Development of Visualization Tools for Sharing Climate Cooling Strategies with Impacted Urban Communities

Abstract

Intensifying heat from warming climates regularly concentrates in urban areas lacking green infrastructure in the form of green space, vegetation, and ample tree canopy cover. Nature-based interventions in older U.S. city cores can help minimize the urban heat island effect, yet neighborhoods targeted for cooling interventions may remain outside the decisional processes through which change affects their communities. This translational research seeks to address health disparities originating from the absence of neighborhood-level vegetation in core urban areas, with a focus on tree canopy cover to mitigate human susceptibility to extreme heat exposure. The development of LiDAR-based imagery enables communities to visualize the proposed greening over time and across seasons of actual neighborhood streets, thus becoming an effective communications tool in community-engaged research. These tools serve as an example of how visualization strategies can initiate unbiased discussion of proposed interventions, serve as an educational vehicle around the health impacts of climate change, and invite distributional and participatory equity for residents of low-income, nature-poor neighborhoods.

1. Introduction

Identifying inequities between population exposures and outcomes is the bread and butter of environmental health. The associations our research generates provide evidence for public health policy or regulatory action to protect vulnerable citizens against harmful exposures. Still, the temptation exists to create even more nuanced evidence of environmental risk while falling short of true research policy translation, interventions, or other actions that stem from recommendations suggested by the evidence. Here we aim instead at translational research that has applications for addressing health disparities originating from the absence of neighborhood-level vegetation, whether owing to systemic underinvestment or benign neglect, with a focus on tree canopy cover to mitigate human susceptibility to extreme heat exposure.

Changing climate has intensified heat and the patterns of damaging precipitation events. However, less riveting harms regularly occur in urban areas lacking green infrastructure, specifically those with minimal urban green space, sparse vegetation, and low rates of tree canopy cover, in the form of extreme heat exposure. Solar radiation concentrates in denser built environments, where extended areas of asphalt, concrete building mass, and impermeable surfaces like sidewalks absorb and reflect infrared solar radiation back to the ground-level atmosphere, compounding these underlying disparities. Historic municipal underinvestment in core city neighborhoods—and residential poverty among those left behind by suburban out-migration—have amplified health disparities rooted in the built environment. These include failure to provide green scaping to reduce temperatures, maintain shade tree canopies that invite physical activity and socialization, construct sidewalks and bike lanes for safer mobility, and invest in basic neighborhood amenities such as benches or seating, traffic crossings, and park repair and maintenance.

Areas of disproportionate heat gain are known as urban heat islands (UHIs) due to their concentrated building mass that absorbs high levels of solar radiation. With few trees providing shade and deflecting incoming radiation, UHIs can harbor daytime temperatures 1–7 °F higher and nighttime temperatures 2–5 °F higher than suburbs with more abundant vegetation [1]. UHIs grow more acute with denser thermal massing, impermeable surface enlargement, and less nighttime cool-off, all typical of aging urban city cores.

UHIs pose health harms to vulnerable subgroups, particularly elderly and low-income residents, resulting in increased heat stroke, ischemic heart disease, hospitalization, and cognitive loss [2,3,4]. Heat gain that accompanies climate change aggravates existing health disparities that frequently track along racial lines. Excessive ambient temperatures in urban neighborhoods add heat-related health risks to city-dwellers already beset by inequitable burdens of poverty, many traced to HOLC redlining [5,6]. The direct benefits of enhanced thermal comfort these interventions create extend to lower mortality rates and reduced energy costs, in addition to greater physical activity and socialization. Studies have shown a range of health benefits from residential proximity to urban nature. Living near urban green space can lower the risk for health problems such as heat stroke [7,8], cancer [9,10], heart disease [11,12], hypertension [13,14], obesity [15,16], diabetes [17,18], pre-term birth [19,20], and asthma [21,22] and can improve mental health [23,24,25].

Nature-based interventions across the U.S.’s older city cores can help minimize the UHI effect. The presence of urban vegetation increases urban albedo via higher reflectivity of sunlight off of surfaces like foliage and the expansion of evaporative ground cover [26]. Simulations of urban climates support the conclusion that increasing albedo, evapotranspiration, and vegetative cover effectively counteracts urban heating. Vegetation can lower temperature by provisioning shade, inducing individuals outside to socialize, itself a mediator of health [27,28], and prompting more engagement in greater physical activity than they might in hotter conditions [29,30]. The cooling effects of urban greening have been amply documented, proven through direct measurements as well as computer simulations [31]. Zardo et al. (2017) has estimated how tree canopy coverage and soil cover as gauged by size and shape constitute adaptation measures that enhance cooling capacity in urban environments [32]. Larger parks of 10 hectares or more and urban forests can lower daytime temperatures by 0.94 °C [33]. Street trees, green roofs [34], and green façades--smaller forms of green infrastructure—also are effective at reducing ambient temperatures [35].

Afforestation strategies hold enormous potential to address the health challenges UHIs pose in low-income neighborhoods like Springfield, Massachusetts’. Street tree plantings can directly reduce radiated heat contributing to UHIs: the presence of tree canopies can reduce the physiological equivalent temperature (PET) by 4 °C to 11 °C of densely built environments [36], as compared to vegetated surfaces receiving full sun exposure. Given that 37% of recorded heat-related deaths since 1991 are attributable to global warming [37], afforestation to counter UHIs is a central tenet of urban temperature-reducing strategies. In fact, the literature is replete with modeled estimates of UHI reductions enabled by tree-based solutions [38,39,40,41]. The numerical and spatial models in these appraisals, however, rarely translate into implementable designs. The few that do fail to consult the communities targeted for these interventions.

This study addresses a proposed intervention to redress climate change-related heat gain that disproportionately impacts poor, nature-deficient residential neighborhoods. Springfield, Massachusetts, serves as a pilot case to display street tree planting as a nature-based intervention helping cities achieve their long-term goal of mitigating urban heat differentials, while also promoting the underlying health and wellbeing of community residents by providing them with a more comfortable, temperate neighborhood environment for pedestrian mobility, exercise, and socialization, as well as lower building energy needs.

This pilot project builds on the integrated design work led by landscape architect Martha Schwartz to develop Springfield, Massachusetts’ linear urban forest (LUF), a research initiative supported by a 2021 Harvard Climate Change Grant. The interdisciplinary research team consisted of Harvard Graduate School of Design landscape designers, Harvard T.H. Chan School of Public Health researchers, US Forest Service biologists, and municipal policy practitioners who collaborated on the original LUF design. The LUF was designed to encompass all streets and run throughout the city. As a living ecological green infrastructure, LUF provisions many eco-system services that address environmental and social challenges, among them city cooling, buffering adjacent river flooding, better management of stormwater runoff, human health protection, urban biodiversity enhancement, decreased energy bills, increased quality of life, enhanced walkability, reduced greenhouse gas emissions, intensified carbon sequestration, and climate equity.

The pilot LUF launched the next step for project realization by supporting the development of dynamic visualization modeling and design prototyping to bring a digitalized green solution to community residents. In this way, community members living in heat-stressed neighborhoods could participate in the decision-making process for reimagining their streets in ways that co-introduce climate mitigation, health, regenerative nature, and equity. This study therefore made use of a community-engaged research (CER) strategy as the strategic framework for developing tools to engage local citizens as co-generators of knowledge regarding neighborhood transformation. CER, or its broader umbrella term, community academic partnerships (CAP) [42], is distinguished by a commitment to partnering with communities to co-create knowledge arising from both researcher and user expertise [43]. By engaging local users in the production of knowledge, researchers aspire to social impact benefiting community members for whom the interventions under discussion are targeted. Collaboration should result in an attenuated “know–do” lag, meaning a shortening of the cycle between research findings and implementation [43].

Community-based research partnerships recognize parity among interventionists and impacted stakeholders in holding seats at the table, different from a strictly research-centered practice where effected community participants are historically excluded. CERs also address procedural equity whereby decisional tools are made available to facilitate full participation in knowledge-creation sessions. Many aspects of community-based participatory research (CBPR) strategies were employed in this collaboration, enabled through the researchers’ partnership with ReGreen Springfield, a trusted community organization whose facilitation of the design charettes had the result of placing community at the center of these sessions, a central tenet of CBPR [38]. Like CBPR, community equity was an explicit goal of the selected research process; moreover, ReGreen Springfield, with its expertise regarding tree presence in these neighborhoods, made for a strong community partner since the design stage of the research. Knowledge co-creation by community members was limited to feedback from visualization of neighborhood intervention, however. CAP and CBPR can assist in translating research knowledge where a lack of prior stakeholder engagement helped sustain disparities in communities now targeted for tree planting [44].

Some examples exist where communities participated in processes that co-produced risk mitigation strategies when faced with climate change. One example of community-based research was a flood vulnerability assessment and risk mitigation in Milwaukee, Wisconsin, with a similar focus on green infrastructure as a partial solution [45]. Visualization tools were applied to graph epistemic disciplines of green and blue infrastructure—tools under scrutiny to counteract urban heat island effects and human health with its increasing vulnerability to climate change risks, including that of extreme heat [46]. However, in these two cases, visualization tools were presented as static mapping without the dynamic visualization of our methods. In another study, Koulouris et al. [47] used augmented reality that takes advantage of locational data to create interactive, real-world scenarios placing humans in the center of urban environments. However, the captured locational data mimicked actual surroundings rather than projecting alternative scenarios, making this a tool of gamification rather than projection. We know of no other study that attempted to visualize nature-based interventions affecting climate risk or human health by interactive means. There were three aims of this research:

• Develop visualization tools in collaboration with the Visualization Research and Teaching Laboratory in Harvard’s Department of Earth and Planetary Sciences, then pilot these visualization technologies with residents of Springfield’s impacted neighborhoods where the LUF is proposed.
• Apply the InVEST Natural Capital Project’s accounting model to estimate the reduction in urban temperature expected from expanding the urban street canopy.
• Conduct design charettes to introduce a new form of pedagogy in community participatory research, including environmental literacy in connecting the health benefits of the presence of nature to reduce heat and asthma burdens and promote physical activity.

This research does not, however, attempt to put forth cost–benefit analysis of this project as engineering, maintenance, or quantified health gains, nor does it measure improvements in air quality or the reduction in CO₂ that may result from full implementation of the project. It does not compare the potential impacts of green infrastructure in Springfield to other cities that have implemented them. However, it does seek to demonstrate how public adoption of nature-based solutions that include climate–health and nature–health implementation goals can be accelerated through integrated assessment methodologies, dynamic visualization modeling, and design prototyping for presenting and soliciting feedback from locally impacted communities.

2. Materials and Methods

Springfield, Massachusetts, an aging industrial city with a legacy of air pollution, was selected as the design canvas for a linear urban forest due to its public health challenges and relatable population size. Springfield ranks number one among U.S. cities for asthma prevalence: the city has the highest number of asthma-related emergency room visits per capita in the U.S., as well as a high rate of asthma prevalence overall. High pollen counts aggravate the susceptibility of Springfield’s residents to asthma [48]. Springfield also ranks tenth among the poorest cities per capita in the U.S. with populations over 100,000, having numerous environmental justice neighborhoods [49]. The city’s residential population, at 21% black and 46% Hispanic [50], has the lowest life expectancy of all cities in Massachusetts [51]. We focused our visualization efforts on the neighborhood known as Upper Hill, which has a high percentage of Hispanic and black households, older citizens, and individuals under the poverty rate, with many of these categories overlapping. The physical environment in the Upper Hill neighborhood is densely populated with aging multi-family homes and limited recreational space.

The study population consisted of approximately 40 individuals who attended one of two Springfield design charette sessions. These individuals ranged in age from 16 to senior citizens. Demographics were equally balanced between males and females, with many couples in attendance. Participants represented a range of racial and economic diversity, although no non-native English speakers attended the sessions. The audience included a mixture of representatives of the neighborhoods selected for greening, members of our non-profit research partner ReGreen Springfield, young adults working in the tree-planting arm of that organization, and representatives of city government. Some charette participants did not live in the exact neighborhoods targeted for tree planting but rather came from the adjacent Forest Park neighborhood, with its mature tree canopy and ample shade that exemplifies the goals of urban greening. A few individuals from other tree-poor neighborhoods attended to see if proposals might transfer to their area. Most attendees actively participated with commentary, providing a representative opinion sample from which to draw data. A survey of participant preferences for specific street tree scenarios was informally conducted through a show of hands and voluntary comments.

2.1. Technology Applications

This study utilized advanced LiDAR scanning technology to capture and digitize the three-dimensional structure of impacted neighborhoods, enabling the creation of highly accurate digital models of streets, landscaping, and existing tree planting conditions. The integration of LiDAR scanning in urban street tree prototyping introduced a novel approach to environmental planning and participatory urban design.

2.1.1. Data Collection and Processing

LiDAR scans were conducted in the Upper Hill neighborhood, where street tree planting prototypes were implemented. Specialists from the Visualization Research and Teaching Laboratory at Harvard University’s Department of Earth and Planetary Sciences collected on the streets of Springfield geolocated point cloud datasets using a VLX LiDAR scanner, manufactured by NavViz, Munich, Germany. This mobile, shoulder-mounted unit integrates LiDAR sensors, cameras, and GPS, allowing for efficient on-foot data acquisition. The system outputs a point cloud density of approximately 8000 points per square meter, achieving a resolution of around 5 mm accuracy. By measuring the two-way travel time of laser pulses and processing calculations based on the speed of light, the scanner generated dense, spatially accurate 3D point clouds. These datasets provided a precise digital representation of neighborhood infrastructure and served as the foundation for subsequent modeling and visualization.

2.1.2. 3D Modeling and Visualization

The collected point cloud data were imported into Blender, where a base mesh was generated to accurately represent buildings, existing tree plantings, and urban infrastructure. To improve the realism of the visualizations, Google Street View imagery was integrated as 360-degree background textures, ensuring that the surrounding neighborhood homes and structures were depicted with a high level of visual detail. The 360 images were precisely matched to the models by aligning their locations with the geospatially referenced LiDAR scans, allowing for an accurate representation of each site. A structured library of digital assets was developed to support scenario visualization. This included a catalog of tree models representing various species at different growth stages, as well as ground vegetation and hardscape elements such as sidewalks, benches, and street fixtures. These assets were integrated into the models to reflect proposed urban forest plans and growth projections.

Four viewpoints were selected for detailed visualization, each serving as a representative section of the study area. For each viewpoint, six distinct planting scenarios were developed, and each scenario was further rendered across four seasonal conditions. This resulted in a total of 96 high-fidelity 360-degree images, which were used in interactive presentations and public engagement sessions. The depicted scenarios ranged from current plantings maturing 50 years into the future to new plantings at spacing intervals of 40, 20, and 15 feet. This last scenario entailed converting the two-way street to one direction and foregoing a row of street parking to accommodate a wider sidewalk median for planting a double row of street trees. A final scenario added vegetation and smaller plantings as an understory to develop a ground-level ecological community around the trees.

2.1.3. Interactive Application Development

To facilitate public engagement and feedback, an interactive application was developed using the Unity game engine. The application featured a user-friendly interface that allowed participants to explore the 360-degree views at the designated viewpoints and toggle between different planting scenarios and seasonal conditions. The interface was designed to provide an intuitive navigation experience, enabling users to compare environmental changes dynamically. Beyond tree planting variations, the application also allowed exploration of additional urban design considerations, such as road layout modifications, linear forest programming, and potential community engagement features. Options included micro parks, small public gardens, and exercise areas integrated along the linear forests, encouraging users to consider broader urban greening strategies. Additionally, the application provided data on the heat island reduction associated with each scenario. By selecting different configurations of tree planting densities and green space allocations, users could observe potential temperature mitigation effects, reinforcing the ecological and social benefits of urban reforestation.

The interactive tool was hosted on AWS, ensuring accessibility for researchers, urban planners, and community members. The structured visualization of urban tree planting impacts across different timescales and environmental conditions provided a robust foundation for participatory decision-making and long-term urban forestry planning.

2.1.4. Species Selection

Tree selection for the visualization tool relied on ten species currently favored by the Springfield Parks Department for total tree composition in the city’s master forest plan. Concerns with species resiliency in urban areas mandate that no more than ten percent of any one species be represented in the planting mix to avoid excessive mortality due to pest or disease. Relying on this guideline, the street tree visualization incorporated six common New England tree species already present in Springfield, with these being red maple, black tupelo, Eastern white pine, white oak, Northern red oak, and black oak (

Table 1. The six primary species used in Springfield’s master forest plan for reforestation.

Type			Height (ft)
Percent	Name	Scientific Name	New	Young	Mature
~10%	Red Maple	Acer rubrum	3–6	15–25	40–60
~10%	Black Tupelo/Gum	Nyssa sylvatica	3–6	15–25	30–50
~10%	Eastern White pine	Pinus strobus	2–4	15–30	50–80
~10%	White Oak	Quercus alba	3–6	15–25	60–100
~10%	Northern Red Oak	Quercus rubra	3–6	15–25	60–75
~10%	Black Oak	Quercus velutina	2–4	15–30	50–60

). Heights at maturity of these species were calculated for accurate representation in street tree canopies. These six species were interspersed in the visualizations, although no adjustments were made in species clustering across the different tree spacing scenarios.

Additional consideration was given to species resiliency to increasing urban temperature due to climate change in planning for the linear urban forest, such that some species not endemic to New England might thrive in warming climate zones at future times. These new species were not represented in the visualization scenarios as their viability has not yet been established. Of note, saplings common to more southern climate zones will be incorporated into the test plots for the urban forest to observe their adaptability to warming northern climates zone such as Springfield’s.

2.2. InVEST Urban Cooling Model Application

Concurrent modeling of estimated cooling potential projected from the original linear urban forest design was performed using the Natural Capital InVEST Model’s Urban Cooling Module [52]. This model generates maps of the heat mitigation index (HMI) to project cooling capacities of urban green spaces based on the land use classes contained in the area under consideration. The InVEST model requires inputs of area land use maps where greening is present or designated and biophysical estimates of annual tree growth, evapotranspiration, and albedo surface reflectivity, measures derived from the scientific literature, as well as a percentage measure of building intensity vis-à-vis vegetated space. For Springfield land use we used MassGIS Data: 2016 Land Cover/Land Use, which combines land cover mapping from 2016 aerial imagery and land use derived from standardized assessor parcel information for Massachusetts at a 1 m spatial resolution [53].

Land maps inputted into the InVEST model typically reflect existing categories of land use. Here instead, the challenge was to model projections of street trees not captured in the LULC maps. The yet-unrealized greening required a work-around of streets designated for the linear urban forest to displace a lane of impervious pavement. This was carried out through a manual reassignment in ArcGIS of a percentage of street parcels from the impervious pavement category to mixed forest. These simulated green interventions appear as new entries in the LULC table as Roads2030, Roads2050, and Roads2070. To these road categories we assigned a decreasing building intensity of 0.86, 0.8, and 0.72, respectively, and a percentage increase in shade and albedo. In this way, we derived future-year estimates of modeled rather than realized cooling.

The remaining model inputs came from ArcGIS mapping or from values contained in the scientific literature. The evapotranspiration raster map used was the MODIS Evapotranspiration product, which uses data on earth land surface from satellite remote sensing obtained at a 1 km² land scale, with evapotranspiration coefficient Kc. Binary measures of greenspace and shading switches are values between 0 and 1, representing the proportion of tree cover (0 for no trees, 1 for full tree cover, with trees >2 m. The default value of 2000 m was applied for average air temperature radius. The reference temperature for a rural area 10 km or more outside of Springfield was 0.59 °C, resulting in a UHI magnitude, or the difference between the rural reference temperature and the maximum temperature observed in the city. The albedo assigned to the land cover map was 2.09 (

Table 2. Inputs to land use categories used to model urban cooling in Springfield through the InVEST Natural Capital Model.

LU Code	Description	Shade 1	Kc 2	Albedo 3	Green Area	Building 4 Intensity
0	Background	0	0	0	0	0
1	Unclassified (cloud, shadow)	0	0	0	0	0
2	High intensity (>80% IA)	0.05	0.37	0.18	0	0.95
3	Med intensity (50–80%)	0.18	0.55	0.19	0	0.3
4	Low intensity (20–50%)	0.33	0.76	0.18	0	0.1
5	Open space (<20%)	0.45	0.93	0.2	0	0.1
6	Cultivated land	0.3	0.7	0.2	1	0
7	Pasture/hay	0.3	1	0.2	1	0
8	Grassland	0.3	1	0.2	1	0
9	Deciduous forest	0.8	1	0.15	1	0
10	Evergreen forest	0.8	1	0.15	1	0
11	Mixed forest	0.8	1	0.15	1	0
12	Scrub/shrub	0.4	1	0.2	1	0
13	Palustrine forested wetland	0.6	1	0.11	1	0
14	Palustrine scrub/shrub wetland	0.4	1	0.11	1	0
15	Palustrine emergent wetland	0.4	1	0.11	1	0
16	Estuarine forested wetland	0.8	1	0.11	1	0
17	Estuarine scrub/shrub wetland	0.4	1	0.11	1	0
18	Estuarine emergent wetland	0.4	1	0.11	1	0
19	Unconsolidated shore	0	0	0	0	0
20	Bare land	0	0.3	0.2	0	0
21	Water	0.3	1	0.06	1	0
22	Palustrine aquatic bed	0.3	1	0.06	1	0
23	Estuarine aquatic bed	0.3	1	0.06	1	0
24	Tundra	0	1	0.2	1	0
25	Snow/ice	0	1	0.75	0	0
26	Roads20	0.05	0.37	0.18	0	0.95
27	Roads30	0.14	0.37	0.22	1	0.86
28	Roads50	0.2	0.37	0.28	1	0.8
29	Road70	0.28	0.37	0.36	1	0.72

¹ Tree Canopy|Multi-Resolution Land Characteristics (MRLC) Consortium, U.S. Geological Service [54]; ² evapotranspiration coefficient (Kc)—[55,56]; ³ Taha et al., 1988 [26]; ⁴ a floating-point value between 0 and 1. This is calculated by dividing the floor area by the land area, normalized between 0 and 1.

).

The estimates of city-wide ambient cooling generated by the InVEST model provided expectations of temperature reduction if the urban forest were to be realized in full. We chose not to model the full linear urban forest as designed because many of the assumptions drawn to project urban streets circa 2070 embodied futuristic scenarios that we felt distracted from the main premise of urban cooling—for example, fully autonomous vehicles and driveway elimination. For this reason, we focused on street tree planting alone.

Several demonstration sessions to beta-test the technology were held prior to the in situ community design charettes in Springfield. Two demonstration pilot sessions were presented to students and professors at UMass Amherst Department of Environmental Conservation, and two pilots were presented in the Harvard Visualization Laboratory Theater with students from Northeastern and Harvard Universities. Participants attending these demonstrations encouraged the use of infographics to enhance visualization by explaining the impact of changes within the street scenarios. From these sessions we added temperature reduction, seasonal changes to the tree canopy, and planned engineering designs for stormwater drainage as a transect below street level.

With the visualization tools ready, we held two community design forums with residents of neighborhoods identified for urban tree planting. Sessions were organized by our non-profit partner, ReGreen Springfield, led by Dr. David Bloniarz, in collaboration with the City of Springfield Public Works Department, which is responsible for implementing neighborhood street tree planting. A brief presentation highlighted the environmental and public health benefits of increased residential contact with nature: temperature and hydrology regulation, particulate filtration, stormwater infiltration, and carbon sequestration, as well as prospective health improvements for impacted residents. Following the visualization sequence, community members responded to the street tree scenarios introduced in their neighborhoods from which we collected feedback. Each participant received a gift card for their time.

3. Results

These digital design charettes demonstrated how climate change impacts of extreme heat and intensive precipitation can be mitigated by repurposing streets for tree planting.

Each of the two community charettes produced a de-identified recording or transcript from the community participatory research in which both the various designs for street tree densification across 2024–2070 were presented to community residents. From this we learned (1) preferences for form and degree of nature-based solutions for climate mitigation, and (2) feedback on the process of community research itself to reiterate and improve heat mitigation projects in other cities.

An initial rendering of the LiDAR-based street diagrams showing street tree densification we shared succeeded in providing an aerial overview of the tree canopy densification (Figure 1). However, the charette participants felt the aerial perspective lacked the immediacy of walking out one’s front door into the street-level environment and experiencing the tree canopy overhead.

Similarly, the wide-angle LiDAR depictions of street-level views distorted the spatial arrangement of buildings, trees, and the street, altering the dimensions of each and rendering these views unusable for public presentation (Figure 2).

The imaging specialists resorted to Google Street View superimposed with a series of street tree plantings that corresponded to scenarios of current plantings maturing 50 years into the future, plus tree plantings at 40-, 20-, and 15-feet spacings. These renderings were well received for their realism and ground-level relatability.

Of the four scenarios presented, charette participants favored street tree planting at 20 feet of distance to maximize the tree canopy without sacrificing a parking lane. This selection corresponds to a modeled 1.75 °F city-wide reduction in ambient temperature. The fact that participants were willing to accept a 0.75 °F increase in temperature is an indication that peri-urban culture remains vehicle-based, without the necessary transit infrastructure to permit alternative infrastructure types, even those such as green infrastructure, which would increase residential comfort and health.

Figure 3 presents still images of the street tree plantings spaced at 20 feet of distance for (a) Wilbraham Road and (b) Albemarle Street. Interactive representations of these scenarios are found in Videos S1 and S2.

The InVEST Urban Cooling Model generated temperature changes for Springfield based on models of a city-wide LUF. From this assumption, we see that the average cooling capacity rises from 0.46 °F under current landscape conditions to 0.54 °F in 2070 under conditions of maturing street trees. Average temperature variation falls slightly from 4.13 °F to 4.09 °F, and average annual temperatures from 3.07 °F to 3.04 °F (

Table 3. Change in temperature output from InVEST Urban Cooling Model 3.87.

Parameter	InVEST Model Springfield-Wide UHI Profile TΔ °C
	2020	2030	2050	2070
Square_Miles	33.08	33.08	33.08	33.08
avg_cooling capacity	0.254C/0.46F	0.281C/0.506F	0.285C/0.513F	0.298C/0.536F
avg_tmp_variation	1.9747	2.292C/4.13F	2.286/4.12F	2.269/4.09F
avg_tmp_annual	1.3847	1.702C/3.07F	1.696C/3.05F	1.690/3.04F
Temp_air_nomix	29.50–30.98	2.45 (0.61C–2.68C) (33.10F–36.82F)	2.45 (0.61C–2.68C) (33.10F–36.82F)	2.45 (0.61C–2.68C) (33.10F–36.82F)
Temp_air (high)	30.75	2.64 °C = 36.75 °F	2.64 °C = 36.75 °F	2.64 °C = 36.75 °F
Temp_air_(low)	30.62	2.18 °C = 35.92 °F	2.17 °C = 35.91 °F	2.15 °C = 35.87 °F
Heat mitigation	0–0.71	0–0.99053	0–0.99053	0–0.99053

).

4. Discussion

Digitally created street images showing how streets can be transformed in the face of rising heat offered community members the opportunity to express preferences and co-create neighborhood-level responses. Creating digital images of neighborhoods allowed them to envision future projections of their actual streets. In doing so, the research team created a new pedagogy for involving citizens and civic leaders, giving them a realistic sense of what adaptations to climate change might look like at the neighborhood scale, thus realizing the project’s innovation potential. In addition to its technical contributions, this work played a vital role in community engagement and education. By combining state-of-the-art technology with a participatory approach, this research provides a replicable framework for urban planning that addresses environmental challenges while fostering community involvement.

Visualizing the impact of urban street trees on cooling cities facilitated on-the-ground collaboration with the Springfield community. Street tree plantings may affect changes to light, shading, sightlines, and resident security, even as they offer heat mitigation and other known physical and emotional health benefits associated with contact with urban nature. Bringing nature-based scenarios to the very residents of these targeted neighborhoods opened a dialogue that sensitized designers of green interventions to residents’ concerns. Through the community design charettes the researchers solicited feedback to the different scenarios. Reactions were very positive to the projected neighborhood greening. No one appeared to favor the status quo after seeing the more tree-rich scenarios. In fact, the audience seemed to recoil upon viewing the seeming barrenness of the current street views that cycled around after the projected tree densifications.

Charette participants explained their voluntary attendance as interest in the design proposals but also as the opportunity to have a participatory voice in a municipal-driven process. Having a third party moderate the discussion also invited participation of individuals typically excluded from such procedures. The presence of city officials meant that feedback to the proposals concerning property security, additional maintenance, and traffic flow was heard directly and could influence final design. For example, participants raised questions about loss of sight lines impacting a sense of safety resulting from tree densification. Another safety concern of residents was sidewalk maintenance resulting from root protrusion, which presents tripping hazards for pedestrians, and the related burden of homeowner liability. The case of tree planting at 15 feet of distance projected onto a widened sidewalk median on residential streets would result in street narrowing. Charette participants roundly voted down this option, rejecting the loss of parking on one side of the street and the conversion of two-way streets to a single direction to accommodate wider tree planting.

The visualization charette elicited many reflective comments from attendees beyond approval and disapproval. One woman familiar with the green-lined streets of Brookline, Massachusetts, said, “When you drive through neighborhoods where there is a complete lack of vegetation, there is a sense of economic degradation. You don’t feel that when you drive along streets filled with trees no matter their economic level.” This comment challenges notions of green gentrification, whereby residents often resist neighborhood nature interventions for fear of increased rent, which greening can regrettably trigger.

Caution tempered some of the enthusiasm otherwise observed during the charettes. A young man spoke in favor of evaluating multiple changes being simultaneously proposed for his neighborhood. “Between solar panels, [tree plantings], the streets changes, there’s a lot to consider. If you go straight to one of those things without double-checking, it could be two times worse than what it was.” Another individual worried that opposition to raking could block such a project. A few individuals pointed out that tree roots would likely upheave the adjacent sidewalk blocks, creating tripping hazards, a frequent result of street tree plantings. Another participant simply attended because he wanted to see if trees might be restored in his neighborhood after their previous removal. When cities are planning infrastructure improvements, residents’ concerns are often not adequately considered. For researchers, community involvement in poorer neighborhoods is often the missing component of translating evidence-based recommendations both as a commonsense matter of public governance and as a question of environmental justice.

Our community participatory research helped accelerate the climate–health and nature–health understanding of nature-based interventions via integrated assessment methodologies, dynamic visualization modeling, and design prototyping by presenting and soliciting feedback from locally impacted communities. The development of tangible state-of-the-art visualization tools permitted community residents to review proposed changes to their neighborhoods. In this way, visualization advanced design efforts beyond sheer prototyping and invited community participation of future interventions as an essential component of climate mitigation that embodied and not just spoke to environmental justice.

Tangible benefits also emerged from this project. First, the City of Springfield has planned three pilot sites to test the survivability rates of the ten tree species when planted along busy Springfield traffic arteries. Second, the visualizations engaged the City Parks and Recreations Department to allow employees to voice the need for increased time and budget dedicated to the street tree maintenance as well as training to deal with homeless individuals who often seek shelter in vegetated areas. Most importantly, the charettes gave community members the chance to be active partners in the design process.

This pilot study has the potential to grow in three areas of interest: (A) expanding the prototype for nature-based interventions, with actual measurements to be taken on both urban cooling and correlated ecosystem services related to urban afforestation. These interventions might include various types of green infrastructure from which to choose as well as offering various cost–benefit points. (B) Demonstrating proof of concept that reimagined urban neighborhoods, visualized through augmented reality and displayed for citizens and community leaders, can redress the social determinants of health and health inequities. (C) Working with city public works and parks departments to develop robust cost–benefit analysis for implementing heat-reducing green infrastructure. Visualizing the LUF may positively activate community engagement, prompt climate concern within vulnerable groups, and stake a claim for environmental justice. Opinions voiced during the digital community charettes and focus groups and received by civic decision-makers can enhance knowledge generated by the extensive modeling currently being developed and become translational. Equity-based urban health is already of interest to health-centered foundations and philanthropies as climate change exacerbates these inequities.

Having tested the pedagogies for engaging citizens, the next steps are to improve imagery to make it more interactive for participants. The technology required to scale this research to cities and neighborhoods beyond Springfield requires the skills of an able modeler and readily available software. Scalability makes this approach adaptable to differing socio-economic conditions, including linguistic barriers, given its visual imagery basis. One way of reaching larger audiences might be to conduct design charettes via Zoom or similar platforms to overcome participant inconvenience or reticence. Scalability to new cities and audiences might include not just street tree plantings but also other forms of green infrastructure such as green roofs, afforestation of open lots, and more creative interventions, ranking participant preference among various design options. The use of digital images can also assist with gauging cost benefits of green interventions within communities, e.g., assessing “willingness to pay” for climate adaptation with green infrastructure.

5. Conclusions

This pilot study tested technologies that invite communities to partake in decisional processes affecting neighborhood change related to a reduction in urban heat islands. The development of tools that invite and facilitate community-engaged research allows neighborhood dwellers to participate in discussions that affect their comfort and autonomy, while recognizing the importance of distributional equity that comes from having a voice in such discussions. Moreover, the impact of changes addressing temperature mitigation have downstream effects on individual health and wellbeing, with cross-generational benefits in communities such as Springfield’s. We hope that this innovative pedagogical model can find traction for use in similar urban communities where nature-based solutions to decreasing heat can rectify decades of municipal underinvestment.