Predicting and Preventing Elephant Poaching Incidents through Statistical Analysis, GIS-Based Risk Analysis, and Aerial Surveillance Flight Path Modeling. (March 2016)

• Record Type: Journal Article
• Authors: Shaffer, Michael J.; Bishop, Joseph A.
• Is Part Of: Tropical conservation science. Volume 9:Number 1(2016)
• Page Start: 525
• Page End: 548
• Publication Date: 2016-03
• Subjects: risk analysis -- geospatial modeling -- African elephant -- poaching -- GIS -- UAV -- Loxodonta; analyse des risques -- modélisation géospatiale éléphant d'Afrique -- braconnage -- GIS -- UAV -- Loxodonta; Rain forest conservation -- Periodicals; Rain forest ecology -- Periodicals; Rain forest conservation; Rain forest ecology; Electronic journals; Periodicals; 577.34
• DOI: 10.1177/194008291600900127
• Languages: English
• ISSNs: 1940-0829

Abstract

The illegal ivory market poses a serious threat of extinction to the African elephant ( Loxodonta africana ). Conservation efforts to protect elephants are challenging due to their vast habitat range. Conservation efforts focused on high risk areas provide a more efficient management method than randomly patrolling entire habitat areas. The use of unmanned aerial vehicles is having a profound effect on conservationists' abilities to detect and stop poachers. This study provides methods for identifying high risk elephant poaching areas and for modeling drone surveillance capabilities. Point pattern analyses were conducted to identify spatial distribution patterns of elephant poaching incidents in the Tsavo National Parks area in Kenya. We performed geospatial analyses on the physical environment to create a risk map based on how roads, water features, and land cover correlate to poaching incidents. We also modeled drone flight paths for surveillance of high risk areas based on aircraft flight characteristics and the horizontal field of view for a selected infrared camera. We conclude that poaching incidents were geographically clustered and followed a deterministic (predictive) process. Poaching incidents correlated with close proximity to roads and water features, and were predominately upon specific land cover types. Drone flight modeling determined the number of flights and duration necessary to cover a specific high risk area. The procedures discussed can be applied using a combination of both GIS software and freely available statistical analysis software. The analysis and risk identification will enable conservation groups with limited budgets to improve the efficiency of their anti-poaching efforts.