Playback point counts and N‐mixture models suggest higher than expected abundance of the critically endangered blond titi monkey in northeastern Brazil

Abstract

Effective management of threatened species requires accurate population size estimation and monitoring. However, reliable population size estimates are lacking for many endangered species. The critically endangered blond titi monkey (Callicebus barbarabrownae) is an endemic primate of the Caatinga biome in Northeastern Brazil.A previous assessment based on presence-only data estimated a minimum population size of 260 mature individuals in 2,636 km 2 , and studies based on visual records suggested very low local relative abundance. However, this cryptic species is known to be difficult to visually detect. We played back recordings of C. barbarabrownae loud calls to count the number of responding groups in 34 sampling sites during 9 consecutive days in a 221-km 2 study area. Repeated group counts at sites were used in N-mixture models, which account for imperfect detection, to estimate the number of groups in relation to dry forest area and distance to villages. We estimated a total of 91 groups in the study area. Considering the mean number of adults per group as three, we estimated a population of 273 adult individuals, resulting in a density of 2.3 individuals/km 2 in the dry forest habitat. Detection probability was four times higher for surveys conducted between sunrise to midmorning than between midmorning to sunset. We also found that C. barbarabrownae abundance increases with increasing dry forest area and increasing distance to the nearest village, indicating the need to promote dry forest restoration in the Caatinga. As our results suggest a larger population of C. barbarabrownae than had been previously estimated for its entire distribution, our results suggest a need for similar assessments in other areas to reliably estimate the total population size. This study demonstrates how playback surveys coupled with N-mixture models can be used to estimate population sizes of acoustically-responsive primates, and thus contribute to more effective conservation management. K E Y W O R D SCaatinga, Callicebus barbarabrownae, dry forest, endangered species, imperfect detection, IUCN Red List

Elusive species and where to find them: assessment of survey protocols for primates habitat selection

“…blond titi monkey ( Callicebus barbarabrownae ); Coelho et al . (2020), Bengal slow loris ( Nycticebus bengalensis ); Oliver et al . (2020), or northern white-cheeked gibbon ( N. leucogeny ); Syxaiyakhamthor et al .…”

“…We could also use the random walk grid survey to study habitat selection of populations easier to detect and count (Plumptre et al 2013), in order to accurately estimate abundance and density by using N-mixture models (Royle 2004) for threatened primate species (e.g. blond titi monkey (Callicebus barbarabrownae); Coelho et al (2020), Bengal slow loris (Nycticebus bengalensis); Oliver et al (2020), or northern white-cheeked gibbon (N. leucogeny); Syxaiyakhamthor et al 2020) and even forest-dwelling communities (Rovero et al 2014;Brodie et al 2018).…”

Historical antecedents and recent innovations in pitheciid (titi, saki, and uakari) feeding ecology

“…In addition to Chiropotes and Cacajao that use very large areas, the most at-risk pitheciid populations appear to be Callicebus spp. inhabiting Atlantic Forests and forest patches in eastern Brazil where populations are impacted by urban development (e.g., Coelho, et al, 2020). It would be tragic to lose populations of pitheciids before we fully appreciate their behavioral and ecological diversity and understand how best to protect them.…”

An endangered native species is being replaced by hybrids: the new normal for primate conservation in the Brazilian Atlantic Forest

“…N-mixture models are hierarchical models that allow estimating abundance and evaluating factors that can in uence the probability of detecting unmarked individuals, correcting for imperfect detection and/or false absences in abundance estimates (Royle 2004). This class of models has proven to be robust for estimating the abundance of several taxa, including primate species (Joseph et al 2009;Froese et al 2015;Senzaki et al 2015;Belant et al 2016;Coelho et al 2020).…”