Shifting landscape suitability for cocaine trafficking through Central America in response to counterdrug interdiction

Abstract

Cocaine traffickers, or ‘narco-traffickers’, successfully exploit the heterogeneous landscapes of Central America for transnational smuggling. Narco-traffickers successfully adapt to disruptions from counterdrug interdiction efforts by spatially adjusting smuggling routes to evade detection, and by doing so bring collateral damages, such as deforestation, corruption, and violence, to new areas. This study is novel for its integration of landscape suitability analysis with criminological theory to understand the locations of these spatial adaptations by narco-traffickers as intentional, logical, and predictable choices based on the socio-environmental characteristics of Central America’s landscapes. Multi-level, mixed effects negative binomial regression models predict the suitability of landscapes for cocaine trafficking across 17 departamentos (the unit of analysis) in Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama from 2007 to 2018. Informed by long-term research in the region, independent variables included proximity to roads, country borders, and international ports, indigenous territories, population density, and protected areas. The year of peak interdiction (measured by kg of cocaine seized) in each departamento was used to analyze spatial shifts in landscape suitability before and after maximum counterdrug interdiction pressure. We find that areas with lower population density and closer proximity to international borders became more suitable following peak interdiction—i.e, they are more likely to be sought out by traffickers seeking to avoid further disruptions from counternarcotic efforts. Additionally, indigenous territories were disproportionately exploited as cocaine trafficking routes following significant interdiction activities by law enforcement. While interdiction may reduce the suitability of targeted locations, it can also unintentionally increase the attractiveness of other locations. Our study pushes criminological theory through its application to a unique space/time context, and it advances land system science by considering landscape suitability for logistical rather than productive uses. Policy implications are clear. Since interdiction resources are limited relative to the overall amount of trafficking activity, knowing which landscape features are viewed as suitable by traffickers can in the short-term guide interdiction deployment strategies, and in the longer term build strategies to mitigate associated harms from trafficking where they are most likely.