Entry for:NZMSS Student Awards
Survival rate is an important demographic parameter that can be used to assess population trends if it is suitably precise. Survival rate of Hector’s dolphins (Cephalorhynchus hectori) at Banks Peninsula, NZ has been estimated from data collected using photo-ID methods. This involves photographing unique marks on individuals (e.g. nicks on the dorsal fin). Some marks appear to be caused by non-fatal encounters with fishing gear. Therefore, conservation measures which reduce fishing effort (e.g. the Banks Peninsula Marine Mammal Sanctuary), have resulted in a decline in the proportion of individuals that are marked (the mark rate). Because of impacts on sample size of individuals in the mark-recapture dataset, we hypothesised that a lower mark rate would decrease the precision of estimates of survival rate. To test this hypothesis, mark rates of 5-15% were simulated by resampling a dataset of 516 marked dolphins collected between 1990 and 2009. A Cormack-Jolly-Seber model was applied to each simulated dataset to estimate survival. This process was repeated to produce 10,000 bootstrap replicates of survival rate and its precision for each mark rate. The median CV of survival rate increased by 42% as mark rate reduced from 15% to 5%, with the most rapid declines in precision occurring below a mark rate of 11%. This study demonstrates that a decline in mark rate can reduce precision of survival rate estimates, and therefore has implications for research directed at detecting population change. Additionally, the methods of the simulation may be applied to other long-term photo-ID studies.