The STEM-Net Dashboard
The STEM-Net Mockup, submitted as part of proposal to San Jose City
Azure Reference Architecture
Board and Moisture Sensor
Azure Backend Screenshot, displaying blob storage
Azure Backend Screenshot, displaying metrics from simulated devices
Azure Backend Screenshot, displaying IO metrics
Azure Backend Screenshot, displaying simulated devices
Team Members of STEM Net
STEM Net Logo
Our project, which we call the Street Tree Monitoring Network (STEM Net), is a prototype software, built with Azure Iot Hub, Azure Cloud, and Azure Maps, that aims to assist urban foresters and stakeholders in reducing urban tree mortality. The software would link a network of soil moisture sensors and provide detailed information to improve urban forestry outcomes. We consulted scientific literature, as well as certified arborists and departmental executives at the City of San José, to inform the design of our project.
Green Infrastructure and Climate Resilience
Thriving in the face of climate change is widely considered the sustainability challenge of the century. Responding to climate change includes both mitigation, or reducing greenhouse gas emissions to reduce the severity of climate change, and adaptation, which involves reducing societal vulnerability to projected changes in climate (National Aeronautics and Space Administration). Extreme heat events, for example, pose a threat to human life and ecological function, and the severity, frequency, and duration of such events is expected to increase with climate change (Norton et al. 2014). The urban heat island effect, which results from heat-absorbing surfaces in urban landscapes, causes cities to be warmer than surrounding rural areas, compounding the effects of climate change-induced heat waves (Norton et al. 2014). In planning for both climate adaptation and mitigation, there is significant interest in “green infrastructure,” which encompasses engineered systems that use ecological processes to achieve sustainability goals (Pataki et al. 2016). Urban forestry, or the planting and management of trees in cities, is one form of green infrastructure that is key to developing climate resilience.
Urban forests provide multiple services that can help achieve sustainability goals. Trees allow cities to harbor remarkable biodiversity, even providing habitat for endangered species (Alvey 2006). Biodiversity provides numerous environmental services, and thus the conservation of biological resources is an important component of sustainable planning (U.S. Climate Resilience Toolkit). Trees and other vegetation also remove gaseous air pollutants by absorption, resulting in improved air quality (Abhijit et al. 2017). Trees also stabilize soil and retain stormwater runoff, and it is estimated that urban forests reduce annual runoff by 2 to 7% (Fazio 2010). Critically, the urban forest, by providing shade and transpiring water, cools urban environments. Thus, urban forests can help reduce the energy consumption and costs associated with climate control, aiding in climate mitigation, as well as ameliorate the effects of extreme heat events, aiding in climate adaptation (Norton et al. 2015, SJCFMP 2021). Many cities seek to increase the extent and density of their urban forests to capitalize on these benefits as they adapt to a warming world.
Table 1: Benefits of Urban Forests
|Climate Mitigation||Reduced energy costs and emissions for cooling||Trees cool urban environments which reduces energy consumption, and thus the energy and emissions costs associated with cooling (Norton et al. 2015).|
|Other Sustainability Benefits||Improved air quality||Trees reduce pollution by absorbing pollutants through stomata (leaf pores) or through surfaces (Abhijith et al. 2017)|
San José, California as a Case Study
We used our hometown of San José, California, a city of over one million people, to inform the design of our project. San José’s warm Mediterranean climate makes the city especially vulnerable to extreme heat events (Norton et al. 2015). Indeed, we personally witnessed such harrowing events in 2020, when record-breaking daytime and nighttime heat crippled the state powergrid and left residents vulnerable to heat-related illness. San Jose has articulated the need to address climate change in its Climate Smart Plan (Climate Smart San José 2018), which recognizes the potential of smart urban forestry in building climate resilience. San José is positioned to benefit immensely from expanding its community forest. Currently San José’s estimated 1.6 million trees generate savings of $77 million annually from reduced cooling costs (Xiao et al. 2013). San José’s large land area and diverse population, as well as its very limited tree management budget in comparison to cities of similar size (SJCFMP), make it a fitting model for understanding and addressing the challenges of urban forestry.
The San Jose Community Forest Management Plan (SJCFMP), which is currently in the drafting stage, analyzes the state of the San Jose urban forest and identifies areas for improvement. In San Jose, there is no central authority to manage city trees and as a result, the governance structure for the management of the urban forest is quite complex, involving multiple city departments as well as private property owners, and nonprofit organizations. The Department of Transportation is responsible for managing many roadside trees, and is also the only city department to have certified arborists on staff (SJ CFMP, SJ Tree Policy Manual). Trees in city parks however, are managed by the Department of Parks, Recreation, and Neighborhood Services (PRNS). There is no funding available to PRNS to plant trees in parks, and the budget to care for trees is extremely limited. As a result, PRNS staff are dependent on partnerships with local nonprofit organizations and community groups to fund and organize tree planting within parks (SJCFMP).
Trees on private property, including trees in front and backyards, in commercial lots, and most street trees along roads, are maintained by each individual property owner in accordance with city regulations, which are enforced by the Department of Transportation (SJCFMP). Property owners are responsible for funding the pruning and watering of private trees. In sum, the urban forest of San Jose, California, as in other cities, is jointly managed by multiple parties.
We spoke with several staff members from the Department of Parks, Recreation, and Neighborhood Services (PRNS) and the Department of Transportation at the City of San José to inform our work: Russell Hansen (certified City Arborist), Nara Baker (certified Assistant City Arborist), Avi Yotam (Interim Deputy Director, Parks Division, PRNS), Riley Knight (Arborist Technician), and Jeffrey Gomez (Parks Facility Supervisor, PRNS). Through these conversations, our review of policy documents, and survey of relevant scientific literature, we have identified one area of urban forestry where Internet of Things technology could improve tree planting outcomes and support vibrant community forests.
Tree Mortality Undermines Urban Forestry Initiatives
There are several challenges cities may face in achieving a successful urban forest program, and tree mortality is one of these challenges. Annual urban tree mortality rates can be as high as 30%, due to a variety of factors (Hilbert et al. 2019), including mortality due to lack of care. We focused on building a solution to assist the logistics of diligently and effectively providing water to newly-planted trees, which is critical for tree survival, and thus for achieving a sustainable urban forest in San Jose (SJCFMP), and in any city.
The tree establishment period, when young trees are developing root systems, is the life stage with the highest risk of mortality (Hilbert et al. 2019). During this period, which typically ranges from 3 to 5 years, supplemental watering must be provided to support tree growth (SJCFMP). This is termed “establishment care.” In semiarid climates like that of San Jose, establishment care is especially critical to the survival of young trees, as trees have the potential to die due to water stress during seasonal dry periods (SJCFMP). Additionally, as climate change increases the frequency and intensity of droughts and other extreme weather events (source), tree failure will continue to be a major challenge to urban forestry efforts (Barona 2015).
Parks are popular sites for tree planting. Unfortunately, in San José, there is no funding available to PRNS to care for newly-planted trees in parks. As a result, the groups that plant trees in parks are responsible for funding and performing follow-up care for the period of time it takes for the trees to establish root systems (SJCFMP 2021). This is termed “establishment care.” However, as the Plan notes, in many cases, this watering often ceases after a few months. Due to the lack of funding to extend irrigation systems or have park staff care for saplings, however, such volunteer watering is the only option available to PRNS (SJCFMP 2021). The Plan notes that “because of the inconsistency of volunteer watering, the City must consider other options to manage park tree planting and establishment care” (SJCFMP 2021).
Residential lots account for 62% of land use in San José and are a key to densifying the city’s urban forestry (SJCFMP). Tree planting campaigns organized by nonprofit organizations to encourage residents and businesses to plant trees on private property often rely on education and consistent reminders to ensure that saplings receive establishment care. It is difficult, however, for involved authorities and nonprofits to monitor the care of such a large number of young trees across a vast land area
Optimal Irrigation Practices for a Resilient Community Forest
Even when watering is provided, it is often inadequate in several ways. The SJCFMP notes that overhead spray sprinkler irrigation systems, which are most commonly used for tree irrigation, do not deliver the quantity of water needed to soak the tree roots. When water is not delivered at depth, trees develop shallow root systems that do not allow them to survive drought, or even grow without irrigation (SJCFMP). Deep watering during establishment allows trees to develop extensive root architecture, resulting in drought-tolerant specimens that can survive without regular irrigation. In the larger context, optimal tree watering produces a less management-intensive and more climate-resilient urban forest. This need for effective irrigation was emphasized by San Jose city staff in our conversations.
The Niche for STEM Net
As a case study of San Jose reveals, tree planting is an important strategy to improve urban sustainability in the face of climate change. For successful urban forestry, providing adequate watering to newly-planted trees is critical, and poses a significant logistical and organizational challenge given that city trees are managed by an array of parties.
Internet of Things technology provides numerous opportunities to urban forest authorities for smart management (Torresan et al. 2021). Soil sensors can provide the data necessary for urban foresters to monitor and ensure the survival of newly-planted trees during establishment, increasing tree survival rates, especially under severe climatic conditions (Pascual et al. 2019). We built the Street Tree Monitoring Network, or STEM Net, to assist urban foresters in monitoring the soil moisture to ensure that trees receive adequate water, thereby helping to improve tree planting outcomes. STEM Net involves the following components:
- A network of soil moisture sensors placed around trees upon planting, providing real-time and historical below-ground soil moisture data to ensure that watering practices are adequate for healthy tree growth;
- an Azure IoT Hub resource to gather telemetry and transmit data to the Microsoft Azure Cloud;
- a front-end public-facing data visualization dashboard to display remote soil moisture observations for urban foresters and to facilitate coordination with parties tasked with providing tree establishment care.
What it does
Historical and current root zone soil moisture observations can help urban foresters identify whether the tree is receiving water, whether the watering is wetting the soil to an optimal depth whether the water is wetting the soil to an adequate radius around the tree, the last time since last watering, potential water stress, or overwatering. We’ve designed this dashboard to be able to integrate rainfall and temperature data in informing this decision making process.
From the public facing dashboard, members of the public as well as city employees or managers from volunteer organizations will be over to view a dot representing a tree on the Azure Maps widget. The user can click on an individual tree to see the most recent moisture sensor data. From this popup window, the user can also look at more information about the tree, like the date it was planted, it’s height, trunk diameter, and other data that the city has already collected about it’s street trees. Also from this popup, the user can view historical moisture data for the tree in the form of a line chart, as well as access a form for submitting an update when the tree has been watered or cared for in any other way.
On the right hand side, there is a pane for the filter options that allows the user to filter by moisture. This is especially important, as it allows the driest trees and plots to be identified quickly, so that they may be attended to.
From the hardware side, a moisture sensor is inserted into the ground through a housing container.The sensor is attached to a board, which sends the moisture data every five seconds to the Auzre IoT Hub. On advice of the San Jose City arborist, we intend on installing two moisture sensors for every tree in different locations and depths, in order to gain the most accurate view of the moisture content of the soil.
How we built it
The backend consists of an Azure IoT Hub, that is capable of receiving data from both mock and physical devices. For the purposes of this hackathon, we provisioned three devices producing mock data and one physical device, as a proof of concept. The telemetry data is streamed from Azure IoT Hub into an Azure blob storage using Azure Stream Analytics, where it is then imported into an Azure SQL Server database. From the SQL database, it can then be queried by our dashboard web application.
The user-facing dashboard is a web application built in ASP.NET Core MVC, with a frontend designed from scratch using Bootstrap in order to conform to Microsoft’s Fluent UI Design standards. The frontend is currently capable of displaying trees on an Azure Maps web control, as well as allowing the user to click on individual trees and view moisture data pulled from the Azure database.
On the hardware side, a capacitive moisture sensor is connected to an ESP32 board. The ESP32 microcontroller is connected to the Azure IoT hub via a WiFi connection. Using the moisture sensor, the microcontroller measures the moisture of the soil and uploads the readings to the IoT hub using the MQTT protocol.
Challenges we ran into
From a design perspective, when we spoke to representatives from the city arborist and parks departments, we heard two primary hardware-based concerns around our sensor: 1) whether or not the battery and service-life of the sensors would last until the end of the establishment period for the trees, and 2) how to deal with vandalism and theft. The hardware implementation is estimated to be able to use a battery that lasts five years without need to recharge, which is the same length of time the soil moisture sensors are rated to last. The establishment period for newly transplanted trees is around three years, depending on the size and species, so we expect that this is good enough for what it is meant to do. To deal with the issue of vandalism, we decided to bury the unit underground, with a 24-inch PVC pipe capped at the top like a christy box being inserted into the ground. The unit can be removed from the housing to be serviced, but it is also hidden such that the sensors are unlikely to be damaged or stolen.
The logistics of installation and maintenance in practice will be a challenge. For private tree owners, we could partner with non-profit tree planting organizations like Our City Forest in order to include our sensors as part of the tree planting package, where installation cost is paid up-front. Internet connectivity could be obtained from a homeowner’s router, public WiFi hotspots, or from cellular data with the inclusion of an additional module.
The final project design challenge is obtaining funding for manufacturing hardware sensors, and hosting the web software. We would be keen to explore public-private partnerships with companies like Microsoft, who would showcase not only their support for the community but also the capabilities of their products as well. In addition, we could apply for government grants aimed at boosting sustainability or aiding the development of forestry, such as CalFire urban reforestation grants.
From an implementation perspective, we ran into several different challenges when building a working product for the hackathon. As we were limited by time and availability, many of the controls on the user-facing dashboard are currently stubs, with only the core functionality of being able to view the moisture telemetry data working as intended. In addition, we have not yet been able to pull data from Azure IoT Hub in real time, instead working through a longer roundabout system of streaming the data to a blob storage container before importing it into the database. Although this process is useful for tracking historical data, we intend to explore our options for real-time data access in the future.
Accomplishments that we're proud of
We are proud that the staff members we spoke to at the city of San José were very excited about the possibilities of STEM Net. Nara Baker, a certified Assistant City Arborist told us, “we are very interested in staying in touch and coordinating with you on potential future implementation! There is certainly exciting potential here!” We too are excited with the potential for implementing our project.
We are proud that our project truly meets the needs of urban foresters and stakeholders in building a more sustainable civic landscape in the face of climate change.
What we learned
This hackathon was the first time that any of us had worked with the Azure IoT platform, and it was a deeply educational experience for all involved. On the dashboard side, we learned how to query a Microsoft Azure IoT Hub to receive message and device data. On the Azure backend side, we learned how to provision mock devices and route messages to be receivable from the frontend. On the hardware side, we learned how to send messages to the Azure IoT Cloud, where it can be monitored and retrieved by a service application.
For many of us, this hackathon also served as the first look into how technology could be applied to resolving real-world issues. The research that he conducted for this project was integral to advancing our understanding of urban forestry, including what solutions already exist and what challenges we could aid in overcoming.
What's next for STEM Network
Staff members we spoke to at the city of San José were very excited about the possibilities of STEM Net. Nara Baker, a certified Assistant City Arborist told us, “we are very interested in staying in touch and coordinating with you on potential future implementation! There is certainly exciting potential here!”
We too are excited with the potential for implementing our project, and we anticipate that we could partner with community tree planting organizations and city governments. In San Jose, the nonprofit Our City Forest funds, grows, and plants trees on a massive scale across the city. We hope to reach out to Our City Forest, as well as other groups that organize tree planting, such as the San Jose Downtown Association and other neighborhood associations, to begin building the STEM Network. The city staff we spoke with believe that these groups would be interested in using our project to coordinate tree care and track tree planting outcomes.
There is increasing consideration among urban forestry advocates for rectifying systemic inequities in urban tree cover. Research has elucidated inequalities in urban forest cover between lower-income neighborhoods and affluent neighborhoods (SJCFMP, Pataki et al. 2016). Communities of color are also more likely to have lower tree density (Pataki et al. 2016). As we implement STEM Net, we want to emphasize that we would need to work with municipalities and community organizations in addressing equity concerns in the siting of tree planting projects, and the engagement of diverse stakeholders. Both of these goals are central to the city of San José (SJCFMP) as well as to policymakers nationwide. We expect that our project would reduce tree mortality, and over time, reduce the cost per tree. By helping to reduce tree mortality rates, we hope STEM Net could assist cities to increase canopy cover even with limited funds and resources. We also anticipate that STEM Net could assist urban foresters in visualizing and analyzing tree cover, tree mortality, and stakeholder engagement, which would help target work in underserved communities.
STEM Net is also designed to work seamlessly with other IoT solutions. As cities become more connected, we envision that existing irrigation systems, as well as other monitoring systems like cameras, diameter-measuring sensors, and so forth can be added to the dashboard, extending its functionality.
STEM Net also has multiple research applications; the data from the soil moisture sensors could assist urban planners, foresters, and scientists in understanding the soil-water dynamics in urban ecosystems, as well as the state of urban forest. The network could also measure soil moisture in potential planting sites and inform the siting of tree planting projects.
Abhijith, K. V., Prashant Kumar, John Gallagher, Aonghus McNabola, Richard Baldauf, Francesco Pilla, Brian Broderick, Silvana Di Sabatino, and Beatrice Pulvirenti. 2017. “Air Pollution Abatement Performances of Green Infrastructure in Open Road and Built-up Street Canyon Environments – A Review.” Atmospheric Environment 162 (August): 71–86. https://doi.org/10.1016/j.atmosenv.2017.05.014.
Alvey, Alexis A. 2006. “Promoting and Preserving Biodiversity in the Urban Forest.” Urban Forestry & Urban Greening 5 (4): 195–201. https://doi.org/10.1016/j.ufug.2006.09.003.
Barona, Camilo Ordonez. 2015. “Adopting Public Values and Climate Change Adaptation Strategies in Urban Forest Management: A Review and Analysis of the Relevant Literature.” Journal of Environmental Management 164: 215–21.
City of San José. n.d. “Climate Smart San José.” City of San Jose. Accessed May 24, 2021. https://www.sanjoseca.gov/your-government/environment/climate-smart-san-jos.
Fazio, James. 2010. “How Trees Can Retain Stormwater Runoff.” U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station. https://doi.org/10.2737/NRS-GTR-194.
Hilbert, Deborah R., Lara A. Roman, Andrew K. Koeser, Jess Vogt, and Natalie S. van Doorn. 2019. “Urban Tree Mortality: A Literature Review.” Arboriculture & Urban Forestry: 45 (5): 167-200. 45 (5): 167–200.
Norton, Briony A., Andrew M. Coutts, Stephen J. Livesley, Richard J. Harris, Annie M. Hunter, and Nicholas S. G. Williams. 2015. “Planning for Cooler Cities: A Framework to Prioritise Green Infrastructure to Mitigate High Temperatures in Urban Landscapes.” Landscape and Urban Planning 134 (February): 127–38. https://doi.org/10.1016/j.landurbplan.2014.10.018.
Pascual, Adrián, Rafael Rivera, Rodrigo Gómez, and Susana Domínguez-Lerena. 2019. “Monitoring Water-Soil Dynamics and Tree Survival Using Soil Sensors under a Big Data Approach.” Sensors 19 (21): 4634.
Pataki, Diane E., G. Darrel Jenerette, Stephanie Pincetl, and Tara LE Trammell. 2016. “THIRTY-NINE. Urban Ecosystems.” In Ecosystems of California, 885–96. University of California Press.
“SAN JOSÉ COMMUNITY FOREST MANAGEMENT PLAN 85% DRAFT.” 2021, 105.
Torresan, Chiara, Marta Benito Garzón, Michael O’Grady, Thomas Matthew Robson, Gianni Picchi, Pietro Panzacchi, Enrico Tomelleri, Melanie Smith, John D. Marshall, and Lisa Wingate. 2021. “A New Generation of Sensors and Monitoring Tools to Support Climate-Smart Forestry Practices.” Canadian Journal of Forest Research, no. ja.
website, NASA’s Global Climate Change. n.d. “Climate Change Adaptation and Mitigation.” Climate Change: Vital Signs of the Planet. Accessed May 23, 2021. https://climate.nasa.gov/solutions/adaptation-mitigation.
Xiao, Q., E. G. McPherson, J. Bartens, C. Wu, J. Simpson, and J. O’Neil-Dunne. 2013. “Urban Forest Inventory and Assessment Pilot Project: Phase Two Report.” Davis, CA: USDA Forest Service, Pacific Southwest Research Station. Submitted to CalFire Fire and Resource Assessment Program.