Changes in Tursiops truncatus Distribution and Behavior in the Drowned Cayes, Belize, and Correlation to Human Impacts

Human interaction greatly influences the behavior and distribution of bottlenose dolphins (Tursiops truncatus). This project focuses on the distribution and behavior of bottlenose dolphins in the Drowned Cayes, Belize. Prior to the 2000s, the area was relatively undeveloped and undisturbed and had minimal human activity. Since the turn of the millennium, development and ecotourism activity has flourished in the area, increasing by more than 800,000 visitors from 1998-2006. Boat-based surveys were conducted in 2015 and were combined with previous survey data collected from 2005-2012 and compared to behavioral survey results from 1999-2000. Total dolphin observation time as a percent of total survey time and average number of dolphins per sighting were 17.2% and 2.7 in 1999-2000 and 10.8% and 1.6 for 2005-2015. The low number of dolphins and the low observation times suggest that the dolphin population in the Drowned Cayes have decreased since the 1990s. Eighty-nine percent of the total observation time for 2015 occurred on days in which there were zero cruise ships in the area suggesting that this decline may be in relation to increased human activity. Furthermore, foraging was the main behavior observed for both 1999-2000 and 2005-2015 data sets, suggesting that the Drowned Cayes area is used as a foraging ground. However, in 1999-2000 the foraging percentage was significantly higher than the 2005-2015 data set, dropping 28.9% and there was a 23.6% increase in traveling behavior between the two data sets. This could be a result of increased human activity. Additionally, survey photographs and results were used in the creation of the first dolphin photo identification database for the country. The guidelines used for photo analysis for photo quality and fin distinctiveness were tested to determine if they are easy to use and give consistent and reliable results regardless of judge. An intraclass correlation model calculated substantial agreement (ICC = 0.7) between judges’ scores, demonstrating consistent results, regardless of experience level. Therefore, the guideline can be used as a standard among multiple researchers