ANALYZING GLOBAL ANTHROPOGENIC IMPACTS ON ATOLL ISLAND CHANGE OVER THE LAST THREE DECADES

Within our lifetime, climate change has the potential to drastically alter coastal resiliency. Atoll island nations are particularly vulnerable to climate change: from increasing ocean temperatures (causing coral die-off), to ocean acidification (decreasing coral resiliency), to increasing SLR. We must understand what will happen to the atoll islands since the land is where people live. However, we lack comprehensive understanding about the impact of humans on these islands and their future survival in the Anthropocene. This uncertainty in predictions hinders local communities’ preparation for the future; we must understand how atoll islands respond and evolve with varying levels of anthropogenic impact. Previous research has found that on specific atolls, human impact can account for over 60% of measured island area change, such that humans are the dominant driver of geomorphic work on these islands. However, given the high spatial heterogeneity of anthropogenic impacts on islands (humans tend to inhabit only one island of the atoll with the rest – upwards of 100 island remain uninhabited) and the focus of studies on a single atoll, we must automate this analysis and apply on a global scale. Utilizing Landsat and Google Earth Engine, we automatically measured atoll island morphometrics (such as atoll island area, perimeter, and centroid location) and tracked the morphometric changes over the last thirty years. By utilizing temporal composites, we removed cloud impacts and georeferencing uncertainties in Landsat imagery. We created an evenly-spaced timeseries of Landsat composites, where we analyzed morphometrics on each atoll island for each atoll through time. We compared changes in atoll island area, perimeter, and other morphometrics for inhabited atoll islands to uninhabited atoll islands. This analysis allowed us to track patterns of human impacts on atolls and identify hotspots of island accretion, erosion, or variability. By using Google Earth Engine, Landsat, and python, we can scale our analysis to the entire globe enabling us to better understand the impact of humans on atoll island change.