PREDICTIVE ACCURACY OF ANNUAL EROSION HAZARD RATES ON MAUI, HAWAII

Beach erosion has direct consequences for Hawaii’s tourist-based economy, which depends largely on the attraction of beautiful sandy beaches. Within the last century, however, 24% of beaches in Oahu and 26% of beaches in Maui have been narrowed or completely lost, threatening tourism and construction development. In order for the counties and state of Hawaii to implement coastal regulations to prevent infrastructure damage, it is necessary to find a statistical methodology that accurately delineates annual erosion hazard rates specific to Hawaii. Data are derived from ortho-photo mosaics of aerial photography described by Fletcher et al. (elsewhere in this volume). To evaluate statistical methods (e.g., Linear Regression, End Point Rate, and Jackknifing), this study determines the predictive accuracy of various calculated erosion rates using (1) temporally truncated data to forecast and hindcast known shorelines, and (2) a comparison between erosion rates with statistical outliers identified using least median squares and a priori outliers (e.g., storms). This study also examines the often-overlooked methodology of weighted least squares (WLS). WLS incorporates unique, but known uncertainties of each historical shoreline point by propagating them into the erosion rate - an important feature in quantifying uncertainties of future shoreline positions. Data from different types of beaches in Maui (e.g., pocket beaches, cuspate headland beaches, and narrow fringing reef-fronted beaches) are compared and results will be presented at the conference.