A brown cow grazing in a green meadow in Colombia
A brown cow grazes in a green meadow in Colombia, where researchers used recently released land cover data to find out how armed conflict influences land use. Credit: iStock.com/Santypan

Colombia is one of the most biodiverse countries on Earth, boasting pristine rain forests, wetlands, and mountain ranges that harbor some species not found anywhere else. More than 50 years of violent, internal conflict profoundly changed the natural landscapes in the country in ways that researchers are only beginning to understand.

In some areas, armed groups made protecting the country’s forests from illegal logging and conversion to pasture impossible. Elsewhere, the conflict helped preserve natural ecosystems: When armed groups moved into rural areas, the local population often left for urban centers, leaving forested areas relatively preserved. Understanding these dynamics is critical for policy makers as they plan for the country’s future in a postconflict era. A peace treaty signed by the government and leftist guerilla groups began the process in late 2016, although since then, the government and some fighters belonging to the Revolutionary Armed Forces of Colombia (FARC), including former FARC leaders, have taken up arms again.

In a new study, researchers looked at how socioeconomic factors interacted to drive land use change in the department of Córdoba, which includes an area of the Caribbean coast where multiple armed groups converged over the past half century.

It’s not the first study to examine the relationship between conflict and land use in Colombia. Previous research has shown that in the Andes, for example, violence drove a rural-to-urban migration and an increase in grazing activities. But all of this past work focused on either the Andes or the Amazon, according to Ronald Gutierrez Llantoy, a professor at Colombia’s Universidad del Norte and an author on the new study. Less was known about the Caribbean region of the country because of a lack of publicly available data.

“Armed conflicts are very rare now in the region, so this was a very unique opportunity to study these kind of dramatic, land use dynamics.”

“There were some restrictions in the data availability for the Caribbean basin,” Gutierrez said. “But recently, some public agencies released some information for research purposes, so we took advantage of that.”

Recently, Colombia’s Institute of Hydrology, Meteorology and Environmental Studies released land cover maps of the Caribbean basin, and the team zeroed in on Córdoba, a department, or governmental unit, in the northwest corner of the country that experienced high rates of violence during the conflict. The team paired the new maps with census and economic data for agricultural and forestry activities for distinct periods between 2000 and 2012. They used statistical analyses to determine how variables such as land cover type, elevation, population, and the number of kidnapped or displaced people interacted to produce the observed changes in land cover at the municipal level over the study period.

“Armed conflicts are very rare now in the [Córdoba] region, so this was a very unique opportunity to study these kinds of dramatic, land use dynamics,” Gutierrez said.

Conflicts and Clusters

The team found that most of the change could be explained by two factors: the relationship between natural land cover and armed conflict and the amount of pasture and livestock. The relationship between land cover and conflict was defined by measures of forest and cropland acreage, the geography of the region, and the number of people who were kidnapped or displaced. The second factor was based on pasture areas, the working-age population, and cattle stock.

“Reality is way more complex than statistical models, but as a first step, generalizing can be helpful.”

The influence of each of these factors varied across 29 of Córdoba’s 30 municipalities (the capital city of Montería was left out of the analysis because of its urban socioeconomic conditions), which the results suggested could be divided into five clusters. (1) Five municipalities fell under the “livestock tradition” cluster, which had low natural land cover and conflict but high levels of pastureland. In those cities, the land cover remained relatively stable over the study period. (2) Five other municipalities saw pasture levels decrease and little armed conflict. The authors found these cities overlapped with pilot projects for land rehabilitation launched in Córdoba over the past 10 years. (3) Two municipalities, Montelíbano and Puerto Libertador, were identified as “livestock and conflict hotspots.” (4) One municipality was identified as a “conflict and deforestation hotspot”: Tierralta, a large chunk of which sits inside the Paramillo National Natural Park. (5) The remaining 16 municipalities had low levels of both factors.

Experts caution that these results should be taken with a grain of salt because it’s not apparent how reliable the underlying data on kidnappings and forced displacement are. “It’s an interesting and important contribution to the conflict-environment/land systems field that opens up for further studies,” said Lina Eklund, an associate professor at Aalborg University in Denmark, whose research focuses on land systems and conflict. She was not involved in the recent study.

“As always, correlation doesn’t mean causation, so even if the study finds a relationship between conflict and land cover change, we need to learn more about the situation to understand what exactly caused the land cover change,” Eklund said.

Statistics can’t tell you how exactly a kidnapping leads to land cover change, she noted. “Reality is way more complex than statistical models, but as a first step, generalizing can be helpful.”

The research’s clustering highlights the variable ways that economic conditions and conflict can influence the environment across geographic locations, suggesting that policies need to be tailored to the local level. “Since this study was carried out at fine resolution—the smallest resolution in Colombia, municipalities—we were able to identify some clusters which were controlled by certain socioeconomic parameters,” Gutierrez said. “That allowed us to suggest some measures that might be applicable to the past conflict stage.”

Conservation After Conflict

Some of the municipalities in the livestock tradition cluster, for example, contain critical wetland ecosystems, which could be negatively impacted by cattle ranching. As a result, the authors call on policy makers to modernize the agriculture and ranching industry there. “We believe that those groups might need some technical support to improve the agricultural practices they’re accustomed to,” Gutierrez said. Investment in agricultural development would benefit both the population employed in the ranching industry, a sector that the authors note is often associated with poor working conditions and social inequality, and the environment.

Ultimately, the study, published in Environmental Science and Policy, reinforces the need for better ecosystem monitoring and environmental governance, responsibilities that often take a back seat to social and economic development after violent conflict.

The researchers are now working to develop models that will help them evaluate how armed conflict influences the hydrology of the region and models that will use these historical data to infer how land use and socioeconomic policies enacted today in the postconflict era will play out.

—Kate Wheeling (@KateWheeling), Freelance Writer


Wheeling, K. (2019), How conflict influenced land use in Colombia, Eos, 100, https://doi.org/10.1029/2019EO136714. Published on 20 November 2019.

Text © 2019. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.