Hello Rocco,

I would just plot up NMS (nonmetric multivariate scaling, an ordination method available in the PC-ORD package) scores or other multivariate axis values as a function of distance from edge. These values at different distances could also be compared among edge types, orientations, etc.

For distance of edge influence (DEI) -- values of which could also be used to compare edge types -- I still like a regression approach where intersection (with appropriate confidence limits) to interior, background levels denotes DEI, as per Burton. 2002. Silva Fennica 36:329–352. A  more sophisticated method is proposed by Harper & Macdonald. 2001. Ecology, 82(3), 649-659, but I have not tried it.

For a good overview of magnitude of edge influence (MEI) and DEI concepts and implications (and needed edge effect research, though mostly with a forest edge focus), see Harper et al. 2005. Conservation Biology 19(3), 768-782.

Good luck and have fun with this.

Phil Burton

Hello Rocco,

I would just plot up NMS (nonmetric multivariate scaling, an ordination method available in the PC-ORD package) scores or other multivariate axis values as a function of distance from edge. These values at different distances could also be compared among edge types, orientations, etc.

For distance of edge influence (DEI) -- values of which could also be used to compare edge types -- I still like a regression approach where intersection (with appropriate confidence limits) to interior, background levels denotes DEI, as per Burton. 2002. Silva Fennica 36:329–352. A  more sophisticated method is proposed by Harper & Macdonald. 2001. Ecology, 82(3), 649-659, but I have not tried it.

For a good overview of magnitude of edge influence (MEI) and DEI concepts and implications (and needed edge effect research, though mostly with a forest edge focus), see Harper et al. 2005. Conservation Biology 19(3), 768-782.

Good luck and have fun with this.

Phil Burton

Dear Phil,

thanks a lot for your answer! I will surely go through your suggestions and try some.

rocco

Dear Rocco, I like Phil Burton's recommendation, but I must add that we will find the most meaning when we use ordination to test for the particular mechanisms: For example, perhaps we group vegetation by its reproductive strategy (vegetative, bat dispersed, wind dispersed), or by palitability to cattle, etc. 

Just some thoughts. 

I would measure out from the edge at various distances, like three meter increments, then characterize what species composes the ecosystem that those points, in terms of percentage basal cover?  

Thank you Liam and Craig.

Indeed, as dealing with herbaceous communities rather than forest edge, I intend to test some species functional categorization (as well as species cover itself, e.g., for alien species) rather than true structural and abiotic parameters.

Hello Rocco,

To quantify the community change according to distance from the edge, you can use multivariate analysis as proposed by Philip Burton or gradient analyses through the evaluation of the Distance of Edge Influence (DEI) and the Magnitude of Edge Influence (MEI) – see Harper et al. (2005) for details.

To my point of view, there are two main categories of methods to evaluate DEI: i) those using a discrete approach, with reference to the hypothetical two-phase response pattern proposed by Murcia (1995) and ii) those using a continuous approach.

Among “discrete” approaches, you can find the two-phase linear regression (e.g. Matlack et al. 1993, Fox 1997), the mowing split window (e.g. Li et al. 2005), the piecewise regression (e.g. Toms and Lesperance 2003).

Among “continuous” approaches, I recommend those of Ewers and Didham (2006) testing a set of models (null, linear, exponential, logisitic, unimodal) referring to ecological hypotheses on the response of environment and communities to edge influence.

Personally, I tested the two-phase linear regression (Alignier and Deconchat 2011) and a continuous approach (Alignier and Deconchat 2013). In both cases, we compared DEI according to edge type.

Finally, I recommend you the special issue on boundaries in Bioscience, 203, vol. 53.

Hoping this will help you,

Audrey

--

Alignier A, Deconchat M (2011) Variability of edge effect on vegetation implies reconsideration of its assumed hypothetical pattern. Applied Vegetation Science 14:67–74

Alignier A, Deconchat M (2013) Patterns of forest vegetation responses to edge effect as revealed by a continuous approach. Annals of forest Science 70:601-609

Ewers RM, Didham RK (2006b) Continuous response functions for quantifying the strength of edge effects. Journal of Applied Ecology 43:527–536

Fox BJ, Taylor JE, Fox, MD & Williams C (1997) Vegetation changes across edges of rainforest remnants. Biological Conservation 82: 1–13

Matlack GR (1993) Microenvironment variation within and among forest edge sites in the eastern United States. Biological Conservation 66: 185–194

Murcia C (1995) Edge effects in fragmented forests: implications for conservation. Trends Ecol Evol 10:58–62

Toms JD, Lesperance ML (2003) Piecewise regression: a tool for identifying ecological thresholds. Ecology 84:2034–2041

Dear Sven Gunter made a similar to what you intend to develop work. I spent the quote from the paper that is published. "Influence of distance to forest edges on natural regeneration of abandoned pastures: a case study in the tropical mountain rain forest of Southern Ecuador". Regards

http://link.springer.com/article/10.1007/s10342-006-0156-0 

Dear Rocco,

I could suggest that the increasing size of quadrat may help you to find out the pattern of plant community changes for one species. It gives the morphological pattern of a community. E.g. If you take a quadrat of A then 2A, 4A, 8A, 16 A, ......etc. From the results you may go for pattern analysis. The graph with certain numbers of peaks. The peaks give certain interpretation to you. 

Lauren Porensky explored definitions of 'edge' and quantification of magnitude and shape:

Porensky, L.M. & T.P. Young. 2013. Edge interactions in fragmented and patchy landscapes. Conservation Biology 27:509-519.

Hi Rocco,

I don’t believe there is any simple answer to your question – one that has long concerned ecologists, and hence this rather long-winded reply. To measure edge effect, first you need to define what you mean by ‘edge’. Depending on their age, some so-called edges or ecotones in forest-grassland mosaics can range from a couple of metres to kilometres. Then you need to first decide which ‘effects’ you wish to measure (e.g. on your grassland) and at which scale and for what purpose. Most edges are transitional zones that tend to be highly dynamic being influenced often by confounding patterns of phenomena such as fire, grazing, topography (even mico-topography). Edge effects on fauna can be problematic without contextual sampling (see Gillison et al. 2003). If you wish to measure effects on vegetation then you need to sample a representative set of transects (e.g. 40x5m or longer) that intercept both communities either side of the ‘edge’. While experimental controls involving impacts are desirable they are largely impractical being long term and generally difficult to relate to complex land use mosaics. Where available, a series of chronosequences that exhibit common environmental drivers can provide a useful sampling framework.

Then comes the question of what to measure. Depending on your purpose and scale, apart from key variables such as impacting agents (fire..etc.), soil data including such above-ground features as litter depth are usually critical. In recent years vegetation transitional sequences involving readily measureable plant functional types and traits  have been found useful as they can provide measureable indices of response to external impacts. These can be combined with data on vascular plant species, vegetation structure and the site physical features.

With reference to your question regarding grasslands: in addition to the literature references provided by other contributors and those below, you may like to check out the methodology in the article (Gillison et al. 2013) that measures vegetational reponse in successional sequences and transitional zones along land use intensity gradients in two continents that include grasslands. These and some other refs may be useful.

Good luck..

andy

Gillison, A.N. (1970). Structure and floristics of a montane grassland forest transition, Doma Peaks Region, Papua.  Blumea18, 72-86.

Gillison, A.N., Bignell, D.E., Brewer, K.R.W., Fernandes, E.C.M., Jones, D.T., Sheil, D., May, P.H., Watt, A.D., Constantino, R., Couto, E.G., Hairiah, K., Jepson, P., Kartono, A.P., Maryanto, I., Neto, G.G., Neto, R.J.V.,van Noordwijk, M., Silveira, E.A., Susilo, F-X., Vosti, S.A. & Nunes, P.C. (2013)  Plant functional types and traits as biodiversity indicators for tropical forests: two biogeographically separated case studies including birds, mammals and termites. Biodiversity and Conservation. 22, 1909-1930.

Gillison, A.N., D.T. Jones, F.-X. Susilo and D.E. Bignell. (2003). Vegetation indicates diversity of soil macroinvertebrates: a case study with termites along a land-use intensification gradient in lowland Sumatra. Organisms, Diversity and Evolution 3:111-126.

Pellissier, V., Bergès, L., Nedeltcheva, T., Schmitt, M.-C., Avon, C., Cluzeau, C., Dupouey, J.-L. (2013), Understorey plant species show long-range spatial patterns in forest patches according to distance-to-edge. Journal of Vegetation Science, 24: 9–24

Santos, G. G. A., Santos, B. A., Nascimento, H. E. M. and Tabarelli, M. (2012), Contrasting demographic structure of short- and long-lived pioneer tree species on Amazonian forest edges. Biotropica. doi: 10.1111/j.1744-7429.2012.00882.x

Bani, L., Massimino, D., Bottoni, L. and Massa, R. (2006). A multiscale method for selecting indicator species and priority conservation areas: a case study for broadleaved forests in Lombardy, Italy.Conservation Biology, 20, 512–526.

Bouroncle, C. & Finegan, B. Tree Regeneration and Understory Woody Plants Show Diverse Responses to Forest–Pasture Edges in Costa Rica. (2011) Biotropica DOI 10.1111/j.1744-7429.2011.00750.x

Cintrón, G., Lugo, A.E.,  Pool, D.J. & Morris, G. (1978) Mangroves of arid environments in Puerto Rico and adjacent islands. Biotropica 10, 110-121.

Reichoff, L., Bohnert, W.& Knapp, H.D. (1978) Die Vegetation des Naturschutzgebietes, “Tote Täler” -Vegetationsdifferenzierung im Übergangbereich zwischen Wald und Rasen. Archiv für Naturschutz und Landschaftforschung 18, 141-150.

Valladares, F., Wright, S.J., Lasso, E., Kitajima, K. & Pearcy, R.W. (2000). Plastic phenotypic response to light of 16 congeneric shrubs from a Panamanian rain forest. Ecology 81,1925–1936.

Harsch, M. A., Buxton, R., Duncan, R. P., Hulme, P. E., Wardle, P., Wilmshurst, J. (2012), Causes of tree line stability: stem growth, recruitment and mortality rates over 15 years at New Zealand Nothofagus tree lines. Journal of Biogeography, 39: 2061–2071.

Dear, thank you all for your replies.

I wish to specially thank Audrey and Andy for their thorough considerations and useful suggestions...the discussion is getting even more intriguing.

This also leads me to hope for a cooperative research.

rocco