Back

Glacier retreat reshapes trophic networks: interaction turnover outpaces species turnover over space-time

Tu, B. N.; Mas-Carrio, E.; Hodecek, J.; Khelidj, N.; Casartelli, M.; Rasmann, S.; Fumagalli, L.; Losapio, G.

2026-04-15 ecology
10.64898/2026.04.13.718237 bioRxiv
Show abstract

As glaciers retreat worldwide, newly exposed terrains are rapidly colonized by plants and their associated animal communities. Although plant-animal interactions are key for biodiversity maintenance and ecosystem functioning, the ecological processes underlying the assembly and development of trophic interactions over space-time remain poorly understood. Here, we investigated the trophic niche structure of plant-arthropod interactions along a 140-year primary succession at Mont Mine glacier foreland (Switzerland). Using arthropod gut-content DNA metabarcoding, we reconstructed trophic interactions at the food web level, revealing numerous previously undetected links among 284 arthropod taxa feeding on 263 plant taxa. Trophic niche overlap among arthropods increased following glacier retreat, indicating decreasing resource partitioning and suggesting increased resource competition. Trophic niche breadth became narrower and diet species richness declined, indicating increased trophic specialization. Notably, changes in trophic interactions occurred more rapidly than shifts in species diversity or community composition of plants and arthropods. These results demonstrate that glacier retreat reorganizes trophic networks beyond simple species turnover, reshaping biotic interactions during ecosystem development. Our findings highlight trophic interactions as sensitive indicators of biodiversity change and suggest that the stability of emerging food webs may be strongly affected as glaciers vanish worldwide.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.