Abstract
While there is widespread recognition of the need to incorporate underserved minority communities into planning and design processes for just climate adaptation policies and programs, the dynamics of facilitating bilingual workshops are still significantly under-explored in the literature. This article addresses the strategies, challenges, and lessons learned from conceptualizing, preparing, and facilitating low-budget bilingual workshops for community engagement in green infrastructure (GI) planning in Tucson, Arizona, United States. The two case studies discussed are connected to multi-stakeholder initiatives aimed at involving historically underserved Hispanic communities in GI planning to address environmental justice issues. We first describe strategies to foster participation and exchanges in bilingual participatory workshops, where limited resources made professional interpretation and translation services unfeasible. We then share four lessons learned from these experiences by drawing on the successes and challenges of our approaches, and examining how cultural conceptions about language—our language ideologies—can impact community engagement in bilingual settings. The two case studies make a compelling argument for taking bilingual design and facilitation seriously to advance procedural and recognition justice in climate resilience planning through co-production in multilingual communities.
INTRODUCTION
As the world becomes increasingly urbanized, advancing sustainability in cities is one of the great challenges of our time. Cities worldwide are experiencing extreme heat and severe storm events that threaten human life, cause flooding, and damage public urban infrastructure and private properties. 1 Heat waves also add pressure to energy systems, with some cities experiencing blackouts, amplifying mortality and morbidity risk 2 and affecting the water supply. 3 In some cities with combined sewer systems, flooding risk is exacerbated by the risk of combined sewer overflows that translate into public health emergencies and property damage. 4
As climate change increases the frequency and severity of extreme heat and flooding events, many cities are turning to nature-based solutions or green infrastructure (GI) as a strategy for climate adaptation. 5 GI can be defined as a network of interconnected systems comprised of natural and human-made features that can enhance the quality of life and urban resilience by providing an array of ecosystem services. 6 In semi-arid regions specifically, these ecosystem services include urban cooling through increased shade, flood mitigation, and aesthetics. 7 A wide range of GI can be found in cities, including rain gardens, bioretention swales, green roofs, and permeable pavements, among other strategies. 8 Despite the increasing adoption of GI in urban areas, researchers and practitioners are mindful of existing challenges, including conflicting regulations, costs, responsibility for long-term maintenance, and the potential for gentrification and community displacement. 9
The potential connection between GI and gentrification directs attention to the importance of co-production as an approach for equitable GI implementation, and other climate adaptation programs more broadly. 10 In short, co-production involves meaningful community participation in the planning and implementation of public services and the integration of local perspectives and knowledge into this process. 11 A central assumption of co-production is that to address complex problems—such as just climate adaptation—we need to create usable knowledge that is scientifically and socially robust. 12 Community participation is thus central to this approach. The existing literature discusses methods, challenges, and opportunities for community participation in knowledge co-production. 13 However, little has been written on the need for and experiences with bilingual facilitation in community engagement, even as practitioners have long discussed how to go about this to ensure meaningful participation of frontline communities in urban climate resilience planning.
This article advances this discussion by presenting strategies and lessons learned in the context of English–Spanish bilingual workshops where funds were not available for professional interpretation or translation services. As applied researchers working in the U.S. Southwest, we often confront the need to organize and facilitate bilingual workshops. To do this, we have learned from each other, community leaders, and city employees committed to incorporating diverse voices to foster dialogue and inform just climate action. The need for bilingual facilitation, specifically in English and Spanish, is critical in this region, where over 40% of the population is of Hispanic origin, and about 23% claim Spanish as their mother tongue. 14 While this article has a strong regional focus, we believe our methodological reflections will be useful to others working on climate resilience planning in multilingual urban communities across the United States and abroad.
COMMUNITY PARTICIPATION IN CLIMATE RESILIENCE PLANNING
Co-production is emerging as a promising approach to more than just urban climate resilience planning. The term has been used to describe how climate adaptation plans and knowledge can be produced collaboratively by decision makers and community participants. The premise is that knowledge elicited through multi-stakeholder participatory workshops can be a better fit to local needs, values, and relations, 15 potentially making climate adaptation strategies more accessible and appropriate for frontline communities. This is often presented as an alternative to top-down approaches to climate mitigation and adaptation that risk perpetuating socio-spatial inequalities. 16
In this context, meaningful and inclusive community participation is a critical first step for procedural justice in climate resilience. However, community participation requires a deep understanding of the underlying power dynamics that produced uneven urban vulnerabilities and that may result in uneven participation in the co-production process itself. 17 Research shows that a lack of participation in planning efforts and disaster management initiatives and lower levels of disaster preparedness are compounding factors affecting uneven vulnerability to climate change. 18 Lack of participation has been attributed to exclusionary planning that ignored resource-poor communities from decision making and consultation 19 and to community distrust in the government due to historical mistreatment and legacies of disinvestment, 20 as well as to consultation and participatory fatigue. 21
Scholars and practitioners concerned with fostering community participation have drawn attention to the count and location of public workshops or forums, making sure they are available and accessible to underserved communities; to the number and types of individuals present, paying attention to the representativeness of different groups, and the level of involvement of individuals in those events. 22 Saliently, accessible language has been identified as a key factor for meaningful and effective community participation. 23 However, not much has been written on experiences with bilingual outreach and facilitation for climate resilience planning and urban planning more broadly. In what follows, we address this gap by describing our experience organizing and facilitating English–Spanish bilingual workshops and examining how cultural conceptions about language—our language ideologies—can impact community engagement in this bilingual setting. 24
STUDY CONTEXT
The Tucson metropolitan area is located in southern Arizona, in the northeastern region of the semi-arid Sonoran Desert and 70 miles (112 km) north of the U.S.–Mexico border. It has about 1 million residents: 49% of the population self-identifies as non-Hispanic White and 38% as Hispanic or Latino. The area’s median household income (in 2022 dollars) is $64,014, with 14.5% living below the national poverty line. 25
Tucson’s climate is characterized by two distinct rainy seasons, occurring during the winter and summer months and separated by extended dry periods 26 (Hoori et al., 2011). Between 1991 and 2023, the mean annual precipitation was 10.62 inches (269.5 mm). 27 Tucson’s annual mean temperature is 71°F (22°C), with summer temperatures regularly surpassing 100°F (38°C). 28 Tucson is the third fastest-warming city in the United States, having experienced a 4.48°F (2.49°C) rise in mean temperature since 1970. 29 Such temperature increases are expected to negatively impact public health. In 2023, neighboring Maricopa County—where heat-related deaths are tracked, unlike other locations—reported 645 heat-related deaths. 30 Such negative public health outcomes are typically inequitably distributed, disproportionally impacting lower-income communities and ethno-racial minorities. In the case of Tucson, lower-income Hispanic communities on the city’s south side have less tree canopy and higher recorded temperatures than other parts of the city and are thus more vulnerable to extreme heat and flooding. 31
The Tucson metropolitan area does not have an extensive subsurface drainage system for stormwater. Instead, roads rapidly convey stormwater, resulting in flooding from intense summer storms. 32 This drainage system offers high potential to capture and use stormwater via Green Stormwater Infrastructure (GSI) to support flood mitigation and increase heat-mitigating urban greenery along streets by routing stormwater for irrigation.
Over the last two decades, GSI practices have become increasingly formalized across Tucson, 33 as evidenced by a series of ordinances and city-run programs aimed at incentivizing and funding GSI construction and maintenance in public and private properties. 34 Concurrently, we have observed growing attention to the role of GSI as a type of infrastructure capable of either reproducing or addressing environmental inequalities across the urban landscape. 35 For example, the City of Tucson implemented a residential rainwater harvesting rebate in 2012. They soon noticed that the program was disproportionally underutilized in the south side of town, an area characterized by lower household incomes and higher minority populations, thus reflecting a distributional justice issue. To reduce the financial, administrative, and language barriers that kept low-income Hispanic households from accessing the rebate, the city partnered with the Sonoran Environmental Research Institute (SERI), a local non-profit, to implement a low-income rainwater harvesting loan/grant program. 36 Attention to equitable GSI coverage continues through the work of the Storm to Shade program, which is funded by the GSI Fund passed by the City and Council in 2019 to support installation, maintenance, and community engagement around public GSI projects. Storm to Shade uses the Tree Equity Score, a composite score that integrates data related to socio-economic and environmental justice considerations to prioritize implementation and maintenance of GSI throughout the city. 37 The Tree Equity Score shows that the south side of Tucson is an area that shall be prioritized for GI installations as well as other forms of GI more broadly.
Practitioners and researchers working on GSI in Tucson recognize that addressing distributional injustices is key to supporting procedural justice. A basic step in this direction is to better engage historically underserved communities in GSI planning and implementation, a process scholars have named co-production. 38 In Tucson, this requires bilingual facilitation in English and Spanish. In what follows, we provide two examples of how we worked bilingually to strengthen community engagement with GSI planning and to collect qualitative data to advance more inclusive climate adaptation measures. Given the objective of this article, we focus our discussion on the participatory methods for community engagement at the core of our projects rather than the broader projects in which they were embedded.
CASE STUDY 1: “A COLLECTIVE VISION FOR RAIN GARDENS”
This project was a partnership between university researchers and practitioners at Watershed Management Group (WMG) and SERI. We worked to articulate a more inclusive vision for the future of GSI in Tucson by expanding the traditional network of actors involved in discussions of GSI. Through peer-to-peer exchanges, this project sought to identify underrecognized priorities and challenges around neighborhood-level GSI efforts in historically underserved Hispanic communities. Given our goal and intended participants, it was obvious from the start that any event needed to be bilingual in English and Spanish, which raised the question of how to design and facilitate such a workshop on a limited budget that didn’t allow for professional interpretation.
Core engagement method
The core engagement method was a half-day visioning workshop called Conversatorio (talking forum). Given that most events around GSI planning happen in English, we chose this Spanish term to indicate that the event was intended as a space to think through conversation in English and Spanish. We relied on WMG and SERI to collate a list of community groups engaged in environmental and social justice issues adjacent to GSI, but who were outsiders to the established English-dominant GSI network. We invited 25 people, but only 14 attended, representing a wide range of community groups. The Conversatorio was scheduled for a Saturday and was held at a venue in the south side of Tucson. Event attendees received bilingual informational materials, a catered lunch, and a gift card to appreciate their time. All materials, such as informational handouts, the agenda, workshop objectives, and instructions, were drafted in English—the shared language of the research team—and translated into Spanish. Due to limited time, we experimented using a free online artificial intelligence (AI) translator for the initial translation, which was reviewed and edited by the project PI, who is bilingual in English and Spanish, familiar with GSI terminology, and had experience with GSI implementation in the region.
The Conversatorio was designed as a consensus process—a space for people to develop a collective vision for implementing rain gardens in Tucson. Thus, participants were presented with three intentionally broad guiding questions: (1) how can rainwater improve the well-being and health in your neighborhood? (2) how should rain gardens be implemented and what considerations should be put in place? and (3) how could businesses and community centers be best engaged to implement rain gardens? In two small groups, based on participants’ language preferences, individuals were asked to write down their top three ideas in response to the three questions. The two groups were balanced in number. Participants then shared their ideas around the table, and a facilitator wrote them down on a board for everyone to see. After creating a complete list, the group reviewed their ideas, condensing repeated items and determining if anything was missing. From a long list, participants discussed and selected their top 8–12 ideas representing a diversity of visions. Each selected idea was then written on an individual sheet of paper in English and Spanish. Participants had 2 hours to work on this.
After lunch, the two groups reconvened to share their ideas in a large bilingual group discussion and worked to sort them into larger common themes. All ideas were placed on a large sticky mat, and the facilitator guided participants as they discussed and organized them into blocks representing different themes. The sticky mat allows ideas to be easily moved back and forth while people discuss. This discussion was facilitated by a bilingual researcher who translated instructions consecutively, invited people to participate in any of the two languages, and translated summaries of the points made by participants as those were expressed. In some cases, bilingual participants translated their points or helped translate those made by others before adding or responding to them, allowing for a conversational dynamic. The bilingual nature of the event, and perhaps the high degree of engagement of those present, made progress significantly slower than planned. After 2 hours, we grouped all ideas into meta-themes but did not have time to create labels for those themes jointly. From a research perspective, this was a shortcoming as we had intended to engage in multi-stakeholder grassroots coding (see Radonic et al., 2023). 39
Overview of results
By engaging a diverse group of stakeholders in dialogue across English and Spanish, we identified nine strategies for implementing rain gardens in underserved neighborhoods in Tucson. These strategies run along a spectrum from direct actions (i.e., increase affordability by providing plant rebates and fund demonstration gardens in underserved neighborhoods) to guiding principles (i.e., identify local neighborhood priorities as input for rain garden design and acknowledge different cultural values). While the strategies themselves were not necessarily novel to existing GSI discussions, finding such a range of ideas in the same space is not common. This diversity of ideas reflects the diversity of experiences of those present, and it made plain the tensions surrounding local GSI implementation. A critical example of this was made evident in participants’ discussions regarding plant preferences, a tangible extension of peoples’ cultural values. While everyone agreed on the importance of respecting different cultural values as a principle, divergent views emerged along language lines when the discussion turned to the manifestation of those values in peoples’ plant choices. In Spanish, some participants emphasized that acknowledging peoples’ values meant incorporating into rain gardens the plants they preferred, especially those that symbolically connected them to their native homes in Mexico, even if they were not native or drought-adaptive. In English, others acknowledged the importance of recognizing and respecting differences but highlighted the need for advocating planting native drought-adaptive vegetation. While these views were not reconciled at the workshop, the Conversatorio identified them as points of tension and put them in dialogue.
CASE STUDY 2: “BUILD YOUR DREAM GARDEN”
This project was a partnership between university researchers and practitioners at SERI and the City of Tucson’s Storm to Shade Program. The project aimed to engage historically underserved communities in the south side of Tucson to identify their plant preferences and understand their expected ecosystem services. The overarching goal was to help inform an equity-centered approach to urban greening by aligning city-led greening efforts, in the context of GSI implementation, to vulnerable communities’ plant-related needs and desires. As mentioned in the previous case study, this is a point of tacit tension around GSI implementation in Tucson.
Core engagement method
The engagement activity at the core of this project was a “Design Your Dream Garden” workshop. Our team created this multi-step game-like method to lead participants in a scaffolded way to reflect on their relations to urban vegetation and identify individual plant preferences and their desired ecosystem services. We held two Dream Garden workshops on two different Saturday mornings in locations on the south side that offered free parking. SERI recruited participants by inviting people involved with other related programs run by them; 30 people attended each workshop. All the materials provided were available in English and Spanish, but given the demographics of the south side, we were prepared to facilitate the event primarily in Spanish. Participants were offered brunch and a gift card for their participation.
As people arrived, the research team informally took notice of peoples’ language preferences, confirming the decision to run the workshop in Spanish. Participants divided themselves into smaller groups as they were invited to sit at one of several large tables upon grabbing a coffee and a breakfast burrito from a local vendor. While all participants spoke Spanish, the small group discussions revealed that a few people were more comfortable expressing complex ideas in English or Spanglish—a language variety that results from conversationally combining English and Spanish. This brought back the need for some bilingual small-group facilitation.
The workshops had four stages: (1) designing their dream garden in Tucson by selecting plants from a catalog of common plant varieties, (2) sharing the design with others, (3) reflecting on the reasons for their specific plant selection with attention to expected ecosystem services, and (4) ranking plants and other garden features in terms of perceived importance. Participants worked individually to design their dream garden using a site map, cut-out icons of vegetation and other garden features we provided, and two plant catalogs. At their tables, participants shared their designs, reflected on their selected plant species, and went on to rank their top five plant preferences and write down reasons for their selection. This produced data in the form of landscape designs, tables listing plant selections and expected ecosystem services, and audio recordings of people’s reflections on all of the above.
The team responsible for designing, planning, and facilitating the two workshops comprised a large multidisciplinary group of practitioners, researchers, and graduate students. More than half of the team were native Spanish speakers or proficient in Spanish. Our bilingual team members served as facilitators, with at least one person assigned to each table to guide participants through the multiple steps and answer questions. Yet those who did not speak fluent Spanish were also present, helping with organization and logistics, and offering their disciplinary expertise as needed. As English was the primary language in which our team communicated, we first developed all workshop materials in English and then translated them into Spanish, which was a time-consuming task.
Overview of results
The dream gardens created by participants and the accompanying materials revealed a strong preference for non-native plants that provide food or medicine for the household, enhance outdoor aesthetics, and reinforce cultural identity. Although a majority of participants reported living in Tucson for over 20 years, the extended time of residency did not result in a preference for desert plants. We found that plant choice is often pragmatic: the most desired plants—citrus, figs, peaches, opuntia cactus, and a diversity of herbs—were selected for their ability to provide food and medicine for the household. At the same time, we found that cultural identity was a strong driver behind these selections. Certain plants trigger pleasant childhood memories or provide food for their traditional cuisines. Including these plants in GSI programs is critical for increasing program adoption in disadvantaged communities and advancing justice in climate change adaptation.
LESSONS LEARNED
Don’t assume; ask about language preference
As the two case studies remind us, bilingual populations may prefer one language over the other in different contexts or for expressing different content. 40 Moreover, not everyone who identifies as Hispanic or Latino/Latinx speaks Spanish fluently or at all. 41 In Case Study 1, given the participants’ involvement in local policy and community organizing, some team members had assumed that everyone would be equally comfortable communicating in English. However, while everyone had some degree of bilingualism, a two-question survey at the beginning of the event revealed that a majority of people strongly preferred either one or the other. In Case Study 2, we anticipated everyone to prefer Spanish and assumed this was indeed the case when people began conversing in Spanish upon arrival. However, small group discussions showed that a few preferred discussing more complex ideas in English or Spanglish.
These cases make evident our own language ideology 42 —the tacit assumptions underlining our views of how Spanish and English should be used, where, and by whom in this border region. Attention to language ideology, in turn, illustrates the importance of directly asking language preference at the start of the event or prior to it, and not assuming a language preference based on people’s ethnic background or national origin.
Be mindful of the power dynamics of language
A central consideration when hosting a bilingual workshop should be power, as different language groups have different amounts of power in a society. In the United States, for example, English is dominant, and in many settings, Spanish speakers continue to be stigmatized. 43 The concept of language ideology is once again useful here. It invites practitioners and researchers to be attuned to the complex interplay between how language is viewed and used, and the hierarchies within which this use occurs in society in response to the socioeconomic and political structures within which speakers are embedded.
Language power differentials can block full participation unless intentionally addressed in the event structure and procedures. 44 At a basic level, this means ensuring that all materials are created in both languages with careful attention to using proper grammar and cultural appropriateness in the non-dominant language. This requires attention to the Spanish variety spoken by participants. About facilitation, when engaging in consecutive translation, we suggest shifting the primary language throughout the event as this indicates that the non-dominant language is equally important, helping to balance power dynamics within the room. Finally, when possible, consider inviting bilingual community partners to assist in translation and/or facilitation while paying for their services. In doing this, it is important to ensure the community partner has experience in facilitation and a strong familiarity with the goals and content of the research. Involving community partners in facilitations helps decenter the English-speaking university researcher as the primary authority figure and, from a linguistic and justice perspective, it is extremely valuable as community partners are more likely to speak the same language/dialect as minority participants.
Small groups are effective, but can they stand alone?
The benefits of small group discussions for active and inclusive teaching and learning have been well established. 45 These case studies illustrate their usefulness for facilitating community participation in bilingual settings while pointing out potential limitations. Based on research goals and the availability of resources, it is critical to determine how to best integrate small and larger group discussions. In Case Study 1, by dividing participants into two small groups by language preference, we were able to foster comfortable spaces for free-flowing dialogue. However, this siloed participants by language following the contours of social and professional networks. This meant that the exchange of ideas across language groups—and hence social and professional networks—was limited to the large group discussion that followed. Thus, to develop a collective vision, small and large group discussions were necessary. In Case Study 2, where the objective was to elicit individual plant preferences and perceptions, small groups were not only sufficient but also more effective. We were able to better engage with all participants, helping them through the multi-step activity in a way that allowed them to maximize the experience of reflecting on what they wanted from their own gardens. This workshop did not require a large-group discussion. However, we could only rely on small group discussions because we had a large research team with people capable of both facilitating and doing it bilingually.
AI does not speak the local dialect
In comparing notes across the two case studies, we are once again surprised by how much time and resources are needed in making a participatory workshop fully bilingual—including workshop materials, facilitation, and later, data analysis (an aspect we are not covering in this article). Thus, we recommend using free AI translators to facilitate the translation process of written materials. These software can provide a very polished draft. However, the AI-translated output should always be reviewed and revised as needed by someone fluent in the language and familiar with the terminology. To show respect and embrace the local dialect, we suggest having a local speaker do a final translation check to ensure appropriateness and reduce stiffness. In Case Study 2, this proved extremely useful as common names for plants and garden features vary regionally.
FINAL REMARKS
Facilitating the workshops bilingually required abundant labor hours from university researchers, students, and community partners, as well as a strong understanding of local cultural dynamics and historical processes. The importance of these efforts cannot be understated. The participatory workshops presented here gathered qualitative data to advance our understanding of how community members in historically underserved low-income Hispanic communities envision GSI development in Tucson, with specific attention to the benefits they most value and their concerns around its implementation. By collaborating with local non-profits and the City of Tucson, we hope our findings can be directly transferred to those in a position of power to develop inclusive policies and advance just climate adaptation. Investing resources in designing, planning, and facilitating effective bilingual engagement processes shows promise in capturing disadvantaged communities’ voices (concerns, needs, and values), which will likely strengthen procedural justice.
Footnotes
AUTHORS’ CONTRIBUTIONS
L.R.: Conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, validation, writing—original draft, writing—review and editing. A.A.Z.-T.: Data curation, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, validation, writing—original draft, writing—review and editing. N.G.: Data curation, formal analysis, investigation, methodology, supervision, validation, writing—original draft, writing—review and editing. M.E.H.: Data curation, formal analysis, investigation, methodology, supervision, validation, writing—original draft, writing—review and editing. K.J.K.: Data curation, formal analysis, investigation, methodology, supervision, validation, writing—original draft, writing—review and editing. C.A.-M.: Data curation, formal analysis, investigation, methodology, writing—original draft. I.E.G.: Data curation, formal analysis, investigation, methodology, writing—original draft. S.M.: Investigation, methodology, writing—original draft. M.L.N.: Investigation, methodology, writing—original draft. F.S.: Investigation, methodology.
ACKNOWLEDGEMENTS
Many people were involved in different capacities throughout the course of these two projects. We would like to thank for their vision and passion Blue Baldwin, Xochitl Coronado-Vargas, and Brooke Bushman at Storm to Shade, Andrea Salazar, Catlow Shipek, Luis Salgado and Joaquin Murrieta at Watershed Management Group, and Molly M. Bryson at the University of Arizona’s Udall Center for Studies in Public Policy. We would also like to acknowledge the work of Imelda Cortez, María Jose Uribe, and Oscar Rodriguez Ponce who helped run some of the workshops. Finally, we are grateful to all participants who shared their experiences and perspectives with us.
AUTHOR DISCLOSURE STATEMENT
The authors have no interest to disclose.
FUNDING INFORMATION
Funding for these projects was provided by Northern Arizona University's Community-Campus Partnership Support Program and the University of Arizona’s Research, Innovation & Impact – International Collaboration Grant.