Aerial surveys for abundance and distribution assessment of the threatened franciscana (Pontoporia blainvillei)

Abstract

Habitat loss and overexploitation are among the primary causes of contemporary high levels of biodiversity loss. Understanding and predicting the multiple effects of human activities on species extinction are challenging tasks that highly depend on robust data on population numbers. Mortality due to bycatch represents the primary driver of franciscana (Pontoporia blainvillei) population reduction. The species is listed as "Critically Endangered" in Brazil, and "Vulnerable" in the IUCN Red List of Threatened Species. Estimating the abundance of the franciscana has been recommended as priority by local and international organizations. The present thesis aimed to develop and implement correction factors for estimating abundance of franciscanas from aerial surveys. Surveys were conducted along the southern and southeastern Brazilian coast, covering the latitudinal range, from the coast up to the 30 m isobath, of the new limits proposed for FMA II and the southern gap in distribution of the franciscana (Florianópolis, SC - Cabo Frio, RJ) as well as the latitudinal range, from the coast up to the 20 m isobath, of the FMA Ia population and the northern distribitional gap (Barra de Itabapoana - Itaúnas, ES). These results corroborate previous studies and indicate that aerial surveys are highly effective to record data on groups' occurrence and total number of individuals; however, if not accounted for visibility and group size bias, abundance estimates can be underestimated by up to ~5 times. Visibility bias (imperfect counts of submerged individuals) is the main factor affecting abundance estimates of franciscanas from aerial survey data. MRDS methods and experiments with concomitant surveys from aerial- and surface-based platforms indicated that observers typically misperceive ~20% of the available franciscana groups (perception bias). Experiments conducted from a helicopter showed that ~60% of the groups will not be available to be counted during a passing mode aerial survey. Bias in group size estimates represents ~30% of the total bias in abundance estimates from aerial surveys. The high speed of the aircraft associated with the differential availability between individuals within the same group are the main factors to affect group size estimates. The abundance estimated for the whole FMA II (𝑁̂ = 6,827, CV = 0.28) as well as to the new proposed subdivisions FMA IIa (𝑁̂ = 1,915, CV = 0.32) and FMA IIb (𝑁̂ = 4,353, CV = 0.24) indicates that mortality due to bycatch is unsustainable. The coastline of FMA II corresponds to one of the most developed and populated region along the species range and therefore a threat to franciscana conservation. The northernmost franciscana population FMA Ia (𝑁̂ = 893, CV = 0.30) is potentially demographically isolated from all franciscana populations. It is proposed that franciscanas in FMA Ia qualify for listing as “Endangered” under the IUCN Red List Categories and Criteria. Improvements in correction factors for aerial surveys and the use of other sample methods (e.g. acoustic methods) should be conducted and integrated analysis performed. In addition, it is encouraged that conservation plans for the franciscana be implemented, along with an abundance and distribution assessment.