I would look at an experimental design called BACI (Before After Control Impact) that was first described by Stewart-Oaten et al. (1986) and further developed by Underwood (1992) .  The design tries to to differentiate changes occurring due impact/recovery from those due to natural variation by monitoring the impact site and control sites before, during and after a disturbance.

I included a recent example of the experimental design applied to a recent recovery study by Roca et al. (2014)

References

Stewart-Oaten, Allan, William W. Murdoch, and Keith R. Parker. "Environmental impact assessment:" Pseudoreplication" in time?." Ecology 67.4 (1986): 929-940.

Roca, Guillem, et al. "Detecting the impacts of harbour construction on a seagrass habitat and its subsequent recovery." Ecological Indicators 45 (2014): 9-17.

Underwood, A. J. "Beyond BACI: the detection of environmental impacts on populations in the real, but variable, world." Journal of experimental marine biology and ecology 161.2 (1992): 145-178.

Hi Konstantin, I would use Sorensen similarity index as long as you focus at the biodiversity level. That is probably the most straightforward measure.

The program Primer does multivariate analysis of communities (http://www.primer-e.com/)

I would look at an experimental design called BACI (Before After Control Impact) that was first described by Stewart-Oaten et al. (1986) and further developed by Underwood (1992) .  The design tries to to differentiate changes occurring due impact/recovery from those due to natural variation by monitoring the impact site and control sites before, during and after a disturbance.

I included a recent example of the experimental design applied to a recent recovery study by Roca et al. (2014)

References

Stewart-Oaten, Allan, William W. Murdoch, and Keith R. Parker. "Environmental impact assessment:" Pseudoreplication" in time?." Ecology 67.4 (1986): 929-940.

Roca, Guillem, et al. "Detecting the impacts of harbour construction on a seagrass habitat and its subsequent recovery." Ecological Indicators 45 (2014): 9-17.

Underwood, A. J. "Beyond BACI: the detection of environmental impacts on populations in the real, but variable, world." Journal of experimental marine biology and ecology 161.2 (1992): 145-178.

Thank you for the replies! I was not precise in my question, but some people use abundance, number of species/taxons. Including similarity indices to compare how community structure transfoms looks promicing. Thank you!

Hi Konstantin,

I would say that the common diversity indices (species number, Shannon,...) are not really relevant of a disturbance according to my studies. I suggest that the composition of community is the best think to study because of the selection of certain species and then functional traits after a disturbance that could imply big change in the ecosystem functioning.

A chronosequence could be interesting to study the possible recovery of a community after a disturbance when we do not have the occasion to be present before the disturbance.

Best,

Dear Apolline,

What parameters would you use given you wish to compare the structure of community in time?

Thank you for your reply!

for the dynamic of the structure I would say a network analysis

Hi Konstantin-

I pretty much agree with Apolline: if you are interested in the effects of disturbance on the community (structure), you will have to look at community sructure - who is there before versus after disturbance...

Beyond that, however, I suggest to rather look at "function", i.e. traits of species, rather than species: even if some species are lost upon disturbance, the community as an entity may still provide the same processes and services, i.e. community functioning might not be changed, if other (new) species exhibit similar traits...

Thus, rather than looking for taxonomic community structure, you should look at functional community structure

Good luck!

martin:-)

Dear Konstantin

There are several biotic indices that can be used for the benthic quality status of water bodies before and after the disturbance, and you can classify your samples according to ecological status defined by the EU Water Frame Directive. In the Mediterranean, the most popular ones are BENTIX, H (Shannon-Weiver Diversity Index), AMBI, M-AMBI  and MEDOCC. The estimation of these indices (except for M-AMBI) is easy and there are also web-sites that you can download a program/exel macro to estimate these indices (except for MEDOCC). For BENTIX, http://www.hcmr.gr/en/articlepage.php?id=141

For AMBI AND M-AMBI, http://www.azti.es

Regards,

Melih 

Hello!

We use the additional indicators: percent of common species, invasive species, labile (eurybiotic) species.

Elena.

Hi there, 

there are several ways to track recovery after disturbance but you should first define what you want to identify. If you want to see how recovery goes throughout time i.e. recovery rate, then you should plot any variable response vs time together with reference data (undisturbed or less disturbed). Knowning succesional steps in your sytems is also neccesary. From the community point of view, most studies reports several variables including; species richness, abundance, biomass, diversity indexes, together with multivarible information usually based on dissimilarity/similarity distance measurements (i.e. bray-curtis, euclidian, etc). But if you want to understand recovery in terms of ecosystem functioning, then you could use some functional approach or indicators that capture ecosystem properties such as ascendency or emergy. However, you must have a clearly identified the disturbed vs the non disturbed situation for comparisons. 

Hope this helps

Aldo  

Hi, Addition to the above I would suggest you to calculate the index of disturbance which is biomass to abundance ratio.

Hi Konstantin,

If you are interested in the changes of species composition and abundance due to the disturbance you can try multivariate analyses. Depending on your sampling design and your data, you have different analyses available. One of them is to perform a Correspondence Analysis (CA) or Detrended CA (DCA) of sites and species abundances. You can then compare the coordinates of sites on the first CA (or DCA) axis along time since disturbance in disturbed and control sites. This comparison gives you a measure of recovery (long studies give the best estimates). A still better option, if you can sample before the disturbance, is to combine a BACI approach with multivariate analyses.

Anyway, there are many possibilities. Maybe you can have a look at our papers (I can send you a pdf if you want):

Pons, P, Clavero, M (2010) Bird responses to fire severity and time since fire in managed mountain rangelands. Animal Conservation 13, 294-305.

Prodon, R, Pons, P (1993) Postfire bird studies: methods, questions, and perspectives. In 'Fire in Mediterranean Ecosystems.' (Eds L Trabaud, R Prodon.) pp. 332-343. (Commission of the European Communities: Brussels)

Hello Konstantin,

have you ever heard of the 'taxonomic distinctness'? This is a diversity index proposed by Clarke & Warwick:

Clarke & Warwick (1998): A taxonomic distinctness index and its statistical properties.  J Appl Ecol 35: 523-531.

Clarke & Warwick (2001): A further biodiversity index applicable to species

lists: variation in taxonomic distinctness. MEPS 216: 265-278.

Clarke & Warwick (2001): Change in marine communities: an approach to statistical analysis and interpretation. 2nd edition. PRIMER-E, Plymouth.

It was shown in the latter reference that taxonomic diversity in terms of taxonomic distinctness decreased with disturbed communities (Ecofisk oilspill and benthic communities). This change was not visible with 'normal' diversity indices.

Would that help you any further?

Best regards

Armin

Good Question, Konstantin.

I have evaluated community recovery for vegetation and wildlife following disturbances including fire, vegetation treatments, and energy development.  The answers preceding mine are very good answers about specific indices or metrics that may be of interest to you. The best disturbance studies involve solid study designs, where pre-disturbance responses of interest are compared to post-disturbance responses; however, the nature of disturbance studies does not always lend itself to good study designs, because we lose randomization (i.e., disturbances occur where and when they do), replication (i.e., just one study site), and often without controls.  At the least it is good to compare disturbed study sites to undisturbed study sites or even evaluate responses to disturbance along a gradient of disturbance (center of disturbance to areas outside of disturbance). Me and my students have had to take these issues into consideration with multiple studies.  Please see my websites for published papers that provide examples of how we dealt with study design issues relative to disturbance and which metrics we used to evaluate disturbances.

Website: http://www.uwyo.edu/esm/faculty-and-staff/beck/index.html

ResearchGate: https://www.researchgate.net/profile/Jeffrey_Beck

Best Regards,

Jeff