Progress 09/01/12 to 08/31/17
Outputs
Target Audience:Azavea's initial target markets aremunicipal government agencies that are responsible for managing street trees. In addition to Parks, Recreation, Stormwater Management, and Urban Forestry Departments, this market includes both special service districts and metropolitan planning organizations (MPOs) across the United States. We also believe there are important untapped markets withuniversity and corporate facilities management, neighborhood associations, environmental nonprofits, electric utilities, golf courses, and military bases. Azavea has reached out to these target audiences through a combination of product webinars, interviews with potential users, direct sales, and partner outreach activities. In June 2017, Azavea partnered with the USDA Forest Service on a webinar entitled, "Citizen Science and Open Data in Urban Forestry: Opportunities and Challenges". While not exclusive to the Urban Forestry Modeling and Prioritization Tools project, the webinar provided a brief overview of the project, as well as information about Azavea's other research and development efforts that support citizen forestry. Changes/Problems:Azavea encountered the following three problems in Phase II that ultimately affected the course of our research. 1. Administrative Delays The administrative review process associated with this grant lasted several months longer than expected and resulted in delayed launch of the project in the first year. Since that time, we have been engaged in a review of our indirect rates that has been delayed several times by the USDA. This required us to take no-cost extensions that we did not need to request. Delaying the project due to funding issues has impacted our ability to devote considerable resources to the project as deadlines and funding for other projects became available. 2. Other Urban Forestry Work OpenTreeMap, our software for collaborative urban tree inventory, is now implemented in over forty cities with other maps in progress. The administrative delays described above required us to balance development of the forest modeling tools with new OpenTreeMap features funded by interested clients. During 2014 and 2015, we also created the software that supported TreesCount! 2015, the decennial street tree census in New York City organized by the New York City Department of Parks and Recreation. The TreesCount software work provided an excellent opportunity to increase knowledge of Azavea in the urban forestry world, but it did distract from the forest modeling work. 3. Research Delays Our Phase I prototype included mortality and growth calculations based on studies limited to certain cities and a specific species list. We conducted a thorough research review and could not find any data that provided more extensive or robust calculations. The Urban Tree Growth and Longevity Working Group and other organizations have identified the need for more consistent data gathered over several years that could support the creation of accurate growth and mortality rates for urban trees. While we felt it was reasonable to extrapolate from the existing studies to create calculations for our Phase II project, the Urban Tree Database provided the more extensive data we had hoped to find at the beginning of the project. However, we expected the researchers to release the database earlier, which ultimately led todelays in how we implemented the mortality and growth rates. What opportunities for training and professional development has the project provided?Many of the staff members involved in this projectattended conferences, including the International Society of Arboriculture annual meeting and the Urban Forestry: Toward an Ecosystem Services Research Agenda workshop held at the National Academy of Sciences. These events provided opportunities to discuss the research with urban forestry experts working for local governments, nonprofits, and the U.S. Forest Service. The project was also mentioned in several webinars we held on Azavea's urban forestry initiatives. How have the results been disseminated to communities of interest?Throughout the grant period, Azavea continued to build our partnerships with key urban forestry groups and consultants including the U.S. Forest Service, the University of Vermont Spatial Analysis Laboratory, NYC Parks, Urban Ecos, and TreeKIT. These partnerships have proven crucial to the commercialization plan for the forest modeling tools. Our partners have already connected us with groups responding to RFPs related to forest modeling, recommended us for collaborations with other forestry groups, invited us to present on the tools at conferences, and proven vital to gathering support for the project in the urban forestry community. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
The Urban Forestry Modeling and Prioritization Tools is a cost effective, web-based software project that assists urban foresters with strategically selecting optimal tree planting locations based on their organization's planting priorities. Urban foresters can then digitally plant trees, customize mortality and replanting rates, and model the environmental and economic impacts of the trees over the following decades. With limited resources to devote to planting and maintaining trees, local governments, community groups, and other organizations struggle with prioritizing planting locations in order to achieve the greatest economical and ecological return on investment. Our tools provide assistance in determining optimal tree planting locations based on a group's individual needs and priorities. Our Phase II research centered on expanding and modifying the Phase I prototype to meet the urban forestry needs of more communities. We rewrote the prototype to cover the continental United States by including default demographic and environmental data sets available at the national level. We implemented a more intuitive user work flow where users may choose to prioritize goals such as planting in densely populated areas or areas with low income but high rate of owner occupied homes. The results can be filtered by geographic area, such as zip code, displayed as only a certain percentage of optimal areas, or overlaid with the locations of existing trees. We integrated the prioritization features with GeoTrellis, a high-performance geoprocessing engine. The resulting maps load in seconds so that users may adjust filters and quickly see changes. One of our primary objectives in Phase II was redesigning the software's user workflow to enhance its utility and ease of use. We recognize that a more intuitive interface is key for creating a system that supports use by individuals who may not be accustomed to working with a map focused website. Our user experience designer created several iterations of wireframes that incorporated proposed changes, which we reviewed both internally and with our external partners. We allocated time specifically for wireframe design, review, and redesign in order to ensure the wireframes had been thoroughly vetted. As a result, we revised the overall page and map layout for integration and consistency with our OpenTreeMap collaborative tree inventory and ecosystems management solution. We added a step-by-step process for creating modeling scenarios that include user-designated tree species, tree diameters, mortality rates, and number of trees planted.Each user-defined scenario provides an overview of cumulative ecosystems benefits over a thirty-year period. The trunk diameter growth rate equations used in the scenario calculations are provided by the Urban Tree Database, a General Technical Report that tracked urban tree growth for 14,000 trees in seventeen U.S. cities over a fourteen-year period. However, because the Urban Tree Database was released at a later date than anticipated, we were not able to integrate canopy growth projections into the forest modeling tools during the course of the grant timeframe as originally planned. In coordination with an advisory board of subject matter experts, Azavea explored implementing regression-based simulations as part of the forest modeling tools, but did not believe thatinformation was available with enough geographic detail to increase the accuracy of results. Through integration with OpenTreeMap, the modeling tools are now implemented in the cloud using Amazon Web Services. A cloud-based deployment system hosted on Amazon Web Services has enabled us to rapidly update and test changes to the forest modeling tools in a stable environment. Integration with OpenTreeMap also supports viewing the modeling tools on touch screen interfaces, including iPads and Android devices. We demonstrated the forest modeling tools to five different U.S. locations and made adjustments to the software and our product development plan based on their feedback. We continue to show the tools to other existing and potential OpenTreeMap clients and have received very positive responses.
Publications
- Type:
Other
Status:
Published
Year Published:
2017
Citation:
https://www.fs.fed.us/research/urban-webinars/citizen-science-open-data/
- Type:
Other
Status:
Published
Year Published:
2017
Citation:
http://blog.opentreemap.org/2017/06/22/launching-successful-open-data-citizen-science-initiative/
|
Progress 09/01/12 to 08/31/13
Outputs
Target Audience: The target end user audience for the Urban Forestry Modeling and Prioritization Tools is the general public. However, our target consumers are municipal governments, regional planning commissions, non-profit organizations including horticultural and urban greening groups, community and neighborhood associations, stormwater and watershed managers, universities, and large parcel owners. Each of these groups could use the web-based tools to more accurately plan tree planting projects, improve the health of the urban forest and deepen citizen engagement. To reach these groups during the first year of the grant, we met with our advisory committee, which is made up of representatives from groups similar to our target audiences, and provided them with information on the progress of our work. We also updated the project website, sent paper mailings and emails to interested audiences, and offered demonstrations of the prototype software. During the second year of the grant, the Urban Forestry Modeling and Prioritization Tools will be integrated into OpenTreeMap, our collaborative platform for crowdsourced tree inventory, ecosystem services calculations, urban forestry analysis, and community engagement. In preparation for that integration, we began discussing the modeling toolkit as part of our outreach for OpenTreeMap during webinars, conference presentations, and trade show events. Changes/Problems: Change in Project Schedule and Rate of Expenditure: The administrative review process lasted several months longer than expected and resulted in delayed launch of the project. 50% of the funds have been withheld until an audit has been completed, but there is a 6+ month wait on the audit. Further, while 50% of the funds should have been released, only $17,000 were released until the issue was finally sorted out in November 2013, more than a year after starting work. The delay in release of funds significantly impeded our ability to implement the project at the planned rate of work. Change in Development Based on Other Urban Forestry Work: OpenTreeMap, our software for collaborative urban tree inventory, is now implemented in over ten cities with other maps in progress. The Urban Forestry Modeling tools will be integrated with OpenTreeMap in a single, integrated cloud-based system. Our success with OpenTreeMap has provided us with both some initial revenue and a wealth of user feedback, but it has required us to focus on both individual tree map implementations, which are difficult to manage on an ongoing basis, and the subscription system, which required resources to develop. We concentrated on building the cloud-based deployment platform in Amazon Web Services first in order to support the launch of the OpenTreeMap system. This required us to slighly alter the schedule for developing the Urban Forestry Modeling tools to focus on cloud-based infrastructure first. We do not believe either issue will prevent us from completing the planned work by the end of the grant term in August 2014. What opportunities for training and professional development has the project provided? During the first year of the grant, the urban forestry team at Azavea met with the advisory committee for the Urban Forestry Modeling and Prioritization Tools. Comprised of forestry experts working for local governments, non-profits, and the US Forest Service, the advisory committee provided essential information on forestry issues. Members of the urban forestry team also attended conferences including the International Society of Arboriculture annual meeting and the Urban Forestry: Toward an Ecosystem Services Research Agenda workshop held at the National Academy of Sciences. We have included information about the modeling tools at urban forestry and technology conferences throughout the first year of the grant. The project was also mentioned in several webinars we held on Azavea’s urban forestry initiatives. Key resources are listed at https://www.opentreemap.org/resources. Through a separate grant, Azavea staff have also attended the Amazon Web Services re:Invent conference and cloud computing workshops. In addition, Azavea hired a gamification consultant to assist with integration of badging, competition and other game elements into the software paltform. How have the results been disseminated to communities of interest? Information about the Urban Forestry Modeling and Prioritization Tools is available at https://www.opentreemap.org/analyze/ and http://www.azavea.com/a/urban-forestry-modeling/. Links to available webinars and blog entries can be found on the Resources page at https://www.opentreemap.org/resources/. We regularly send email updates to individuals who have signed up to receive notifications about new features. We have also provided several demonstrations of the software to interested organizations. What do you plan to do during the next reporting period to accomplish the goals? Due to a lengthy administrative review process, we were delayed in the launch of our Phase II grant. We intend to proceed with the original work plan outlined in our grant narrative, pending any further delays based on administrative auditing and funding issues. Key tasks include: Finalize national scale data sets Implement the revised user experience based on completed wireframes Move modeling tools to cloud-based deployment system Develop regression-based simulation Develop administrator dashboard Implement test instance Enable use of local data sets
Impacts
What was accomplished under these goals?
The Urban Forestry Modeling and Prioritization Tools is a cost effective, web-based software project that assists urban foresters with first strategically selecting tree planting locations in order to maximize impact on the environment and local communities and then modeling the environmental and economic impacts of those trees over the following decades. With limited resources to devote to planting and maintaining trees, local governments, community groups, and other organizations struggle with prioritizing planting locations in order to achieve the greatest economic and ecological return on investment. Our tools provide assistance in determining optimal tree planting locations based on a group’s individual needs and priorities, enabling cities to more effectively allocate limited resources for planting and maintaining the urban forest. Our research has resulted in the following outcomes and impacts connected to the major goals outlined for the project. 1. Support the ability for local organizations to upload and add their own datasets to the system while still providing national scale. By providing several key national scale data sets, we can support the deployment of the modeling tools anywhere in the United States. We have compiled a draft list of data sets including information from the United States Census, the American Community Survey, Landsat temperature data, and various public and open lands organizations. These data sets are available at a national scale and will be included in the standard Urban Forestry Modeling and Prioritization Tools web interface so that any organization in the country may create prioritization heat maps without needing to upload additional data. The modeling tools will also support the ability for organizations to upload local data sets to support increased accuracy of the prioritization models. The most common request from our prospective users has been the ability to add urban tree canopy assessments although we expect organizations will wish to upload other local data sets related to land use, soil types, impervious surface, and other factors. We will implement this feature in the second year of the grant. 2. Redesign the user workflow as well as the cartography design. During creation of the Phase 1 prototype, we realized that the user experience had become more complex than originally intended due to the large number of factors users could select from as part of the decision-making process. During the first year of the Phase II grant, we redesigned the user workflow to support a more intuitive use of the map interface, easier data entry, and more simplified methods for adjusting factors used in the modeling calculations. Specific enhancements include adjusting the color display, improving the appearance of map overlays, and arranging the workflow into a step-by-step process that incorporates the map into the user experience. Users also can now limit the prioritization calculations to a defined geographic area such as a zip code, land use type, or surface type. These workflow improvements will enable organizations to more effectively identify optimal planting locations based on factors essential to the organizations’ goals and mission. 3. Extend the sophistication and flexibility of the tree planting simulation options. During our Phase I grant, our advisory committee emphasized that the tree planting simulation options provided incredibly useful data for organizations wishing to demonstrate the economic and ecological benefits of trees. By modeling the growth, mortality, and benefits of trees over a thirty year period, the Urban Forestry Modeling and Prioritization Tools give organizations essential information on the ideal species to plant in key locations. As we reviewed the prototype with more stakeholders, we learned that there are some additional variables that groups want to consider in the planting design process. We adjusted the wireframes and design of the simulation engine to support user customization of replanting options, growth and mortality rates, tree planting by land use category, and other factors. The resulting system will enable organizations to create models that meet their unique needs. 4. Integrate regression equations to model spatially varying relationships in tree species growth. The Urban Forestry Modeling and Prioritization Tools prototype currently calculates growth rates for tree species based on a study of Chicago's Urban Forest Ecosystem that was completed by the Northeastern Forest Experiment Station in 1994. While this has been effective for the Philadelphia prototype, our subsequent research indicates that integrating regression equations into the modeling component will enhance its ability to model tree species growth over extended periods and facilitate a more accurate measurement of related ecosystem benefits. Regression equations are used to model spatially varying relationships, such as the levels of tree and canopy growth as a result of tree species, geographic location, climate, soil condition, and similar variables. We will implement regression equations in the second year of the grant. 5. Implement a cloud-based deployment option. The Urban Forestry Modeling and Prioritization Tools prototype application was designed to operate on commodity Linux servers, but based on our market research and feedback from existing customers of OpenTreeMap (our collaborative platform for urban forestry inventory and analysis), we believe that the prototype must be adjusted to support deployment in a cloud-based subscription system. In the first year of our grant, we began the process of transitioning our urban forestry tools to a cloud-based system using the Amazon Web Services (AWS) platform. We selected AWS after considering cost, range of services, global network of data centers, support for auto-scaling, billing, and other features. We believe AWS provides the most robust infrastructure, and we have transitioned OpenTreeMap, our other urban forestry software tool for collaborative urban tree inventory, onto the Amazon system in a way that will enable us to integrate the modeling and prioritization tools into the same framework. 6. Implement touchscreen capabilities. Contemporary web applications need to be able to operate effectively on both a conventional desktop/laptop workstation and a tablet device. Our revised wireframes adjust the design of the software to support use via touch screen interfaces on tablet computers. Our advisory committee has reviewed the wireframes and provided feedback on the possible use of the software on tablet devices. We will implement the revised design, including support for tablets, in the second year of the grant.
Publications
- Type:
Websites
Status:
Published
Year Published:
2013
Citation:
Urban Forestry Modeling Page on OpenTreeMap Product Web Site - https://www.opentreemap.org/analyze/
|