Spectral determination of concentrations of functionally diverse pigments in increasingly complex arctic tundra canopies

As the Arctic warms, tundra vegetation is becoming taller and more structurally complex, as tall deciduous shrubs become increasingly dominant. Emerging studies reveal that shrubs exhibit photosynthetic resource partitioning, akin to forests, that may need accounting for in the "big leaf" net ecosystem exchange models. We conducted a lab experiment on sun and shade leaves from S. pulchra shrubs to determine the influence of both constitutive (slowly changing bulk carotenoid and chlorophyll pools) and facultative (rapidly changing xanthophyll cycle) pigment pools on a suite of spectral vegetation indices, to devise a rapid means of estimating within canopy resource partitioning. We found that: (1) the PRI of dark-adapted shade leaves (PRIo) was double that of sun leaves, and that PRIo was sensitive to variation among sun and shade leaves in both xanthophyll cycle pool size (V + A + Z) (r(2) = 0.59) and Chla/b (r(2) = 0.64); (2) A corrected PRI (difference between dark and illuminated leaves, Delta PRI) was more sensitive to variation among sun and shade leaves in changes to the epoxidation state of their xanthophyll cycle pigments (dEPS) (r(2) = 0.78, RMSE = 0.007) compared to the uncorrected PRI of illuminated leaves (PRI) (r(2) = 0.34, RMSE = 0.02); and (3) the SR680 index was correlated with each of (V + A + Z), lutein, bulk carotenoids, (V + A + Z)/(Chla + b), and Chla/b (r(2) range = 0.52-0.69). We suggest that Delta PRI be employed as a proxy for facultative pigment dynamics, and the SR680 for the estimation of constitutive pigment pools. We contribute the first Arctic-specific information on disentangling PRI-pigment relationships, and offer insight into how spectral indices can assess resource partitioning within shrub tundra canopies.