ABSTRACT1. Worldwide, there are concerns about the potential impact of a growing tourism industry on the behaviour and ecology of sharks. Several shark species are key attractions in the scuba diving industry in both the Gala´pagos Marine Reserve (Ecuador) and the Sanctuary of Fauna and Flora Malpelo (Colombia).2. In this study, the reactions of five species of shark to the behaviour of scuba divers were investigated in the Gala´pagos and Malpelo.3. Four discrete categories of shark reaction (evasion, spontaneous approach, alert and no reaction) were identified and analysed against five categories of diver behaviour (direct approach, camera flash, sudden movement, noise and simple presence), two categories of observation strategy (still and movement) and the distance of the focal diver group to the sharks.4. The type of reaction in the sharks was determined by the behaviour of the divers, their distance to the fish and, especially, by their observation strategy. Shark reactions varied between species and locations and divers acted more intrusively towards those species they were not afraid of, and which they could access more easily due to the conditions of their environment.5. 'Direct approach', 'sudden movement', the observation strategy 'moving' and distances closer than 4 m between divers and sharks elicited stronger behavioural responses by the species targeted. Recommendations are given for the preferred actions divers should employ in order to diminish the effects on the target species analysed in this study.

, was directed at determining the abiotic environmental factors governing the distribution of the bottlenose dolphin (Tursiops truncatus) and the pan-tropical spotted dolphin (Stenella attenuata) in Golfo Dulce. The spotted dolphin was associated with significantly deeper zones (t-test =-9.761; p < 0.001, n = 202) and with higher salinity (t-test =-3.538; p = 0.001; n = 202) than the bottlenose dolphin. The combination of environmental variables in the distribution areas of each species in Golfo Dulce was different, both spatially (F = 9.724, df = 12; p < 0.001) and seasonally (F = 9.735; df = 12; p < 0.001). Of all the environmental variables considered, Multiple Discriminant Analysis showed depth to be the main factor differenciating the distribution areas of the two species. The relationship between group size in each species, depth, and the location of their distribution areas in relation to the shore were analyzed and evidence regarding these aspects and variations in foraging strategies and prey composition was proposed to explain the differences found in their spatial distributions. Seasonal variations in salinity in the distribution areas and their inverse relationship with the time of the year were analyzed in relation to their possible combined effects on prey composition during different seasons in order to explain the seasonal variations in distribution patterns.

Distribution and environmental conditions related to the behavior in the bottlenose dolphin (Tursiops truncatus) and the spotted dolphin (Stenella attenuata) (Cetacea: Delphinidae) in Golfo Dulce, Costa Rica. Habitat characteristics influencing behavior in animal species vary locally. The influence that a particular environmental characteristic can have on a species depends not only on other variables, but on morphological, physiological and social conditions of that species. in this study, developed from June 1996 to July 1997, i studied whether specific behaviors are related to particular distribution areas and environmental factors in the bottlenose (Tursiops truncatus) and the spotted dolphin (Stenella attenuata). The study area was covered along oblicuous linear transects, and the behavior of single groups was observed from 15 min to 5 h. Environmental factors such as depth, temperature, salinity and distance from shore, among others, were considered. For the bottlenose dolphin, foraging/feeding activities showed exclusive coincidence with river mouths, coral reef and mangrove areas, while social and milling activities where seen close to feeding areas. Traveling occurred along different points parallel to the coast, with a low percentage of cases across the gulf (16.56 %), suggesting that the bottlenose rarely crosses from one side to the other. in the spotted dolphin, several behaviors were observed simultaneously in the schools and it was not possible to associate areas with particular behaviors. The lack of significant relationships among activities and particular environmental variables (ANOvA tests) is attributed to three aspects: (a) transitions among activities generally occurred into a low variable area, (b) dolphins often traveled along large areas without changing activities and (c) environmental conditions in Golfo Dulce are homogeneous. in the two species the highest average in the number of individuals per group corresponded to the category of active socializing, followed by traveling, passive socializing and feeding. in the case of the bottlenose dolphin, the smallest group size was associated with feeding activities (ANOvA, F= 2.624, p=0.037, n= 156, df= 4), while in the spotted dolphin the smallest group size corresponded to milling activities (ANOvA,

ABSTRACT1. Understanding spatial interactions between specific sources of anthropogenic impact on cetaceans and the ecological conditions intrinsic to these mammals, has several applications for conservation and management. Here cetacean habitats are classified into vulnerability levels by combining human development metrics and cetacean characteristics, allowing areas identified as more vulnerable to be given management priority.2. Ten bio-physical and anthropogenic factors were identified as key shapers of cetacean habitats and, using the Gulf of California as a case study, 18 variables were defined to measure those factors. Principal component analysis (PCA) and GIS were used to distill the relationships across all variables, in order to obtain a vulnerability index. Aggregates of variables were also investigated to map the spatial distribution of anthropogenic impacts and cetacean characteristics (species richness, relative abundance of encounter, residence condition, diet specificity and threat status). For this latter purpose, separate indices representing five 'categories' (pollution from land sources, urban development, potential for accidents, resource depletion and fragility of cetacean areas) and three 'vulnerability components' (risk of hazards, intrinsic resilience and damage) were calculated.3. The methodology clearly differentiated the study area according to levels for each 'category' and 'vulnerability component', as well as to levels of vulnerability of cetacean areas in relation to human activities; and identified those variables with the greatest contributions, without recourse to the subjective criteria common in other approaches. The analyses were conducted using environmental units generated by the Marine Ecological Zoning Plan for the Gulf of California (OEMGC, by its Spanish acronym) which tied model outputs to this planning framework. The variables selected were easily obtainable through governmental and research institutions which make them suitable for adaptive management purposes.