we also highlight where closer integration of science and management could have resulted in opportunities for amplifying implications of lessons learned.

Case study 1

Lesson learned: preplanning postfire rehabilitation improves resource prioritization and implementation

In Case Study 1, Burns District BLM, Oregon aimed to reduce medusahead, increase perennial bunchgrasses, and reduce future fire risk after the 2007 Bartlett Mountain Fire, which burned 13,076 ha (32,312 acres). Treatments to the burned area included seeding, dormant season grazing, and targeted herbicide application. Restoration seeding included a mixture of native perennial bunchgrasses and crested wheat-grass (Agropyron cristatum [L.] Gaertn, drill seeded in the fall of 2007, and aerial seeded in the spring of 2008). Dormant season (i.e., fall and winter) livestock grazing began in 2012 within a 3,141-ha (7,762-acre) pasture and periodic imazapic (pre-emergent herbicide) applications at 0.44 liters per hectare (6 ounces per acre) of formulated product at a rate of 105.17 g · acid equivalent (ae) · ha^-1 (0.09 lb ^˙ae ^˙acre^-1) were made along roadsides throughout the project area to reduce fuels and control primary weed spread corridors. Two of three photo-monitoring locations showed a post-treatment decrease of medusahead cover and ongoing recovery of perennial bunchgrasses in response to dormant season grazing and seeding efforts. More monitoring sites would have been needed to capture the variation between the effects of dormant season grazing, drill, and aerial seeding. Ecological site type was a major determinant of the efficacy of dormant season grazing and seeding to promote desired perennial plants in the medusahead-invaded rangeland. Perennial plants reestablished on ecological sites with deep and porous soils, and flat and clayey sites experienced lower reestablishment. Additionally, the seeding, grazing, and roadside herbicide treatments demonstrated some management success for reducing wildfire spread. This area burned again in 2014, as a part of the Buzzard Fire complex (which burned about 116,145 ha [287,000 acres]), and one of the ignition sites was stopped in a roadside fuel break where medusahead had been previously treated with imazapic in the dormant season grazing pasture.

In a separate 7,000-ha (17,300-acre) area burned by the 2014 Buzzard Fire Complex, managers applied imazapic in 2015 to control medusahead at a rate of 105.17 g · ae · ha^-1 (0.09 lb ^˙ae ^˙acre^-1). However, inconsistent herbicide application across the treatment area in 2015 did not result in medusahead control and led to a follow-up treatment in 2016, which performed better. The area was also seeded post fire during the spring of 2015 (aerial), fall of 2016 (drill), and winter of 2015/early 2015 (broadcast). Burns BLM staff learned that treating medusahead infested areas with imazapic immediately post wildfire greatly improved treatment results.

Table 1

Summary of case studies presented, including name and organization of original presenter at 2020 Invasive Annual Grass Workshop organized by the High Desert Partnership, the SageCon Partnership, and Oregon State University

Opportunity: need for researcher/manager integration

This case study highlights an opportunity to tighten the research-management relationship; researchers could work with managers to develop postwildfire rehabilitation and monitoring plans for testing multiple management strategies. Managers and researchers could team up before wildfire season to proactively develop “on-deck” ecological-site-specific management strategies and related hypotheses for postfire implementation plans providing BLM staff with tailored management and researchers with more robust experimental designs (see Wollstein et al. this issue for a description of an applicable proactive planning framework).¹⁶ Collaboratively planning monitoring sites would have helped disentangle the grazing, seeding, and herbicide treatment effects. Unfortunately, management project needs exceed the capacity of academic researchers, and involving researchers inevitably requires more time from both parties. Although most managers do not have easy access to researchers at a university or research station, they likely interact with local cooperative extension agents from universities as well as federal and state agencies, who engage in applied research. Extension agents translate science into management for stakeholders, have established relationships with land managers, and are located beyond the university campus, making them more accessible to partner with land managers. An extension agent/land manager team could provide the scientific community with management-scale feedback on science-based practices, while the land managers receive defensible and improved land management strategies. Extension agents could facilitate broader communication of lessons learned from case studies and enable more information sharing among land managers.

Lastly, BLM Resource Management Specialists indicated concerns about the potential for development of herbicide resistance in the project area from repeated roadside applications of a particular pre-emergent herbicide to control medusahead and reduce fuels. The 2020 Invasive Annual Grass Workshop provided Burns District BLM staff and other participants from around the West the opportunity to learn about Rejuvra (indaziflam), which is registered for invasive annual grass control on rangelands. A single application provides multiple years of annual grass control¹⁷ and, importantly, offers a different mode of action than other pre-emergent herbicides. Strategies of repeated applications of herbicides to control weed vector pathways or for fuels reduction should consider the potential for herbicide resistance and use mixtures or rotate among chemistries offering different modes of action whenever possible.¹⁸

Case studies: capturing complexity and capitalizing on collaboration

These four case studies (Table 1) exemplify many of the challenges managers face when implementing science-based invasive annual grass management in a complex environment. Each case study provides a wealth of lessons learned and opportunities for future change, and the 2020 Invasive Annual Grass Workshop provided an opportunity to share them with fellow managers and researchers. While workshops represent an effective method for communication among colleagues, they cannot convey the wealth of knowledge and expertise of current natural resource professionals. Furthermore, to tailor workshops to the specific information needs of individual managers and their unique circumstances would be challenging. Managers need more timely information than can be delivered via periodic workshops. CCAST is an example of a platform to share case studies such as these described here but has limitations: case studies can only be shared by CCAST members, and the search ability is limited to tag words. An easily searchable database documenting management efforts, site characteristics, weather conditions, and treatment outcomes could improve the efficacy of future restoration projects and quicken the pace of knowledge transfer.

Currently, this database does not exist, but we have a system for documenting land treatments on primarily BLM land using the Land Treatment Digital Library (LTDL),^27,28 and the integration of those treatments within a land management planning tool in the Land Treatment Exploration Tool (LTET).²⁹ Indeed, LTDL houses an inventory of BLM land treatments searchable by treatment type, spatial location, and planning or implementation using specific identifiers (e.g., project number). However, LTDL does not include information about site characteristics, weather conditions, or treatment outcomes associated with land treatment projects, and currently is focused predominately on BLM lands. As in any library, LTDL is only as good as the information recorded and submitted for archiving, so the completeness and quality of records can vary within the database. LTET integrates data from LTDL, the Soil Survey Geographic Database (SSURGO), and National Resources Conservation Service (NRCS) ecological site descriptions. It contains information on climate and drought histories, US Fish and Wildlife Service information for planning and consultation, and a 12-month drought forecast to identify when treatment success might be favorable as well as drought predictions for the next year. Although the LTET is a useful tool for planning future treatments, it was not designed to capture information related to project outcomes.

Building a comprehensive database to document management outcomes is challenging because assembling such a database requires substantial coordination between different federal, state, local, and tribal agencies, and buy-in from a rangeland community lacking implementation capacity and that can be fiercely independent. The creation of this database requires sufficient investment in additional capacity to focus on the coordination, communication, population, and maintenance of a formal system for learning from adaptive management efforts. While perhaps ideal, this would be a long-term process. So where does that leave us in the meantime?

Working with what we have

We have useful and relevant tools to understand the efficacy of invasive annual grass treatments under a variety of site and environmental conditions. We can leverage the LTDL and LTET, by pairing them with the other remotely sensed products like the Rangeland Analysis Platform (RAP)^30,31 or Rangeland Cover Map (RCMAP)³² to inform adaptive management efforts. For example, if we are interested in controlling invasive annual grasses in an area of rangeland located within or near the Burns District BLM, we can query the LTDL or LTET for relevant treatments in the geographic area of interest. An authorized user would query LTDL or LTET for Case Study 1 (Table 1) to retrieve the site characteristics, boundary, year, treatment type, implementation details, and any discoverable documentation accompanying the treatment or project such as planning documents and monitoring reports. Unfortunately, only about 21% of treatments/projects within LTDL have documentation on effectiveness monitoring, because there is a lack of reported documentation to the LTDL.³³ The LTDL and LTET provide a shapefile to download containing the boundaries of treatment or project areas, which when paired with RAP provides plant cover and precipitation data before, during, and after treatment. For example, using the treatment area boundaries in the downloaded shapefile associated with a Burns BLM District 2017 application of pre-emergent herbicide to query RAP for plant functional group cover and precipitation values indicated (Fig. 1):

This information indicates pre-emergent herbicide treatments were, at least initially, effective for reducing annual plant cover and increasing perennial plant cover. However, we need additional years of data to fully understand the longer-term effects of herbicide and seeded treatments. Also, although the data from the LTDL/LTET, and RAP provided a general evaluation of treatment outcomes, it does not substitute for on-the-ground monitoring and observations conducted by resource management professionals. Therefore, we suggest contacting project leads and other resource professionals whenever possible. Furthermore, combining multiple decision-support tools could prove problematic for time-limited managers who may be unfamiliar with the LTDL/LTET, RAP, or RCMAP. Managers are unlikely to adopt practices that are overly complex, less accessible (i.e., requiring authorized use), or are not reliably advantageous.¹¹ Despite these challenges, we are fortunate to be managing in the digital information age,where digital decision support and remotely sensed tools exist free of charge, providing information compatible with the scale of management.

Figure 1.

Output from a downloaded shapefile of treatment areas from the Land treatment Data Library (LTDL) from Case Study 1 in the Burns BLM district. The 2020 perennial herbaceous plant cover layer from the Rangeland Analysis Platform (RAP). The treatment area was used to create the cover estimates in the graph showing the percent foliar cover of annual (red) and perennial (green) herbaceous vegetation through time, from 1985 to 2020. The blue bars represent annual precipitation in inches.

Conclusion

Management case studies present unique opportunities to translate best available science into effective on-the-ground restoration. However, translating science conducted within a narrow range of temporal and spatial conditions into large-scale management presents implementation challenges and often surprising, and sometimes disappointing results. Although awareness and implementation of cutting-edge science can improve management success, researchers often overlook the practical side of scaling up to land management levels and sometimes a reality check is needed. Whenever possible, management and research should complement each other to cultivate closer collaboration between science and land management by integrating research into management opportunities and testing multiple land treatments at management scales. Management benefits from using the most current information to tackle the complex and ever-changing problem of invasive annual grass management, and reality checks provided by managers accelerate scientific breakthroughs from the study plot to the open range. In rangelands combining research and management is the exception, not the rule. With more managers than researchers, most management projects are not conducted near a research facility already conducting similar work. However, remotely sensed tools continue to improve in accuracy and reliability, offering a unique opportunity to bridge the historic gap between researchers and managers.^30,31 Researchers typically restrict the number and scale of sites by necessity, but landscape level analyses conducted by researchers might help land managers better understand where exactly on the landscape a treatment might succeed (or fail). Furthermore, extension agents may help play a critical role in bridging this gap as key management partners conducting applied research.

To advance the learning portion of the adaptive management process used by rangeland managers, we need a tool such as a dynamic management database capturing the variability and successes or failures of past treatments. Such a user-friendly, comprehensive tool is in the future, but in the meantime, managers should use currently available tools such as the LTDL, LTET, SSURGO, and RAP or RCMAP to infer treatment effectiveness across spatial and temporal differences. These tools can inform implementation and hopefully increase a manager's rate of success in conducting land treatments. Lessons learned from management case studies should be accessible through workshops, multiagency and discipline collaborations, and scientific journals that accept case studies with negative or null results (e.g., AoB Plants, or Ecological Solutions and Evidence) and are open access so managers can access these publications to tackle the complex problem of invasive annual grass management.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests. DDJ and VMS are Guest editors for this Special issue but were not involved with the review or decision process for this manuscript. HEQ is an employee of Bayer R&D Services, LLC. The authors certify that they have no financial interest in the subject matter discussed in the manuscript. The content of sponsored issues of Rangelands is handled with the same editorial independence and single-blind peer review as that of regular issues.