Integrating snake distribution, abundance and expert‐derived behavioural traits predicts snakebite risk

Martín, Gerardo and Erinjery, Joseph and Gumbs, Rikki and Somaweera, Ruchira and Ediriweera, Dileepa and Diggle, Peter J. and Kasturiratne, Anuradhani and de Silva, Hithanadura J. and Lalloo, David G. and Iwamura, Takuya and Murray, Kris A. (2022) Integrating snake distribution, abundance and expert‐derived behavioural traits predicts snakebite risk. Journal of Applied Ecology, 59 (2). pp. 611-623. ISSN 0021-8901

Abstract

Abstract: Despite important implications for human health, distribution, abundance and behaviour of most medically relevant snakes remain poorly understood. Such data deficiencies hamper efforts to characterise the causal pathways of snakebite envenoming and to prioritise management options in the areas at greatest risk. We estimated the spatial patterns of abundance of seven medically relevant snake species from Sri Lanka, a snakebite hotspot, and combined them with indices of species' relative abundance, aggressiveness and envenoming severity obtained from an expert opinion survey, to test whether these fundamental ecological traits could explain spatial patterns of snakebite and envenoming incidence. The spatial intensity of snake occurrence records in relation to independent environmental factors (fundamental niches and land cover) was analysed with point process models. Then, with the estimated patterns of abundance, we tested which species' abundances added together, with and without weightings for aggressiveness, envenoming severity and relative abundance, best correlate with per capita geographic incidence patterns of snakebite and envenoming. We found that weighting abundance patterns by species' traits increased correlation with incidence. The best performing combination had three species weighted by aggressiveness and abundance, with a correlation of r = 0.47 (p <0.01) with snakebite incidence. An envenoming severity and relative abundance‐weighted combination of two species was the most strongly associated with envenoming incidence (r = 0.46, p = 0). Synthesis and applications. We show that snakebite risk is explained by abundance, aggressiveness and envenoming severity of the snake species most frequently involved in envenoming cases. Incorporating causality via ecological information of key snake species is critical for snakebite risk mapping, helping to tailor preventive measures for dominant snake species and deploying the necessary antivenom therapies.