Forests across the eastern United States provide essential benefits, including timber, clean water, wildlife habitat and carbon storage. However, these systems face increasing pressures from changing conditions, market uncertainty, evolving land-use patterns and urbanization demands. Addressing these challenges requires better information, improved tools and stronger connections between science and real-world decision making. The Promoting Economic Resilience and Sustainability of the Eastern U.S. Forests (PERSEUS) project continues to make progress toward meeting these needs.

During this third year of the project, PERSEUS advanced the development of digital tools and data systems that improve how forests are measured, monitored and understood. New approaches combining field measurements, airborne and satellite data, and advanced modeling are helping create more detailed and accurate representations of forest conditions across large areas. These efforts are

supporting the development of “wall-to-wall” mapping products and integrated data platforms that allow users to explore forest information in more accessible and meaningful ways.

At the same time, the project made progress in modeling how forests may change in the future. By combining forest growth models, landowner decision-making models, stakeholder-informed scenarios and AI-enhanced models, PERSEUS is building a framework to explore how forests may respond to different environmental, economic and policy conditions. These tools help illustrate potential outcomes and trade-offs, supporting more informed planning and management decisions.

Engagement with stakeholders remains a central component of the project. This year included expanded outreach through workshops, conferences and direct collaboration with landowners, industry and government partners. Feedback from these groups is actively shaping the design of tools, models and data products to ensure they are relevant, practical and easy to use. Efforts included continued collaboration with policymakers and national organizations to support broader conversations about the future of forest inventory and management.

PERSEUS also continued to invest in education and workforce development. Students, interns and early-career researchers participated in hands-on training, collaborative research and professional development opportunities across institutions. New and updated courses, workshops and online learning modules are helping prepare a workforce with the skills needed to apply digital technologies in forestry.

Together, these efforts are building a stronger foundation for modern, data-driven forest management. By integrating advanced technologies with stakeholder needs and practical applications, PERSEUS is helping to improve the sustainability, resilience and economic value of forests across the eastern U.S. while supporting the people and communities that depend on them.

PERSEUS will provide scientifically sound information, outreach and educational opportunities that lay the foundation for a paradigm shift in forestry toward data-driven, artificial intelligence (AI)-supported forest management systems that increase both the provision of ecosystem services and operational efficiency. PERSEUS will develop measurement and monitoring tools, data-driven decision systems and management practices through a unified effort of automated measurement, integrated multi-objective modeling, engaged data-driven management and a digitally competent mindset in students and professionals.

• Objective 1 Measurement Automation: 
  • Develop scalable automated data capture systems with integration of multimodal,
    multiplatform and multitemporal data to assist stakeholder data-driven forest management practices and decisionmaking.
• Objective 2 Model Integration and Application:
  
  • Create a multimodel ensemble that is locally calibrated and capable of projecting forest ecosystem services under a range of conditions.
• Objective 3 Informed and Engaged Mindset:
  • Encompass stakeholder actions, motivations and values to inform the development and implementation of optimized data-driven management systems that increase the capacity of forest production and ecosystem services.
• Objective 4 Digitally Competent Mindset:
  • Contribute to measurable national impacts through the development of educational programs for digitally competent and capable practitioners, managers and researchers who can pipeline into successful academic, industry or governmental careers.

The PERSEUS project is organized with a leadership team consisting of the project director (Purdue University [Purdue]) and two co-directors (University of Georgia [UGA] and University of Maine [UMaine]). The PERSEUS project contains four broad objectives, and each objective has an objective lead per university to ensure project goals are met.

The University of Southern Maine’s Data Innovation Project (DIP) conducts the PERSEUS external evaluation. The evaluation is guided by an evaluation-specific logic model and utilizes mixed methods to collect data to support formative and summative outcomes related to four domains: Implementation and Sustainability of the Project; Collaboration, Coordination and Engagement; Research Capacity and Productivity; and Education and Workforce Development. In Year 3, evaluation-related data collection took place via meeting feedback surveys (i.e., May 2025 Annual Retreat, December 2025 All-Hands Virtual Meeting), a ripple-effect mapping exercise (during Annual Retreat), and a partnership self-assessment deployed to PERSEUS leadership and objective leads.

During the 2025 in-person retreat, the evaluation team led a ripple-effect mapping exercise with approximately 70 participants. This exercise collected information on the team members’ perspectives on the key highlights, achievements and successes they have experienced through PERSEUS to date. The top cross-objective highlights included collaboration and teamwork, networking, communication, beneficiary engagement, and tool/product creation.

The partnership self-assessment measured key indicators of successful collaboration and provided information on opportunities to maximize collaborative potential. The domains of the assessment included synergy, leadership, efficiency, use of resources, administration/management, decision-making, satisfaction and benefits/drawbacks to participation. Strengths of the project included the project’s plans for achieving goals and implementing strategies that are likely to work for the whole project; leadership’s ability to be inclusive and create a welcoming work environment; well-organized activities, including meetings and projects; and decision-making processes. Areas for improvement going forward included making connections to key stakeholders and coordinating communication with people and organizations outside of the three universities. PERSEUS team members continue to have a clear understanding of the project’s goals, their role on the project, their expected outcomes and how their contribution fits into the project.

In Year 4, the evaluation team will collect more data related to all four evaluation domains and continue to participate in the monthly Sustainable Agricultural Systems (SAS) evaluation Community of Practice meeting led by the University of Wisconsin-Madison, Division of Extension

AUTOMATED (RESEARCH)

Develop, evaluate and integrate sensors, multistream data and AI algorithms to create a set of digital tools and wall-to-wall coverage data that will provide refined, near real-time, and spatially explicit measurements. Novel data acquisition systems and data analytics will be developed to better measure and monitor timber and fiber production, greenhouse gas mitigation (carbon sequestration, wildfire fuel assessment) and other ecosystem services for every acre across the eastern United States.

Task 1.1 iForester (months 1-36): Develop an integrated AI-assisted iForester (measurement tool —ground truth/inventory measurements) smartphone application that automatically inventories
major tree species for the Eastern U.S. Forest with two key functions: (1) AI-assisted species
recognition using tree bark and (2) Light Detection and Ranging (LiDAR)-Red Green Blue (RGB)
image-enabled measurements of key tree biometric measures.

Task 1.2 StemMapper (months 1-36): Create a LiDAR-based and AI-assisted StemMapper (proximal and near proximal remote sensing-derived models and products) to provide stem-level inventory at the stand and tract level with large-scale automated inventory for multi-objective, data-driven management for forestry professionals.

Task 1.3 Data Coverage (months 1-54): Generate multiscale data products (e.g., fiber, habitat, carbon) with higher spatial and temporal resolution for every acre across the eastern United States to facilitate both local- and regional-level management optimization.

Deliverables Year 3

• Smartphone App
• Hardware Integration
• Data Analytics
• Data Fusion
• Scaling

Accomplishments in Year 3

Task 1.1 iForester: Year 3 saw an additional 365 downloads of the application. Development focused on improving smartphone-based measurement algorithms for diameter at breast height (DBH) and tree height. Researchers advanced a depth- and AI-based DBH estimation approach, combining smartphone depth sensing with AI models to improve measurement accuracy and robustness in field conditions. Code improvements and debugging were completed for the DBH measurement tool, and the application is now considered stable for this methodological approach. In parallel, smartphone-based tree height estimation method was developed, which uses short vertical video scans and photogrammetry techniques to derive height measurements from mobile devices. Additional work focused on expanding the overall mobile application framework that will ultimately support iForester. Development of a tree grading application interface also progressed during this period, which will allow users to grade their standing or downed logs and estimate their economic value.

Task 1.2 StemMapper: Multiple LiDAR data acquisition platforms, including backpack-based systems, unmanned platforms and crewed airborne sensors, were tested and refined during the reporting period. The backpack LiDAR system, equipped with cameras, GPS/INS navigation and onboard data logging systems, was deployed for field data collection and continues to
be improved through efforts focused on increasing image acquisition frequency, trajectory accuracy and real-time mapping performance. Researchers are actively developing improved Simultaneous Localization and Mapping (SLAM)-based trajectory enhancement and path-planning methods to address positioning challenges associated with GPS signal loss under forest canopy.

Additional work focused on improving sensor calibration and integration between LiDAR and imaging systems, which is essential for producing accurate 3D forest structure data. Development of calibration workflows and LiDAR image alignment methods continued. In parallel, testing of a crewed airborne LiDAR platform equipped with high-resolution RGB and hyperspectral sensors began following system maintenance and payload integration.

Task 1.3 Data Coverage: A major milestone was the development and publication of a statewide biomass and forest carbon dataset for Maine at 20-meter resolution, along with additional LiDAR-calibrated biomass datasets derived from the Howland Forest experimental site. These datasets integrate multiple LiDAR acquisitions and field measurements to generate improved biomass and carbon estimates and have been made publicly available.

Backpack and Unmanned Aircraft System (UAS) LiDAR scans collected by Purdue’s systems were merged with UMaine’s airborne RIEGL LiDAR acquisition from 2025 and analyzed with Purdue’s processing pipeline. Together, these datasets form a unique multiscale benchmark for evaluating tree segmentation and biomass modeling approaches.

Plans for the Coming Year, April 2026-March 2027

Deliverables Year 4

• Smartphone App
• Hardware Integration
• Data Analytics
• Data Fusion
• Scaling

Task 1.1 iForester: Focus will be on integrating existing prototype applications into a unified smartphone-based forestry measurement system. Current standalone tools for DBH, height estimation and grading will be consolidated into a single application framework. Additional testing and validation will be conducted to improve measurement accuracy across a wider range of forest conditions, including refinement of diameter-at-height measurements and photogrammetry-based height estimation methods. Researchers will also continue evaluating AI-based segmentation approaches and measurement algorithms to enhance robustness and processing efficiency in field conditions. Feedback on performance and user experience in dense, mixed-species conditions will inform ongoing refinements and support progress toward improved iForester application releases, with select components potentially deployed to collaborators and project partners for pilot testing following internal validation.

Task 1.2 StemMapper: Planned efforts include continued improvement of backpack-based LiDAR sensing systems, such as enhanced image acquisition frequency, improved SLAM-based trajectory estimation, and refined calibration between LiDAR and imaging sensors. Researchers will also advance image methods and super-resolution approaches to increase the effective density of point clouds collected from unmanned platforms.

Additional airborne LiDAR system testing and payload integration are planned, alongside expanded field deployments to support model development and validation. Ongoing work will apply advanced geolocation calibration methodologies across additional plots and integrate airborne LiDAR capacity, including emerging sensor technologies, with backpack and UAS data. This integration will enable alignment of proximal and remote sensing products across measurement scales and support the creation of high-quality, multiplatform datasets needed to train AI models for automated tree detection and stem-level inventory generation.

Task 1.3 Data Coverage: Researchers will expand multiscale data collection, integration and evaluation efforts to support the development of high-resolution forest structural datasets. Planned activities include additional airborne LiDAR acquisitions across research sites, including both leaf-on and leaf-off conditions, alongside continued terrestrial laser scanning (TLS) and backpack LiDAR data collection. These efforts will support cross-platform dataset integration and the development of unified datasets that link TLS, airborne LiDAR and field inventory measurements for improved tree segmentation, biomass estimation and structural analysis.

Further work will focus on scaling plot-level measurements to broader spatial datasets by integrating airborne and satellite-based LiDAR systems, including Global Ecosystem Dynamics Investigation (GEDI), as well as incorporating additional data sources such as 3DEP, National Agriculture Imagery Program (NAIP)/Vexcel imagery and Sentinel-2. Data processing pipelines will be refined to support scalable implementation across larger geographic regions, with progress toward operational deployment of wall-to-wall mapping products and public data hosting. Multiple full inventory seasons will be completed and archived to support ongoing model calibration and validation. Advancements in geolocation calibration methods will be finalized and disseminated, while new sensor technologies
and field tools will be deployed operationally and integrated into standard data collection protocols.

In parallel, efforts will include evaluation of existing large-scale forest datasets to assess their suitability for regional and stand-level applications. An assessment of TreeMap 2016 and 2022 using independent Forest Inventory and Analysis (FIA) field plot data examines both statistical accuracy and spatial predictive performance, including bias, root mean square error (RMSE), concordance correlation coefficient, variance retention and scale-dependent accuracy. Preliminary findings suggest that while TreeMap may be appropriate for regional biomass estimation, limitations remain for fine-scale spatial mapping of forest structure.

Integrated Multi-Objective (Research)

Leverage Objective 1 to construct and apply an integrated framework for modeling current and future forest ecosystem service trends due to various changes such as insect outbreaks and extreme weather for multiobjective optimization at the landowner scale, while providing multistakeholder simulations and tradeoff analyses of forest management at the regional scale. Develop a generalized simulation/optimization framework to inform regionally appropriate data-driven resilience solutions.

Task 2.1 Landowner Optimization (months 13-48): Link available forest data (from PERSEUS and existing datasets such as FIA) to the integrated multimodel ensemble to optimize ecosystem services at a local scale (1–1,000,000 ha).
Task 2.2 Broad Simulation (months 13-48): Co-develop (with stakeholders) a dynamic simulation system to present broadscale (>1M ha) assessments of alternative policy and market scenarios, while facilitating fine-scale assessments of tradeoffs among management and ecosystem services for
regional decision-making.
Task 2.3 Value Chain (months 25-60): Develop and refine methods to investigate potential efficiencies through forest management activities in the eastern United States.
Task 2.4 Data Visualization (months 1-60): Develop a cloud-based data warehouse to allow key project data to be stored, visualized and shared with stakeholders.

Deliverables Year 3

• Optimization Model Ensemble
• Simulation Model Ensemble
• Efficiency Simulations
• Geospatial Information System

Accomplishments in Year 3

Task 2.1 Landowner Optimization: Analysis of long-term forest inventory data was conducted to examine relationships among landownership, climate, harvesting strategies and tree species diversity. This work links PERSEUS data with existing forest inventory datasets and provides a
basis for understanding how management decisions influence forest conditions across ownership types. Building on this foundation, a harvest choice model was developed using a multinomial logit framework to estimate landowner decision-making behavior as a function of stand characteristics, ownership and economic factors. This modeling approach enables simulation of management responses under alternative policy and market conditions and represents a key component of the broader optimization framework. These efforts mark the initiation of the optimization model ensemble and establish critical inputs for subsequent integration with scenario development and simulation modeling.

Another project is the exploration of determining forest health using ground-based and aerial-based data collection. One subproject collected canopy foliar data, leaf hyperspectral measurements, chemical analyses and herbivory quantification. This included the development of an open-source Python tool and webpage interface for high-throughput spectroscopic analysis. Another subproject developed super-resolution imaging models for multispectral and hyperspectral analysis from UAS and aircraft imagery. Both subprojects have the downstream tasks of disease classification and evaluating forest health at local to regional scale.

Task 2.2 Broad Simulation: Core models, including the Forest Vegetation Simulator (FVS), LANDIS-II, Woodstock Optimization Studio and the harvest choice model, were calibrated and evaluated using FIA data across thousands of plots. Comparisons of merchantable timber estimates across inventory cycles and projected conditions supported validation and stakeholder engagement. Nonspatial simulation capabilities advanced through development of a state-level modeling framework based on FIA data to evaluate management alternatives. An online interface is being developed to allow users to explore outputs from predefined scenarios, with ongoing efforts to expand functionality toward user-defined scenario analysis.

Spatial simulation efforts progressed through enhancements to LANDIS-II and development of complementary modeling approaches. Work included inverse parameterization using FIA data, refinement of biomass succession processes, and improvements to computational performance to support multi-scenario analyses. Parallel efforts are advancing an empirical landscape modeling framework that integrates inventory and management datasets to simulate landscape-scale outcomes under alternative conditions.

Supporting model development included compilation of species-specific parameters, processing of repeated-measure FIA datasets, and initial model implementation. Scenario development progressed in parallel, providing a structured framework for integrating climate, policy, and market drivers into simulation analyses.

Task 2.3 Value Chain: Year 3 saw the initiation of the economic and operational modeling foundations needed to evaluate efficiency across the forest products supply chain. Efforts centered on linking forest management decisions to downstream impacts on timber supply and product flows. A harvest choice model was developed to estimate landowner decision making as a function of economic, spatial and ownership factors, providing a basis for understanding how management behavior influences value chain dynamics. These efforts were complemented by analysis of management intentions survey data, which characterized silvicultural approaches across ownership classes and forest types. This information supports translation of management decisions into timber supply outcomes, forming a critical input to value chain modeling and efficiency analysis. Additionally, development of a transportation-focused modeling framework started to assess efficiencies in the forest products supply chain. This includes evaluating operational and regulatory factors influencing transportation costs, capacity, and sustainability, as well as examining safety considerations within timber transportation systems. The resulting optimization model will integrate forest inventory data, mill demand, and transportation networks to inform movement of wood from forests to mills across multiple regions.

Year 3 also saw the development of a natural-language interface to assist with developing hypothetical scenarios. The approach has an automated pipeline combining large language models (LLMs, such as ChatGPT, Gemini, etc.) with social, economic and geometric data, and optimization methods. Collectively, the platform enables selecting a particular plot of land and then asking for modifications yielding a particular objective. The LLM and optimization methods inspect all other plots available and determine the modification to the target plot that most likely accomplishes the design goal. Initial development focused on urban plots (having green and gray infrastructure), but extending to large forest plots is planned. This effort builds off the project’s ability to determine urban gray and green infrastructure as well as simulation weather/climate affect of altering tree canopy at scale.

Task 2.4 Data Visualization: Advanced the geospatial information systems and data infrastructure needed to support integration, analysis and dissemination of PERSEUS data. Continued development of the PSAE/Areal platform established it as the primary cloud-based environment for storing, visualizing, and sharing project data. The platform integrates FIA plot data, airborne LiDAR, and high-resolution imagery, providing a centralized system for geospatial analysis and stakeholder access. Visualization development was further informed by stakeholder-driven insights. Findings from the Delphi assessment of digital technology needs identified key categories of decision-support
requirements and tool characteristics, which are being incorporated into the design of visualization outputs and future online platforms.

Additional efforts demonstrated the application of visualization tools for comparing outputs across modeling frameworks, supporting interpretation and communication of multimodel results.
Contributions to national-level discussions on forest inventory modernization further emphasized the importance of scalable data infrastructure and visualization capabilities. Supporting data systems, including spatio-temporal data repositories, continue to be maintained to ensure accessibility and interoperability of large geospatial datasets.

plans for the coming year, april 2026-march 2027

deliverables year 4

• Optimization Model Ensemble
• Simulation Model Ensemble
• Efficiency Simulations
• Geospatial Information System
• Online Optimization Platform

Task 2.1 Landowner Optimization: Continue working on advancing landowner optimization models and integrating them with scenario-based analyses. Building on recently published work examining relationships among landownership, management, and forest conditions, further development will transition the harvest choice model to a fully operational state. The optimization model ensemble will be refined to incorporate scenario framework outputs, enabling evaluation of landowner decision making under alternative future conditions. These enhancements will support multi-objective optimization at the landowner scale, providing a foundation for integrating economic, ecological and management considerations within the broader modeling framework.

Task 2.2 Broad Simulation: The multimodel comparison will be extended to include additional growth modeling frameworks and validated using updated FIA datasets to improve robustness and comparability of outputs. Advancements in model performance will enable expansion of simulation domains to broader geographic extents, supporting regional analyses across the eastern U.S. The established scenario framework will be coupled with the simulation ensemble to project forest outcomes under varying climate, policy, and market conditions through 2055. Integration of landowner decision-making models will further enhance simulation capabilities.

Task 2.3 Value Chain: The harvest choice model and management intentions survey data will be used to evaluate potential efficiencies associated with forest management activities across ownerships and regions. Timber supply projections generated from the simulation model ensemble will be linked with stumpage prices, transportation costs, and mill capacity data to assess value chain dynamics at broader spatial scales. These analyses will support evaluation of how management decisions and market conditions influence the flow of forest products and overall system efficiency. The natural-language interface to assist with developing hypothetical scenarios will be expanded to include urban, peri-urban, and large forest plots.

Task 2.4 Data Visualization: The PSAE/Areal platform will continue to serve as the primary environment for integrating, visualizing and sharing project data. New visualization modules will be developed to support presentation of multimodel comparison results, scenario-based projections and wall-to-wall mapping products. Supporting data infrastructure will be updated to incorporate new outputs from modeling and simulation efforts, ensuring accessibility and interoperability of datasets. Visualization design will be guided by stakeholder-informed insights, including identified preferences and decision-support needs, to improve usability and relevance.

Engaged, CLIMATE SMART (Extension)

Engage stakeholders to develop data-driven management practices that can improve the sustainability and resilience of forest ecosystems in the eastern United States, based on outcomes from Objectives 1 and 2.  Use-inspired, co-production model of research and Extension will be deployed to facilitate both (a) the successful development of the simulation/optimization system and (b) the actual adoption of data-driven management practices by stakeholders to build environmentally and economically sustainable forests, especially on private lands in rural states.

Task 3.1 Stakeholder Perceptions (months 1-24): Develop and conduct a survey of foresters and forest landowners concerning current forest practices and environmental challenges, and their interests in ecosystem services and alternative future scenarios.

Task 3.2 Scenario Development (months 1-37): Co-develop (with stakeholders) “what-if” scenarios to examine current and future risk perceptions from future weather patterns, available markets, or
pests and the degree of support for forest management strategies.

Task 3.3 Focused Outreach (months 25-60): Assess end-user satisfaction through staggered, in-person or virtual participatory workshops with stakeholders across subregions to assess project outcomes and the general usability of the developed tools.

Task 3.4 Technology Application (months 25-60): Conduct training sessions for stakeholders on tool and system use.

Deliverables Year 3

• Identify Stakeholder Needs
• Conduct What-If Scenarios
• Remote Training Events

Accomplishments in Year 3

Task 3.1 Stakeholder Perceptions: A large-scale survey of forest landowners was developed and implemented across all three regions. The survey was administered to 7,500 landowners, and 1,676 usable questionnaires were returned (effective response rate: 24%) and used for subsequent analysis. An analysis was completed to assess landowners’ perceptions of emerging digital forestry tools, their general willingness to adopt or allow the use of these technologies on their property, and the key factors influencing adoption decisions. An additional landowner cluster analysis has also been completed to identify landowner typologies across all three regions.

Complementing the large-scale forest landowner survey across the three regions, a second survey focusing on large landowners was also administered and completed to evaluate and compare how landowners in Maine, with more than 1,000 acres of forestland, view and utilize digital technologies to manage their forests. The survey was administered to 147 landowners, and 37 usable questionnaires were returned (response rate: 25.2%).

A forest business survey was completed for the southeastern U.S. Initially, 4,286 questionnaires were sent by mail to forest business contacts in Georgia. Despite the lower-than-expected response rate, the 82 responses represent a wide range of forest businesses in the southeastern states, including primary (59%) and secondary wood-using firms (41%). A majority of them were primary wood-processors (i.e., sawmills, comprising 40%).

In addition to these stakeholder surveys, two systematic literature review projects were completed. The first is a systematic literature review of forest ownership in North America and the Nordic countries, identifying a diversity of objectives, values and behaviors among landowners. The results were used to synthesize forest ownership typologies that were then assessed as tools for engaging forest owners in adaptive forest management for increased resilience against changing environmental conditions. The second project was a systematic literature review to examine engagement of forest products firms with ecosystem services. The review identified several motivational factors associated with forest companies’ engagement with ecosystem services and evaluated corporate strategy of addressing environmental sustainability to manage risks and compliance, especially guiding firm-level decision making to safeguard ecosystem services.

Task 3.2 Scenario Development: Scenario development focused on establishing a strong foundation for stakeholder-informed modeling and decision support. A comprehensive review of the literature on risk perceptions and stated preference elicitation techniques was completed to guide the design of survey instruments and ensure robust capture of landowner decision-making factors. Ongoing collaboration with modeling efforts supported iterative review of tools under development, including feedback on form, function and usability. Input was also provided to help refine scenario structures to better reflect stakeholder needs and real-world applications. Building on these efforts, initial “what-if” scenarios were developed using landowner survey results across regions. These early scenarios provide a baseline for exploring management and policy alternatives.

Task 3.3 Focused Outreach: Feedback on outreach materials was actively solicited through more than 10 meetings with both project participants and external stakeholders, informing the development and iterative refinement of an Extension toolkit. This toolkit is now being promoted for internal use to support consistent and effective communication of PERSEUS tools and outcomes.

Outreach activities included the development of a coordinated outreach plan and continued dissemination of materials through strategic venues such as landowner conferences, professional workshops, training events and major industry gatherings. Engagement with the broader forestry community was further supported through presentations and exhibitions.

Efforts also extended to high-level stakeholder and policy engagement. Meetings with congressional staff and coordination of federal appropriations strategies advanced visibility. Engagement with USDA Forest Service leadership and participation in national advisory efforts contributed to discussions on modernizing forest inventory infrastructure. Industry partnerships were strengthened through continued dialogue with key stakeholders, resulting in positive feedback and interest in ongoing collaboration.

Additional outreach included organizing and hosting a national conference focused on AI applications in forestry at UGA, which brought together researchers, practitioners and industry partners for presentations, demonstrations and participatory discussions on emerging tools. Presentations at regional and national meetings further expanded reach within practitioner and industry communities. These combined efforts have positioned the project to continue expanding partnerships, outreach impact and adoption of digital forestry tools into Year 4.

Task 3.4 Technology Application: A major milestone was the AI Technology for Forestry Conference, which served as a large-scale training event featuring hands-on demonstrations of PERSEUS tools and providing continuing forestry education credits to participants. In addition, the PERSEUS annual meeting included pre-meeting workshops on UAS operations, plot measurement techniques, drone demonstrations and training on the Data to Science (D2S) platform, directly engaging users with project-developed measurement and modeling technologies.

Efforts also supported pathways for operational deployment. Technical consultation was provided to state-level partners to define requirements for hosting wall-to-wall forest attribute mapping products, representing a concrete step toward broader agency adoption of PERSEUS-derived technologies. Coordination across institutions was further supported through shared training materials and tool overviews to align technology transfer efforts.

Work is ongoing to formalize and expand training opportunities, including development of a structured training plan, pilot testing with key stakeholders and dissemination through strategic venues. Feedback is actively being collected to refine training approaches and improve usability.

PLANS FOR THE COMING YEAR, APRIL 2026—MARCH 2027

Deliverables Year 4

• Identify Stakeholder Needs
• Conduct What-If Scenarios
• Small-Group Workshops
• Remote Training Events

Task 3.1 Stakeholder Perceptions: Efforts will focus on advancing analysis, expanding data collection and deepening understanding of stakeholder perceptions to better inform outreach and technology adoption. Key manuscripts, including the large landholder technology survey and systematic literature review, will be completed and advanced through publication. Findings from these efforts, along with prior published work, will be synthesized to provide a comprehensive characterization of stakeholder perceptions and decision-making factors, directly informing the design of outreach and training activities.

Comparative analyses will examine differences in perceptions and adoption intentions across landownership categories, particularly between smaller and larger ownership groups. Further research will also investigate the role of trust in both active and passive technology use.

A new integrative effort will be initiated to strengthen connections between research and real-world applications. Specifically, this will be carried out through a “walk-the-land” project that brings together measurement, modeling, stakeholder perception and outreach components to demonstrate forestry technologies directly on landowner properties, while collecting real-time user feedback that will be used to inform improvement of the existing technologies, as well as the
development of new technologies. Continued coordination across objectives will also support cross-disciplinary integration and ensure that stakeholder insights are effectively incorporated into tool development, outreach strategies and training programs.

Task 3.2 Scenario Development: Scenario development efforts will focus on advancing the scenario framework, strengthening integration with modeling tools, and refining outputs based on stakeholder input. The scenario framework manuscript will be completed and submitted, formalizing the conceptual and methodological foundation for scenario-based analysis. Ongoing collaboration with modeling efforts will support the translation of scenario narratives into quantitative indicators that can be incorporated into Objective 2 models. This includes continued iterative work to refine model inputs, identify meaningful metrics and improve the structure, functionality and visualization of scenario outputs. Resulting projections will quantify potential forest futures under each of the defined scenarios.

Further analysis of management intentions survey data will be conducted to better characterize how landowners may respond under different scenario conditions. The scenario framework will also be refined to emphasize key geographic considerations and STEEP (social, technological, environmental, economic and policy) factors most relevant to stakeholders. Cross-regional validation of scenarios will be conducted in collaboration with project partners, and results will be shared through stakeholder workshops to gather feedback on scenario plausibility and relevance to practitioner decision making. These efforts will ensure that scenarios remain grounded in both empirical data and real-world application needs.

Task 3.3 Focused Outreach: Focused outreach efforts will emphasize continued and expanded stakeholder engagement, strategic communication and broader dissemination of PERSEUS-related information.

Hosting the Applications & Solutions in Digital Forestry International Conference (June 3-5, 2026) will serve as a major focused outreach event. Outreach activities will also include presentations at professional venues and expanded engagement through a multistate digital forestry coalition. A webinar series and regional workshops will be used to share results, increase awareness of tools and engage diverse audiences. These efforts will include dissemination of scenario development findings and initial scenarios to key stakeholders, supporting dialogue around relevance and application.

Targeted outreach materials will be developed and refined based on stakeholder insights, including findings from prior survey efforts. Extension resources will be expanded to include materials tailored to varying levels of expertise, from introductory to advanced users. Feedback from workshops and outreach events will be actively collected and analyzed to inform continuous improvement, with refinement of outreach strategies and planning for final-year activities continuing throughout the reporting period.

Additional activities include continuing engagement with USDA Forest Service leadership on forest inventory modernization; presenting PERSEUS results at the Cooperative Forestry Research Unit, Society of American Foresters and other professional venues; advancing the multistate digital forestry coalition through engagement with Idaho and Arizona State University partners; and developing targeted outreach materials informed by Delphi and landholder survey findings.

Task 3.4 Technology Application: Efforts will focus on available Objective 1 and Objective 2 products, as well as expanding training opportunities, supporting operational deployment and ensuring consistent delivery of PERSEUS tools across stakeholder groups. The Applications & Solutions in Digital Forestry International Conference will serve as a key platform for information sharing and engagement with forestry professionals. Additional training opportunities will be developed and disseminated through strategic outlets, including expanded continuing education offerings aligned with PERSEUS tool demonstrations.

Targeted training sessions will be conducted for stakeholder groups to support access to and use of PERSEUS data products, with a focus on practical application
in management and decision-making contexts. Ongoing technical consultation will support state-level deployment of mapping products, advancing pathways for operational use by agencies and partners. Coordination of training materials across institutions will ensure consistent messaging and alignment in how tools and technologies are presented. Feedback from training events will be actively collected and used to refine content, delivery approaches and user experience.

DIGITAL COMPETENCE IN STUDENTS AND PROFESSIONALS (EDUCATION)

Develop a digitally competent mindset in students and professionals for data-driven natural resources management. Actively train or retrain diverse cohorts of students and professionals through immersive learning experiences and online learning opportunities to modernize a skilled workforce. Target recruitment at underrepresented and underserved rural communities to enhance the voices engaged in forestry and to upskill those typically overlooked for such opportunities.

Task 4.1 Learning Communities (months 1-60): Design project-focused learning communities that provide opportunities for data-science/engineering undergraduates to work on forestry issues and for forestry undergraduates to learn digital technologies.

Task 4.2 Interns and Fellows (months 1-60): Provide undergraduate internships and graduate fellowships for students interested in digital forestry, recruiting and effectively engaging students with all aspects of the research project.

Task 4.3 Curriculum Development (months 1-60): Develop curriculum aimed at expanding the digital awareness of students engaged in forestry via a wide selection of transdisciplinary courses corresponding to the data pipeline in Global Information System (GIS) science, UAS and remote sensing, and data science. Develop a professional master’s in digital natural resources.

Task 4.4 Online Certificate (months 25-60): Develop online, cross-institutional digital forestry curricula for certificates including the core digital forestry curriculum — for example, undergraduate, graduate, bridging to professional science master’s degree, and professional training, workforce development, Extension.

Deliverables Year 3

• Establish Learning Communities
• Recruit Internships/Fellowships
• Propose New Digital Forestry Curriculum
• Online Certificate/Modules

Accomplishments in Year 3

Task 4.1 Learning Communities: The graduate student learning community continued to expand through regular cross-institutional meetings and collaborative activities. Students participated in hands-on training and knowledge exchange opportunities, including workshops held in conjunction with the annual meeting, which provided experience in field measurement techniques, UAS operations and use of the D2S platform.

Virtual all-hands meetings further supported engagement through dedicated breakout sessions focused on student collaboration and community building across institutions. Cross-institutional manuscript development also advanced, with graduate students and postdoctoral researchers forming working groups to provide peer feedback and support publication efforts.

Additional learning opportunities were provided through participation in a national conference focused on AI applications in forestry, where students
engaged directly with forestry professionals, federal agency staff and industry partners. These experiences strengthened connections between the academic learning community and the broader practitioner network, supporting continued development of a collaborative and applied learning environment.

Task 4.2 Interns and Fellows: Continued support of graduate students, postdoctoral researchers and interns was provided across institutions, contributing to workforce development and research advancement. Ongoing support was also provided for multiple graduate research assistants and postdoctoral fellows working across Objectives 1-3. Contributions included LiDAR data coverage and geolocation research, modeling and scenario development, and stakeholder-perception studies. Fellows and students were actively engaged in manuscript development, with multiple publications in progress spanning data integration, modeling and stakeholder-focused research.

Student and fellow training was further supported through participation in annual meeting workshops, where participants gained hands-on experience with UAS operations, plot measurement techniques and the D2S platform. These activities provided opportunities to build technical skills while engaging in cross-institutional collaboration.

Internship programs also expanded across partner institutions. At UGA, interns contributed to both data collection and technology development efforts, including large-scale image acquisition to support AI-based tree species identification and the development of virtual reality (VR) applications for forestry education and outreach. These efforts resulted in new training tools, conference demonstrations and ongoing dissemination through educational programming and manuscript development.

At Purdue, internships supported PERSEUS-related research and development activities, while additional students contributed through senior design projects focused on improving the backpack system. Collectively, these efforts strengthened experiential learning opportunities and supported the development of technical and applied skills aligned with digital forestry workforce needs.

Task 4.3 Curriculum Development: At UMaine, the Enhanced Forest Inventory and Analysis course — focused on big data applications in forestry — remains a core component of the graduate certificate program and contributes directly to the broader digital forestry curriculum. Although the course was not offered during the reporting period, materials were actively updated to incorporate new data sources and methods developed through the project, including advances in LiDAR and integrated data platforms. The existing graduate certificate in forest biometrics and data science further supports this effort by providing an operational framework that integrates GIS, remote sensing and data science training.

Additional curriculum development efforts were supported through professional master’s and workforce training programs aligned with digital forestry. Purdue’s Digital Ecology and Natural Resources (DENR) professional master’s program was approved to provide interdisciplinary training in data science, remote sensing and applied natural resource management, serving as a key pathway for preparing the next generation of professionals. The first Digital Data Acquisition Camp (DDAC) was hosted at Purdue and offered hands-on, field-based training in emerging technologies such as UASs, LiDAR and sensor integration. These programs provide practical, applied learning experiences that align closely with PERSEUS objectives and workforce development goals.

At UGA, progress was made toward establishing a new graduate-level certificate focused on AI applications in forestry. The proposed program integrates coursework in ethics, modeling and forestry with applied problem-solving, where students use AI to address real-world challenges. The proposal has been advanced for university-level review and represents a significant step toward expanding formal training opportunities at the intersection of artificial intelligence and forestry.

Task 4.4 Online Certificate: Progress continued toward the development and coordination of online certificate programs and asynchronous learning opportunities aligned with digital forestry workforce needs. The UMaine’s graduate certificate in forest biometrics and data science remains active and continues to incorporate PERSEUS-developed methods and data sources, representing initial implementation of online certificate and learning module development. These efforts contribute to expanding accessible, flexible training pathways for students and professionals.

Professional development offerings were further advanced through continuing education programming. The AI Technology for Forestry Conference was approved for continuing forestry education credits, demonstrating a scalable model for delivering applied, technology-focused training to working professionals. Additional discussions at project-wide meetings supported coordination of online curriculum development across institutions, with a focus on aligning content and identifying opportunities for integration.

Complementary efforts include the continued expansion of asynchronous online learning platforms. The Online Learning in Applied Forestry (OLAF) system (olaf.uga.edu) provides Society of American Foresters (SAF)-approved continuing education courses, including a completed remote sensing course supported through the project, with additional modules in artificial intelligence and related topics under
development. These offerings provide accessible, self-paced training opportunities that extend the reach of digital forestry education beyond traditional academic settings. 

PLANS FOR THE COMING YEAR, APRIL 2026—MARCH 2027

Deliverables Year 4

• Establish Learning Communities
• Recruit Internships/Fellowships
• Propose New Digital Forestry Curriculum
• Online Certificate/Modules

Task 4.1 Learning Communities: During Year 4, efforts will continue to strengthen and expand the cross-institutional PERSEUS graduate student learning community through coordinated engagement, collaboration and applied learning opportunities. Planned activities include dedicated student programming at the annual meeting and support for student participation in the Applications & Solutions in Digital Forestry International Conference, providing opportunities for networking, professional development and direct interaction with practitioners and industry partners.

Cross-institutional research collaboration will remain a priority, with continued support for joint manuscript development, data sharing and collaborative analysis among graduate students. Additional efforts will explore opportunities to engage undergraduate students, particularly in data science and engineering, through interaction with PERSEUS datasets and platforms.

Further development of structured learning opportunities will include advancing plans for credit-bearing coursework that supports student engagement in project activities. This includes the potential implementation of a one-credit independent study-style course to formalize student participation and align learning outcomes with ongoing research and professional development efforts.

Task 4.2 Interns and Fellows: Focus for Year 4 is on sustaining and expanding support for graduate students, postdoctoral fellows and interns across participating institutions. Continued support will be provided for graduate research assistants and postdoctoral fellows contributing to PERSEUS research priorities, including remote sensing, forest biometrics and data science. Fellows will remain actively engaged in cross-institutional collaboration through joint manuscript development and professional development activities.

Recruitment efforts will target new graduate students aligned with project priorities, with an emphasis on increasing participation from underrepresented groups and underserved rural communities to support broader workforce development goals.
New interns will be recruited to contribute to ongoing initiatives, including VR applications and other technology-focused efforts, with an emphasis on candidates bringing strong technical and coding skills. These experiences will provide hands-on training opportunities while supporting project deliverables and innovation.

Task 4.3 Curriculum Development: Efforts will focus on implementing updated course offerings, expanding online learning modules and strengthening cross-institutional coordination. The Enhanced Forest Inventory and Analysis course will be offered on its regular cycle with updated content incorporating PERSEUS data sources, methods and tools developed over the first three years. Course materials will include hands-on exercises using integrated data platforms and project datasets to provide applied learning experiences.

Continued refinement of DENR curriculum components will emphasize integration of data science, remote sensing and applied forestry to better reflect emerging digital forestry workforce needs. Complementary workforce training initiatives such as the DDAC will also be refined to enhance hands-on instruction in UAS operations, LiDAR and field-based data collection, further strengthening connections between formal coursework and experiential learning.

Online and asynchronous learning opportunities will continue to expand. New course modules focused on AI and precision forestry are under development, contributing to a growing suite of accessible training resources for students and professionals. These efforts will support broader dissemination of digital forestry concepts and tools while reinforcing a coordinated, multi-institutional approach to curriculum development.

Task 4.4 Online Certificate: Expanding and coordinating online learning and professional development opportunities will be aligned with PERSEUS methods and tools. Continued integration of project-developed content into existing certificate programs will support the growth of digital forestry training pathways.

Additional continuing education opportunities will be pursued through professional accreditation, including expanding credit offerings for conference sessions and training events. Cross-institutional coordination will support the development of aligned online curriculum components, leveraging complementary programs and platforms to ensure consistency in content and delivery.

Efforts will also explore the development of modular, standalone training resources focused on key digital forestry topics, including UAS, LiDAR, data integration and scenario analysis. These modules are intended to increase accessibility for working professionals and extension audiences, supporting broader adoption and application of PERSEUS tools and approaches.

Publications Acknowledging PERSEUS Funding

Ardohain, C., Fei, S. 2025. The impacts of training data spatial resolution on deep learning in remote sensing. Science of Remote Sensing. 11: 100185. https://doi.org/10.1016/j.srs.2024.100185.

Bettinger, P. 2025. Forest harvest scheduling: From linear programming to heuristic search. Cham, Switzerland: Springer Texts in Business and Economics. https://doi.org/10.1007/978-3-031-89432-9.

Bettinger, P., Maier, F. 2025. Field Note: A case study of generative AI responses concerning infrared reflectance of evergreen coniferous and broadleaf deciduous trees during summer. Journal of Forestry. https://doi.org/10.1007/s44392-025-00023-2.

Bettinger, P., Sandoval, S., Merry, K., Lowe III, R.C., Rasheed, K. 2025. Approaches for simulating alternative futures of complex forested landscapes: A review. Environmental Development. 56: 101285. https://doi.org/10.1016/j.envdev.2025.101285.

Carpenter, J., Jung, M., Goel, A., Fei, S., Jung, J. 2025. Species classification of northern hardwood forest inventories from terrestrial laser scans and airborne LiDAR. Frontiers in Forests and Global Change. 8:1500178.  https://doi.org/10.3389/ffgc.2025.1500178.

Choi, D.H., Morton, I.S., Darling, L.E., Wang, J., Thapa, B., Price, E.P., Zaya, D.N., Fei, S., Hardiman, B.S. 2025. Detecting patches of invasive shrubs using high-density airborne LiDAR data and spectral imagery. Urban Forestry & Urban Greening, 107: 128764. https://doi.org/10.1016/j.ufug.2025.128764.

Choi, D., Darling, L., Ha, J., Shao, J., Song, H., Fei, S., Hardiman, B.S. 2025. Understanding the effects of spatial scaling on the relationship between urban structure and biodiversity. Urban Forestry & Urban Greening. 138: 104441. https://doi.org/10.1016/j.jag.2025.104441.

dos Santos, R., Shin, S.Y., Manish, R., Zhou, T., Fei, S., Habib, A. 2025. General Approach for Forest Woody Debris Detection in Multi-Platform LiDAR Data. Remote Sensing. 17: 651. https://doi.org/10.3390/rs17040651.

Firoze, A., Fei, S., Aliaga, D. 2026. Where are the city trees? Monitoring urban trees across the U.S. using generative AI. Communications of the ACM. 69:50-59. https://doi.org/10.1145/3762636.

Hanafy, H., Shin, S., Eissa, A., Hany, Y., Park, S., Fei, S., Habib, A. 2025. General framework for the georeferencing and interpretation of multi-resolution LiDAR data for fine-scale forest inventory. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 48: 567-674.  https://doi.org/10.5194/isprs-archives-XLVIII-G-2025-567-2025.

Huang, Y., Fei, S. 2025. Temperate forest tree species classification with winter UAV images. Remote Sensing Applications: Society and Environment. 37:101422. https://doi.org/10.1016/j.rsase.2024.101422.

Jung, M., Choi, J., Carpenter, J., Fei, S., Jung, J. 2025. Individual tree biomass estimation using single-scan terrestrial laser scanner with efficient projection-based deep learning. Journal of Forestry. https://doi.org/10.1007/s44392-025-00065-6.

Lu, Y., Chiu, J., Khanal, N., Chen, S.C., Guo, Q., Liu, D., Fei, S., Chen, Y.V. 2025. VidFin: A video-based monocular UAV pipeline for automatic forest inventory in diverse scenes. Smart Agricultural Technology. 12: 101504. https://doi.org/10.1016/j.atech.2025.101504.

Merry, K., Bettinger, P., Merritt, C., Lowe, III, R.C., da Silva, B. 2026. AI technology use in North American forestry. Journal of Forestry. https://doi.org/10.1007/s44392-026-00079-8.

Park, S., Fei, S., Habib, A. 2025. UAV-assisted scan planning for improved forest inventory using a mobile backpack LiDAR system. Computers and Electronics in Agriculture. 239, 111147. https://doi.org/10.1016/j.compag.2025.111147.

Salcido, E.L., Soucy, A.R., De Urioste-Stone, S., Johnson, N., Tiwari, M., Zhao, J., Abrams, J., Daigneault, A., Ma, Z., Simons-Legaard, E. and Weiskittel, A., 2025. A Delphi assessment of digital technology needs in the Eastern U.S. forest sector. Society & Natural Resources, pp.1-18. https://doi.org/10.1080/08941920.2025.2589901.

Sandoval, S., Montes, C.R., Bullock, B.P. 2026. Modeling dominant height using stand information, soil characteristics, and water balance variables for loblolly pine in the Western Gulf region of the USA. Forestry. 99(2): 1-11. https://doi.org/10.1093/forestry/cpaf060.

Shao, J., Choi, D.H., Liu, J., Tian, X., Thapa, B., Lee, S., Habib, A., Fei, S. 2026. A three-stage framework for stand-level automated stem volume estimation in temperate forests using mobile laser scanning. Remote Sensing of Environment. 335: 115246. https://doi.org/10.1016/j.rse.2026.115246.

Thapa, B., Hardiman, B.S., Fei, S. 2025. Flower color index for detecting and monitoring warm-colored flowering across scales. International Journal of Applied Earth Observation and Geoinformation. 145: 104978. https://doi.org/10.1016/j.jag.2025.104978.

Thapa, B., Darling, L., Choi, D., Hardiman, B.S., Fei, S. 2026. Optimal spectra for deciduous tree species identification in urban areas: Insights from hyperspectral remote sensing. International Journal of Remote Sensing. 47: 3044-3061. https://doi.org/10.1080/01431161.2026.2626094.

Tiwari, M., Siry, J., Abrams, J., Bettinger, P. 2026. Linking environmental sustainability in forest companies to ecosystem services: A systematic review and outlook. Ecosystem Services. 79: 101842. https://doi.org/10.1016/j.ecoser.2026.101842.

Vatandaslar, C., Bettinger, P., Merry, K., Stober, J., Lee, T. 2025. Semi-automatic stand delineation based on very-high resolution orthographs and topographic features: A case study from a structurally complex natural forest in the southern USA. Forests. 16(4): 666. https://doi.org/10.3390/f16040666.

Warner, C., Wu, F., Gazo, R., Benes, B., Fei, S. 2025. Environmental sensitivity in AI tree bark detection: Identifying key factors for improving classification accuracy. Algorithms. 18: 417; https://doi.org/10.3390/a18070417.

Wei, X., Hayes, D.J., Schwalm, C.R., Fisher, J.B., Huntzinger, D.N., Ma, L., Vargas R., and Brunsell, N.A. (2025). Climate constrains the enhancement of CO2 fertilization on forest gross primary productivity. Environmental Research Letters, 20(6), 064013. https://iopscience.iop.org/article/10.1088/1748-9326/add177.

Wei, X., Hayes, D., Weiskittel, A., & Zhao, J. (2025). Warming-driven shifts in dominant tree species potentially reduce aboveground biomass in northeastern United States forests. Forest Ecology and Management. 580, 122536. https://doi.org/10.1016/j.foreco.2025.122536.

Xiang, W., S. Fei, and S. Zhang. 2025. Single shot high-accuracy diameter at breast height measurement with smartphone embedded sensors. Sensors. 25: 5060. https://doi.org/10.3390/s25165060.

Yin, Z., Fei, S., Zhang, S. 2026. Tree diameter at breast height measurement with smartphone sensors. Optics and Lasers in Engineering. 201: 109684. https://doi.org/10.1016/j.optlaseng.2026.109684.

Zhao, C., Fei, S., Habib, H. 2026. Integrated scan simultaneous trajectory enhancement and mapping (IS2-TEAM) for fine resolution forest inventory using BackPack LiDAR. Remote Sensing of Environment. 334: 115212. https://doi.org/10.1016/j.rse.2025.115212.

Zhao, C., Hyung S., Manish R., Shin S., Park S., Fei S., Habib, A. 2025. Framework for the georeferencing and processing of BikePack LiDAR data for urban tree mapping. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 48: 1753–1760. https://doi.org/10.5194/isprs-archives-XLVIII-G-2025-1753-2025.

Zhao, J., Daigneault, A., Wei, X., Salcido, E., Weiskittel, A. (2025). Forest type, landowner practices, and climate shape tree species diversity in Maine, USA. Forest Ecology and Management. 593: 122919. https://doi.org/10.1016/j.foreco.2025.122919.

Zhou, X., Li, B., Benes, B., Habib, A., Fei, S., Shao, J., Pirk, S. 2025. TreeStructor: Forest reconstruction with neural ranking. IEEE Transactions on Geoscience and Remote Sensing. 63: 4408419. https://doi.org/10.1109/TGRS.2025.3558312.

Conference Presentations Referencing PERSEUS Research

A Delphi Assessment of Decision Support Tool Needs in the Eastern U.S. Forest Sector. E. Salcido, D. Soucy, S. De Urioste-Stone, N. Johnson, M. Tiwari, J. Zhao, J. Abrams, A. Daigneault, Z. Ma, E. Simons-Legaard, A. Weiskittel. International Association for Society & Natural Resources Conference, June 2-5, 2025, Vancouver, Canada.

A Delphi Assessment of Information and Digital Technology Needs in the Eastern USA’s Forest Industry.

E.L. Salcido, A.R. Soucy, S. De Urioste-Stone, N. Johnson, M. Tiwari, M. Zhao, J. Abrams, A. Daigneault, M., E. Simons-Legaard, A. Weiskittel. International Association for Society and Natural Resources Conference, June 8-12, 2025, Vancouver, Canada.

Advancing Carbon Monitoring, Reporting, and Verification. D.J. Hayes. Michigan State University Forest Carbon and Climate Program Learning Exchange Series, April 22, 2025.

AI for Urban Visual Computing. Daniel Aliaga. Conference on Computer Vision and Pattern Recognition, June 3-7, 2025, Denver, CO.

Airborne LiDAR-Informed Forest Inventory: Case Studies in Maine’s Working Forests. D.J. Hayes, A.R. Weiskittel. SilviLaser Conference, September 30, 2025, Quebec City.

Application of AI and Digital Technology for Scalable Carbon and Biomass Inventory. Songlin Fe. October 2025, Indiana University, Bloomington, IN.

Application of AI and Digital Technology for Scalable Carbon and Biomass Inventory. Songlin Fei. SinoEco Seminar Series, October 2025.

Application of AI in Digital Forestry. Songlin Fei. May 2025, University of Georgia, Athens, GA. Application of Digital Forestry in Forest Health Monitoring. Songlin Fei. North Central Forest Pest

Workshop, September 2025, Ludington, MI.

Applications of Digital Technologies in Forestry. E. Kronenberger, Z. Ma, J. Jung, M. Jung, B. Hancock, A. Habib. Ohio River Valley Woodland and Wildlife Workshop, March 21, 2026, Hamilton, OH.

Applying AI and Digital Technology in Urban Forests. Songlin Fei. Partners in Community Forestry, November 2025, Las Vegas, NV.

Areal: A Cloud-Based Platform for Collaborative SAE Research and Development. K. Legaard, K. Bundy. PSAE Team Meeting, March 2025.

Areal: A Cloud-Based Platform for Collaborative SAE Research and Development. K. Legaard, K. Bundy. PSAE Team Meeting, June 2025.

Assessing the Accuracy of TreeMap Vegetation Data Using Field Measurements from Talladega National Forest. J. Tu. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, University of Georgia, Athens, GA.

Automating Forest Stand Delineation with AI Techniques Using Aerial Imagery and Airborne LiDAR. A. Sankhe. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, University of Georgia, Athens, GA.

Building Instance Segmentation for Dense Urban Settlements. A. Firoze, R. Yeh, D. Aliaga. Association for the Advancement of Artificial Intelligence Conference on Artificial Intelligence, January 2026, Singapore.

Current and Future Role of GIS and Digital Forestry for Forest Management Organizations. Songlin Fei. 15th Southern Forestry and Natural Resource Management GIS Conference, December 2025, Athens, GA.

Enhancing Forest Biomass Estimation Using Airborne LiDAR Across Diverse Forest Types. X. Wei, D. Hayes, G. Daniels, A.R. Weiskittel. American Geophysical Union Annual Meeting, December 2025, New Orleans, LA.

Enhancing Forest Inventory and Management with AI and Digital Technology. Songlin Fei. Kentucky-Tennessee Society of American Foresters, January 2026, Lexington, KY.

Forest Companies’ Willingness to Invest (WTI) in Ecosystem Services Projects. M. Tiwari, J. Siry, J. Abrams. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, University of Georgia, Athens, GA.

Forest Product Companies’ Intention to Invest in Ecosystem Services. M. Tiwari, J. Siry, J. Abrams. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, University of Georgia, Athens, GA.

GNSS Through the Trees Presentation. K. Merry, P. Bettinger. Southern Forestry and Natural Resource Management GIS Conference, December 6-7, 2025, Georgia, Athens, GA.

Harnessing Reinforcement Learning for Forest Growth Modeling Under Uncertainty. C. Merritt. Society of American Foresters Annual Convention, October 22-25, 2025, Hartford, CT.

Human-Nature System Interactions Under a Sustainable Future: Concepts, Modeling, and Application. X. Zhao, A. Daigneault, E. Sinha, C. Reyer, M. Luo. American Geophysical Union Annual, December 2025, New Orleans, LA.

Identifying Sources of Non-Passive Carbon Sequestered in Forests. B. Sohngen, R. Kimura, J.B. Kim, A. Golub, A. Daigneault, E. Davis. American Geophysical Union Annual Meeting, December 2025, New Orleans, LA.

Insights on Family Forest Owners’ Willingness to Adopt Digital Forestry Technologies. C. Rodrigo, Z. Ma. Institute for Digital Forestry Mini-Symposium, February 28, 2026, West Lafayette, IN.

Insights on Family Forest Owners’ Willingness to Adopt Digital Forestry Technologies. C. Rodrigo. Institute for Digital Forestry Mini-Symposium, February 28, 2026, West Lafayette, IN.

Integrating Climate Change, Socioeconomic Drivers, and Ownership Dynamics in a Forest Sector Model: Implications for Timber Supply and Biodiversity in the Northeastern United States. J. Zhao, A. Daigneault. NAREA Conference and Workshop, June 10-11, 2025, Burlington, VT.

Introduction to Artificial Intelligence and Machine Learning. F Maier, K. Rasheed. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry Conference, May 13, 2025, Athens, GA.

Large Landholder Technology Survey Results. E.L. Salcido. PERSEUS All-Hands Meeting, November 13, 2025.

Leveraging Spaceborne LiDAR to Assess Understory Fuel Loads. T. Myrie. Society of American Foresters Annual Convention, October 22-25, 2025, Hartford, CT.

Leveraging Spaceborne LiDAR: A Quantitative Analysis of Fuel Loads Below the Forest Canopy. T. Myrie. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, Athens, GA.

Mapping Forest Productivity Using Mapped Soil and Geospatial Variables Across the Georgia Landscape. Zhao, P. Bettinger, K. Merry, B. Bullock, D. Dickens, S. Kinane. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry Conference, May 13, 2025, Athens, GA.

Modeling Forest Growth in an Uncertain Future. C. Merritt, P. Bettinger, S. Kinane, and F. Maier. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, University of Georgia, Athens, GA.

Multimodel Comparison Results for Maine Forests. A.R. Weiskittel. National Council for Air and Stream Improvement, September 29-October 1, 2025, Athens, GA.

OLAF: Online Learning in Applied Forestry Education Tool. K. Merry, P. Bettinger, F. Maier, K. Rasheed, Amouhadi, C. Merritt. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry Conference, May 13, 2025, Athens, GA.

Predicting Tree Mortality in Southern Pine Forests Using Advanced Modeling Techniques. D.M. Senevirathne, S.-I. Yang, D. Zhao, B. Bullock, S. Kinane. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry Conference, May 13, 2025, Athens, GA.

Promoting Economic Resilience and Sustainability of the Eastern U.S. Forests (PERSEUS) - Tools in Development. E. Kronenberger, S. Fei, A. Daigneault, K. Merry. Southern Group of State Foresters, Winter Meeting and Biochar Workshop, February 3-6, 2026.

Promoting Economic Resilience and Sustainability of the Eastern U.S. Forests (PERSEUS) - Tools in Development. E. Kronenberger, Z. Ma, J. Jung, M. Jung, B. Hancock, A. Habib. Indiana Society of American Foresters Winter Meeting. March 5, 2026, Brown County State Park, IN.

Promoting Economic Resilience and Sustainability of the Eastern U.S. Forests. E. Kronenberger, Z. Ma, Boby, K. Merry, E. Jackson, L. Farlee. Indiana Association of Consulting Foresters Summer Meeting, August 7, 2025, Morgan-Monroe State Forest, IN.

Smartphone GNSS Uncertainty in Forested Environments. K. Merry, P. Bettinger. Society of American Foresters Annual Convention, October 22-25, 2025, Hartford, CT.

Stakeholder-Driven Carbon MMRV in Maine’s Working Forests. D.J. Hayes, X. Wei, B. Cook, A. Daigneault, Finley, Y. Shafirin, A.R. Weiskittel, J. Zhao. American Geophysical Union Annual Meeting, December 2025, New Orleans, LA.

Super-Resolution of Multispectral Imagery in Forestry. H. Gu, J. Couture, M. Crawford. Institute for Digital Forestry Mini-Symposium, February 28, 2026. West Lafayette, IN.

Tree Species Classification Using Hyperspectral Data. S. Mohammadi, M. Crawford, J. Couture. Institute for Digital Forestry Mini-Symposium, February 28, 2026. West Lafayette, IN. Award: Most Interdisciplinary Research Award.

Understanding Wood-Using Companies’ Perceptions of Ecosystem Services. M. Tiwari, J. Siry, J. Abrams, Bettinger, K. Merry. Society of American Foresters Annual Convention, October 22-25, 2025, Hartford, CT.

Urban Visual Computing. Daniel Aliaga. Conference on Graphics, Patterns and Images (SIBGRAPI), September 30, 2025, Salvador, BA, Brazil.

Using Aerial LiDAR Data for Estimating Current Forest Conditions Within the Talladega National Forest of the Southern United States. S. Sandoval, P. Bettinger, K. Merry, J. Stober. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry Conference, May 13, 2025, Athens, GA.

Value-Optimizing Forest Stand Delineation Through Evolutionary Algorithms. B. Alaila. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry, May 13, 2025, Athens, GA.

Who Owns the Woods Shapes What Grows: Harvesting, Climate, and Tree Diversity. J. Zhao, A. Daigneault, A.R. Weiskittel. Society of American Foresters National Convention, October 22-25, 2025, Hartford, CT.

Wood-Using Companies’ Intention to Invest in Payment for Ecosystem Services Programs: A Theory of Planned Behavior Approach. M. Tiwari, J. Siry, J. Abrams, P. Bettinger. Warnell Graduate Student Symposium, February 5-6, 2026, University of Georgia, Athens, GA.

Other Products

Activities

• April Center for Research on Sustainable Forests Update, Cooperative Forestry Research Unit Spring Meeting, Orono, ME.
• July The Application of AI in Plantation Management. Plantation Management Research Cooperative Annual Advisory Committee Meeting.
• July 2025. Artificial Intelligence and Other Advanced Technology Solutions for Eastern United States Forestry: Insights from an Outreach Activity of the PERSEUS Project, Online Sustainable Agriculture Systems Community Meeting.
• July 2025. Enhancing Forest Variable Estimation Using Satellite LiDAR and Spatial Correlation: A Case Study in the Talladega National Forest Presentation, Plantation Management Research Cooperative Annual Advisory Committee Meeting.
• July 2025. GEDI Applications in Estimating Understory Biomass Presentation, Plantation Management Research Cooperative Annual Advisory Committee Meeting.
• July 2025. Mortality Modeling: An Approach Using Segmented Models for Loblolly Pine Stands Presentation, Plantation Management Research Cooperative Annual Advisory Committee Meeting.
• July Using LANDIS-II as a Tool for Large Scale Modeling for Evaluating Harvest Prescription Impact Presentation, Plantation Management Research Cooperative Annual Advisory Committee Meeting.
• August 2025. Backpack LiDAR Data Collection Demonstration, City of West Lafayette Parks and Recreation Department.
• August Invasive Plant Identification and Forestry Data Tool, Indiana Consulting Foresters Summer Meeting.
• September 2025. Backpack LiDAR Data Collection and Applications Workshop, North Central Forest Pest Workshop.
• November 2025. PERSEUS Project Outreach Presentation and Survey Distribution, Indiana Annual Woodland Owner Conference.
• November 2025. PERSEUS Project Overview and LiDAR Applications in Forestry Presentation, Indiana Society of American Foresters Pesticide Training Program.
• December 2025. PERSEUS Project Outreach and Engagement Presentation, American Geophysical Union Annual Meeting.
• January Data to Science Platform and 3D Visualization Tools Workshop.
• February Center for Research on Sustainable Forests and Digital Forestry Overview, UMaine Joint Development Institute, Orono, ME.
• February 2026. Center for Research on Sustainable Forests Update, CRSF Advisory Board Presentation, Orono, ME.
• February 2026. Online Learning in Applied Forestry Education Tool Presentation, Southern Group of State Foresters Winter Meeting, Lexington, KY.
• February 2026. PERSEUS Project Overview and Applications, Southern Group of State Foresters Winter Meeting, Lexington, KY.
• February 2026. PERSEUS Project Tools and Extension Applications Presentation, Southern Regional Extension Forestry Unit Leaders Meeting.
• March 2026. Extension Toolkit and Forestry Applications of PERSEUS Tools Presentation, Indiana Society of American Foresters Winter Meeting.
• March 2026. Extension Toolkit and Forestry Applications of PERSEUS Tools Presentation, Society of American Foresters Southeastern Section Annual Meeting.
• March PERSEUS Project Outreach and Applications, Ohio River Valley Woodland and Wildlife Workshop.
• March 2026. PERSEUS Project Outreach Using Extension Toolkit Presentation, Northeastern Society of American Foresters Winter Meeting.
• March 2026. PERSEUS Scenario Development: Plausible Futures for Eastern U.S. Forests, PERSEUS Objective 2/Objective 3 Joint Meeting.
• March UAV LiDAR data collection and Data to Science Platform Demonstration.

Datasets

Ayrey, E., Hayes, D., Wei, X., Shao, G., Weiskittel, A., Fei, S., Zhao, J., Zhang, B. 2025. Forest Aboveground Biomass for Maine, 2023. ORNL DAAC, Oak Ridge, Tennessee. https://doi.org/10.3334/ORNLDAAC/2435.

Wei, X., Hayes, D., McHale, G., Zhao, J., Weiskittel, A., Daigneault, A. 2026. PyFIA: Analyzing and Visualizing Forest Attributes Using the United States Forest Inventory and Analysis Database. Carbon Balance and Management, 21(1), 18. https://doi.org/10.1186/s13021-025-00364-7.

Wei, X., Hayes, D., Weiskittel, A., McHale, G., Howe, P., Zhang, C., Shao, G. 2025. Aboveground Biomass for Howland Forest, Maine, 2012-2023. ORNL DAAC, Oak Ridge, Tennessee. https://doi.org/10.3334/ORNLDAAC/2434.

Surveys

PERSEUS Management Intentions Survey Dataset: 926 forest management scenarios across 31 states, with management regime characterizations for 14 forest types by ownership class and geographic region.

Scenario Survey Dataset: 1,746 respondent records with influence scores and impact rankings for 10 forest future drivers across the Northeast, Southeast, Central Hardwoods, and Lake States regions. Includes the validated 2x2 scenario framework parameters.

Events

• May AI Technology for Forestry Conference. University of Georgia.
• May PERSEUS Annual Meeting. University of Georgia.
• May 2025. Second Annual pyFIA Code Jam. G. McHale, D.J. Hayes, X. Wei. University of Maine.
• February Institute for Digital Forestry Mini-Symposium. Purdue University.