4.6. Linking the FVI to observable stress and mortality
Connecting the FVI to the physical expression of vulnerability, such
as reduction in leaf area, growth decline, and mortality is a critical
step that needs to be explored in detail. For this initial application, we
chose to look for direct visual evidence of canopy die-off and mortality
in association with positive FVI areas in Google Earth. A randomsample
of 132 points was established across Oregon and Washington in forest
areas not impacted by abrupt disturbances. At each point, the corresponding
800-m FVI pixel boundary was extracted and imported into
Google Earth for visual interpretation. Then, using the time series of
high resolution Google Earth images over the study period, presence
of stress and mortality (dead and dying trees) was recorded at each
plot based on visual interpretation. Pixels that contained widespread
canopy die-off and/or mortality, indicated by the presence of dead or
dying trees spread across the pixel area, or that contained large patches
of mortality, indicated by groups of adjacent dead or dying trees, were
recorded as stressed. All other pixels were recorded as background
mortality. Correspondingly, each pixels slope (FVI value) and p-value
were extracted. Since the p-value declares a FVI value statistically significant,
we plotted the proportion of observations labeled as stressed
against p-value.
5. Results
5.1. FVI