The increasing availability of species distribution data and climatic conditions worldwide at resolutions up to 1km2 has fuelled a growing body of literature on the impact of global change on biodiversity1. The climatic conditions described in widely used climatic databases, such as Chelsa2 and Worldclim3, characterize long-term average atmospheric conditions from interpolated measures at standardized weather stations, whose sensors are typically located at ~2 m above a mown lawn and characterize free-air (macroclimatic) conditions. Mounting evidence increasingly reveals, however, large differences between such macroclimatic conditions and the conditions actually experienced by organisms (microclimates) due to local variations driven by terrain, wind and vegetation4. Ignoring this variability has led to conflicting predictions of climate change impacts on species distributions4–6. To fully understand and better predict how biodiversity relates to climate and its changes, microclimates need to be integrated into ecological research4,5. As we move to finer scales, however, other factors than microclimates, including local environmental conditions (e.g., soil, land use) and biotic interactions, are assumed to play an increasing role7. Shifting from global to local scale thus raises the question of the ecological relevance of the variables that operate at such scales, and their availability8,9.
The issue of geographic scale is intimately associated with the size of the organisms10. A large species requires a relatively large ‘physical space’, which often translates into a larger environmental niche11. Its niche may hence be best described by environmental factors at a coarser resolution than that of smaller species, occupying smaller physical spaces that differ from each other in terms of microclimate and substrate characteristics10. In plants, micro-environmental variations are modified at fine scale by the presence of large species themselves, generating a variety of small physical spaces that only smaller species can occupy12. The herbaceous layer plays, for example, an important role in the modulation of the availability of micro-environmental niches for the bryophyte layer. There is a trade-off between the shade provided by herb cover, protecting the underlying moss layer from desiccation, and the competition with dense herbaceous layers13. On the other hand, bryophytes can facilitate seed retention, maintain high moisture levels, stabilize soils, and protect seeds and seedlings from consumers. Conversely, a thick moss layer may also hamper seed germination and even compete with forbs13. Species composition and vegetation structure thus integrate abiotic and biotic factors, and may hence potentially be better predictors of individual species occurrences than abiotic factors alone14.
We took advantage of RechAlp, a platform for transdisciplinary research in the Western Swiss Alps15, which offers an unprecedented biotic and abiotic inventory in a single mountain region, to assess our ability to identify the fine-scale drivers of plant species distributions at 2m resolution (Fig. 1a). Predictors of bryophyte and tracheophyte species composition included 68 abiotic variables of soil, microclimates, land cover and topography at 2-25m resolution. We further evaluated the contribution of one group (bryophytes vs tracheophytes) as a predictor of the community composition of the other group, considering species composition and architecture, which directly reflect biotic interactions and indirectly integrate variation in abiotic conditions (Fig. 1b).
Variance partitioning analyses showed that climatic variation at 25m resolution is a significant predictor of plant species distributions (p-value of randomization test <0.001), but captured only 4% of the unique variation in bryophyte species composition (6% as individual explained variance when only accounting for climate), and 2% (8%) in tracheophyte species composition (Fig. 2a, b). While climate data at 1km resolution was found to be the most influential driver of spatial patterns in tracheophyte species distributions across the Alps16, the minor role played by climatic data in the present study suggests that even a resolution of 25m is too coarse to account for fine-scale species distributions patterns.
In line with growing evidence that the entire ecological space needs to be considered when modeling species distributions16, non-climatic abiotic factors alone uniquely added another 7% (p<0.001) and 2% (p<0.001) of explained variance in bryophyte and tracheophyte species composition, respectively (Fig. 2a, b). Still, despite the use of a comprehensive matrix of 68 abiotic predictors of climatic, topographic, edaphic and land cover data, environmental variables altogether explained only 16% of bryophyte species composition and 24% of tracheophyte species composition (Fig. 2a,b).
The use of tracheophyte species composition as a predictor of bryophytes species composition allowed for a substantial increase of explained variance of 35% (p<0.001, median adjusted r2 across 999 randomizations=0.0029, evidencing that the substantial increase of r2 thanks to the tracheophyte matrix is not caused by overfitting). Schaffers et al14 similarly reported that plant community composition consistently outperformed environmental conditions in predicting arthropod assemblages. Plant community composition integrates and synthesizes the habitat conditions across a number of causal factors and could be interpreted in terms of unmeasured (hidden) abiotic factors and direct interactions between the tracheophyte and bryophyte layers. Although the two effects are difficult to disentangle, variation in architectural types of the herb canopy accounted for a significant, but low proportion of the explained variance of bryophyte community composition (6%, p<0.001; Fig. 2c), tending to suggest that direct canopy effects play a comparatively minor role.
In turn, the much more limited contribution of bryophyte architectural types (2% p<0.001) and species composition (8%, p<0.001; Fig. 2d) to tracheophyte composition suggests that actual effects of the bryophyte layer on the tracheophyte layer are marginal and that bryophytes species are poorer indicators of local ecological conditions for tracheophytes as compared to the substantial indicator value of tracheophytes for bryophytes. This calls for a comparative study on the congruence of abiotic factors derived from tracheophyte and bryophyte species indicator values, respectively.
Altogether, our results suggest that, even in an open-air laboratory such as the Western Swiss Alps, we are far from having a comprehensive picture of the factors driving plant species distribution at fine scales, hampering our chances to accurately predict future species distributions in the context of climate change and calling for actual measurements of in-situ environmental factors at fine resolution (Fig. 1c).
Pradervand et al17 reported that improving the resolution of topographic data up to 2m was not sufficient to improve the accuracy of species distribution models. We suggest that, at least for small-sized organisms like bryophytes that rely on microhabitat conditions, micro-topography, microclimatic variables and light intensity at even finer scales of tens of centimeters, would be crucial (Fig. 1d). For organisms, which are less than a few mm long, microscale processes occur at spatial scales below 10 cm18. In these conditions, even local microclimates such as those investigated here do not necessarily reflect the near-surface climatic conditions that are experienced by small organisms19. Ongoing efforts to generate high resolution data of soil and near-surface temperature worldwide1 offer a promising perspective of improvement, but will have to be complemented by a new array of environmental variables at very fine spatial scales, to which new technologies, such as drones20, may usefully contribute.