Again, I think we need a fast, robust, repetitive, and synoptic way to assess human driven disturbance over large areas. That means we have to use remote sensing data. A response to Andreas. I wonder if you can find any "reference state" for a forest community without natural or anthropogenic disturbance in the real geographic world. This is not just a theoretical concept but a very practical one stemming from the recognition of the primary role of anthropogenic disturbance pervading the dynamics of systems like forests. I do not think that large-scale wind throws or biotic disturbances like insect attacks (e.g. bark beetle infestations) can be considered "usual" disturbances. In trying to discriminate between "natural" and human driven disturbances that can apparently overlap if observed at a single scale of observation, we have to consider the differences between large scale processes, like you mentioned (but I would add climate change), and human driven processes, which are characterized by different frequency, magnitude and operational scales. In particular, human driven processes are more "localized" in the scale of operation than the others, and are more frequent and systematic in occurrence. Thus if you want to discriminate between the two kinds of disturbance, you need a cross-scale approach for disturbance recognition (in space and time). Because of such differences in the operational scale, you would not run the risk of "automatically" classifying as 'anthropogenic' disturbances" what is due, for instance, to climate change or insect infestations. This also because you should interpret the disturbance pattern not "in isolation" as it could be for an automatic classification, but rather in the context of other available information on the map like the pattern of human presence and activity, for instance, roads, settlements, urban sprawl, and agricultural practices.

See for example some papers on cross-scale analysis of disturbance "Disturbance patterns in a socio-ecological system at multiple scales" or "Source/sink patterns of disturbance and cross-scale mismatches in a panarchy of social-ecological landscapes".

What sort of disturbances are you looking at? That will help you decide what and how to measure it. If you are looking at something like logging, measures such as patch size and distance to edge might be important. If you are looking at development, then housing density is a common measure. If you are looking at human intrusion, such as hiking or driving, then passage rates (# of people/cars per unit time) could be a measure. Some papers to look at might be Peres et al. 2006, Trends in Ecology and Evolution 21:227-229, Ramierez-Marcial et al. 2001, Forest Ecology and Management 154:311-326, Liir et al. 2007, Forest Ecology and Management 250:34-46. There are also a couple of books on the subject;

Alternatives to deforestation: steps towards sustainable use of the Amazon rain forest, and Tropical forests in transition: ecology of natural and anthropogenic disturbance processes.

we analysed "anthropogenic disturbance" in terms of patch configuration and age of forests. To analyse the influence of fragmentation on species richness two approaches were adopted: first we applied species/area analysis using the total species richness and then we focused on the different effects due to the ecological characteristics (traits) of the forest plant species. Plant traits were chosen for their significance in plant community assembly processes in fragmented landscapes, i.e. traits associated with the response of plants to the loss of habitat and linked environmental changes

(response traits). Habitat quality, in terms of suitability for forest species, was found to be important in explaining the presence of species of high conservation value, but patch age (as an indicator for habitat quality) played a major role too.

There is a good paper about the topic in Trends in Ecology and Evolution (TREE): Peres, C.,Barlow, J and Lawrence, W. F. 2006. Detecting anthropogenic disturbance in tropical forests. TREE. 21 (5): 227-229.

I guess that there is a two-fold challenge with the assessment of 'anthropogenic disturbances'. First you may need to define a 'reference state' of your plant communities without disturbances in order to isolate the disturbance effects from other environmental factor. Then, you might have to distinguish between 'natural' and 'anthropogenic' disturbances. There is perhaps useful method paper you can find at http://institute.redlands.edu/emds/manuscripts/pdf/pnw_2004_hessburg001.pdf by Hessburg et al. (2004) based on experiences from temperate zones.

Indian forest communities have generally a high anthropogenic influence. If you are looking for overall diversity , ie flora as well as fauna, factors such as logging/lopping, amount of NTFP collection, human presence, etc are significant factors.

I want to bookmark this question, How can I d this? :))

In this regard you must know the previous state of that forest area, through secondary data and previous works by the department. just go through it you will automatically came to know the variations in it and make it the difference with year (how far it take to change). it may give you an ideology about the consequence.

from this point you may start your work based on the recommendations by above mentioned.

Hi, I am doing a research into forest management (logging) a how it influence ground-dwelling spiders, I have study sites on clear cuts and control plots without management. I will measure also Ph of soil and canopy openness for calculations.

Dear Ludmila Cernecka, Ground dwelling spiders also need an cover of canopy above their web, please measure the temperature of the surface area and within the habitat of the spiders, may temperature play some role in habitation rather that Ph.

Thx K.Mophin Kani, it is a very good suggestion to measure also temperature.

I would like to second the suggestion by Dr. Cernecka about measuring change in temperature. Temperature can be a very good indicator of whole ecosystem change (for solid thermodynamical reasons). See and .

Dear Ludmila. It is important to note that if temperature difference between sites is significant then it can be an important variable to consider.

It is always difficult if not impossible to look for a "reference" state without disturbance (a kind of pristine state?) as it is always difficult to decide how far the bio-ecological clock must be reset. On the other hand, looking for a reference state in another "similar" system is also very difficult (with a lot of inference problems) as all system are unique and dynamic with their own memory and the "reference" state is changing too along its own trajectory. I think that, operationally, in order to measure anthropogenic disturbance in a forest one has to consider that natural forest processes are usually slow processes, whereas humans accelerate all processes and often determine sudden and abrupt changes like, for instance, in land cover due to land use conversion. So one way to estimate anthropogenic disturbance is looking at the difference in the frequency and magnitude of change of some state variable like, for instance, NDVI (normalized difference vegetation index) from remote sensing. NDVI is a very robust indicator of photosynthetic activity that can be used to discriminate among fast and slow processes in a synoptic manner. With a few exceptions due to natural fires and weather extremes from climate change that, however, can be identified as they operate at different spatial-temporal scales respect to humans also determining gaps and fragmentation in forests, the use of entropy-related indices based on NDVI time-series could be a way to approach the problem of identifying human disturbances. There are some papers in the literature dealing with anthropogenic disturbance estimates across scales (see, for example, "Highlighting order and disorder in social-ecological landscapes...." on Landscape Ecology)..

I believe there is simple and common sense approach to this question, when we are talking about forest we are more concerned about trees than other things. In forest tree felling for anthropocentric reasons is a common phenomenon. Disturbance in forest could be assessed by calculation of number of tree stems cuts to those standing live in given quadrate to arrive at disturbance index. Secondly a complex way of getting a good picture of overall effect on forest would be to arrive at Regeneration versus harvest value. When these studies are related with overall productivity, regeneration and nutrient dynamics would give better insights into ecosystem functioning in such fragile areas.

I agree with Giovanni that the 'reference' state of a forest community (without disturbances) is dynamic due to environmental change; and it is a theoretical concept, since disturbances are always part of forest dynamics. However, it might be helpful to use this concept if you are interested in 'anthropogenic disturbances' in particular. However, I do not agree with the statement that natural disturbances in forest ecosystems are usually slow processes. Think of abiotic disturbances like large-scale windthrows or biotic disturbances like insect attacks (e.g. bark beetle infestations). With your suggested method this disturbances would be automatically classified as 'anthropogenic' disturbances.

I think you from all the suggestions and references mentioned by previous researchers, you can tailor fit procedures that you think will be most appropriate and practical in your area. I will just share with you some of my experiences dealing with measures of anthropogenic disturbance: One is using forest vegetation as a measure like determining proportions of native and introduced species (e.g. trees). Another is counting tree stumps from tree harvesting and natural fallen logs (which usually indicates minimal tree harvesting). Also measuring diameter at breast height and tree distances will give an idea whether the forest is an old growth or secondary. In my area, I also considered investigating faunal composition. I observed that greater proportion of forest restricted animals are present in less disturbed forests. On the other hand, proportion of species associated with anthropogenic disturbance is higher in human altered forests. You can establish baseline information on specific forest parameters and continue forest monitoring so that you can better observe and account for forest dynamics especially those that are unique in your study area. Glad that you are also into forest conservation.

I think we need a fast, robust, repetitive, and synoptic way to assess human driven disturbance over large areas. That means we have to use remote sensing data. A response to Andreas. I wonder if you can find any "reference state" for a forest community without natural or anthropogenic disturbance in the real geographic world. This is not just a theoretical concept but a very practical one stemming from the recognition of the primary role of disturbance in the dynamics of systems like forests. I do not think that large-scale wind throws or biotic disturbances like insect attacks (e.g. bark beetle infestations) can be considered "usual" disturbances. In trying to discriminate between "natural" and human driven disturbances that can apparently overlap if observed at a single scale of observation, we have to consider the differences between large scale processes, like you mentioned (but I would add climate change), and human driven processes, which are characterized by different frequency, magnitude and operational scales. In particular, human driven processes are more "localized" in the scale of operation than the others, and are more frequent and systematic in occurrence. Thus if you want to discriminate between the two kinds of disturbance, you need a cross-scale approach for disturbance recognition (in space and time). Because of such differences in the operational scale, you would not run the risk of "automatically" classifying as 'anthropogenic' disturbances" what is due, for instance, to climate change or insect infestations. This also because you should interpret the disturbance pattern not "in isolation" as it could be for an automatic classification, but rather in the context of other available information on the map like the pattern of human presence and activity, for instance, roads, settlements, urban sprawl, and agricultural practices.

See for example some papers on cross-scale analysis of disturbance "Disturbance patterns in a socio-ecological system at multiple scales" or "Source/sink patterns of disturbance and cross-scale mismatches in a panarchy of social-ecological landscapes".

You may be interested in the Special Section on “Habitat disturbance and tropical rainforest mammals” which I edited in 2000 (Conservation Biology 14(6): 1574-1703). This special section is comprised of 12 papers on different types of anthropogenic disturbance (e.g., logging, road building and road networks, forest fragmentation, land-cover changes, colonization, hunting, and others) and their effects on tropical mammals. Some of the papers address more than one type of disturbance and provide information on how disturbances interact. Other relevant issues such as species richness, abundance, biomass, and diversity, population viability, community changes, forest structure, seed dispersal, population monitoring, indicator species and biotic indicators of disturbance, forest restoration, and protected area effectiveness are also considered. The special section includes work undertaken in 9 countries in Africa, Asia and America. This group of papers emphasizes the situation of tropical rainforest mammals, which can serve as sensors of the rhythms affecting tropical rainforest, but the scheme is general and can be used as a framework in the study of disturbance and the conservation of natural ecosystems and their biota.

I will attach all the papers, if I can learn how to use this.

Carrillo, E., G. Wong, and A.D. Cuarón. 2000. Monitoring mammal populations in Costa Rican protected areas with different hunting restrictions. Conservation Biology 14(6): 1580-1591.

Escamilla, A., M. Sanvicente, M. Sosa, and C. Galindo-Leal. 2000. Habitat mosaic, wildlife availability and hunting in the tropical forest of Calakmul, Mexico. Conservation Biology 14(6): 1592-1601.

Fa, J.E., J.A. García Yuste, and R. Castelo. 2000. Bushmeat markets on Bioko Island as measure of hunting pressure. Conservation Biology 14(6): 1580-1591.

Is there a way in which I can include several file attachments together?

Wilkie, D., E. Shaw, F. Rotberg, G. Morelli, and P. Auzel. 2000. Roads, development, and conservation in the Congo Basin. Conservation Biology 14(6): 1614-1622.

Malcolm, J.R. and J. Ray. 2000. Influence of timber extraction routes on Central African small mammal communities, forest structure, and tree diversity. Conservation Biology 14(6): 1623-1638.

Laidlaw, R.K. 2000. Effects of habitat disturbance and protected areas on mammals of Peninsular Malaysia. Conservation Biology 14(6): 1639-1648.

Chiarello, A.G. 2000. Density and population sizes of mammals in remnants of Brazilian Atlantic Forest. Conservation Biology 14(6): 1649-1657.

Lopes M.A and S.F. Ferrari. 2000. Effects of human colonization on the abundance and diversity of mammals in eastern Brazilian Amazonia. Conservation Biology 14(6): 1658-1665.

Medellín, R.A., M. Equihua, and M.A. Amín. 2000. Bat diversity and abundance as indicators of disturbance in Neotropical rainforests. Conservation Biology 14(6): 1666-1675.

Cuarón, A.D. 2000. Effects of land-cover changes on mammals in a Neotropical region: a modelling approach. Conservation Biology 14(6): 1676-1692.

... and last, but not least:

Galindo-González, J., S. Guevara, and V.J. Sosa. 2000. Bat- and bird-generated seed rains at isolated trees in pastures in tropical rain forest. Conservation Biology 14(6): 1693-1703.

Sorry, for not being able to send all the files in a single communication.

Again, I think we need a fast, robust, repetitive, and synoptic way to assess human driven disturbance over large areas. That means we have to use remote sensing data. A response to Andreas. I wonder if you can find any "reference state" for a forest community without natural or anthropogenic disturbance in the real geographic world. This is not just a theoretical concept but a very practical one stemming from the recognition of the primary role of anthropogenic disturbance pervading the dynamics of systems like forests. I do not think that large-scale wind throws or biotic disturbances like insect attacks (e.g. bark beetle infestations) can be considered "usual" disturbances. In trying to discriminate between "natural" and human driven disturbances that can apparently overlap if observed at a single scale of observation, we have to consider the differences between large scale processes, like you mentioned (but I would add climate change), and human driven processes, which are characterized by different frequency, magnitude and operational scales. In particular, human driven processes are more "localized" in the scale of operation than the others, and are more frequent and systematic in occurrence. Thus if you want to discriminate between the two kinds of disturbance, you need a cross-scale approach for disturbance recognition (in space and time). Because of such differences in the operational scale, you would not run the risk of "automatically" classifying as 'anthropogenic' disturbances" what is due, for instance, to climate change or insect infestations. This also because you should interpret the disturbance pattern not "in isolation" as it could be for an automatic classification, but rather in the context of other available information on the map like the pattern of human presence and activity, for instance, roads, settlements, urban sprawl, and agricultural practices.

See for example some papers on cross-scale analysis of disturbance "Disturbance patterns in a socio-ecological system at multiple scales" or "Source/sink patterns of disturbance and cross-scale mismatches in a panarchy of social-ecological landscapes".

To follow-on from Alessandro's closing comment, I think that at some point you would want/need to do botanic sampling in order to characterize forest structure at the local and/or microhabitat levels. Within that sampling, you could also collect data that would be indicative of anthropogenic disturbance and reflect *how* humans were disturbing forest habitats, because forest structure and anthropogenic disturbance of that forest structure are obviously inimately linked.

My doctoral research was on black lemur socioecology in northwestern Madagascar, and one of the key issues I was interested in was how anthropogenic habitat disturbance might affect black lemur socioecology. To examine this, I used the sort of "blended" botanic sampling I described above.

The late tropical botanist/ecologist Alwyn Gentry used as rule-of-thumb that 1000 sq. metres was an adequate sample area to indicate plant species diversity (see: Gentry, A.H., 1982. "Patterns of neotropical plant species diversity"; pp. 1-84 in: Evolutionary Biology, vol. 15 (M.K. Hecht, B. Wallace, and G.T. Prance, eds.). New York: Plenum Press). Gentry's usual method was to lay out 10 botanic transects across a study area, each measuring 50 x 2 metres. Because the site at which I worked in Madagascar was a mosaic of succesional and disturbed secondary forest, I modified Gentry's botanic transect method and instead utilized 10 botanic sample plots each measuring 10 x 10 metres (which still gave me a total area sampled of 1000 sq. metres). Botanic sample plot position was chosen to best reflect forest structure charactereistic of the home range areas of each of my four study groups of black lemurs -- sample plots were placed so that three plots were located in the home range area of each study group, three of the sample plots were placed in areas of home range overlap between three different study group pairs, and at least one sample plot was placed in the "core area" (i.e., area of exclusive use) of each study group's home range.

I recorded the following data in each botanic sample plot (and you could revise/modify these variables to best suit the Indian forests in which you are working):

i) all plants, and the number of individuals of different species, > 2.5 cm dbh (diameter at breast height, where 1.37 m = breast height) rooted in the sample plot were counted and identified (where possible);

ii) the number of larger trees (i.e., dbh = 10-20 cm; 20-30 cm, and > 30 cm) within the sample plot (where sampled trees have buttresses, diameter is measured above the butresses);

iii) the number of saplings < 2.5 cm dbh within the sample plot;

iv) the number of palm trees (including understory species) within the sample plot;

v) the number of "traveller's palm" (Ravenala madagascariensis) within the sample plot -- this endemic Madagascar tree species is frequently found in disturbed and forest edge areas (clearly, this is a variable you would need to revise -- perhaps there are analogous colonizing species in India that do very well in disturbed forest patches);

vi) the number of individuals of each of four different species of "screw pine" (Pandanus spp., Pandanaceae) within the sample plot (Pandanus also do quite well in disturbed areas);

vii) the number of cinnamon plants (Cinnamomum aromaticum, Lauraceae) in the shrub-layer of the forest -- this plant is not native to Madagascar, so presence of this exotic species was taken as an indicator of habitat disturbance;

viii) the number of tree stumps and/or trees with cut-marks on them -- this provided a direct indicator of anthropogenic habitat disturbance;

ix) the abundances of lianas, bamboos, and epiphytes were scored on a 5-point ordinal scale: 0 = absent; 1 = rare; 2 = occasional; 3 = frequent; 4 = abundant (e.g., light-loving lianas could show high abundance in anthropogenically disturbed forest habitats, such as forest edges);

x) estimated average height of the forest canopy in the sample plot; and,

xi) the estimated percent canopy cover in the sample plot (this can be quickly scored a couple of ways; one would be to use a camera with a fish-eye lens to capture a hemispherical view of the forest canopy -- a "quick-and-dirty" way to achieve the same ends would be to use binoculars and view the forest canopy through the objective lens end of the binoculars).

I also employed multiple measures of plant species diversity, again in light of the habitat heterogeneity with which I was faced at my study site. I used the Shannon-Weiner diversity index, which varies between 0 (monospecific stand, zero species diversity) and 1; increasing values of the Shannon-Weiner diversity index reflect greater species diversity. I also used Margalef's index, which is equal to ( the number of species recorded - 1) / ln (the total number of individuals of all species recorded); the value of Margalef's index increases, in general, with greater species diversity, particularly under conditions where relatively few individuals of a great many species are recorded. Finally, I also used the Berger-Parker index 'd' = (the number of individuals of the most abundant species)/(the total number of individuals of all species recorded); that is, the Berger-Parker index 'd' is the % abundance of the most abundant species. A tremendous reference that provides applications and worked examples of these diversity measures is Anne Magurran's book "Measuring Biological Diversity" (2004, Blackwell Science, Ltd., Oxford -- a partial pdf version of the book -- chapters 2-4, plus references -- is available at: http://www2.ib.unicamp.br/profs/thomas/NE002_2011/maio10/Magurran%202004%20c2-4.pdf ); another great source by Magurran is her 1988 book, "Ecological Diversity and Its Measurement" (Princeton University Press, Princeton).

Lastly, to return to my earleir comment about gaining data on *how* humans are disturbing forest habitat, you should also consider including an ethnobotanic interview component to your study. By speaking with local people, particularly any plant specialists that people might recognize in their communities (for the use of medicinal plants, for example), you could gain a wealth of data on which plant species might particularly be under human pressure, be it for ethno-medicinal use, or for preferred materials for building, weaving, etc. Such targeted exploitation of certain species by humans would be a factor to consider in interpreting whatever patterns of anthropogenic forest disturbance you end up uncovering.

I agree with others that there should be a reference forest site without human disturbance as a comparison for what is natural. I have been working with others on a sampling method for studying biodiversity changes in disturbed areas using sound recordings. The soundscape of a forest can offer a wonderful data set of species richness and forest health measured by the number of species heard over time within a sound file. Interestingly also, is that vocal animals will fill different frequency ranges, where insects may call at higher frequencies than frogs. This theory is called the acoustic niche hypothesis. A study done by Krause and Gage in the forests of California found that vocal species were much less diverse in disturbed areas than non-disturbed areas. If trapping animals is a logistic problem or expensive, sound recorders (especially Wildlife Acoustics SM-2s) are relatively cheap and very efficient. There are a number of sound analysis tools out there too but feel free to contact me if you are interested.

Have you had much success in finding an answer to this question yet? My research team is also grappling with this question as we're trying to develop an "index" of sorts to reflect degrees of disturbance for the different forest fragments we study. Would be happy to discuss further if you think that would be helpful?

Thanks allot 'Sarah Paige' for your proposal. I found many good suggestions and references related with the topic, but still I am not fully satisfied. I want to show these disturbances of the forests surrounded by the human population in the real sense. I will be very thankful if you & your team help me to do it.

I don't know if it will help, but there are a number of papers out there that have come up with different ways to quantify anthropogenic disturbance outside protected areas. May you can adapt that to your project? See my 2004 paper in Landscape Ecology (Wiersma, Nudds, Rivard) for one example of calculating a "human footprint" and relating it to species losses. Other work (cited in my paper) has looked at human population density (see Woodroffe and Gisberg) as well as forward citations to each.

Thanks Yolanda! I will take a look at your piece. The concept of "human footprint" is something I hadn't thought about. I wonder if my search terms were just inaccurate.