Quantitative Assessment of Remote Sensing Capabilities to Improve Monitoring of Giant Kelp Condition

Kelp forests are critical ecosystems that provide habitat for marine species, protect coastlines, and contribute to carbon sequestration. These ecosystems face increasing threats from climate-driven stressors such as storms, marine heat waves, and ocean acidification, as well as pollution and overgrazing. Effective conservation requires monitoring approaches to detect early signs of ecological stress before kelp canopy loss starts to occur. This study evaluates whether kelp physiological condition, specifically tissue nitrogen content, is detectable from remotely sensed imagery. to link spectral indices from UAVs and Sentinel-2 satellite imagery to field measurements of percent nitrogen in giant kelp (Macrocystis pyrifera) tissue at two Santa Barbara Channel reefs. Models using satellite imagery outperformed those using UAV imagery, with the best predictions resulting from including normalized difference vegetation index (NDVI) with local upwelling conditions (BEUTI). We reason that the higher resolution in UAV imagery is more sensitive to sea surface roughness, driving higher error in the model predictions. Our results demonstrate that remote sensing can contribute to kelp condition estimations, offering a scalable, cost-effective complement to traditional diver surveys. As climate change intensifies marine heat waves, disrupts upwelling patterns, and increases storm frequency, robust monitoring strategies are increasingly critical for understanding kelp responses and informing adaptation.

Acknowledgements

UC Santa Barbara Bren School: Dr. Alexandra Phillips, Assistant Teaching Professor; Dr. Hunter Lenihan, Professor

UC Santa Barbara Marine Science Institute: Dr. Chris Jerde, Associate Researcher; Dr. Kyle Emery, Assistant Researcher

Woods Hole Oceanographic Institute: Dr. Tom Bell, Assistant Scientist