Date of Completion


Document Type

Honors College Thesis



Thesis Type

Honors College

First Advisor

Brendan Fisher

Second Advisor

Eric Hochberg

Third Advisor

Daniel Erickson


coral, remote sensing, Hawaii, AVIRIS, coral cover, thermal stress


Coral reefs are declining at unprecedented rates across the world, threatening ecosystem goods, services, and functionality. Previous research suggests thermal stress is a major driving factor of coral decline, however, these studies primarily draw conclusions from small-scale, in situ data collection. This study represents the first effort to quantify large-scale change in coral cover across the Hawaiian Islands over an extended time period using remote sensing. Hyperspectral imagery from the Airborne Visible-Infrared Imaging Spectrometer (AVIRIS) was used to quantify benthic cover of reef ecosystems across the Main and Northwestern Hawaiian Islands in 2000 and 2017. A total of 320.28 km2 of benthic area was mapped and analyzed across Kauaʻi, Oʻahu, Lanaʻi, Maui, and French Frigate Shoals. Average coral cover dropped from approximately 23% to 8% over the 17-year time period, representing a relative decline of more than 65%. Various thermal stress metrics were obtained through NOAA’s Coral Reef Watch bleaching area alert data and compared to the coral cover data. Adjusted R2 values ranged between 0.02 and 0.043, indicating weak correlations between bleaching area alert metrics and absolute change in coral cover. This study offers preliminary evidence that Hawaiian coral reefs are declining not only on a small-scale, but across entire reef ecosystems. However, thermal data alone cannot explain the decline in coral cover observed. Although the question of what causes this decrease in cover remains, hyperspectral remote sensing may be the missing link. Only once we understand coral change at the reef scale can we hope to conserve these ecosystems.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.

Available for download on Saturday, May 17, 2025