Coral Reef Color: Remote and In-Situ Imaging Spectroscopy of Reef Structure and Function

Abstract:

Coral reefs are threatened at local to global scales by a litany of anthropogenic impacts, including overfishing, coastal development, marine and watershed pollution, rising ocean temperatures, and ocean acidification. However, available data for the primary indicator of coral reef condition — proportional cover of living coral — are surprisingly sparse and show patterns that contradict the prevailing understanding of how environment impacts reef condition. Remote sensing is the only available tool for acquiring synoptic, uniform data on reef condition at regional to global scales. Discrimination between coral and other reef benthos relies on narrow wavebands afforded by imaging spectroscopy. The same spectral information allows non-invasive quantification of photosynthetic pigment composition, which shows unexpected phenological trends. There is also potential to link biodiversity with optical diversity, though there has been no effort in that direction. Imaging spectroscopy underlies the light-use efficiency model for reef primary production by quantifying light capture, which in turn indicates biochemical capacity for CO₂ assimilation. Reef calcification is strongly correlated with primary production, suggesting the possibility for an optics-based model of that aspect of reef function, as well. By scaling these spectral models for use with remote sensing, we can vastly improve our understanding of reef structure, function, and overall condition across regional to global scales. By analyzing those remote sensing products against ancillary environmental data, we can construct secondary models to predict reef futures in the era of global change. This final point is the objective of CORAL (COral Reef Airborne Laboratory), a three-year project funded under NASA’s Earth Venture Suborbital-2 program to investigate the relationship between coral reef condition at the ecosystem scale and various nominal biogeophysical forcing parameters.