Forest inventory data provide a new method for mapping forest canopy and understory vegetation densities
Canopy Cover in Marcell Experimental Forest.
Forest overstory and understory vegetation layers have different effects on hydrologic processes and water supply. Several physically based hydrologic models recognize the different influences that overstory versus understory canopies exert on hydrologic processes, yet most inputs to such models consist only of total leaf area index (LAI) rather than leaf area index differentiated by strata.
Researchers set out to provide improved input datasets for hydrologic modeling that distinguish overstory and understory canopy layers based on forest inventory data. Three pre-existing methods were applied for estimating overstory leaf area index, and one new method for estimating both overstory and understory leaf area index, to measurements collected from a probability-based plot network established by the USDA Forest Service’s Forest Inventory and Analysis (FIA) program, for an area encompassing two large watersheds in Montana.
Plot-level leaf area index estimates were combined with spatial datasets (such as biophysical and remote sensing predictors) in a machine learning algorithm (random forests) to produce annual gridded leaf area index datasets. Leaf area index estimates were compared to Landsat-based leaf area index maps. Methods that estimate only overstory leaf area index tended to underestimate leaf area index relative to Landsat-based leaf area index. This new method, which estimated both overstory and understory layers, was most strongly correlated with Landsat-based leaf area index. A new method for partitioning forest vegetation data into overstory and understory density correlates well with existing Landsat leaf area index data. During 1984-2019, interannual variability of understory density exceeded that for overstory density.
This year-to-year variability may affect partitioning of precipitation to evapotranspiration vs. runoff at annual timescales. Distinguishing overstory vs. understory leaf area index components is anticipated to improve the ability of leaf area index-based analyses and models to simulate how forest change influences hydrologic processes. Because both overstory and understory data are collected from Forest Inventory & Analysis (FIA) plots on a consistent, ongoing basis, this dataset represents an opportunistic data source for future water resources modeling and assessments.