Experimental community assembly: when do niche-assembly processes matter?

This post was contributed by Kristina Riemer who is a PhD candidate in Dr. Ethan White’s lab 

With the introduction of neutral theory in Hubbell (2001), ecologists began to discuss the relative importance of species identities, including traits, in ecological communities. The two papers that we discussed were intended to capture some of the experimental research that has been done on the importance of such niche-based processes on community assembly, which is unfortunately relatively rare. The first paper contains the results from laboratory-based experiments using protist microcosms, while the second paper presents results from experiments done in the field using grassland plants.

  • Jiang, L. and S.N. Patel. 2008. Community assembly in the presence of disturbance: microcosm experiment. Ecology 89:1931-1940. (pdf)
  • Fargione, J., C. Brown, and D. Tilman. 2003. Community assembly and invasion: an experimental test of neutral versus niche processes. PNAS 101:8916-8920. (pdf)

While the study organisms and experimental methods in these papers were very different, the research questions and results were similar. The authors in these papers were determining whether and when niche-assembly processes, as opposed to neutral processes, were important in their systems. Niche-assembly processes occur when the traits or differences between organisms are influential in the assembly of their communities. Jiang and Patel (2008) focused on how different levels of disturbance affected the influence of traits, while Fargione, Brown, and Tilman (2003) addressed if functional traits influenced the ability of new species to become established in a community.

Jiang and Patel (2008) introduced 10 species of protists, in stable groups of two species, to individual microcosms randomly. For each of five randomly chosen sequences, one group of two species was added each week for five weeks. In the control microcosms, all 10 species were added simultaneously. The species composition of each microcosm was then determined every week for the following six weeks. Additionally each sequence underwent three disturbance regimes: low, medium, and high. This disturbance was in the form of sonication of increasing percentage of the microcosm medium. Each sequence and disturbance regimes combination was replicated three times.

The authors determined that niche-assembly processes were more important in high disturbance regimes than in low. Microcosms that experienced high disturbance had communities that were more similar, which led to the conclusion by Jiang and Patel (2008) of less alternative stable states in these communities. They explained that this was likely due to the greater environmental filter created by disturbance, which filtered out those species that did not have traits that conferred disturbance resistance. Therefore, niche-assembly processes were important in the assembly of these communities. Conversely, communities with low disturbance levels had more variable species composition. Traits that provide advantages in disturbance were not as important in assembly of these communities as colonization history was. Therefore, neutral-(or disturbance-) assembly processes were more important than niche processes in low disturbance communities.

These results were very similar to those in a paper by Chase (2007), which we discussed previously in this seminar. Chase (2007) also conducted an experimental study in which the similarity of species composition of communities exposed to varying levels of disturbance was determined. This was a field-based study, though, using species in freshwater ponds where the disturbance was drought. The author also concluded that communities with greater disturbance had more similar species composition than those that weren’t.

Screen Shot 2013-12-20 at 5.25.29 PM

In addition to niche and neutral processes, the results from Jiang and Patel (2008) shed light on the intermediate disturbance hypothesis. This hypothesis states that species diversity is greatest at intermediate levels of disturbance. The intermediate disturbance hypothesis has not been supported in most studies, including in the paper currently being discussed. In Jiang and Patel (2008), only the control treatment had the greatest species richness in the intermediate disturbance regime. This was not the case for any of the other sequences. The authors suggested that the intermediate disturbance hypothesis might only hold for communities which have simultaneous introduction of species, such as in their control, and not sequential introduction, like the other sequences. Because simultaneous introduction rarely occurs in natural systems, it seems unlikely that most systems would exhibit the intermediate disturbance hypothesis.

Screen Shot 2013-12-20 at 5.25.59 PM

In response to the paper by Jiang and Patel (2008), we discussed how to properly frame and focus ecological papers. One criticism of this paper is their attempt to discuss two hypotheses in a single paper, the influence of niche and neutral processes on community assembly and the intermediate disturbance hypothesis. In reference to the former hypothesis, alternative stable states were also mentioned. It is not clear from the title of the paper, and only slightly more clear in the abstract, that the authors are addressing two hypotheses in this paper. We suggest that a better approach may be to choose a single concept and address it more thoroughly than to attempt to pack too much information and too many ideas into one paper.

In Fargione, Brown, and Tilman (2003), experiments on grassland plots were carried out to determine the influence of functional traits on community assembly. At the Cedar Creek LTER, particular numbers of randomly chosen grassland species were planted on 3m x 3m plots of land. After three years, which allowed for these plant species to become established, 27 new species of plants were introduced. The biomass and ground cover increase of these introduced species was determined two years later. All of the plant species were categorized based on traits including resource use into the following four functional groups: legumes, forbs, C3 plants, and C4 plants. The purpose of the experimental study was to determine if it is more unlikely for introduced species of a particular functional guild to become established if there are many resident species of that guild already present due to competitive exclusion. If this does occur, it indicates that nonneutral processes are important in the assembly of these communities.

The first key result was that, regardless of functional groups, greater resident species richness resulted in less introduced species richness. When we examined the figure that presented this data, we questioned the strength of this relationship. The negative relationship between introduced and resident species richness seemed to be driven primarily by the greatest resident species richness value (n = 24). The relationship did not hold for the remaining resident species richness values. The second problem that we identified with this result was that is has no bearing on the research question being asked. The focus of the paper is to use functional traits to determine the relative influences of niche and neutral assembly processes on these communities. Because the functional groups were not distinguished in this result, it tells us nothing about niche or neutral processes. This result only informs us that, when there are more resident species regardless of their traits, there will be fewer species that are able to become established. This is an intuitive conclusion because there are limited resources and, if there are more species present, there are less resources available for newly introduced species.

Screen Shot 2013-12-20 at 5.26.47 PM

The second, more informative result presented in Fargione, Brown, and Tilman (2003) was that the resident species of each functional guild had the “most negative impact” on the introduced species of their own guild. This seems to show that there are niche processes involved in this community’s assembly, because species tended to prevent introduced species with similar traits from being able to become established. This was shown, somewhat unclearly, in a table which I modified for clarity below. Resident species of each guild had the most negative values with the same introduced guild, and these values are in red boxes. If only neutral processes were occurring, each resident functional guild would have the same influence on all of the introduced functional guilds, including its own, and that was not the case.

Screen Shot 2013-12-20 at 5.27.42 PM

While the two papers that we discussed had experiments that used different study systems and organisms, different methodologies, and different questions, they both addressed the relative influence of neutral and niche processes on community assembly. Using protist microcosm experiments, Jiang and Patel (2008) determined that niche processes are more important in high disturbance regimes because disturbance is an environmental filter that prevents non-disturbance resistance species from becoming established. In low disturbance, neutral processes such as dispersal are more important and result in greater variation in community composition. Fargione, Brown, and Tilman (2003) used the influence of functional traits to determine whether niche processes were occurring in the assembly of grassland communities. Because greater abundance of resident species with functional traits prevented functionally similar introduced species from becoming established, it seemed that niche processes were important in their system. 


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s