Improving analytical approaches for estimating forest deer density

• Grant number: 22/06502-7
• Support Opportunities: Scholarships abroad - Research Internship - Master's degree
• Start date: August 29, 2022
• End date: February 24, 2023
• Field of knowledge: Biological Sciences - Ecology - Applied Ecology
• Principal Investigator: José Maurício Barbanti Duarte
• Grantee: Jeferson Lucas Sousa Freitas
• Supervisor: Philip Andrew Stephens
• Host Institution: Faculdade de Ciências Agrárias e Veterinárias (FCAV). Universidade Estadual Paulista (UNESP). Campus de Jaboticabal. Jaboticabal , SP, Brazil
• Institution abroad: Durham University (DU), England
• Associated to the scholarship: 21/02087-2 - Who is the red Mazama in Rio Doce State Park, Minas Gerais? A genetic non-invasive approach and integrative taxonomy, BP.MS

Abstract

Deer are key ecological components of many ecosystems and while some species are declining, others are considered overabundant. Either to increase threatened populations or to reduce overabundant ones, reliable estimates of population density are essential to inform management effectively. Population estimates of elusive species, such as small neotropical forest deer, are difficult to obtain because the most common methods, based on aerial or land-based direct sightings, have not proved effective. As an alternative, a viable option is the identification of faecal samples to species level, followed by counting the faeces in a pre-determined area of species occurrence, often with the assistance of a faecal detection dog. One method used for this purpose is the faecal standing crop (FSC), for which the formula is D = ((N/(Wd*L))/P)/(R*Min(T,A)), where: N = number of samples, L = length of transect in meters, Wd = faecal detection dog sampling strip width in meters, P = dog detection effectiveness, R = species defecation rate per day, T = faecal mean persistence time in days and A = faeces maximum age for dog detection in days. To date, however, the propagation of error from the estimated parameters in the formula has not been formalised. This project aims to improve analytical approaches for estimating forest deer density using the FSC method. This will be done by: 1) correcting the value of L to avoid double-counting owing to the search methodology; 2) incorporating and propagating uncertainty in the parameters Wd, P and A; and 3) simulating theoretical scenarios to verify the proposed changes in the FSC method. By doing so, we hope to gain a better understanding of how each parameter influences the final density value and to obtain a more reliable estimate, enabling more robust inferences. (AU)