Trojan Horse on the Great Plains: Landowner Thresholds, Coping Capacity, and Management of Kentucky Bluegrass,

ABSTRACT Kentucky bluegrass, an invasive cool-season grass in the US northern Great Plains, embodies the Greek story of the Trojan horse in which an apparent gift initially seems to bestow benefits but ultimately becomes harmful and subversive. Its competitive nature and early spring growth give it an advantage over native cool season grasses. This early and prolific growth provides forage for livestock but eventually leads to a decrease in other plant species and to negative impacts to multiple ecosystem services. Developing effective landscape-scale management techniques for invasives like Kentucky bluegrass depends on understanding the perceptions, coping capacity, and management abilities of private landowners. Through a mail survey conducted in North Dakota, this research provides insight into the cognitive processes that motivate agricultural landowners to manage Kentucky bluegrass and investigates psychological factors that constrain or enable the implementation of management behaviors. The results showed that less than half of the landowners surveyed had a management goal for Kentucky bluegrass, and among those with such goals, the perceived capacity to cope with the species was moderate to low. The majority of landowners who wanted to decrease the abundance of Kentucky bluegrass on their rangelands did not believe they could effectively do so. This highlights the need for the rangeland science and extension community to develop effective and feasible management techniques and conservation programs that integrate research findings to provide viable solutions. Participatory processes that include diverse interested and affected parties and rights holders can provide social learning environments conducive to the joint exploration of problems and refinement of actionable solutions. Through the focus on Kentucky bluegrass in the northern Great Plains, our research approach and recommendations also provide transferable insights for invasive species management and conservation in other social-ecological systems.


Introduction
Managing invasive species to mitigate their negative social and ecological impacts is a global priority in agricultural systems and for broader environmental conservation ( Paini et al. 2016 ; ✩ Current address: Kiandra Rajala, US Fish and Wildlife Service, Science Applications Program, East Lansing, MI 48823-6360, USA.✩✩ This research was funded by the USDA-ARS cooperative agreement 58-3064-6-003 and the USDA Natural Resources Conservation Service Conservation Effects Assessment Project and was a contribution from the Long-Term Agroecosystem Research (LTAR) network.LTAR is supported by the USDA.The findings and conclusions in this article are those of the author(s) and do not represent the official views of the US Fish and Wildlife Service.Ricciardi et al. 2017 ;Shackleton et al. 2019 a).Effective invasive species management, whether focused on eradication or coexistence, hinges on both subjective (e.g., people's perceptions of the invader) and objective (e.g., engagement and impact of management actions) human dimensions ( Shackleton et al. 2019 a).Perceptions about an invasive species are rarely uniform.They can be polarized and intractable given high uncertainty, complexity, and variation in social and ecological costs and benefits ( Woodford et al. 2016 ).Along with perceptions of the invasive species, actionable management also depends on people's perceptions of available management options and their capacity to implement them ( Bruskotter et al. 2009 ;Wachinger et al. 2013 ;Shackleton et al. 2019 b;Tebboth et al. 2020 ).
Most grassland ecosystems are already imperiled through land use changes like conversion to crop production ( Wright and Wimberly 2013 ), and invasive grasses pose additional threats to the https://doi.org/10.1016/j.rama.2023.07.006 1550-7424/Published by Elsevier Inc. on behalf of The Society for Range Management.This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ) plant, animal, and human communities that depend on grassland ecosystem services ( Bengtsson et al. 2019 ;Perkins et al. 2019 ;Gaskin et al. 2021 ).In the grasslands and rangelands of the North American Great Plains, invasive grasses are a growing ecological and agricultural concern ( Coon et al. 2020 ;Gasch et al. 2020 ;Gaskin et al. 2021 ).Invasive grasses in rangelands also present unique challenges given the difficulty of managing them in isolation of existing native vegetation, as well as the potential for some resource users to perceive or experience benefits from the new species ( Gaskin et al. 2021 ).Much of the Great Plains is privately owned and managed for agricultural production of crops and livestock, positioning private landowners as frontline gatekeepers or catalysts in the spread of invasive species ( Wilcox et al. 2018 ;Rajala et al. 2021 ).Landowners' capacity to cope with and adapt to invasive species through changing management is an important aspect of social resilience ( Marshall et al. 2011 ).Little is known, however, about how agricultural landowners throughout the Great Plains perceive and manage invasive grasses ( Coon et al. 2020 ).
While perhaps one of the most well-known and widespread grasses on suburban North American lawns, Kentucky bluegrass ( Poa pratensis L.) has become an aggressive invasive species in rangelands of the northern Great Plains ( DeKeyser et al. 2015 ;Gasch et al. 2020 ;Palit et al. 2021 ).Despite its moniker, Kentucky bluegrass (which we also refer to as bluegrass) is a Eurasian native, which was introduced to the United States in the 1600s and valued by colonists for its forage palatability, high yield, and attractiveness as a lawn grass ( DeKeyser et al. 2015 ).In the southern and eastern United States, Kentucky bluegrass is still favored as a productive, nutritious, and grazing-tolerant grass species that has also been used widely as a turfgrass and for erosion control ( Bush 2002 ).Kentucky bluegrass is a cool-season, midstature grass with structural characteristics similar to its native counterpart in the northern Great Plains, green needlegrass ( Nassella viridula [Yrin.]Barkworth).Yet Kentucky bluegrass is more competitive and able to initiate growth earlier in the spring than native cool season grasses ( DeKeyser et al. 2013 ).As Kentucky bluegrass cover increases, other plant species decrease ( Tatina 1994 ;Miles and Knops 2009 ;DeKeyser et al. 2013 ), initiating a series of cascading effects that impact multiple ecosystem services ( Toledo et al. 2014 ).
Kentucky bluegrass exemplifies aspects of the Trojan horse story from Greek mythology in which an apparent gift that initially seems to bestow benefits ultimately becomes harmful.When bluegrass initially invades, it may be valued by some rangeland users as an early source of forage for livestock ( Rajala et al. 2021 ).As its coverage increases > 35%, detrimental impacts such as native species loss, altered biological feedback cycles, reductions in pollinator abundance, and reduced adaptive capacity of livestock management operations also accelerate ( Toledo et al. 2014 ;Printz and Hendrickson 2015 ).
Landscape-scale management effort s in the northern Great Plains can be enhanced by understanding how private landowners perceive Kentucky bluegrass and appraise their ability to cope with and actually manage the species.Perceptions of invasive species often vary on the basis of a complex interplay of personal characteristics, traits of the invasive species itself, and people's knowledge and experience of species' impacts, all of which are embedded within a larger social-ecological context ( Sharp et al. 2011 ;Shackleton et al. 2019 b).We focus on landowners' primary land use, exposure to Kentucky bluegrass, general familiarity with bluegrass, and beliefs about bluegrass impacts as key factors that shape perceptions of Kentucky bluegrass and its risks ( Sharp et al. 2011 ;Wachinger et al. 2013 ).These perceptions are related to a landowner's tolerance for bluegrass.Once the psychological tipping point or tolerance threshold is crossed, landowners may be motivated to form management goals to control or reduce Kentucky bluegrass ( Devine-Wright and Howes 2010 ; Zajac et al. 2012 ).
Engaging in Kentucky bluegrass management, whether to increase, maintain, or decrease the species, is a form of goal-oriented behavior driven by a landowner's management goal for the species and behavioral intentions to implement specific management practices ( Fishbein and Ajzen 2011 ;Bamberg 2013 ).However, the pathway from forming a goal to enacting a behavior is subject to many potential constraints.Differing management goals present their own challenges in execution; thus intention-implementation gaps often exist ( Ajzen 2002 ;Bamberg and Moser 2007 ;Bamberg 2013 ).Selecting and implementing Kentucky bluegrass management behaviors in pursuit of goal attainment involves cognitive evaluations of self-efficacy, attitudes about available behaviors, and beliefs about the efficacy of possible behavioral options in relation to one's management goal for the species ( Fishbein and Ajzen 2011 ;Bamberg 2013 ;Wachinger et al. 2013 ;Coon et al. 2020 ).
We use the results of a mail survey to investigate how agricultural rangeland landowners in the northern Great Plains of North Dakota perceive and manage Kentucky bluegrass.This research provides insight into the cognitive processes that motivate agricultural landowners to manage Kentucky bluegrass and investigates psychological factors that constrain or enable the implementation of management behaviors.First, we examine landowners' tolerance thresholds for Kentucky bluegrass and explore how this threshold relates to goals to manage the species.Second, we explore factors that may be associated with landowners whose tolerance threshold for Kentucky bluegrass has been exceeded.Third, we assess landowners' capacity to carry out their management goal for Kentucky bluegrass.Finally, we identify landowners' current and intended use of management practices for Kentucky bluegrass on the basis of their management goal for bluegrass.This sequential exploration of landowner perceptions about both Kentucky bluegrass and potential management options to control or reduce it helps identify key leverage and intervention points for rangeland scientists, managers, and conservation outreach professionals to target future effort s.With a f oundation grounded in prior research on cognitive processes and psychological factors that constrain or enable behaviors, we encourage the application of this approach to inform other invasive species management and conservation efforts.

Study area
The northern Great Plains is within the temperate steppe ecoregion of North America and has a semiarid continental climate, with evaporation typically exceeding precipitation.The region's long cold winters and short hot summers sustain a variety of cropping systems and livestock production based on rangeland, pastures, and hay production.Agricultural production of cattle in the northern Great Plains account for roughly 2% of the national total ( USDA-NASS 2023 ) and are an important source of local livelihoods and state and regional revenue ( USFWS 2019 ; NDDA [North Dakota Department of Agriculture] 2023.).The grasslands that support this livestock production are also of conservation concern as they provide an important suite of ecosystem services and critical habitat for migratory birds, breeding waterfowl, pollinators, and iconic North American mammals such as bison (Bison bison) and pronghorn (Antilocapra americana) ( Perkins et al. 2019 ;USFWS 2019 ).
Despite the ecological and agricultural importance of northern Great Plains grasslands, more than 530 0 0 0 ha of grass-dominated cover were converted to crop production in this region between 2006 and 2011 ( Wright and Wimberly 2013 ).This change in land use was precipitated by increases in crop prices and facilitated by changes in technology and climate that have allowed certain crops to now be grown successfully in the region.Current projections suggest that the northern Great Plains of the future will have greater winter and spring precipitation and warmer temperatures, creating longer growing seasons ( Conant et al. 2018 ).The future climate will likely maintain the current trend of land use change toward cultivation and enable increases of current invasive plant species ( Walthall et al. 2013 ).Data from the US Department of Agriculture (USDA) −Natural Resources Conservation Service (NRCS) National Resources Inventory show that Kentucky bluegrass foliar canopy cover is increasing throughout the Great Plains, with the greatest cover in the northern Great Plains of North Dakota (Kentucky and Canada bluegrasses are present on 86% ( ±3.7) of nonfederal rangeland in North Dakota; NRI 2018 ).
We conducted our study in nine counties of North Dakota creating a west-to-east transect across the state's four primary ecoregions: Missouri Slope, Missouri Coteau, Drift Prairie, and the Red River Valley ( Figure 1 ).Golden Valley, Billings, Stark, and Morton Counties lie within the Missouri Slope and are characterized by the buttes and rough terrain of the Little Missouri Badlands, as well as the semiarid rolling plains of the Missouri Plateau ( Bryce et al. 1996 ).Predominant land uses in these western-most counties include livestock grazing on the native shortgrass rangelands and production of spring wheat and alfalfa ( Bryce et al. 1996 ;USDA-NASS 2019 ).Burleigh, Kidder, Stutsman, and Barnes Counties span the Missouri Coteau and Drift Prairie, which contain many permanent and seasonal wetlands throughout rolling terrain, providing productive waterfowl habitat and fertile soils.Land use in these central counties transitions from livestock production on rangelands to intensive dryland farming of spring wheat, soybeans, corn, sunflowers, and alfalfa farther east ( Bryce et al. 1996 ;NASS 2019 ).Cass County, the eastern-most portion of the study area, is in the flat and fertile Red River Valley, characterized by highly productive farmland producing primarily soybeans, corn, wheat, and sugar beets ( Bryce et al. 1996 ;NASS 2019 ).

Sampling, survey design, and administration
Our study focused on private landowners who owned at least 1 0 0 0 acres (405 ha).We used data from the 2012 Census of Agriculture to approximate the total number of landowners who owned at least 1 0 0 0 acres in our study area counties ( USDA-NASS 2014 ).From this estimate, and on the basis of expected response rates, we designed a sampling scheme targeting 1 278 landowners to achieve a ±5% margin of error for true population values assuming maximum heterogeneity in responses ( Dillman et al. 2014 ).We created our sampling frame for landowner selection using the 2019 public property records from each county in the study area.To achieve a geographical diversity of perspectives across land use types in our study area, we randomly sampled 142 eligible landowners from each county.After US Postal Service address validation, and because two counties had fewer than 142 eligible landowners, our final sample consisted of 1 219 landowners.
We developed a self-administered survey based on previous research on human dimensions of invasive species (e.g., Rajala 2019 ) and ecological literature about Kentucky bluegrass (e.g., Toledo et al. 2014 ).We refined the survey on the basis of interviews ( n = 20) and pretests ( n = 23) with North Dakota landowners and subject matter experts.We administered our survey in November and December of 2019 using a slightly modified tailored design approach of five mailings to encourage participation ( Dillman et al. 2014 ).All respondents received an introductory letter (d 1), survey packet with prepaid return envelope (d 7), and thank you/reminder postcard (d 14).Respondents who had not yet completed the survey or contacted us to opt out of the study received a replacement survey packet (d 36).Finally, all remaining nonrespondents who had not opted out received a final reminder letter (d 50).This research was approved by the Virginia Tech Institutional Review Board (19-312) and the US Office of Management and Budget (0518-0048).

Data collection Demographic and land ownership questions
Underlying individual factors may help explain people's perceptions about invasive species ( Shackleton et al. 2019b ).We examined demographic and land ownership characteristics expected to be salient to experiences of Kentucky bluegrass and its effects.We asked landowners to provide demographic information, including their age, race, gender, highest level of education completed, work status, and net annual household income.We also asked landowners whether or not their primary occupation was a part of, or related to, the farming/ranching industry.We constructed a variable to represent the total number of acres in a landowner's operation by adding the number of acres they owned with the number of acres they leased from others and removing the number of acres they leased out.
We asked landowners how many years they had owned their land and created a variable to represent years of land ownership as a proportion of their adult life (yr owned land/[age-18]).We also asked if they inherited or acquired the land from their family, whether they reside on their land full time, and how many hours per week they typically spend managing, operating, or working on their land.We asked landowners to estimate the proportion of their annual income earned from activities on their land.We grouped landowners into four categories according to their primary land use: livestock operation, crop operation, crop and livestock operation, unspecified agriculture, or some other land use.

Kentucky bluegrass questions
For our survey, sections related to Kentucky bluegrass, we first provided landowners with a drawing of bluegrass from the USDA-NRCS Plants Database and a description of the species.We intentionally used neutral language in this description of bluegrass to minimize potential priming effects and social desirability bias through words like "invasive" (e.g., Larson 2005 ): • Kentucky bluegrass ( Poa pratensis L.) is most known as a lawn and turf grass.A variety of this cool-season, perennial grass grows in the grasslands and rangelands of North Dakota.• Kentucky bluegrass is sometimes called June grass because it matures early in the growing season and usually dries out and produces seed in June.• We want to know your opinions about Kentucky bluegrass even if it does not currently grow on your land.
Familiarity with Kentucky bluegrass -General knowledge and familiarity with an invasive species are attained through both direct and indirect experiences and can mediate beliefs and attitudes about the species ( Kull et al. 2012 ;Verbrugge et al. 2013 ).We used a single item asking landowners to indicate their familiarity with Kentucky bluegrass using a 5-point Likert-type scale (1 = Not familiar at all , 2 = Slightly familiar , 3 = Moderately familiar , 4 = Very familiar , 5 = Extremely familiar ).
Exposure to Kentucky bluegrass -Exposure to an invasive species through direct experience can influence awareness and concern and provides personal experiences formative to beliefs about the species' impacts ( Johnson et al. 2011 ;Wachinger et al. 2013 ;Shackleton et al. 2019 b).We measured landowners' perceptions of their exposure to Kentucky bluegrass using three questions.First, we asked landowners to indicate whether Kentucky bluegrass was currently present on their rangelands ( No , Unsure , or Yes ).For landowners who indicated Kentucky bluegrass was present, we asked them to provide two estimates: 1) the percentage of land on which Kentucky bluegrass was present (extent) and 2) the percentage of the overall plant community that was Kentucky bluegrass, where bluegrass was present (proportion).For estimates of extent and proportion, landowners who self-reported that they did not have bluegrass on their rangelands were assigned a value of 0 and landowners who self-reported they were unsure about the presence of bluegrass were omitted from the analysis.
Tolerance threshold for Kentucky bluegrass -We focused on the tolerance threshold at which Kentucky bluegrass becomes unacceptable (e.g., Zajac et al. 2012 ).We asked landowners to provide an overall evaluation of the current abundance of bluegrass on their rangelands using a 5-point Likert-type scale where 1 = Much too low , 2 = Too low , 3 = About right , 4 = Too high , and 5 = Much too high .We created a binary variable to reflect a tolerance threshold where 0 = Much too low , Too low , and About right and 1 = Too high and Much too high .
Kentucky bluegrass management goal -In stage models of behavior change, forming a goal intention is the first step in a series of motivational and cognitive transitions toward enacting a behavior ( Bamberg 2013 ).Once a goal intention is formed, a person begins the cognitive process of considering the pros and cons of different possible behaviors to achieve their goal ( Bamberg 2013 ;Steg et al. 2013 ).We asked landowners to indicate which of four options best describes their management goal for Kentucky bluegrass on their rangelands: increasing the abundance of bluegrass, maintaining the current abundance of bluegrass, or decreasing the abundance of bluegrass.Landowners could also indicate that they had no goal for the abundance of bluegrass.
Beliefs about Kentucky bluegrass-Beliefs about invasive species may be informed through direct experiences and personal observations or other forms of indirect learning and influence ( Shackleton et al. 2019 b).Beliefs about the species' harmful or beneficial impacts and its attractiveness have been identified as influential drivers of perceptions for a wide range of invasive species ( Shackleton et al. 2019 b).We measured landowners' beliefs about 13 potential impacts from Kentucky bluegrass invasion using a 5point Likert-type scale where 1 = Definitely not true , 2 = Probably not true , 3 = Unsure , 4 = Probably True , and 5 = Definitely true .We developed the statements from both the ecological literature and landowner interviews.Belief statements were phrased as potentially desirable or undesirable outcomes to vary the connotation of items rather than to match current scientific opinion.Beliefs focused on impacts to forage for livestock, wildlife and pollinators, other grassland vegetation, soil health, grassland aesthetics, and the ability to cope with drought.
Perceived behavioral control over managing Kentucky bluegrass-Regardless of motivation or intentions, enacting a behavior is not always within a person's complete volitional control ( Ajzen 2002 ).We used a single indicator to assess a landowner's perceived control over performing management behaviors, in general, to attain their desired outcomes for Kentucky bluegrass.We asked landowners to indicate their agreement or disagreement with the follow-ing statement: Given the current amount of Kentucky bluegrass on my rangelands, I can effectively manage it at the levels I desire ( 1 = Strongly disagree , 2 = disagree , 3 = Neither agree nor disagree , 4 = Agree , 5 = Strongly Agree ).
Attitudes about controlling or reducing bluegrass-Attitudes are favorable or unfavorable evaluations of a behavior that can encompass instrumental and experiential aspects of the behavior ( Fishbein and Ajzen 2011 ).We measured landowners' attitudes about controlling or reducing Kentucky bluegrass on their rangelands using a 7-point semantic differential scale with the following bipolar adjective pairs related to cost (expensive-inexpensive), desirability (undesirable-desirable), ease (difficult-easy), effectiveness (ineffective-effective), possibility (impossible-possible), necessity (unnecessary-necessary), feasibility (infeasible-feasible), and benefit (detrimental-beneficial).
For beliefs about effectiveness, current management, and future intentions, we asked landowners to consider five potential practices to manage Kentucky bluegrass and provided them the option to write in other management practice(s).These five practices were selected on the basis of research on Kentucky bluegrass (e.g., DeKeyser et al. 2015 ;Ereth et al. 2017 ;Gasch et al. 2020 ) and other invasive grasses (see overview in Archer et al. 2023 ), as well as interviews with landowners and subject matter experts in the study area: • intensive livestock grazing in early spring (e.g., April and May) • prescribed fire (i.e., burning rangelands) • targeted application of chemical herbicides • mob grazing (i.e., ultrahigh animal density for short durations) • multispecies grazing (e.g., sheep, goats, cattle) Beliefs about effectiveness of management practices-Beliefs about likely outcomes and consequences of a behavior are formative considerations in behavioral intentions and actual behavior ( Ajzen 2002 ).We asked landowners to indicate how effective they thought each management practice is at controlling or reducing Kentucky bluegrass using a 5-point Likert-type scale (1 = Not at all effec tive, 2 = Slightly effective , 3 = Moderately effective , 4 = Mostly effective , 5 = Completely effective ).
Current management of Kentucky bluegrass-Landowners indicated whether or not they had used each practice specifically to manage Kentucky bluegrass on their rangelands recently and/or in the past using the following response options: never, ONLY more than 2 yr ago, ONLY within the past 2 yr, more than 2 yr ago AND within past 2 yr.We created a binary variable to account for landowners' current use within the past 2 yr where 1 = Yes (ONLY within the past 2 yr, more than 2 yr ago AND within the past 2 yr) or 0 = No (never, ONLY more than 2 yr ago).
Future intentions to manage Kentucky bluegrass-Landowners indicated how likely or unlikely they were, in the next 12 mo, to use each practice specifically to manage Kentucky bluegrass on their rangelands using a 5-point Likert-type scale ( 1 = Extremely unlikely, 2 = Unlikely, 3 = Neither likely nor unlikely, 4 = Likely, 5 = Extremely likely).

Data analysis
We examined the relationship between landowners' tolerance for Kentucky bluegrass and their management goal for bluegrass using chi-square tests and adjusted standardized residuals to identify which goals contributed to a significant chi-square value.We then characterized landowners who have and have not exceeded their threshold for Kentucky bluegrass by comparing landowners' primary land uses, familiarity with bluegrass, exposure to bluegrass, and beliefs about bluegrass.We employed chi-square tests to compare the categorical primary land use variable and used adjusted standardized residuals to determine which land uses signif-icantly contributed to the outcome.Because chi-square tests compare observed counts to expected counts, we report whether the counts in a cell are more or less than expected based on the standardized residual.We used Mann-Whitney U tests to examine the differences in abundance of bluegrass, familiarity, and beliefs between landowners who had and had not exceeded their tolerance threshold.
To understand landowners' coping capacity and actual management of Kentucky bluegrass, we limited our analysis to landowners who expressed a management goal to either maintain or decrease the abundance of bluegrass.We excluded landowners who self-reported that they did not have Kentucky bluegrass on their rangelands.Among landowners who self-reported having bluegrass on their rangelands and had goals to maintain or decrease the species, we examined perceived behavioral control to manage bluegrass, attitudes about controlling/reducing bluegrass, and beliefs about the efficacy of bluegrass management practices.Given that coping capacity and goal attainment may differ per management goal, we compared the responses of landowners who wanted to maintain bluegrass with those who wanted to decrease bluegrass using Mann-Whitney U tests.Finally, we examined landowners' current use of practices and behavioral intentions to manage Kentucky bluegrass.We compared management actions based on landowners' goal to either maintain or decrease bluegrass using chi-square tests with adjusted standardized residuals.For comparisons with expected cell values < 5, we employed Fisher's exact test.

Results
From the 1 219 landowners randomly selected to participate in our study, 55 were ineligible (e.g., undeliverable address, deceased, had sold land), 136 opted out, and 373 completed the survey.Although eligibility for the study was established as landowners who owned at least 1 0 0 0 acres, 25 landowners who completed the survey indicated they owned < 1 0 0 0 acres.We implemented an additional inclusion criterion based on total acres operated (total acres operated = acres owned + acres leased in -acres leased out).Of the 25 landowners who owned < 1 0 0 0 acres, we included 17 landowners who had total operations between 598 and 4 640 acres in our analysis (see Rajala et al. 2021 for additional details).Accounting for this inclusion adjustment and the 55 ineligible landowners, we obtained a survey response rate of 32% ( n = 365) based on a final adjusted sample size of 1 156 landowners.
For our analysis, we only included landowners who indicated they were involved in making most of the day-to-day decisions about their land ( n = 327).Of these primary decision makers, 85% ( n = 277) completed the Kentucky bluegrass sections of the survey and 15% ( n = 50) opted out.Due to individual item nonresponse, the number of landowners in each analysis varies.
Landowners who opted out of the Kentucky bluegrass sections differed significantly from those who responded based on how they primarily used their land (X 2 (3) = 47.71,P < 0.001, Cramer's V = 0.39, n = 306).Compared with landowners who provided usable responses, significantly more landowners who opted out (79%) used their land primarily as a crop operation (adjusted residual = 6.73).Significantly fewer landowners (8%) used their land primarily as a livestock operation (adjusted residual = −4.09)or crop and livestock operation (2%, adjusted residual = −3.04).For these reasons, this sample predominantly reflects livestock producers and likely under-represents crop farmers.We caution against generalizing results to all landowners, particularly those who primarily use their land for crop production.

Responding landowner characteristics and estimates of Kentucky bluegrass
The proportion of responding landowners in each county ranged from 8.3% (Stutsman County) to 13.4% (Morton County).Most landowners were at least 60 yr old (75%), white (98%), and male (92%) who, on average, had owned their land for the majority (71%) of their adult life.Most (66%) had inherited or acquired some portion of their land from their family and currently lived on their land full time (76%).Around one-third (32%) of landowners reported completing a 4-yr college degree or higher level of education.The majority of landowners (61%) were working full time, and almost all (86%) reported that their primary occupation was part of, or related to, the farming/ranching industry.Just under half of landowners (46%) reported net annual household incomes of $10 0 0 0 0 or more.On average, landowners spent 46 h per week (standard deviation [SD] = 29) managing, operating, or working on their land and 75% of landowners earned at least half of their annual household income from their land.Landowners' primary land uses were split between livestock operations (39%), crop operations (28%), mixed crop and livestock operations (20%), and unspecified agriculture or other land uses (13%).
Around half of landowners (51%) reported having Kentucky bluegrass on their rangelands while the rest indicated that they did not have bluegrass (22%) or were unsure about its presence (27%).Overall, landowners estimated that Kentucky bluegrass was present on an average of 20% of their rangeland (median = 10%; IQR: 0-30%) and that bluegrass comprised an average of 22% of the rangeland plant community (median = 15; IQR: 0-32.5%).The subset of landowners who reported having bluegrass on their rangelands estimated bluegrass extent as present on an average of 29% of their rangeland (median = 20%; IQR: 10-40%) and bluegrass proportion as an average of 32% of the plant community where bluegrass was present (median = 25%; IQR: 12.5-50%).

Kentucky bluegrass thresholds and management goals
Almost three quarters (71%) of landowners indicated that the abundance of bluegrass on their rangeland was within their tolerance threshold (i.e., not too high) while the remainder (29%) indicated that the abundance of bluegrass on their rangelands had exceeded their threshold (i.e., was too high).Overall, the majority of landowners (57%) had no management goal for Kentucky bluegrass currently on their rangelands, 23% wanted to decrease bluegrass, 19% wanted to maintain what they currently had, and 2% wanted to increase bluegrass.
Landowners' management goals were strongly related to whether or not bluegrass had exceeded their tolerance threshold ( X 2 [2] = 109.59,P < 0.001, Cramer's V = 0.73 n = 204).For landowners who had not exceeded their bluegrass threshold, significantly more than expected (29%) reported having a management goal to maintain the abundance of bluegrass currently on their rangeland (adjusted residual = 4.03) or had no goal for bluegrass (67%, adjusted residual = 5.61).Significantly fewer of these landowners than expected reported the goal to decrease the current abundance of bluegrass (4%, adjusted residual = −10.42).In comparison, among landowners who indicated the abundance of bluegrass had exceeded their tolerance threshold, significantly more than ex-Figure 3. Kentucky bluegrass beliefs for landowners who rated their abundance of Kentucky bluegrass as within their tolerance threshold (i.e., not too high) or as exceeding their threshold (i.e., too high).An asterisk ( * ) indicates beliefs that were significantly different ( P < 0.05) between landowners who were within their tolerance threshold for bluegrass and those who had exceeded their threshold for bluegrass.pected (73%) reported the goal of decreasing the abundance of bluegrass (adjusted residual = 10.42) and significantly fewer than expected (3%) reported the goal to maintain bluegrass abundance (adjusted residual = −4.03)or had no goal (24%, adjusted residual = −5.63).

Factors related to tolerance threshold for Kentucky bluegrass
Primary land uses were significantly different between landowners who had and had not exceeded their tolerance threshold for Kentucky bluegrass ( Figure 2 A; X 2 [3] = 8.75, P = 0.033, Cramer's V = 0.20, n = 209).More than half of the landowners (54%) who indicated that the abundance of Kentucky bluegrass was too used their land primarily as a livestock which was significantly more than expected (adjusted residual 2.6).Fewer landowners than expected (5%) primarily used their land for agriculture (unspecified) or other land (adjusted residual = −2.02).Landowners who were tolerant of bluegrass consisted of fewer people than expected who used their land primarily as a livestock (35%; adjusted residual = −2.57)and more than expected (15%) in the unspecified agriculture/other land use group (adjusted residual = 2.02).
Landowners who had their threshold for Kentucky bluegrass reported significantly greater familiarity with Figure 2 B; Mann-Whitney U: z = −6.219,P 0.001, n = 224).Most landowners (70%) who had exceeded their threshold bluegrass reported that they were very to extremely familiar Kentucky bluegrass (mean [M] = median = 4), whereas the majority of landowners (72%) who had not exceeded their threshold were not at all to moderately familiar with the species (M = 2.88, median = 3).who indicated that Kentucky bluegrass had exceeded their tolerance threshold reported having more bluegrass on their land in terms of both extent (see Figure 2 C; Mann-Whitney U: z = −5.074,P < 0.001, n = 164) and proportion (see Figure 2 D; Mann-Whitney U: z = −5.505,P < 0.001, n = 159) than landowners who were tolerant of the species.Landowners who reported that bluegrass was too high estimated that the species was on over as much of their rangeland acres (M = 36% of rangeland; median = 25%) and comprised more than twice as much of the plant community (M = 37% of plant community; median = 30%) compared with those who had not exceeded their tolerance threshold.
Landowners who were tolerant of bluegrass were largely unsure about bluegrass impacts beyond beliefs that bluegrass probably is a valuable source of early season forage for livestock (M = 3.73), is preferred by livestock during its primary growing season (M = 3.47), and reduces soil erosion (M = 3.81; Figure 3 ).Landowners who indicated Kentucky bluegrass was too high had stronger beliefs about most potentially desirable and undesirable impacts of bluegrass; however, they were more unsure if bluegrass was preferred by livestock over other grass species during bluegrass's primary growing season (M = 2.89).Landowners who had exceeded their threshold for bluegrass also believed that bluegrass probably did not increase the number of wildlife species (M = 2.15), soil quality (M = 2.47), and water availability (2.51) in rangelands.They also did not believe that bluegrass improves the natural beauty of rangelands (M = 1.83) or maintains its nutritional value once mature (M = 1.83).Landowners who had surpassed their bluegrass threshold believed that bluegrass probably decreases forage available for livestock over the entire growing season (M = 4.37), outcompetes other grasses and forbs (M = 4.31), decreases their ability to cope with drought (M = 3.81), reduces the number of pollinator species in rangelands (M = 3.81), and makes rangelands less pristine (M = 4.12).

Coping capacity
A landowner's management of an invasive species is determined by not only their perceptions of the invader but also perceptions of available management options and their capacity to enact a response.Therefore, we next examine the coping capacity of landowners who self-reported that they did have, or were unsure if they had, Kentucky bluegrass on their rangelands and expressed a management goal to either maintain or decrease the abundance of bluegrass (32% of analysis sample, n = 88).

Perceived behavioral control
Landowners' perception that they could effectively manage bluegrass on their rangeland was fairly evenly distributed, with 39% agreeing or strongly agreeing that they could effectively manage the species and 35% disagreeing or strongly disagreeing (M = 2.97, median = 3).However, landowners with the goal to decrease the abundance of bluegrass on their land expressed significantly lower perceived behavioral control than those who wanted to maintain the abundance of bluegrass (Mann-Whitney U: z = 3.77, P < 0.001, n = 88).The majority of landowners (58%) who wanted to maintain the current abundance of Kentucky bluegrass on their land agreed or strongly agreed that they could effectively manage bluegrass, followed by 30% who said neither, and 12% who disagreed or strongly disagreed (M = 3.42, median = 4).In contrast, the majority of landowners (54%) who wanted to decrease the current abundance of bluegrass disagreed or strongly disagreed that they could effectively manage bluegrass at their desired level, followed by 23% who neither agreed or disagreed, and 23% who agreed or strongly agreed (M = 2.6, median = 2).

Attitudes about controlling or reducing bluegrass
Overall, landowners held a range of attitudes about controlling or reducing Kentucky bluegrass on their rangelands.Landowners were neutral about the feasibility (M = 3.94, SD = 1.47) of controlling or reducing bluegrass but indicated that it would be slightly beneficial (M = 5.07, SD = 1.38), slightly desirable (M = 4.71, SD = 1.67), and between neutral and slightly possible (M = 4.57; SD = 1.42).Landowners considered controlling/reducing Kentucky bluegrass slightly difficult (M = 2.90, SD = 1.58), between neutral and slightly expensive (M = 3.48, SD = 1.45), and between neutral and slightly ineffective (M = 3.59, SD = 1.40).
Attitudes about controlling or reducing Kentucky bluegrass differed between landowners with the goal to maintain the current abundance of Kentucky bluegrass on their rangelands and those with the goal to decrease bluegrass ( Figure 4 ).For all attitude items, neutral ratings dominated the responses from landowners who wanted to maintain bluegrass, whereas those who wanted to decrease bluegrass expressed more polarized views.Compared with landowners who wanted to maintain bluegrass, landowners who wanted to decrease bluegrass considered controlling/reducing bluegrass significantly more desirable (Mann-Whitney U: z = −3.97,P < 0.001), possible (Mann-Whitney U: z = −1.97,P = 0.049), necessary (Mann-Whitney U: z = −4.94,P < 0.001), and beneficial (Mann-Whitney U: z = −5.03,P < 0.001) but also significantly more difficult (Mann-Whitney U: z = 3.76, P < 0.001) and expensive (Mann-Whitney U: z = 1.99,P = 0.046).There were no differences in landowners' attitudes regarding how effective or feasible it was to control/reduce bluegrass based on landowners' management goal for bluegrass.

Beliefs about effectiveness of practices
Landowners expressed relatively low confidence in the effectiveness of potential management practices to control/reduce Kentucky bluegrass.Landowners rated intensive livestock grazing in the early spring (M = 2.63, SD = 0.99), mob grazing (M = 2.69, SD = 1.13), multispecies grazing (M = 2.49, SD = 1.12), and targeted application of herbicides (M = 2.43, SD = 1.21) between slightly and moderately effective.Landowners believed prescribed fire was only slightly effective (M = 2, SD = 1.05) at controlling/reducing Kentucky bluegrass.While landowners were provided with the option to write in a management practice for Kentucky bluegrass and indicate their beliefs about its effectiveness, no landowners wrote in additional practices.There were no significant differences in landowners' beliefs about effectiveness based on their management goal for bluegrass.

Current and intended Kentucky bluegrass management
Of the 88 landowners who had a management goal to maintain or reduce bluegrass, less than half (46%) were currently (i.e., within the past 2 yr) employing practices to manage Kentucky bluegrass on their rangelands.Among these landowners, early intensive livestock stocking was the most commonly used practice (43%), followed by mob grazing (17%) and herbicides (8%).Only 3% of landowners reported currently using multispecies grazing to manage Kentucky bluegrass, and no landowners were currently using prescribed fire.Several landowners indicated that they currently used multiple practices; early intensive stocking and mob grazing was the most common combination (9%).One landowner with the goal to decrease bluegrass wrote in that they were currently using twice over rotational grazing to manage Kentucky bluegrass.Significantly more landowners than expected with the goal to decrease Kentucky bluegrass were currently using early intensive stocking (55% of landowners), compared with only 31% of those with the goal to maintain bluegrass (X 2 [1] = 4.608, P = 0.032, Cramer's V = 0.246, n = 76).There were no differences among the other practices currently used to manage bluegrass between landowners who wanted to maintain or decrease the abundance of bluegrass.
More than half (61%) of landowners with a management goal for Kentucky bluegrass indicated they were likely or extremely likely to manage Kentucky bluegrass on their rangelands in the next 12 mo.Similar to the practices currently used, early intensive stocking was the most common practice landowners intended to use to manage bluegrass (53% of landowners), followed by mob grazing (24%) and herbicides (11%).Few landowners intended to use multispecies grazing (3%) or prescribed fire (1%) to manage bluegrass.Of landowners who indicated intentions to use multiple practices, the most common combination was early intensive stocking and mob grazing (17%).One landowner with the goal to decrease bluegrass wrote in that they were extremely likely to manage Kentucky bluegrass using twice-over rotational grazing.
As seen with practices currently used, landowners' intentions to manage Kentucky bluegrass through early intensive stocking was the only management intention that varied by management goal (X 2 [1] = 4.454, P = 0.035, Cramer's V = 0.238, n = 79).Almost two-thirds of landowners (64%) who wanted to decrease Kentucky bluegrass indicated they were likely or extremely likely to use early intensive stocking in the next 12 mo, whereas only 41% of those who wanted to maintain bluegrass were likely or extremely likely to use the Apart from early intensive stocking, most landowners were decidedly unlikely to use each practice to manage Kentucky bluegrass on their rangeland in the next 12 mo ( Figure 5 ).Nearly all landowners (95%) were unlikely or extremely unlikely to use prescribed fire or multispecies grazing (85%) to manage Kentucky bluegrass.

Discussion
Each year billions of dollars are spent in the United States to control invasive plant species ( Westbrooks 1998 ;Pimentel et al. 2005 ).Yet many control campaigns only gain momentum once the invasion is advanced and negative impacts have become widespread and severe (e.g., Roberts et al. 2018 ;Epanchin-Niell et al. 2010 ).Control effort s often lack, or significantly lag in.engagement with the human and social dimensions of the invasion ( Shackleton et al. 2019 a and2019b).Kentucky bluegrass embodies the Trojan horse.It is a seemingly beneficial gift to livestock producers that may increase overall forage productivity without disrupting the flow of other ecosystem services; however, if left unchecked, undesirable impacts can quickly accelerate to the detriment and surprise of landowners and managers.To effectively steer the Kentucky bluegrass invasion, managers of large tracts of land and the coordinated efforts of many landowners are needed.But notably, our results suggest that both motivation and ability to proactively manage Kentucky bluegrass may be lacking among agricultural landowners.In parallel, the climate of the northern Great Plains is already bound by extremes, which are projected to become more pronounced and potentially more favorable to Kentucky bluegrass in the future through the lengthening of the growing season accompanied by greater winter and spring precipitation ( Shafer et al. 2014 ;Conant et al. 2018 ).
Kentucky bluegrass is likely not a management priority for most large-scale agricultural producers in North Dakota's rangelands.In our sample, only half of responding landowners reported having Kentucky bluegrass on their rangelands, most were tolerant of the current abundance of bluegrass, and a majority indicated that they had no management goal for the species.Landowner self-reports of bluegrass presence contrast National Resources Inventory data that report the presence of Kentucky and Canada bluegrass ( Poa compressa, an invasive closely related to Kentucky bluegrass) of 86% ± 3.7% in North Dakota (USDA-NRI 2018 ).Landowners' overall tolerance of bluegrass and lack of management goals for the species also conflict with the common beliefs held by many research and extension personnel that most landowners are intolerant of bluegrass and managing to reduce it (personal communication with presurvey subject matter experts).This reflects a knowledge and understanding mismatch between private landowners working on and managing the landscape and those doing research and extension.As recognized with other invasive grasses, people aren't likely to prioritize, let alone manage for, a species they are not aware they have ( Johnson et al., 2011 ;Kelley et al., 2013 ).
Despite potentially low salience overall, a subgroup of responding landowners indicated that Kentucky bluegrass had exceeded their tolerance threshold and that they were more likely to have management goals to decrease the abundance of the species.Compared with landowners who were tolerant of bluegrass, those who considered Kentucky bluegrass as too high consisted of more livestock producers and reported having greater familiarity and exposure to the species.Primary land uses can shape personal experiences leading to beliefs about Kentucky bluegrass impacts (e.g., Shackleton et al. 2019 b); however, they may not be sufficient to explain sensitivity to and the perceived threat of ecological change ( Sorice et al. 2022 ;Sorice et al. 2023 ).Although all landowners tended to believe that Kentucky bluegrass had probable benefits as early-season forage for livestock and through reducing soil ero-sion, landowners who had exceeded their tolerance threshold for bluegrass held stronger beliefs that bluegrass produced a variety of undesirable outcomes (e.g., decreased forage available for livestock, decreased ability to cope with drought, reduced the number of pollinator species in rangelands, loss of pristine rangelands).Shifts in landowners' tolerance and beliefs toward increased perceived threat of bluegrass align with expected ecological impacts associated with increased stages of bluegrass invasion.
Comparing landowners' estimates of bluegrass extent and its proportion of the total plant community, based on their tolerance or intolerance of Kentucky bluegrass, indicates alignment between landowners' experiential knowledge and expert opinion that Kentucky bluegrass has increasingly detrimental impacts over 35% foliar cover.Recent research by Toledo et al. (2023) found that in grasslands of central North Dakota, pastures that were completely invaded by Kentucky bluegrass could not meet nutritional requirements for cow-calf pairs throughout the grazing season without supplementation.These authors suggest that Kentucky bluegrass pastures may not meet metabolizable energy and protein requirements throughout the growing season because of decreases in forage quality as the grazing season progresses.In addition to "ground truthing" a likely Kentucky bluegrass threshold with complementary sources of knowledge from different ways of knowing, the connection between landowners' experience with increased exposure to bluegrass and the results of recent empirical findings highlight a promising opportunity for peer-to-peer learning among landowners.However, this must be balanced with the consideration that responding landowners in our study varied greatly in their estimates and, potentially, their ability to identify Kentucky bluegrass.
Less than half of landowners in our sample had a management goal for Kentucky bluegrass.Among landowners with the goal to either maintain or decrease Kentucky bluegrass, perceived capacity to cope with the species was moderate to low.Coping capacity was lower for landowners with the goal to decrease bluegrass than for those with the goal to maintain, potentially nullifying intentions to reduce the species.In fact, the majority of landowners who wanted to decrease the abundance of Kentucky bluegrass on their rangelands did not think they could effectively do so.While landowners who wanted to maintain bluegrass were largely ambivalent about many facets of controlling or reducing the species, landowners with the goal to decrease bluegrass considered the instrumental aspects of control (e.g., beneficial, necessary, desirable) more positively but evaluated the experiential aspects of control (e.g., cost and difficulty) as more negative.Further, landowners may not believe that there are any effective management practices for Kentucky bluegrass.This underscores a strong need for the rangeland science community to innovate around effective and feasible Kentucky bluegrass management techniques, including potential adaptations to current management to better utilize bluegrass, and promote findings that provide viable options.
Given low perceived capacity to cope with bluegrass, it is not surprising that a substantial implementation gap exists among landowners with management goals for Kentucky bluegrass and those who are currently, or intend to, manage the species on their rangelands.Emerging research highlights prescribed fire as a promising management tool for Kentucky bluegrass −invaded areas, which also has potential to mitigate drought effects on forage resources and maintaining animal performance ( Spiess et al. 2020 ).Rangeland scientists consider returning fire to the landscape critical for not only maintaining or restoring grassland ecosystems in the northern Great Plains but also buffering rangeland enterprises against climate extremes ( Spencer et al. 2015 ).However, no landowners in our study were currently using prescribed fire and nearly all (95%) were unlikely to use the practice in the next 12 mo.Echoing our findings, Bendel et al. (2020) found that the perceived threat low levels of Kentucky bluegrass invasion did not warrant behavioral changes among respondents related to prescribed fire.However, behavioral research suggests that as issues become more salient, landowners might be more receptive to information that might influence their management decisions (e.g., Prochaska and DiClemente 1983 ).

Potential pathways forward
There is a clear disparity between Natural Resources Inventory data about bluegrass presence and the associated concerns of rangelands scientists and responding landowners' identification of and perceptions about the species on their land.The responsibility for closing this gap may at first appear to fall on either scientists to "better communicate" the science and on land managers to "better seek out and apply" the science, depending, perhaps, on the readers' identification with either group.However, both reinforce the attractively simple but flawed information-deficit model ( Simis et al. 2016 ).This false dichotomy also fails to account for the recognition that actionable knowledge is less a "thing" to be received and then used and more a "process of relating" involving social learning and the transformation of scientific information and experiential information into knowledge applicable and actionable within a specific context ( Stern et al. 2021 ).Such information conversions focused on Kentucky bluegrass knowledge can be supported through intentional interactions emphasizing multiple forms of trust development when agency and extension personal consult with landowners.Going further, participatory processes including diverse compositions of scientists, managers, landowners, interested and affected parties and rights holders can provide social learning environments conducive to joint exploration of problems and refinement of actionable solutions.Codesigning research and solutions with end users incorporates benefits of humancentered design to innovate sustainable solutions (e.g., Sorice and Donlan 2015 ) and promises not only better research products but also faster discovery, dissemination, and uptake of new approaches specifically in agricultural contexts ( Wilmer et al. 2018 ;DeLong et al. 2022 ).
While most North Dakota landowners are not currently amenable to fire, recent research has found that they are receptive to incorporating scientific research into land management decisions ( Bendel et al. 2020 ).Within the social learning space created through a community of practice, such as a burn association, prescribed fire demonstrations and evidence from researchers and landowners related to cost-effectiveness and favorable impacts of fire on forage production may be more impactful than agency-led outreach alone.Development of novel conservation programs can nudge engagement with prescribed fire and burn associations, especially for landowners with little to no previous exposure.Focusing on the social sciences to improve two-way communication, monitoring, and trust would build on existing programs and make them more effective at improving long-term conservation and stewardship behavior ( Vollmer-Sanders et al. 2016 ).
Another pathway for Kentucky bluegrass management points toward the continued exploration of nonfire management practices that can control Kentucky bluegrass, recognizing the near impossibility of eradication.For example, herbicides have shown some efficacy but are not as economically feasible or ecologically advantageous given costs, long residual times, and impacts to surrounding vegetation ( Cornish and Burgin, 2005 ).Alternatively, the current engagement and interest of some rangeland landowners in mob grazing to manage Kentucky bluegrass may provide a fruitful area for more participatory research, building on existing experiments at the USDA-Agricultural Research Services-Northern Great Plains Research Laboratory.While potentially disheartening to rangeland scientists and managers interested in restoring fire to the northern Great Plains for a suite of additional ecological benefits, a step back from fire may allow for more authentic participatory exploration and co-creation of solutions that better account for existing values, beliefs, and management needs of North Dakota landowners (e.g., Sorice and Donlan 2015 ;DeLong et al. 2022 ).Additionally, a sole focus to simultaneously overcome landowner ambivalence about Kentucky bluegrass, as well as entrenched negative beliefs about fire, may slow actors in the social-ecological system into a state of nonresponse.However, participatory processes focused on novel Kentucky bluegrass management solutions also do not preclude the emergence of attitudinal shift and actionable knowledge related to prescribed fire in combination with, or isolation from, other practices.Potential participatory pathways for Kentucky bluegrass management include nurturing and supporting a social norm of prescribed fire through the development of conservation programs.Prescribed Burn Associations can be effective experiential and normative agents starting with willing landowners and using social proof and social networks to promote further diffusion of the practice ( Twidwell et al. 2013 ).

Implications
Increasingly, many conservation and natural resource organizations are recognizing the limited ability of scientists and managers to resist ecological transformations and thus broadening the range of response options to include acceptance and intentional effort to direct the transformation; that is, Resist-Accept-Direct (e.g., Schuurman et al. 2022 ).Despite the recent acceleration of Kentucky bluegrass invasion in the northern Great Plains, there is still an opportunity to proactively respond to bluegrass and shape the novel ecosystem emerging in its wake.Through understanding landowner perceptions, tolerance thresholds, and coping capacity, a new ending to the Trojan horse analogy may be possible in which landowners and rangeland managers are neither overrun by the surprise invasion nor overwhelmed by their own effort s to burn the wooden horse down but instead can collaboratively salvage and upcycle the materials into something valuable.
Maintaining bluegrass below the approximate 35% threshold, before deleterious impacts to productivity and other ecosystem processes accelerate, will only be possible through bolstering the adaptive capacity of landowners to recognize and respond to this ecological change.Although challenging, this also presents a serious call to action for rangeland researchers, managers, and conservation outreach and extension professionals toward innovation informed by and accounting for relevant human and social dimensions.We encourage reflexivity and examination of traditional research and science communication approaches toward more collaborative and participatory processes that can yield actionable Kentucky bluegrass knowledge and sustainable management pathways for the northern Great Plains social-ecological system.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Figure 1 .
Figure 1.Map of the nine North Dakota study area counties.Bars show the proportion of farmland, per county, that was used as grassland/pasture (black bar) or cropland (white bar) based on the 2017 Census of Agriculture ( USDA-NASS 2019 ).US Environmental Protection Agency Level III ecoregion boundaries( Bryce et al. 1996 ) are color coded and listed by common name ( NDGF 2019 ).

Figure 2 .
Figure 2. Comparison of landowners'A, primary land use; B, familiarity with Kentucky bluegrass; C, estimate of Kentucky bluegrass extent; and D, estimate of Kentucky bluegrass proportion based for landowners' who indicated that Kentucky bluegrass was within their tolerance threshold (i.e., not too high) or had exceeded their threshold (i.e., too high).For estimates of extent ( C ) and proportion ( D ), the mean is indicated by the diamond within each boxplot.

Figure 4 .
Figure 4. Landowners' attitudes about controlling/reducing Kentucky bluegrass based on whether landowners had the management goal to maintain or decrease bluegrass.An asterisk ( * ) indicates attitude items were significantly different between landowners with the goal to maintain or decrease bluegrass at P < 0.05 level in Mann-Whitney U tests.

Figure 5 .
Figure 5. behavioral intentions to use each practice to manage Kentucky bluegrass on their rangelands in the next 12 mo.Includes all landowners with the goal to either maintain or decrease bluegrass.