How to deal with ground truthing affected by human-induced habitat change?: Identifying high-quality habitats for the Critically Endangered Red Siskin.

Ada Sánchez‐Mercado,Brian Coyle,Michael J Braun,José Rafael Ferrer‐Paris,Jhonathan Miranda,Samuel Toro,Kathryn M Rodríguez‐Clark,Arlene Cardozo‐Urdaneta

doi:10.1002/ece3.3628

Abstract

Species distribution models (SDM) can be valuable for identifying key habitats for conservation management of threatened taxa, but anthropogenic habitat change can undermine SDM accuracy. We used data for the Red Siskin (Spinus cucullatus), a critically endangered bird and ground truthing to examine anthropogenic habitat change as a source of SDM inaccuracy. We aimed to estimate: (1) the Red Siskin's historic distribution in Venezuela; (2) the portion of this historic distribution lost to vegetation degradation; and (3) the location of key habitats or areas with both, a high probability of historic occurrence and a low probability of vegetation degradation. We ground‐truthed 191 locations and used expert opinion as well as landscape characteristics to classify species' habitat suitability as excellent, good, acceptable, or poor. We fit a Random Forest model (RF) and Enhanced Vegetation Index (EVI) time series to evaluate the accuracy and precision of the expert categorization of habitat suitability. We estimated the probability of historic occurrence by fitting a MaxLike model using 88 presence records (1960–2013) and data on forest cover and aridity index. Of the entire study area, 23% (20,696 km2) had a historic probability of Red Siskin occurrence over 0.743. Furthermore, 85% of ground‐truthed locations had substantial reductions in mean EVI, resulting in key habitats totaling just 976 km2, in small blocks in the western and central regions. Decline in Area of Occupancy over 15 years was between 40% and 95%, corresponding to an extinction risk category between Vulnerable and Critically Endangered. Relating key habitats with other landscape features revealed significant risks and opportunities for proposed conservation interventions, including the fact that ongoing vegetation degradation could limit the establishment of reintroduced populations in eastern areas, while the conservation of remaining key habitats on private lands could be improved with biodiversity‐friendly agri‐ and silviculture programs.

Highlights

One of the most promising applications of species distribution modeling (SDM) for conservation management is ranking areas by estimated habitat quality (Kramer-Schadt, Revilla, & Wiegand, 2005)
Species are not always present where high occurrence probabilities are predicted (Peterson et al, 2011). This mismatch between modeled predictions and field observations may result from problems with the Species distribution models (SDM) itself, such as conceptual errors, limitations in variable selection (due a poor understanding of factors driving species distribution or use of outdated presence information with respect to environmental predictors used (Peterson et al, 2011)
We combined SDM and Random Forest model (RF) results to examine the drivers of model mismatch, addressing three basic questions about species habitat availability: (1) How extensive was the historic distribution of Red Siskins in Venezuela? (2) How widespread is habitat loss as measured by vegetation degradation? and (3) Where are remaining key habitats, or areas with both high historic occurrence probability and low landscape transformation? In addition, to demonstrating the usefulness of our approach for identifying key habitats for a threatened, elusive, and poorly studied species, we aimed to examine implications for the threat status of the species and consider the consequences of our results for the design of effective strategies for reintroduction, including habitat conservation, which may be needed to achieve self-sustaining populations of the Red Siskin

Summary

| INTRODUCTION

One of the most promising applications of species distribution modeling (SDM) for conservation management is ranking areas by estimated habitat quality (Kramer-Schadt, Revilla, & Wiegand, 2005). If land cover transformation is gradual species detection is still possible where habitat has been partly degraded but not lost Such “snapshot detections” of species occurrence may generate a misleading picture of relative habitat quality, which in turn could have disastrous consequences if, for example, the model is used to identify areas for the reintroduction of captive- bred endangered species (Lahoz-M onfort, Guillera-Arroita, & Wintle, 2014). In such situations, a more nuanced, nonbinary approach to field validation is essential. We combined SDM and RF results to examine the drivers of model mismatch, addressing three basic questions about species habitat availability: (1) How extensive was the historic distribution of Red Siskins in Venezuela? (2) How widespread is habitat loss as measured by vegetation degradation? and (3) Where are remaining key habitats, or areas with both high historic occurrence probability and low landscape transformation? In addition, to demonstrating the usefulness of our approach for identifying key habitats for a threatened, elusive, and poorly studied species, we aimed to examine implications for the threat status of the species and consider the consequences of our results for the design of effective strategies for reintroduction, including habitat conservation, which may be needed to achieve self-sustaining populations of the Red Siskin

| METHODS

Findings

| DISCUSSION