Giant Kelp (Macrocystis pyrifera) is an important primary producer along the west coast of North America. It provides critical habitat to a wide range of marine organisms. While satellite sensors can easily quantify canopy area of kelp, using similar techniques to gauge the physiological health of these macroalgae has proven more difficult. Bell et al. (2015) devised an algorithm that effectively estimated the chlorophyll to carbon ratio (Chl:C)—a proxy for kelp health—using AVIRIS imagery. A comparison of AVIRIS imagery of the Isla Vista kelp bed sampled in 2013 and 2015, before and during the recent El Nino- associated west coast ‘warm anomaly’, indicates a decline in Chl:C in 2015 (student t-test p<0.01). The lower Chl:C of Isla Vista observed in 2015 could be caused by environmental stressors associated with El Niño such as increased sea surface temperature, decreased nutrient availability, and disturbance. However while AVIRIS imagery shows great potential in mapping kelp forest health, as an airborne sensor its availability is inconsistent over time, making it less ideal for continuous kelp forest monitoring. We therefore attempt to extend this method of determining Chl:C based on reflectance values to Landsat 8 satellite imagery. We found that although USGS Landsat 8 atmospherically corrected reflectance data does not accurately estimate kelp health, simulated Landsat 8 data from an AVIRIS image does. This suggests that although the spectral resolution of Landsat 8 is much lower than AVIRIS, with sufficient atmospheric correction the satellite will be able to classify kelp health.

Presentation Slides: https://www.scribd.com/document/320955597/Using-Remote-Sensing-Techniques-to-Measure-Chl-C-in-the-Santa-Barbara-Channel#fullscreen&from_embed