Source: UNIVERSITY OF RHODE ISLAND submitted to NRP
DESIGN AND IMPLEMENTATION OF A GEOSPATIAL DECISION SUPPORT SYSTEM FOR MANAGING STATE? OWNED FORESTS IN RHODE ISLAND
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1017930
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Oct 17, 2018
Project End Date
Sep 30, 2020
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF RHODE ISLAND
19 WOODWARD HALL 9 EAST ALUMNI AVENUE
KINGSTON,RI 02881
Performing Department
Natural Resources Science
Non Technical Summary
Approximately 55% of Rhode Island (RI) is covered with forest (Butler, 2014). Effective management of publicly-owned, operable timber harvest lands is a key responsibility for the Rhode Island Department of Environmental Management (RIDEM) Division of Forestry. To date, geospatial data are available and used as a component in the management of some elements of the RIDEM forestry resources. However, RIDEM is looking to expand the application of geospatial technologies applied for similar purposes by other states in the Northeast to develop high-level input to plan for harvest and regrowth operations supporting ecological resource management activities as well as forest health assessments.Rhode Island has a rich collection of geospatial datasets that are well-suited for inventorying and assessing the condition of RI's forest resources. These data originate from a variety of sources. The goal of this project is to integrate these data along with other relevant forestry-specific information and develop a series of easy-to-use online mapping applications that will facilitate the planning activities of the RIDEM Division of Forestry and the RIDEM Division of Fish and Wildlife. This data compilation and resulting decision support toolbox will make it possible for RIDEM forest managers to utilize best available geospatial data in a readily accessible format to assist in their decision making.Smart forest resource management and planning requires the consideration of many important variables. At this time, the critical data needed to manage forests are in GIS format. However, most forest management decision-makers do not have ease-of-use access to these geospatial data resources and the analytical capabilities to visualize and manipulate these data.Our project will develop an easy-to-use web mapping interface that will provide access to relevant geospatial data by decision-makers who do not have the time or resources to learn desktop GIS software. By developing the most accessible and easy to use interfaces to accurate models incorporating the most current and best available geospatial data we can maximize understanding of the condition of our forest environment thereby preserving and enhancing critical ecological resource areas.Literature Cited:Butler, Brett J. 2014. Forests of Rhode Island, 2013. Resource Update. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 4 p
Animal Health Component
60%
Research Effort Categories
Basic
(N/A)
Applied
60%
Developmental
40%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
12306992060100%
Goals / Objectives
1. Collaborate with RIDEM biologists and forest ecologists to identify characteristics, variables, and regulatory parameters for harvesting timber within state-owned forest land. Forest land will ultimately be categorized as "harvestable with no restrictions," "non-harvestable," or "harvestable with restrictions." Develop and document repeatable geospatial analyses to classify forest mapping units into operational harvest outputs. Develop interactive online maps depicting harvest potential. Develop summary statistics of acreage within each category.2. Make use of an enterprise GIS to store, manage, and distribute content. Publish web maps, journals, and story maps to communicate results to resource managers. Publish tabular reports and summaries of harvest potential by management area.3. Explore using input of up-to-date, high resolution true-color and multispectral imagery collected with sUAS technology to augment pre/post-harvest geospatial data collection, interpretation, classification of harvest/no-harvest areas, and forest health assessment.
Project Methods
RI has developed one of the richest, most comprehensive, complete, and high resolution statewide geospatial databases in the United States. RI state foresters and wildlife biologists are interested in incorporating these data into their decision-making framework to address major issues within the RI forests including: climate vulnerability, resilience and adaptation; invasive species identification, monitoring and mitigation; fire prevention and fuelwood management; and insect and pathogen monitoring and management. Managing for early successional and young forests to promote wildlife diversity is also an important goal (Rittenhouse et al., 2017).The state's small area makes conducting geospatial analytics on a statewide scale attainable and repeatable. We will work in cooperation with state foresters and wildlife biologists to define current management areas, determine minimum harvest mapping unit size and identify and extract parameters from statewide biological and physical geospatial datasets that will identify constraints for harvest, restricted harvest, and no harvest areas. LiDAR-based elevation data will be an important source dataset to extract detailed slope information. Other important collateral inputs to this analysis are soils, wetlands, hydrography, roads, and trails.We will develop and document a transferable and stepwise analytical workflow model to apply constraints and parameters, resulting in a classified operable forest online map depicting restricted, partially restricted, and no-harvest areas. Interactive web maps of classified operable harvest potential will be produced along with tabular output summarizing harvest potential by management area. Using enterprise GIS capability, preliminary results with RIDEM forestry and wildlife managers will be shared and feedback incorporated into subsequent runs of the analytical model. The enterprise GIS database provides for a flexible, scalable, and distributed data model for forest assessment (Rogers, 2018). Time permitting, we will explore the acquisition, interpretation, and classification of high-resolution sUAS to support pre-/post-harvest conditions and forest health assessment. The resulting sUAS products will be incorporated into the enterprise GIS system.Literature Cited:Rittenhouse, C.H., Rittenhouse, T.R., and T.H. Worthley. 2018. Silvicultural Approaches to Increase Wildlife Diversity in Forest Wetlands. The University of Connecticut. McIntire Stennis Accession No: 1004150Rogers, L.U. 2018. UN Forest Assessment Model. University of Washington. McIntire?Stennis Accession No: 101581

Progress 10/17/18 to 09/30/20

Outputs
Target Audience:Forestry and wildlife managers with the Rhode Island Department of Environmental Management Divisions of Forest Environment and Wildlife Management. These include forest managers and ecologists specializing in general forestry practices, fire ecology, urban forestry, disease, pest management, and forest health. Wildlife managers and ecologists specializing in amphibian, aquatic, avian, and small/large mammal habitats. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Development of the RI State-owned Lands Operable Forest Viewer application created professional development opportunities for URI Geospatial Extension Program staff to explore, learn, implement, and keep pace with the rapidly developing functionality of ArcGIS Online web mapping applications using ArcGIS Online Story Map Series, ArcGIS Dashboards, and Web AppBuilder applications. In addition we were able to experiment with the workflows of generating and exporting ArcGIS Pro ModelBuilder geoprocessing workflows and collaborate with RI DEM to import, install, and duplicate these processing workflows on their systems. These skills and workflows are directly transferrable to workshop and certificate program modules within our URI Geospatial Extension Training programs. We were also able to extend this technical knowledge through one-on-one interactions with RI DEM GIS technical staff during the implementation and application, data, and geoprocessing model transfer phases of the project. How have the results been disseminated to communities of interest?Results of the RI State-owned Lands Operable Forest Viewer application were shared in real-time with RI DEM Divisions of Forest Environment and Wildlife Management staff throughout the ongoing phases of development. When it came time to transfer the components of the application to them, we used digital export package functionality within the ArcGIS Pro and ArcGIS Online platform environment to share the application functionality, content, and modeling workflows. We worked with RI DEM GIS staff to install and implement the components of the application directly on their online GIS platform infrastructure. RI DEM now has the capability to use the application for pre-harvest management activities, update the geodatabase as new geospatial inputs become available, and re-run the geospatial processing models. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Goal 1 The operable forest web mapping application for RI State-owned forest management areas is based on assessments of the most current forest habitat cover in RI Department of Environmental Management (RI DEM) forest management areas, regulatory prohibitions to harvest such as wetland habitat, setbacks to roads, important forest habitat, species of concern, and excessive terrain. Also included are restricted harvest areas where cutting is limited due to proximity to trails and streams. The interface is map-based, interactive, web- accessible, easy-to-use, and allows generation of printed maps and tabular summary reports. We worked closely with RI DEM, to incorporate RI-specific regulatory parameters guiding timber harvest and successfully deliver the RI State-owned Lands Operable Forest Viewer application, input data layers, and geospatial processing model package to them. Via teleconference, we introduced the final application to RI DEM, trained them how to use it, worked with them to move the application onto their internal online GIS portal and provided technical assistance to them in getting it installed and operating. RI DEM is now equipped with a tool that will enhance forest pre-harvest planning and management and is communicating internally on how to incorporate use of the application into their planning workflows and decision-making. Feedback from our RI DEM partners on the final product was very positive and representative of the kind of tools that RI DEM is seeking to develop around improved approaches to managing their forest assets based on sound science, quality data and analytics. RI DEM managers indicated making use of the application to visualize site conditions prior to carrying out timber clear and selective cutting operations, with the goal of promoting understory growth to improve wildlife habitat. Knowing in advance what areas may be restricted due to proximity to streams, and prohibited because of wetlands and existence of steep slopes, is very useful to them before meeting with field crews for onsite assessments. Goal 2 and 2a. Application Overview - The RI State-owned Lands Operable Forest Viewer application is internally-accessible via web browser to members of the RI DEM online organization GIS platform. The visual application is developed within the Esri ArcGIS Online platform using the Story Map Series application templates to embed the content pieces - basic map viewers and dashboards. The Create Report functionality was developed using the Esri ArcGIS Online Web AppBuilder application. The analytical engine was developed with the ArcGIS Pro Desktop ModelBuilder geoprocessing module. The map viewer is organized into a series of tabs: Summary, Map Input Layers, Dashboard: Operable (harvestable) Lands, Dashboard: Non-Operable (non-harvestable) Lands, Create Report, and Project Team. Summary tab - introduces the application. Map: Input Layers tab - is an interactive and printable web map with click on/off scale-dependent map layers, pan/zoom functionality, map legend, and access to a 24-map basemap gallery Base Layers - RIDEM Forest Management Areas (2020); Upland Forest Habitat (2010) Restricted Harvest (Flag) Layers - where cutting intensity is limited. 2015 Trails buffered by 50 ft; 1997 Streams buffered by 50 ft Visual Reference Layers - 2019 RI State Natural Heritage - rare, threatened, and endangered species proximity of occurrence areas; 2019 Young Forest Habitat - by height class Non-Operable (non-harvestable) Layers - RIDEM Priority Forest - delineated important forest areas, managed for old growth, Elevation (2011) - from 2011 LiDAR (2018 LiDAR was not yet publicly available at project conclusion) raster layer containing both harvestable, terrain and non-harvestable areas, where slope is considered excessive; 2014 Wetlands - buffered by 25 ft; 2018 Roads - Rhode Island E-9-1-1 program roads buffered by 50 ft. Goal 1a. and c. Dashboard: Operable (harvestable) Lands tab - is web map-centric with a legend, pan/zoom, and click on/off, and scale-dependent layer functionality. The web map is surrounded by a dashboard of metrics tabulated by forest management area (chosen in pull-down menu). Operable areas are colored dark green, non-operable areas appear in dark gray. Dashboard metrics include acres of operable (harvestable) forest, restricted harvest acres flagged in buffered proximity to streams and trails. Also included are bar graphs summarizing the operable forest and upland forest habitat acreage by management area. Dashboard Non-Operable (non-harvestable) Lands tab - is web map-centric with a legend, pan/zoom, and click on/off, and scale-dependent layer functionality. The web map is surrounded by a dashboard of metrics tabulated by forest management area (chosen in pull-down menu). Operable areas are colored dark green, non-operable areas appear in dark gray. Dashboard metrics include acres of non-operable (non-harvestable) forest, according to presence of roads, excessive slope, wetlands, and/or priority forest areas. Goal 2b. App: Create Report tab - allows the user to generate a printable report by management area (selectable by pulldown), or by user defined area (interactive) which includes a map with selectable layers. The report contains a summary of total acres of management area, count and acres of operable forest polygons, count and acres of overlap with potential natural heritage occurrence areas. Also included are the individual polygon counts with management area, forest type, presence/absence of restricted harvest stream or trail flag buffers, and area summaries. Team tab - List of URI and RI DEM Partners Goal 1a., b., and d. The back-end geospatial analytical functionality that combines geospatial inputs resulting in visual, mapped and numerical, tabular summaries for harvestable, non-harvestable, and restricted harvest areas was developed using the Esri Model Builder geoprocessing module within ArcGIS Pro Desktop software.Model Builder allowed us to capture our analytical inputs and operational relationships in a visual, well-documented diagram-based, object-oriented, workflow.The model takes input geospatial data layers and carries out approximately 80 geoprocessing operations and generates 40 intermediate data layers. This object-based development environment allows RI DEM to easily follow along with processes and workflows, update the application when new input data become available, change parameters to incorporate new criteria or modify the model to keep pace with changes in regulatory parameters.We worked directly with the GIS Specialist at RI DEM to transfer the final package files between our respective organizational online GIS portals and to get the application running on the internal RI DEM system.We continue to support RI DEM as they have questions on modifying the interface, changing input parameters, or need assistance incorporating updated input layers as they become available. Goal 3 During the FY20 project period we continued to discuss ways in which we could collaborate with RI DEM staff on using sUAS technology to augment forest management pre/post-harvest planning and assessments as well as forest health monitoring.We are in discussions with the RI DEM Forest Health Program to deploy a multi-spectral sensor to detect potential onset of infestations of the southern pine beetle in a large nature preserve dominated by pitch pine managed by The Nature Conservancy. The pandemic situation created logistical challenges associated with re-opening health and safety guidelines, maintaining physical distancing, and the ability to carry out in-person cooperative extension programming activities. As a result of current restrictions and due to staffing and shifting of schedules at RI DEM we were unable to arrange for these activities during FY20 but remain hopeful that we can collaborate in FY21 through our URI Cooperative Geospatial Extension Program.

Publications

  • Type: Websites Status: Published Year Published: 2020 Citation: RI State-owned Lands Operable Forest Viewer. 2020. University of Rhode Island Geospatial Extension Program and Rhode Island Department of Environmental Management. Forestry management pre-harvest planning decision support tool using the ArcGIS Online Story Map Series template, Basic Map Viewer, Web AppBuilder, and ArcGIS Pro ModelBuilder geoprocessing module.


Progress 10/17/18 to 09/30/19

Outputs
Target Audience:Forestry and wildlife managers with the Rhode Island Department of Environmental Management, Division of Forest Environment. These include forest managers and ecologists specializing in general forestry practices, fire ecology, urban, disease, pest management, and forest health. Wildlife managers and ecologists specializing in amphibian, fish, bird, small and large mammal habitats. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?As we work towards refining the input data and finalizing development of this application we intend to hold a training exercise for RIDEM foresters in order to demonstrate the functionality of the system. How have the results been disseminated to communities of interest?We are communicating our incremental progress in development of this web-based application by providing direct internet access to our partners. They are able to access the results of any changes we commit to the application in real-time. We are communicating progress and results through live demonstrations at our meetings. What do you plan to do during the next reporting period to accomplish the goals?1) When ready, we are planning to incorporate the results for the adapted RIDEM land ownership boundary into the current version of the Operable Forestlands application. 2) In consultation with RIDEM, we will be determining which elements of the application will remain publicly-accessible and which will be restricted to RIDEM internal guidance access only based on data sensitivity (e.g., if locations of rare, threatened, or endangered species are incorporated). 3) Develop a more refined Slope Layer from the most recent and complete Statewide LiDAR-derived Digital Elevation Model 4) Compare 1:5,000 scale Streams Layer with National Hydrography data to see if these data would provide enhanced detail and value-added benefit to the application. 5) Develop web maps, journals, and story maps to enhance communication of results to RIDEM forest managers 6) Conduct a training exercise to demonstrate functionality of the application. 7) Using application results, publish tabular reports and summaries of harvest potential by management area. 8) Consult with RIDEM foresters to determine where to apply UAS image data collection to enhance pre/post-harvest and forest health operations.

Impacts
What was accomplished under these goals? We have been collaborating with forest resource managers at the RI Department of Environmental Management's (RIDEM) Division of Forest Environment to develop an operable forestlands web mapping decision support system. We are focusing on selecting geospatial data inputs, criteria and parameters for determining harvestable and non-harvestable areas within their designated management areas. We are integrating these data and assembling them into a web mapping application. We held meetings in Fall 2018 and Spring 2019 at RIDEM Division Forest Environment and at the University of Rhode Island during the Summer of 2019. We've agreed to move forward with the development of an online operable forestlands mapping application that includes the following elements: 1) Adoption and adaptation of elements from the NH Good Forestry parameters in developing criteria for the RI Operable Forestlands Mapping application within RIDEM managed areas. 2) Decision to adapt the DEM land ownership boundary layer and attributes to be compatible with the geospatial boundaries and database schema of other layers in the application. 3) Selection of the esri ArcGIS Enterprise and ArcGIS Online development platforms. Elements of the Operable Forestlands application can be developed and hosted on both the URI (publicly-accessible content) and RIDEM (public and sensitive data, password protected) ArcGIS Online Organizations. 4) Based on the system capabilities and limitations we have discussed, RIDEM Forestry will consider/propose where to apply UAS image data collection for pre/post-harvest and forest health applications. We have completed the development of the first stage prototype of an enterprise GIS web mapping application that is being used as the platform to test and refine the system as well as to field comments and suggestions. We have included the following initial data layers: •Forest Management Areas: State Conservation Areas 2018: DEM will update these for consistent naming among property holdings and supervisory division. Some additional properties are also being added. •Forest Habitat Layer: We are using the URI-developed 2010 Forest Habitat Layer with wetland/upland, deciduous/coniferous, forest/shrubland classes. •Young Forest Layer: Developed by URI. Determination being made if this can be or is appropriate to be used in this application. •Natural Heritage Areas (generalized): Used the point to flag area (generalized to area) to the presence of a Natural Heritage occurrence or habitat. •Slope: Currently obtaining slope characteristics from Soils Layer (>35%). Plans are to use LiDAR elevation data to generate slope for DEM Managed areas. Exclude areas >20% slope. Make determination on appropriate patch size for analysis/output. Potentially include Forest Type patch size used by URI. •Wetlands: Using National Wetlands Inventory data layer. •Soils: Drainage classes including hydric soils and poorly drained/very poorly drained categories •Streams: Using 1:5,000 scale stream layer, coded with stream order. Use 5K stream features code with IWQMA stream order also attribute with width < 10 ft. or >10 ft. Also check against new National Hydrography data. Evaluate a potential enhancement - explore the creation an editing app that could be used to add attributes of stream width. •Important Forest Layer: DEM has this layer in rough draft format and will provide refinement which will include designated important non-harvest areas. •Roads: From RIDOT 2016 (considering E911 Roads as an alternative). •Trails: Include as viewable layer on final maps. We are in the process of working cooperatively with the RIDEM Forest Management Team on getting feedback on the input layers, criteria, and parameters. Whether or not to include distance buffers, and by what distance, around specific layers is under consideration. RI is in the process of purchasing a new statewide LiDAR elevation data layer (Digital Elevation Model) collected in 2018. We are exploring the public usage of these data. RIDEM is finalizing the property boundaries and attributes of their forest management areas. URI is keeping pace with the latest enterprise and online GIS software developments and committing any new software updates to be used as the basis of supporting this application.

Publications

  • Type: Websites Status: Under Review Year Published: 2019 Citation: https://arcg.is/1XXP4K