Ecological Archives A022-024-A9

Douglas J. McCauley, Kevin A. McLean, John Bauer, Hillary S. Young, and Fiorenza Micheli. 2012. Evaluating the performance of methods for estimating the abundance of rapidly declining coastal shark populations. Ecological Applications 22:385–392.

Appendix I. Role of behavioral biases in shark surveys.

Much consideration has been given to the biases that may be created when using human observers to inventory icthyofauna (Kulbicki 1998, Watson and Harvey 2007). Observer bias is highly relevant when surveying sharks, which may perceive observers as predators, competitors, or potential prey. Our work with diver-based surveys (belt transects and point counts) reveals that these biases are not static but may vary (1) within a single survey, (2) across multiple surveys, and (3) between sites.

At unfished Palmyra qualitative observations suggest that the shark Carcharhinus amblyrhynchos exhibits a mild attraction to divers at the beginning of a dive, but rapidly loses interest and disperses from the survey area as the dive progresses. This initial behavioral attraction could affect the outcome of density estimates: surveys conducted at the start of a dive might provide higher estimates of shark density than paired replicates conducted at the end of dive. In our survey test protocols at Palmyra, point counts always preceded belt transects. This selection of ordering may have contributed to the elevated field density estimates that point counts yielded relative to belt transects. Outputs from our simulation (that operates without the confounding influence of behavioral biases of sharks for divers) indicate, however, that structural differences between belt and point counts are the primary cause of these differences. To reduce the confounding effects that this positive anthrotaxis may have on field derived abundance estimates, it may be advisable to begin shark surveys 5–15 min after divers initially enter the water. Delayed survey start times may be particularly useful at remote sites where sharks are especially naïve of humans and attraction to divers is pronounced.

Such small-scale oscillations of intrigue and curiosity abatement may also become manifest at larger temporal scales. Sharks at remote sites that are unaccustomed to humans may display an initial strong attraction to divers upon first contact that may diminish after sustained human activity. We began our surveys of sharks at Palmyra during the first year of intense researcher activity at the Palmyra Atoll field station. Diving activity prior to this point at Palmyra had been relatively light. During the several month duration of this study we conducted replicate surveys every several weeks at the same reef sites. To examine whether sharks’ interest in divers may have saturated and waned during the course of this study, we examined trends in the relationship between shark density (all sharks pooled) and time (expressed as Julian calendar days). Shark density estimates generated using diver-based point count surveys (our most data rich diver survey) showed a mild but significant decline over the duration of the study (R² = 0.11, P = 0.03). A decline of this kind could be the result of sharks becoming habituated and less interested in the regular presence of divers at the same reef site, or it could be the outcome of a biologically meaningful reduction in shark density over time. Examinations of the relationship between diver-independent video surveys showed no such declines over time (R² = -0.03, P = 0.61) suggesting that perhaps the density reductions observed during point counts may in fact have been caused by changes in shark behavior towards survey divers. If this type of bias is common in other shark species as well, it could contribute to the overestimation of shark densities in remote and rarely frequented sites. The use of stationary unbaited video surveys provides a useful means to test for and work around such biases.

Large vertebrates, marine and terrestrial, exhibit very different behaviors in environments where they are hunted (Kilgo et al. 1998, Cole 1994). Because a common application of shark surveys is to compare sites with shark fishing to unfished sites, this potential bias also deserves consideration. Interestingly, during research surveys we conducted at Tabuaeran Atoll, a fished atoll 350 km from Palmyra, the only two C. amblyrhynchos that we observed during > 80 hours of underwater observation fled from us immediately after we entered the water. Avoidance behaviors need not be this extreme to influence density estimates. Mild increases in wariness by sharks for divers could push them just beyond the defined boundaries of survey areas reducing the frequencies at which they are encountered. Differences in inter-site behavior of this type might further depress density estimates of sharks that are already rare at fished sites.

Overall, biases of such as discussed above need to be carefully examined when designing protocols to survey coastal sharks and when interpreting abundance outputs from these studies working with these subjects.


Cole R.G. 1994. Abundance, size structure, and diver-oriented behaviour of three large benthic carnivorous fishes in a marine reserve in Northeastern New Zealand. Biological Conservation 70:93–99.

Kilgo, J. C., R. F. Labisky, and D. E. Fritzen. 1998. Influences of hunting on the behavior of white-tailed deer: implications for conservation of the Florida panther. Conservation Biology 12:1359–1364.

Kulbicki M. 1998. How the acquired behaviour of commercial reef fishes may influence the results obtained from visual censuses. Journal of Experimental Marine Biology and Ecology 222:11–30.

Watson, D. L., and E. S. Harvey. 2007. Behaviour of temperate and sub-tropical reef fishes towards a stationary SCUBA diver. Marine and Freshwater Behaviour and Physiology 40:85–103.

[Back to A022-024]