Assessing conservation status of remnant mammal populations in a Neotropical cloud forest

Abstract

Neotropical cloud forests are some of the most biodiverse and yet threatened ecosystems. Terrestrial mammals play key roles delivering numerous ecosystem services within such systems including seed dispersal and nutrient cycling. Yet like the habitat in which they reside, some mammal species are also highly threatened; sensitive to deforestation and hunting. Protected areas are a cornerstone of conservation policy and National Parks have notionally one of the highest status designations, yet in many instances such designation may not be effective in conserving biodiversity. This study assessed the efficacy of Cusuco National Park in preserving Neotropical cloud forest mammal fauna in northwest Honduras, Central America. A comprehensive list of all non-volant mammal species was assembled, reflecting a highly biodiverse ecosystem. The park’s more protected core zone was more biodiverse with high mammal abundance than its less protected buffer zone. Species Distribution Models (SDMs) suggested that anthropogenic pressures were the principal driving factors in determining species ranges throughout the park rather than local environmental variation. General Additive Models (GAMs) were used to assess temporal trends in mammal populations revealing all groups, most notably large hunted species, declining dramatically in recent years with near complete mammalian defaunation of the park predicted as early as the mid-2020s. Analysis of camera trapping surveys indicated that large mammalian species detections were negatively associated with proximity to scientific research stations. Additionally smaller species were negatively associated with proximity to survey transects used for monitoring and surveillance of mammal tracks and signs. These results suggest a need for conservationists to be mindful of the impact of their activities on the systems they aim to protect. Analysis of satellite imagery captured widespread deforestation within the park with the highest rates of loss in recent years. Under ‘business-as-usual’ it is predicted that 20% of forest within the park will be lost by 2040. Areas of greatest forest cover are most likely to be lost especially in the northeast of the park’s buffer zone. Spatial predictions of the probability of both mammal occurrence and deforestation risk should empower government, law enforcement, forest managers and conservation organisations to target conservation resource investment into enforcing legal protections in-and-around villages associated with greatest risk of illegal activities

Date of Award	Jul 2020
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Northern Ireland Department for the Economy
Supervisor	Neil Reid (Supervisor) & William Montgomery (Supervisor)