Abstract:
The whale shark was first described scientifically in 1828 by Andrew Smith. But even though
it is the world’s largest fish (Rowat & Brooks, 2012, p. 1019; Bradshaw, et al., 2008, p. 1895)
there had been only 320 credible reports of whale shark sightings until 1985 (Stevens, 2007,
p. 4). However, since then “a huge increase in recreational diving and boating activity
around the world” (Stevens, 2007, p. 5) has led to the discovery of several whale shark
aggregation sites with a reliably predictable abundance of whale sharks. At most of these
sites, whale shark tourism has become a popular and economically important activity.
The non-consumptive use of the whale shark has fostered the conservation of the species
and its habitat on a local scale e.g. the declaration of marine protected area to protect
important foraging grounds of the whaleshark. But the intrusion of humans into the habitat
of the whale shark can also have adverse effects, especially if the whale shark aggregation
represents a common pool resource. Since the knowledge of the ecology and biology of
whale sharks is still very incomplete, it is also unclear how tourism affects whale sharks and
which factors have a major influence on the extent of the negative impacts.
This knowledge gap is also one of the biggest obstacles to the effective regulation of whale
shark tourism, which is reflected in the inconsistent regulations and standards at the
different whale shark tourism destinations.
In this thesis, a data-driven approach to the regulation of whale shark tourism is presented
using the example of the whale shark tourism in the South Ari Marine Protected Area
(SAMPA) in the Maldives. 964 whale shark encounters are evaluated and four random forest
models are developed to identify the most important factors influencing the positive course
of a whale shark encounter. Based on the results of the models and successful aspects of
regulations from other whale shark destinations a regulation proposal for the whale shark
tourism in the SAMPA is developed.
