Assessment of Factors Influencing Human-elephant Conflict in Kajiado County

Abstract

Increasing human populations and changes in land use have brought increased competition for space and resources between people and wildlife in the Amboseli Ecosystem. Among the large mammal species, elephants are one of the most vulnerable to land use change due to their home range and seasonal migrations. The ensuing interactions due to sharing resources are frequently negative and the individuals involved progressively develop negative attitudes towards elephants, viewing them as a risk to their survival. The aim of this research was to assess some of the factors driving human-elephant conflict in Imbirikani Group Ranch for improved management and coexistence of elephants and local communities. The specific drivers analyzed were land use and cover (LULC) changes as well as human perceptions toward human-elephant coexistence. LULC changes were analyzed in a 32-year epoch from 1981 to 2019 from Landsat images for 1987, 1999, 2013 and 2019. The data was analyzed through supervised classification with QGIS 3.12 using the maximum likelihood classification algorithm. Percentages of land area were computed to make comparisons of the changes in LULC over time. In addition, a multi-stage sampling design was applied whereby the study area was divided into strata based on the administrative sub-locations. Household survey was conducted on 320 sampled households selected through a multistage sampling design and primary data gathered by interviewing selected households through semi-structured questionnaires. The households were interviewed on the held attitudes and perceptions on the nature and extent of human elephant conflicts. Further, field visits and observatory data were gathered to unveil farms and property damages caused by Elephants. Focus group discussions were also carried out to gain more insights on available resources mapping and human elephant conflict. Collected data were summarized by descriptive statistics including frequencies and arithmetic means and results presented based on thematic areas such as types of conflict, seasonality of attacks, crops affected, and costs of HEC incidences. A logistic regression model was generated to assess explanatory power of variables while Pearson’s chi-squared tests were computed to establish associations between hypothesized independent variables and the dependent variable of attitudes toward elephant conservation. Shrublands experienced the highest change with a steady decrease of almost 30% over the study period. Bareland increased by 16%. Grassland increased by 7.51%. Woodlands experienced a decline of 11.2%. Irrigated agriculture showed an increase of 12.48%. xiv Crop raiding was the most prevailing form of HEC with the behaviour of raiding elephants being described as habitual from specific elephant groups and nocturnal. The type of crop grown was the most significant predictor for HEC. Education level, gender and location had a significant influence on perceptions and attitudes toward human-elephant coexistence (p = 0.0132, p = 0.0037 and p > 0.0001, respectively). Age (p = 0.1377) and type of livelihood (p = 0.1272) had no significant relationship with the participants’ perceptions and attitudes. Additionally, 61% believed elephants were beneficial but 82% felt bad about free roaming wildlife while the majority, 61%, had a negative attitude toward wildlife authorities. Understanding the spatio-temporal distribution of conflicts, cost implications and nature of crop raiding behaviour in addition to community knowledge, attitudes, and perceptions on the same, should be integrated into management decisions to guide planning of conservation interventions. Integrated spatial planning and land use zoning should be executed in collaboration with the community. Moreover, elephant repellent crops should be grown in HEC hotspot areas and mechanisms to disincentivize agricultural expansion in wildlife habitats and movement corridors established for improved conflict mitigation.