Climate mismatches with ectomycorrhizal fungi contribute to migration lag in North American tree range shifts

Abstract

Climate change will likely shift plant and microbial distributions, creating geographic mismatches between plant hosts and essential microbial symbionts (e.g., ectomycorrhizal fungi, EMF). The loss of historical interactions, or the gain of novel associations, can have important consequences for biodiversity, ecosystem processes, and plant migration potential, yet few analyses exist that measure where mycorrhizal symbioses could be lost or gained across landscapes. Here, we examine climate change impacts on tree-EMF codistributions at the continent scale. We built species distribution models for 400 EMF species and 50 tree species, integrating fungal sequencing data from North American forest ecosystems with tree species occurrence records and long-term forest inventory data. Our results show the following: 1) tree and EMF climate suitability to shift toward higher latitudes; 2) climate shifts increase the size of shared tree-EMF habitat overall, but 35% of tree-EMF pairs are at risk of declining habitat overlap; 3) climate mismatches between trees and EMF are projected to be greater at northern vs. southern boundaries; and 4) tree migration lag is correlated with lower richness of climatically suitable EMF partners. This work represents a concentrated effort to quantify the spatial extent and location of tree-EMF climate envelope mismatches. Our findings also support a biotic mechanism partially explaining the failure of northward tree species migrations with climate change: reduced diversity of co-occurring and climate-compatible EMF symbionts at higher latitudes. We highlight the conservation implications for identifying areas where tree and EMF responses to climate change may be highly divergent.

Keywords: climate change; ectomycorrhizal fungi; mycorrhizal symbiosis; range shifts; species distribution modeling.

Fig. 1.

Geographic changes in climate envelopes for tree species and EMF across North America. We analyzed how the suitable climate boundaries of 50 tree species (A) and 402 EMF species (B) change at southern and northern locations (2.5% and 97.5% percentiles, respectively) between present and future climates (2070 RCP8.5). The four quadrants show different climate suitability scenarios under climate change (illustrated with red and blue bars): expansion further north and south overall (Top Left); shift toward higher latitude at both northern and southern boundaries (Top Right); shift toward lower latitude at both northern and southern boundaries (Bottom Left); or contraction toward lower latitudes at northern boundaries and higher latitudes at southern boundaries (Bottom Right). Center points are mean latitude difference distribution values for each tree or EMF species summarized across longitude bands ± 1 SD (gray error bars).

Fig. 2.

Geographic hotspots of suitable climate overlap between tree-EMF species pairs. We analyzed how climate change affects the cumulative area where EMF SDMs predict overlapping climate suitability with co-occurring tree species. Predicted EMF richness maps reflect the sum of overlapping climate suitability between a given tree species and all their EMF partners. Maps show family-level trends as the sum divided by the number of tree species per family. Brighter colors indicate potential interaction hotspots—areas predicted to support a greater number of co-occurring and climatically compatible tree-EMF symbioses. Maps show tree-EMF hotspots under the present and future climate scenarios. Plots on the right show how predicted EMF richness changes with latitude for each climate scenario, with trend lines showing mean composite richness extracted across longitude bands ± 1 SD (shaded area).

Fig. 3.

Climate change effects on the total suitable climate overlap between tree-EMF pairs. (A) Total area of suitable climate overlap across all tree-EMF pairs. Note that the Y-axis is scaled by a square root transformation. (B) Changes in the extent of tree-EMF suitable climate overlap between current and future climate projections. The Top distribution shows that climate change may shrink the geographic extent of shared climate niches for 35% of tree-EMF pairs overall. The Bottom plot shows how this distribution varies among tree families. (C) Mean change in suitable climate overlap across the top 50 EMF genera for each tree family. Orange colors indicate negative change, and purple colors indicate positive change.

Fig. 4.

Climate envelope mismatches between tree-EMF pairs. (A) We quantified mismatch as the absolute latitude difference in climate suitability between tree-EMF pairs at their southern and northern boundaries (2.5% and 97.5% percentiles of latitudinal distributions, respectively). (B) Latitude change in suitable climate mismatch between tree-EMF pairs at southern and northern boundaries. The Top distributions show how climate change is predicted to increase the latitudinal mismatch of climate suitability across all tree-EMF pairs. The Lower distributions show how this varies among tree families.

Fig. 5.

Reduced diversity of climate-compatible EMF partners corresponds with lower tree migration potential. (A) We predicted EMF richness for each tree species by summing the climate suitability models of all co-occurring EMF partners based on overlap with their host tree climate niche. We then calculated the species pool of climate-compatible EMF partners at the southern, central, and northern climate positions of tree hosts. The line plot shows how the predicted EMF species pool richness varies across these positions. Colors indicate tree families, small points are tree species, and large points are tree family averages. (B) We used large-scale forest inventory measurements to estimate tree species migration potential by calculating latitude differences between seedlings vs. adult trees at their southern and northern extremes (2.5% and 97.5% percentiles of occurrence records, respectively). The four quadrants represent the possible migration scenarios: seedlings at higher latitudes than adult trees at both northern and southern boundaries (Top Left); seedlings expand further north and south than adult trees (Top Right); adult trees expand further north and south than seedlings (Bottom Left); and seedlings at lower latitudes than adult trees at both northern and southern boundaries (Bottom Right). Center points are mean latitude difference distribution values for each seedling–adult tree species comparisons summarized across longitude bands ± 1 SD (gray error bars). (C) Comparison of tree species migration potential and predicted richness of climate-compatible EMF partners. Center points are mean values summarized across longitude bands ± 1 SE (error bars).