Locally Sourced Seed is a Commonly Used but Widely Defined Practice for Grassland Restoration

Globally, grasslands are among the most threatened ecosystems (Hoekstra et al. 2005), and in North America, grasslands face challenges on many fronts, including conversion, fragmentation, and encroachment by invasive species. North America's grasslands have experienced dramatic conversion for row-crop agriculture (Wright and Wimberly 2013; Gage et al. 2016; Comer et al. 2018). In fact, conversion is nearly complete in some tallgrass systems, and loss continues in many parts of the Great Plains (Lark et al. 2015). Consequently, this loss has severely fragmented what grasslands remain. Fragmentation of grasslands threatens both wildlife and plant communities by reducing local population viability through loss of gene flow or by providing insufficient resources to fulfill life history needs (Johnson et al. 2010; Frankham et al. 2017), and climate change is likely to intensify the impact of these threats (Perkins et al. 2019).

Addressing these challenges will likely require complete reconstruction or restoration of degraded grasslands. Here, we use the more commonly used term restoration to refer to the practice of reestablishing native grassland plant communities on disturbed, often plowed, land. Restoring grassland is an important strategy to maintain or increase population viability for many species in increasingly fragmented landscapes. Furthermore, increasing or maintaining connectivity is one of the most frequently cited strategies to enhance species adaptation to climate change through range shifts (Heller and Zavaleta 2009). The increasing challenges facing North America's grasslands suggest that ecosystem structure and function might be diverging from historical patterns (Fuhlendorf et al. 2018; Perkins et al. 2019). Restoration goals often focus on creating species assemblages that mimic precolonization ecological communities and processes, but Gann et al. (2019) suggested that reference models or conditions for restoration should be based on the condition of the site before degradation, which may or may not represent precolonization conditions and should also consider environmental change. For example, instead of striving to restore a crop field using tallgrass prairie diversity and function goals based on precolonization conditions, practitioners (we define here as anyone conducting grassland restorations in the United States or Canada) should aim for conditions that mimic current diversity and function and allow for adaptation to changing conditions. Rowe (2010) advised that grassland restoration can be complex, lengthy, and expensive and often uses a variety of methods that depend on the goals of the restoration. Therefore, from site preparation to established grassland, incorporating strategies to mitigate the impacts of climate change or other anticipated future conditions will be important for restoration success.

One step in the process that can have a large impact on future success is the development of the seed mix. Creating a seed mix involves many considerations, including cost, seed availability, source of the seed, and how many and which species to use. The cost and availability of seed are constant constraints for any practitioner. However, within these constraints, numerous options may exist for which species to include in the mix and the origin of that seed. Seed source options for grassland restoration include cultivars of native species, native species grown in production plots, local ecotype grown in production plots, and wild harvested seed. Within the conservation community, practitioners generally accept that using local ecotype seed, whether from production plots or wild harvested, will produce the best outcomes (Hamilton 2001; Wilkinson 2001).

The justification for using local seed is that seed sourced locally will be better adapted to the biotic and abiotic conditions of the restoration site than nonlocal seed. There is evidence that this is the case for many species (Bucharova et al. 2017), but the pattern is not always consistent (Carter and Blair 2013). Kawecki and Ebert (2004) suggested that reciprocal transplant studies are the best way to evaluate local adaptation. In these studies, researchers identify two distinct populations for a given species and plant seeds from both populations at each location to determine which population has the greatest fitness (e.g., survival and seed set) at each site. Local plants outperforming nonlocal plants at both sites is considered compelling evidence for local adaptation (Kawecki and Ebert 2004; Leimu and Fischer 2008). However, in a meta-analysis of reciprocal transplant studies, Leimu and Fischer (2008) found that while 71% of studies had evidence of local plants outperforming nonlocal plants at one transplant location, local plants only outperformed nonlocal plants at both locations in 45% of the studies. Additionally, as the condition of remnant grasslands degrades due to fragmentation or invasive species, the ability of native plant communities to maintain viable local adaptation may suffer due to genetic drift or inbreeding as gene flow is reduced (Schlaepfer et al. 2018). Finally, a changing climatic regime might further decrease the viability of isolated populations that have limited ability to disperse or adapt in a place with limited gene flow (Breed et al. 2013).

One of the challenges in understanding the extent and importance of local adaptation in grassland plants is the variability in how practitioners define “local,” whether that is by distance, ecoregional zone, soil type, etc. As scientists tackle these questions, for results to be useful to practitioners, it is important that scientists and practitioners have a shared understanding of what locally sourced seed means. Moreover, even though the use of local seed is commonly promoted, it is not clear how widely this practice is used in restorations and, if it is being used, whether changing conditions like fragmentation and climate change are also being considered in seed sourcing decisions. We sought to survey practitioners to better understand how commonly locally sourced seed is used in restorations and, for those using local seed, to understand how practitioners defined local and their justification for focusing on local seed.

To evaluate the seed sourcing strategies and justifications used by practitioners, we developed an electronic survey through the online platform Survey Monkey (www.surveymonkey.com). Our target audience for the survey was practitioners, which we defined here as anyone conducting grassland restorations in the United States or Canada. Because the focus of the survey was grassland restorations, we focused the geographic scope of the survey on the Midwest and Great Plains. However, we recognized that there are many important types of grasslands outside of central North America, and including these areas could enhance our ability to learn about a wider diversity of seed sourcing practices and justifications. Therefore, we did not limit the geographic scope of the survey within the United States and Canada.

We used Dillman et al. (2009) to guide the development of the questions and the way that the questions were presented. In the survey, we asked questions regarding whether seed sourcing policies existed, the nature of those policies, the justification for those policies, and any other climate adaptation strategies considered in restoration practices (see Text S1, Supplemental Material, for full survey design and questions). Respondent answers about whether local seed sourcing policies guided their work determined the number of questions that they were asked. If individuals reported no organizational, state/local government, or informal policies around using local seed sources, they completed the survey with seven questions. Individuals that reported either state/local government or informal policies completed the survey with 13 questions, and individuals that reported organizational policies completed the survey with 14 questions. The survey software automatically directed respondents to the appropriate set of questions based on their responses to questions 5, 15, and 23 (Text S1, Supplemental Material). None of the questions required answers. Individuals were allowed to skip questions that they did not want to answer as they were presented.

To determine the geographic distribution of responses, we included multiple-choice options for the states and provinces most likely to be conducting grassland restoration, and we included an “Other” option to allow for additional states or provinces. When respondents used the Other option to include multiple states across which they work, we included in our summary all the states listed by each respondent. To summarize where policy for locally sourced seed existed, we grouped responses by policy type (e.g., organization, state/local government, individual, or none). The organizational policy type summary included the number of respondents stating that their organization or agency had an internal policy for locally sourcing seed (Text S1, Supplemental Material, yes to question 5). The state/local government policy type summary included the number of respondents stating that state or local government had a policy for locally sourcing seed (Text S1, Supplemental Material, yes to questions 11 and 15). We also included respondents who had an internal organizational policy in the summary for state/local government policy if their state or local government also had local seed sourcing policies in place. The individual policy type summary included the number of respondents stating that even though no organizational or state/local government policy existed, they considered local seed sources in their restorations anyway (Text S1, Supplemental Material, yes to question 23). Finally, to determine the number of respondents who did not have organizational, state/local government, or individual policies, we summarized the number of respondents answering no to all of the above questions (Text S1, Supplemental Material, no to questions 5, 15, and 23). To elucidate which types of organizations and agencies have enacted policies for local seed sourcing, we summarized only the respondents who answered yes to having internal policies for seed sourcing by affiliation (e.g., non-profit, local government, etc.).

To understand how practitioners defined locally sourced seed, we asked an open-ended question about the definition of local. Because respondents generally answered either conceptually or using a defined distance from their restoration site, we summarized conceptual and distance responses separately. We grouped responses based on similar terms used in the definition. For example, all answers referring to ecoregion were grouped together. We also directly asked whether policies dictated specific distances for locally sourced seed and if so, what those distances were.

Many organizations define locally sourced seed using a maximum distance, and we wanted to understand how that maximum distance varied. We directly asked respondents with a multiple-choice question if their organization had a distance from which seed was considered local. To understand what motivated agencies to promote the use of locally sourced seed, we asked about the factors that influenced the seed sourcing policy with a multiple-choice question that included an Other option. If climate change was a factor in their decision, we asked a follow up open-ended question asking them to describe how they considered climate change. We summarized these results the same way as the open-ended question regarding the definition of local by grouping responses based on similar terms used in the definition. Finally, to understand how people considered climate change in their restoration work, we asked whether they used any other climate adaptation strategies. Here, we also used a multiple-choice question with an Other option. We were approved to conduct the survey through The Nature Conservancy's review process for research with human subjects.

The survey opened on November 1, 2017, and was open for 3 mo. We sent the survey link and completion requests directly to 42 contacts from the following states and provinces: California, Colorado, Illinois, Iowa, Kansas, Manitoba, Minnesota, Missouri, Montana, Nebraska, North Dakota, Oklahoma, Ohio, Ontario, Oregon, Saskatchewan, South Dakota, Texas, and Wisconsin. We also distributed the survey link and completion requests through the following networks: Minnesota Department of Natural Resources' Local Technical Teams, U.S. Fish and Wildlife Service's Prairie Network, Prairie Reconstruction Initiative (https://sites.google.com/view/prairiereconinitiative/home), and the Prairie Ecologist blog (https://prairieecologist.com/). In the email to contacts and networks, we encouraged people to pass it on to other practitioners. To analyze the results of the survey, we report the frequency of responses. Because not all respondents answered all questions, we used the total number of responses for individual questions in our analyses. Therefore, we calculated the frequency of response by dividing the number of people choosing a given response by the total number of responses for that question. For open-ended questions, we collated all the answers, grouped the answers by similarity of response, and reported the frequency of the grouped responses.

We had 494 responses to the survey, and 76% of the respondents answered all the questions presented to them. Respondents took an average of 7 min to complete the survey. The geographic distribution of responses was much wider than we anticipated. We had 40 U.S. states and 5 Canadian provinces represented in our responses, with the bulk of the responses coming from the Midwest and Great Plains regions (Figure 1; Table S1, Supplemental Material). Policy for sourcing seed locally was common, with only 3.7% of people stating that no organizational, state/local government, or individual policies exist and 47.8% of people stating that their organizations had some type of local seed sourcing policy (Figure 2; Tables S2–S5, Supplemental Material). Additionally, 91.2% of people without formal organizational or state/local government policies for sourcing seed locally reported considering local seed sources anyway. Within the positive organizational policy responses, state governments and non-profit organizations were more likely to have policies regarding the use of local seed (Table 1; Table S6, Supplemental Material).

When defining what local means within the context of their policy, respondents either used descriptive terms or a numerical distance from the restoration site. For sake of interpretation, we summarized the descriptive and numerical responses separately (Figure 3; Tables S7–S9, Supplemental Material). Both types of responses had a wide range of variability. Some of the more common responses included ecoregion, state/province, or nearby counties when using descriptive terms (Figure 3A). While governmental boundaries are easily defined, this does not account for the possible differences in the scale of how organizations define less distinct regions such as ecoregion or landscape. When using numerical distances, we observed that larger distances, such as 101–200 or 201–400 km, were the most common response (Figure 3B). When asked explicitly about whether their policy included a maximum distance for sourcing seed, we found that trends were similar to how practitioners defined local using distance (Figure 3B), but over 50% of respondents used a criterion other than distance for determining source, and 26% responded that seed from any distance was allowable (Tables S10–S12, Supplemental Material). In both cases, the larger distances from the source site were the more common responses.

Finally, we were interested in how climate change factored into restoration decisions for seed sourcing. When asked which factors were considered when developing the locally sourced seed policy, less than 20% of respondents said climate change was considered (Figure 4; Tables S13–S15, Supplemental Material). However, practitioners commonly considered local adaptability (80%), preserving local gene pools (76%), and increasing establishment success (62%). In the Other category, five other factors had at least five mentions: availability of seed, invasive species management, pollinator habitat, cost, and regulations.

For respondents who factored climate change into their seed sourcing policy, we had an open-ended question for them to explain how it influenced their policy. We only had 67 responses to this question, and 50 of them had a complete response. Of those 50 responses, the most common strategy was to safeguard species by maintaining local adaptations (Table 2; Tables S16–S18, Supplemental Material). Expanding seed zones and focusing on southern seed sources were also common strategies. Finally, we asked if other climate adaptation strategies were used in the restoration process, and less than 15% said no (Figure 5; Tables S19–S21, Supplemental Material). The most commonly used strategies were maximizing the number of species in a seed mix and maintaining or increasing functional diversity in seed mixes.

Our survey respondents revealed that local sourcing of seed was their dominant strategy for grassland restoration. Even when no formal organization or state/local government policy existed to source seed locally, most practitioners were considering local seed sources for their restoration mixes. However, the definition of local was widely interpreted. Although the word local implies a seed source near the restoration site, when asked what they considered local, respondents reported relatively large areas or distances (e.g., ecoregion or hundreds of kilometers). This could be an artifact of how we framed the question. We suspect that respondents reported the maximum distances that they would consider local. In fact, after asking them to describe their definition of local, the survey did ask whether a maximum distance policy existed. Therefore, when considering seed for their projects, respondents might prefer sources closer to a restoration site if available. Regardless, the concept of obtaining seed from local sources was familiar to and common among the wide diversity of practitioners surveyed.

The practice of local seed sourcing is consistent with many recommendations in the restoration ecology literature and, as noted in the survey, is often dictated by policy (Clewell and Rieger 1997; Wilkinson 2001). Local adaptation can enhance the success of grassland plants (Joshi et al. 2001; WeiBhuhn et al. 2012), and over 50% of city government, state government, and nonprofit practitioners said they have organizational policies requiring the use of locally sourced seed for restorations. However, the field and experimental evidence suggests that the success of local seed sources is not universal (Galloway and Fenster 2000; Leimu and Fischer 2008; Carter and Blair 2013), and given the level of anthropogenic disturbance in grassland landscapes in North America (Comer et al. 2018) and the pace of climate change (Dobrowski et al. 2013), practitioners may need to reconsider their seed sourcing strategies.

Climate change was not a driving factor for developing seed sourcing strategies among respondents. The prevalence of strategies such as preserving local gene pools and maintaining local adaptation is consistent with the commonly reported practice of locally sourcing seed for restorations. Practitioners were generally most concerned about current establishment success and factors that they perceived to be important to establishment success. Although cost and availability of seed were not given as options, they were both written in as driving factors in seed sourcing strategies for the Other category, and these constraints may have been in practitioner's minds as they were answering the question. Including cost and seed availability as choice options or asking questions to tease apart seed source strategy from logistical constraints would have been informative.

Although climate change did not rank highly for practitioners in this survey, Havens et al. (2015), Breed et al. (2018), and others proposed that many seed sourcing strategies facilitate long-term adaption to climate change for grassland restoration. These strategies include assisted migration of species northward outside their current range (McLachlan et al. 2007), expanding or shifting seed zones (Galatowitsch et al. 2009), and climate matching or identifying seed sources that currently experience the projected climate of a restoration site (Galatowitsch et al. 2009; Breed et al. 2018). Probably one of the most suggested strategies is simply facilitating gene flow within a species' current range either locally or regionally (Broadhurst et al. 2008; Aitken and Whitlock 2013; Breed et al. 2013; Carter and Blair 2013; Bucharova et al. 2019). Although many strategies have been suggested and some are likely being tried, most survey respondents were not yet focused on climate change strategies, suggesting that there is room for more outreach and implementation of climate change strategies for seed sourcing in grassland restorations, and some organizations are developing planning tools to help practitioners consider climate in their seed source decisions (e.g., the Climate Smart Restoration Tool https://climaterestorationtool.org/csrt/).

Given the evidence for local adaptation (Leimu and Fischer 2008) and the need to consider climate in seed sourcing decisions (Breed et al. 2018), we recommend that practitioners build on the foundational practice of local seed sourcing using strategies such as regional mixing of seed sources from different populations (e.g., regional admixture; Bucharova et al. 2019). One of the potential barriers to regional admixture and potentially even to the development of policies around any seed sourcing strategy is the reliable availability of local seed. A regional admixture approach requires seed from multiple populations for each species included in a restoration mix. One strategy that practitioners have been using to overcome the limitations on seed availability and diversity in seed mixes (both species and population) is the development of seed collaboratives, where multiple agencies pool seed acquisition and harvesting resources (e.g., Northwest Minnesota Native Prairie Seed Consortium, https://lccnetwork.org/sites/default/files/Resources/Northwest%20Minnesota%20Native%20Prairie%20Seed%20Consortium_0.pdf). Partner collaboration in seed sourcing can reduce costs, increase efficiencies, and improve restoration outcomes.

Finally, although the survey was distributed through many different agency and organizational outlets, it is likely that practitioners passionate about grassland restorations were more likely to respond, potentially biasing results by sampling similar restoration philosophies. There was also a bias in responses from the upper Midwest, specifically Minnesota and Iowa. These two states have strong restoration cultures with a wide variety of agencies and organizations conducting grassland restorations. Despite these limitations, the overall number and geographic scope of responses bolsters the information we gained from this survey. Our results provide a snapshot of the current policies guiding grassland practitioners and insight about the need for more outreach on the consideration of climate change in seed sourcing strategies.

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Text S1. Full text of the front matter and questions for a survey that we sent to grassland restoration practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S1 (28 KB DOCX)

Table S1. Responses to the question “In which U.S. state or Canadian province do you work?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S2. Responses to the question “Does your organization have internal policies regarding the use of locally sourced seed in grassland reconstruction/restoration projects?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S3. Responses to the question “Does the state or local government have policies or laws regarding the use of locally sourced seed or the distance from which the seed may be obtained for use in grassland reconstructions/restoration projects?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S4. Responses to the question “Does the state or local government have policies or laws regarding the use of locally sourced seed or the distance from which the seed may be obtained for use in grassland reconstruction/restoration projects?” from a survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S5. Responses to the question “If no state or internal policies formally exist about locally sourcing seed for grassland reconstructions/restorations, does your office consider local seed sources anyway when possible?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S6. Responses to the question “What type of organization do you work for?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S7. Responses to the question “How does your organization define locally sourced for prairie restorations/reconstructions? Please describe your organization's definition of local” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S8. Responses to the question “If the state or local government requires locally sourced seed in grassland reconstructions/restoration projects, how do they define locally sourced? Please describe their definition of local” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S9. Responses to the question “If you do consider locally sourcing seed despite no policy, briefly describe how you define local” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S10. Responses to the question “Does your organization have internal policies regarding the distance from which seed may be obtained for use in prairie reconstruction/restoration projects?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S11. Responses to the question “Do the state or local governments have policies regarding the distance form which seed may be obtained for use in prairie reconstruction/restoration projects?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S12. Responses to the question “Does your office have a distance from which seed is considered for use in prairie reconstruction/restoration projects?” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S13. Responses to the question “What factors influence the development of the locally sourced seed policy? (check all that apply)” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S14. Responses to the question “What factors influenced the development of the locally sourced seed policy? (check all that apply)” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S15. Responses to the question “If you are not required by any policy to source seed locally but do this anyway, what factors influence this decision? (check all that apply)” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S16. Responses to the question “If climate change was considered in developing policies around locally sourcing seed for grassland reconstructions/restorations briefly describe how it factored into the policy development” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S17. Responses to the question “If climate change was considered in sourcing seed locally for grassland reconstructions/restorations briefly describe how it factors into decisions” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S18. Responses to the question “If climate change was considered in developing policies around locally sourcing seed for grassland reconstructions/restorations briefly describe how it factored into the policy development” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S19. Responses to the question “Which climate adaptation strategies does your organization use when implementing grassland reconstruction/restoration? (check all that apply)” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S20. Responses to the question “Which climate adaptation strategies does the state or local government use when implementing grassland reconstruction/restoration? (check all that apply)” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

Table S21. Responses to the question “Which climate adaptation strategies does your office use when implementing grassland reconstruction/restoration? (check all that apply)” from the survey that we sent to practitioners (anyone conducting grassland restorations in the United States or Canada) about local seed sourcing policies in 2017.

Available: https://doi.org/10.3996/JFWM-21-079.S2 (126 KB DOCX)

We thank the Cox Family Fund for Science, the many practitioners who took the time to complete this survey, D. O'Neal and J. Norland for guidance on the survey design, the anonymous reviewers, the Associate Editor, and the copy editor who provided valuable feedback to improve the manuscript.

Any use of trade, product, website, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government.