Small-parcel forests are more vulnerable to climate change and natural disasters than large forests. Such vulnerabilities are especially challenging for Minnesota because more than 125,000 landowners own small-parcel forests. Moreover, various land management decisions have been fragmenting private forests into small parcels, posing even higher risks to Minnesota forestry and the economy. To advance the effective management of private forest lands, it is critical to monitor forest fragmentation and degradation e.g., remote sensing.

Yet, existing remote sensing data are challenging for business-ready decision-making. Conventional satellite imagery may fail to detect forest fragmentation if the divide between fragments is finer than image resolutions. Instead, the 3-D structure information from LiDAR is more effective. For example, the USGS 3DEP aircraft LiDAR provides statewide information at high spatial resolution (<100 feet) suitable for capturing small-parcel forests. However, the aircraft LiDAR data only contains outdated snapshots and can’t capture continuous changes in time. Meanwhile, the technical terminology of the USGS 3DEP LiDAR output, e.g., point clouds and canopy height statistics, challenges decision-making for non-technical experts. Thus, we aim to enhance the public usability of LiDAR data and provide real-time, accountable, and business-ready information about forest fragmentation and degradation.

To facilitate efficient forest management, we are developing a real-time interactive web dashboard for statewide forest fragmentation and degradation with business-ready and accountable information. This dashboard will highlight where, when, and how much forests statewide are fragmented and degraded. Our project highlights two main innovations:

1. Translating technical LiDAR data into business-ready information at high spatial resolution. Our dashboard will directly show the forest boundaries and their area changes as an intuitive illustration of forest fragmentation and degradation. We will develop algorithms to derive forest boundaries from LiDAR data. The web dashboard includes an interactive map so that users can accurately retrieve the information at a finer than 1-acre resolution.

2. Statewide continuous updates. We are going to incorporate the high-resolution spatial details from USGS 3DEP LiDAR snapshots into NASA’s satellite continuous statewide LiDAR measurements using machine learning. This will yield us the time series of statewide forest boundaries from 2018 to the present at a monthly/quarterly scale. By tracking the changes in forest boundaries, we will deliver direct measurements of forest fragmentation and degradation in history and real-time.

Our project outcomes directly relate to the public purpose of protection, conservation, preservation, and enhancement of the state’s natural resources. For natural resource management, our proposed dashboard will show business-ready information about where, when, and how much forests statewide are fragmented and degraded. This outcome directly benefits land management through rapid detection and historic tracing of changing forests due to human activity and natural disasters. We offer affordable and up-to-date information on private land which is often inaccessible for field surveys. For federal and public agencies, e.g., the Forestry Resource Assessment Team at the Department of Natural Resources, our dashboard will serve as a handy tool to pinpoint regions with significant changes and strategically procure new aircraft LiDAR data.