PS 72-123
Greendown signal, simply a signal of local habitat variability or indicator of leaf physiology changes?
Climate change can cause shifts in forest phenology which affect ecosystem processes. Remote sensing data provide a standardized approach for assessing changes in green-leaf phenology across spatiotemporal scales. A pronounced feature of 30-m resolution Landsat data across North America is complex spatial variation in forest phenology; some canopies remain green throughout the growing season while others “greendown”. However, the factors controlling the stability of summer greenness remain uncertain. We hypothesized that abiotic factors (e.g. aspect, slope and irradiance) and/or leaf properties (e.g. changes in leaf-water and/or leaf-N availability) control greendown. To assess these hypotheses we collected leaf samples at the beginning of the growing season (BOS) and end of the growing season (EOS) from sunlit leaves of 64 Quercus alba trees (2 samples/tree) in a deciduous forest that spanned a wide range of variation in canopy greenness. Leaf δ13C and δ15N (proxies for water and N availability, respectively) and %N (proxy for chlorophyll content) were measured using an elemental analyzer interfaced with an isotope ratio mass spectrometer. Spectral measurements were taken for each leaf to derive biochemical constituents and reflectance. Slope and aspect were calculated using a 2-m DEM and irradiance using a 30-m DEM. We used regression analysis to assess the relationship between greendown and the measured leaf variables for each sampling period. We also assessed the relationship between greendown and Δ (difference between BOS and EOS values) of the leaf variables, and abiotic factors.
Results/Conclusions
Greendown was positively correlated with aspect (r=0.36, p<0.01), slope (r=0.44, p<0.01) and negatively correlated with irradiance (r= -0.55, p<0.01). There were no significant relationships between greendown and measured leaf variables (including Δ-leaf) except for BOS leaf-visible reflectance (r=0.41, p<0.01) and EOS leaf-near-infrared reflectance (r=0.21, p<0.01). We found positive correlations between topographic features and some leaf variables, such slope and BOS δ15N (r=0.27, p<0.05) and slope and EOS δ15N (r=0.27, p<0.01). These results suggest that local topography influences resource availability (e.g., nitrogen, light) which determines the stability of summer greenness and by inference photosynthetic activity. Leaf-level properties appear to have less influence on the rate of greendown than abiotic factors.