Potential global distributions of an important aphid pest, Rhopalosiphum padi: insights from ensemble models with multiple variables

J Econ Entomol. 2025 Jan 13:toae237. doi: 10.1093/jee/toae237. Online ahead of print.

Abstract

Species distribution modeling is extensively used for predicting potential distributions of invasive species. However, an ensemble modeling approach has been less frequently used particularly pest species. The bird cherry-oat aphid Rhopalosiphum padi L. is an important pest of wheat (Triticum aestivum L.) worldwide and causes 30% yield losses. Here, we developed a series of ensemble models with multiple variables to predict the habitat suitability of this pest at a global scale. The current suitable habitat for R. padi is mainly distributed in East Asia, South Asia, Europe, southern North America, southern South America, eastern Australia, and New Zealand. The highly suitable regions are primarily distributed in east of China, Japan, most of North America, southeastern South America, most of Europe, and southeastern edge of Australia. In future scenarios, the suitable habitats will undergo a significant contraction overall northward, and no moderately nor highly suitable habitats are predicted for this pest in other areas. Our findings indicate that a high risk of R. padi outbreaks currently exists for the highly suitable regions mentioned above, especially with wheat cultivation, but the capacity of R. padi to cause such outbreaks will weaken in the future. Climate-associated factors are significantly more important than land use, elevation and host-plant factors, and the BIO11 (mean temperature of the coldest quarter), in particular, predominated in shaping projections of R. padi's distribution. The predicted distribution pattern and key ecological factors affecting this pattern identified herein could provide important guidance for developing management policies targeting this economically important pest.

Keywords: Triticum aestivum; Aphididae; climate change; land use; species distribution modeling.