Measuring them all: Individual-based functional spatial patterns in mountain grasslands

Prepared by Luciano L. M. De Benedictis, Stefano Chelli, Roberto Canullo, and Giandiego Campetella

What happens if you measure traits and coordinates for each plant individual in a grassland? In 1947, Alex S. Watt already recognized that a plant community is ultimately made up of individual plants, their characteristics, and spatial relations with each other, but considered an individual-level analysis impracticable.

Still, that question drove us to sample the mountain grassland vegetation in a strict reserve managed by our university since 1970, to an unusual level of detail. For each individual in ten 50 × 50 cm quadrats we recorded the position and three widely studied traits: the height of the individual, the area of a leaf and the ratio between leaf area and its dry mass (specific leaf area).

This sampling design allowed us to apply statistics from point pattern analysis and study the spatial patterns of these traits at scales of a few centimetres, test them against null models, and detect the specific scales at which non-random patterns occurred. At the same time, the abundance of trait data allowed us to estimate how much trait variation there is within species, compared to between species. While the former source of variation is often ignored, we also wanted to see if the results of our analyses changed whether we considered species identity or not: after all, each individuals can be described by its own traits, and if each has the potential to cover most of the range of variation of a trait in the community, species identity would be of secondary importance.

The analysis of trait variability showed the importance of variation within species: on average, a species can cover a large portion of the trait distribution of all species, and of that of another species. For some traits, the variation within species is almost as high as between species.

Our results from point pattern analyses show that individuals very close together tend to be smaller and of similar size; at distances above 8 cm, instead, they tend to be different but overall larger, and specific leaf area also differed more, reflecting more varied leaf strategies. Whether we considered pairs of individuals of the same species or only of different species, the results were very similar, confirming this hypothesis. Surprisingly, if we kept the coordinates, but replaced the trait values by species means, as it is often done, we lost almost all patterns!

Given the small scale, comparable to the size of these individuals, we think competition for space is an important mechanism causing the patterns, but we found evidence to support both interactions between plants and small-scale heterogeneity of soil conditions.

We can’t recommend to everyone to adopt our laborious sampling design, but it has allowed us to conclude that plant individuals are arranged non-randomly down to very small scales, and that if you’re interested in these patterns, you should consider intraspecific variability.

Reference:

• Watt, A. S. (1947). Pattern and process in the plant community. Journal of Ecology, 35(1/2), 1–22. https://doi.org/10.2307/2256497

This is a plain language summary for the paper of De Benedictis et al. published in the Journal of Vegetation Science (https://doi.org/10.1111/jvs.70029)