PRESENTERS

• Meghan Halabisky, University of Washington Remote Sensing and Geospatial Analysis Lab
• Anthony Stewart, PhD Student, University of Washington
• Andy Robertson, GeoSpatial Services at Saint Mary’s University of Minnesota

ABSTRACTS

Meghan Halabisky
Accurate, un-biased wetland inventories are critical to monitor and protect wetlands from future harm or land conversion. However, most wetland inventories are constructed through manual image interpretation or automated classification of multi-band imagery and are biased towards wetlands that are easy to detect directly in aerial and satellite imagery. Wetlands that are obscured by forest canopy, occur ephemerally, and those without visible standing water are, therefore, often missing from wetland maps. To aid in detection of these cryptic wetlands, we developed the Wetland Intrinsic Potential tool, based on a wetland indicator framework commonly used on the ground to detect wetlands through the presence of hydrophytic vegetation, hydrology, and hydric soils. Our tool uses a random forest model with spatially explicit input variables that represent all three wetland indicators, including novel multi-scale topographic indicators that represent the processes that drive wetland formation, to derive a map of wetland probability.

Anthony Stewart
Inland wetlands disproportionately contribute to the soil organic (SOC) carbon pool by storing 20-30% of all SOC despite occupying only 5-8% of the land surface. However, difficulty identifying wetland areas under perennial forest canopy increases uncertainty in estimates of SOC stocks across watershed to regional scales. We used a machine learning approach that utilized aerial lidar-derived hydrologic and topographic metrics to characterize the landscape surface and identify areas of potential wetland formation for three study areas in the Pacific Northwest which represent an east-to-west (low-to-high) precipitation gradient. This approach produces a spatially explicit and continuous model of wetland probability as a range from wetland to upland across a landscape. We then collected soil samples and measured SOC stocks along the wetland-to-upland probability gradient and used the probability along with surficial geology corresponding to geomorphology to model SOC stocks.

Andy Robertson
GeoSpatial Services has been focused on landscape level wetland inventory and functional assessment for over two decades. Throughout that time we have explored a variety of data development and classification tools for creating derived data that supports comprehensive resource monitoring and assessment. This presentation will describe several modelling efforts based on machine learning algorithms, object analysis, derived elevation surfaces, network analysis and high-resolution optical imagery. These techniques include: potential wetland landscapes, potentially restorable wetlands, derived and hydro modified surface hydrology and cumulative impact assessment based on proximity to oil field development and potential contamination. These tools will be presented in the context of wetland program plan development supported by EPA CWA Section 404 grants.