Wild Vanilla and pollinators at risk of spatial mismatch in a changing climate

Charlotte Watteyn^1,2,3,4*Tobias Fremout⁵Adam P. Karremans²Koenraad Van Meerbeek^1,4Steven Janssens^3,6Sander de Backer³Monika M. Lipińska^2,7Bart Muys^1,4

• ¹Division Forest, Nature and Landscape, Department of Earth and Environmental Sciences, Faculty of Sciences, KU Leuven, Leuven, Belgium
• ²Lankester Botanical Garden, University of Costa Rica, Cartago, Cartago, Costa Rica
• ³Botanic Garden Meise, Brussels, Belgium
• ⁴Plant Institute, KU Leuven, Leuven, Belgium
• ⁵Alliance Bioversity and CIAT, Rome, Sicily, Italy
• ⁶Department of Biology, Faculty of Sciences, KU Leuven, Heverlee, Belgium
• ⁷Department of Plant Taxonomy and Nature Conservation, Faculty of Biology, University of Gdańsk, Gdańsk, Pomeranian, Poland

Climate change is expected to drive substantial shifts in species' geographic ranges. Species-specific responses of interacting species, such as plants and their pollinators, may lead to a spatial mismatch in their future distributions, disrupting these interspecific interactions. The crop wild relatives (CWRs) of the tropical cash crop vanilla hold valuable genetic resources for use in crop breeding, but their persistence is dependent on the presence of their pollinators, and at risk due to several anthropogenic pressures including climate change. To contribute to the safeguarding of this wild Vanilla gene pool, the present study aims at better understanding the effects of climate change on Vanilla species and their pollinators, and to identify potential spatial mismatches between both. Focusing on the Neotropical realm, we used MaxEnt species distribution models (SDMs) to predict potential changes in the range overlap between Vanilla and their pollinators by 2050 under the SSP2-4.5 and SSP3-7.0 climate change scenarios. We were able to compile enough occurrence records to generate SDMs for 11 Neotropical Vanilla CWRs, of which data on pollinator identity was available for four animal-pollinated species. Our models showed varying results among Vanilla species, with some predicted to undergo a net contraction (-1% to -53%) and others predicted to experience a net expansion (+11 to +140%), while the area of suitable habitat for all pollinators was predicted to decline (-7% to -71%). Our models predict a decline in range overlap between animal-pollinated Vanilla species and their pollinators under climate change, and this spatial mismatch was more pronounced for species reliant on a single known pollinator (-60% to -90%). Furthermore, the proportion of overlapping ranges located within protected areas is predicted to shrink for all species if no action is taken. Based on these findings, we propose priority areas for in situ and ex situ conservation to safeguard Vanilla's genetic resources.