Hi Alena, in reality there is no good or bad way to measure diversity, it is all relative to your research question, what you really want to give more weight to (i.e., common species, uncommon species, etc). However, most measures, but species richness, have serious problems in reflecting the 'true' diversity and are difficult to compare. One very clever alternative is to use instead Hill numbers, which are related to these traditional ways of measuring diversity (Shannon, Simpson, etc) and deal with the comon problems they have. You can read a number of recent papers that describe in more detail what the Hill numbers are, but I would start with Jost 2006 Oikos 113:363 and Jost 2007 Ecology 88:2427 (and check Jost's web page www.loujost.com). Hope this helps.

Please have a look at this useful ResearchGate link.

https://www.researchgate.net/post/what_is_the_deference_between_Shannon_Wiener_diversity_Index_and_Simpson_diversity_Index

Please have a look at this link also.

https://www.researchgate.net/post/Does_anyone_have_experience_using_the_Shannon_Wiener_Diversity_Index_Score

Dear Simon Wengrat and Arvind Singh, thank you very much for your replies.

Please have a look at this link and PDF attachment also.

https://www.researchgate.net/post/Which_biodiversity_index_is_more_reliable_better_and_how_can_it_be_applied_to_a_mixed_vegetation_How_might_I_differentiate_diversity_indices_and_IVI

Dr. Arvinnd Singh, nice answers.

If I had to choose one I would use species richness; however, I believe that it is better to use more than one.

Dear Alena, reference is giant but to start you should see the

Anne E. Magurran Ecological Diversity and Its Measurement

First edition published by Croom Helm Ltd 1988 Reprinted by Chapman & Hall 1996, Springer-Science+Business Media, B.Y.

Call me, please, by RG mailing for PDF.

As well you should precise what type of b. you have to do.

Below I give you a part of the very good interpretation from Encyclopedia of biodiversity_2013 - p. 399-410: (PDF I'll give you the same way like previous):

Biodiversity, Definition of Ian R Swingland,

University of Kent, Canterbury, UK r 2013 Elsevier Inc. Glossary Biodiversity/biological diversity Species, genetic, and ecosystem diversity in an area, sometimes including associated abiotic components such as landscape features, drainage systems, and climate. Diversity indices Measures that describe the different components of biodiversity, such as species richness (alpha diversity), beta and gamma diversity, endemicity, and higher taxon richness. Ecosystem diversity Diversity of habitats, ecosystems, and the accompanying ecological processes that maintain them. Endemicity State of a species or other taxon being restricted to a given area, such as a specific habitat, region, or continent. Flagship species Charismatic or well-known species that is associated with a given habitat or ecosystem and that may increase awareness of the need for conservation action. Genetic diversity Genetic variety found within or among species; this diversity allows the population or species to adapt and evolve in response to changing environments and natural selection pressures. Keystone species Species that has a disproportionately greater effect on the ecological processes of an ecosystem, and whose loss would result in significantly greater consequences for other species and biotic interactions. Organismal (species) diversity Number and relative abundance of all species living in a given area. Species richness Absolute number of species living in a given area (also called alpha diversity), giving equal weight to all resident species. Use values Values that are obtained by using a natural resource, such as timber, fuelwood, water, and landscapes. These include direct, indirect, option, and nonuse values.

Some detiles you can find in my several articles (majority have English versions), like

Community structure in cryophilic infusoria. Article Community structure in cryophilic infusoria.

and some other.

Andrey

Я для описания рыбного населения использовал следующие индексы:

Шеннона (Shannon, Weaver, 1949), Пиелу (Pielou, 1966), Животовского (Животовский, 1980), Симпсона (Simpson, 1949), Маргалефа (Margalef, 1958),

Для сравнения близости видового состава выборок взят индекс Сьёренсена (Sörensen, 1948). Для сравнения бóльшего числа списков видов – его обобщённая форма – индекс биотической дисперсии Коха (Koch, 1957).

Hi Alena

I think this is s good idea for you to read this paper as follow, you can download it easily.

Choosing and using diversity indices: insights for ecological applications from the German Biodiversity Exploratories

Kind Regards

Hi Alena, in reality there is no good or bad way to measure diversity, it is all relative to your research question, what you really want to give more weight to (i.e., common species, uncommon species, etc). However, most measures, but species richness, have serious problems in reflecting the 'true' diversity and are difficult to compare. One very clever alternative is to use instead Hill numbers, which are related to these traditional ways of measuring diversity (Shannon, Simpson, etc) and deal with the comon problems they have. You can read a number of recent papers that describe in more detail what the Hill numbers are, but I would start with Jost 2006 Oikos 113:363 and Jost 2007 Ecology 88:2427 (and check Jost's web page www.loujost.com). Hope this helps.

This article might be useful on calculating species diversity using R:Article BAT - Biodiversity Assessment Tools, an R package for the me...

Hi

In my opinion. you calculate all the alpha diversity indices. which is better value in your work area later on. There are many software available for calculating diversity indices. then you can decide on the diversity values you have achieved and then compare it with the statistical methods.

Dear Alena,

Shannon Information (Shannon-Wiener is a widely spread incorrect name of this index) is good to compare old studies where only this index was usednwith a newer results. You do not give details of your data set so my advice cannot be specific. The most intuitive measure is species richness and if your data set permits it, it can be estimated with a reasonable precision using monte-carlo approach (rarefactions, etc.). You can download software (free) such as EstimateS, PAST.3, EcoSim. EstimateS and Ecosim have very good manuals to understand what you really are calculating; the most useful estimates are Chao 2 and Jackknife 2.

Hill numbers are very good for comparisons as already mentioned in above posts by Jaime Pinzon. Among simpler approaches, the reverse of Simpson's dominance is a good measure too, it shows how many species are easy to find in a community.

No one index gives a picture of biodiversity and species richness gives the least information. Indeed read Magurran (2004 ) or Gaston and Spicer or you could look at my papers on Research Gate (esp. Feest et al. 2010) and in Ecological Indicators.

So the answer is probably that it depends on your question!

Dear Sir

Shannon and Weiner index is formulated for comparing the diversity of two or more ecological conditions or same ecologicals condtions on temporal scale. However the value of the index is likely to increase in if consider more area, it is important to note that very dissimilar habitats can not be compared as they will certainly have difference. many people use it for assessment of pollution but the population can be changed due other activities also like over exploration or use of biodiversity of a particular area/ degradation of habitat / any type of disturbance of the habitat. many ecosystems are naturally gifted with lots of species while some are with less depending upon natural conditions. that does not mean the ecological entity with lesser number or diversity of species is polluted. if these areas are subjected to assessment of indices t will certainly indicate the change may be by any reason. In case of pollination, the species which are tolerant to particular type of pollution will only multiply leading to dominance hence less value of index. The value is highest when more number of species are present and represented by equal numbers. It should be used carefully.

Hill numbers are probably the best way to go (see Jaime Pinzon's post), and you can calculate them easily in R using the 'vegetarian' package. Hill numbers don't have all the issues that other diversity indices have.

Dear Alena, a common way is to use biodiversity profiles because they consider simultaneously richness, Shannon, and Simpson indices. Here you can find 2 papers dealing with this.

To plot profiles and measure biodiversity in an instant of time as the area under the curve you can use the R package BioFTF. In this way, you get a scalar that takes into account different biodiversity indices simultaneously:

Article BioFTF: An R Package for Biodiversity Assessment with the Fu...

This is based on Patil and Taille biodiversity profiles.

To assess changes in biodiversity profiles from one time observation to another using the volume, you can read this new paper:

Article A functional approach to Hill's numbers for assessing change...

This is based on Hill's biodiversity profiles.

Regards

threat index

Dear colleagues,

Thank you all very much for your replies and for the information from everyone. 

This is an important question with many possible answers. In addition to the excellent answers already given, there is a bit more to add.

Here is a selection of measures of diversity of species:

number of clusters (colonies): Each member of a species has the potential to enter into an existing colony or start a new colony. For example, wrens tend to be anti-diverse, typically starting each season by trying to attract more than one mate. By contrast, chickadees encourage diversity by typically beginning each spring by selecting one mate to start a family. In effect, the number of colonies of chickdees is potentially greater than the number of colonies of wrens.

number of offspring: It well-known that each member of a family has distinguishing features that separate that member from its siblings. Hence, parents that give birth to large numbers of offspring provide a natural transition to greater diversity of a species.

Hello, Shannon has always been considered a good index, it can bring a good idea about species diversity by assigning equal weights to rare and common species. However, their comparative power in communities has unsatisfactory properties, because not always half the value of the index corresponds to half of the value in species. So I suggest you use the most recent proposed metrics in the update, which uses an adaptation of the Hill numbers outlined by Ane Chao in 2016.

Here's the article on how to calculate diversity metrics:

http://chao.stat.nthu.edu.tw/wordpress/paper/120_pdf_appendix.pdf

or

Article iNEXT: An R package for rarefaction and extrapolation of spe...

The suitability of these indices is mainly context related. But bear in mind that Shannon might not be appropriate sometimes as it scales with species richness. You can also explore Pielou’s evenness, which is a standardized version of the Shannon index (i.e. Shannon index divided by the natural logarithm of the species richness). The vegan R package is excellent resource for exploring these indices in details.

Simpson's Diversity Index that considers species richness and evenness may also be used.

Read this, It contains several sample cases!

May these articles (see files) be useful for you.

Alena, Jaime has made some good relevant points. There are many indices of diveristy in the qualitative community ecology and biological diversity literature. As we know many of these indices attempt to accommodate a fair weighthing of both the rare and common species in a community. Also when we talk about species and community diversity, we need to be aware that species diversity encompasses three aspects; species richness, species heterogeneity and and species equitability. Therefore, depending on the actual alogorithms used, there may be biases with respect to these dfifferent aspects of community/species diversity. I would recommend you to consult the book by Professor Anne E. Magurran, Professor of Ecology and Evolution at the University of St Andrews, titled "Measuring Biological Diversity", ISBN: 978-1-118-68792-5 264 pages April 2013, Wiley-Blackwell. It covers and reviews the relative usefullness of the different indices. It will be a good place for you to start from and then based on your objective (s) as Jaime advises, you can make a decision on which index or indices you want to settle for.

I agree with Alen. No universal index provides the real diversity and species richness. Hence according to your scope you have to select appropriate index. indices can be divided in to species richness indices, species dominance indices and species evenness indices. see maguran.

These are some highly cited papers which measured different diversity indices. Have a look....

http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2435.2010.01695.x/full

http://www.pnas.org/content/106/23/9322.short

Article Empirical Relationships between Species Richness, Evenness, ...

https://hal.archives-ouvertes.fr/halsde-00293183/

I would like to thank everyone very much for the helpful replies. I really appreciate it.

this link may help you

Assessing the suitability of diversity metrics to detect bio...

The Shannon index has been widely criticized, since May (1975) has shown that it is a measure very insensitive to the characteristics of the abundance distribution of the species. This author defends the employment of Simpson and Berger-Parker. Taylor (1978) is inclined to the use of  because of its great discriminatory capacity and the fact that it is not very affected by the sample size, in addition to being less sensitive than Shannon or Simpson to the abundance of the dominant species. However, as we have seen, it can not discriminate between samples with S and N equal. In summary, we can say that The indices that reflect the elements of species richness are better discriminatory between samples but are more affected by the sample size than the series of measures of dominance / uniformity. The indices , Berger-Parker S and Margalef have the advantages of being simple to calculate and easy to interpret. It is important to consider before the choice of an index, on the one hand, the conceptual aspects such as sensitivity, the influence of the abundance model, or the assumptions of the methods; and on the other, the methodological problems of the sampling itself that refer to the quality of the estimate, such as the existence of biases or the influence of the sample sizes. The problems of sampling are important and the usual assumption of randomness even doubts of estimators considered unbiased but developed under this assumption. In general, it can be recommended that the following suggestions be taken into account, when possible: Make sure that samples are, if possible, of equal sizes, and large enough to be representative. Draw the range / abundance graphs to have the first indication of the differences in the distribution forms of the data (logarithmic, geometric, lognormal or broken stick model). Calculate simple indexes to initially have a quick measure of diversity, but use deeper ones in the final analyzes. Replicate as much as possible the samplings to apply statistical tests in case it is possible Pielou (1966c), on the other hand, has tried to evade the problems of the indices by means of the method of sequential aggregation or accumulation of the indices, which also allows obtaining a robust H 'with an associated variance estimate (with the same advantages than the jackknife application.). for this, it accumulates the sample units by calculating the indices (in this case H ') until by visual inspection of its behavior it detects the stabilization of the curve, except for the normal random variations. In this method, however, the shape of the curve depends on the order of accumulation, so Heyer and Berven (1973) extend the method by repeating the procedure for different random orderings of the accumulation sequence in order to obtain an improved estimate, as well as a standard error. Finally, to improve the estimates of the indexes, the methods of resampling (Jackknife and bootstrap) are being used recently.

Dear All,

I would like to share this new method for measuring species diversity, and ranking and clustering ecological communities

Regards

Fabrizio

on Researchgate:

Article Unsupervised classification of ecological communities ranked...

free online for one month at:

https://www.sciencedirect.com/science/article/pii/S1470160X18301651

with R script (Appendix A. Supplementary data with data and script)

In general cases the Shannon-Weiner index is more commonly used. It is an information statistics index, which means it assumes all species are represented in a sample and that they are randomly sampled. Here all species present in a community are equally considered for measuring the index.

But the Simpson index is a dominance index because it gives more weight to common or dominant species in the community. Here, the species which are represented by only few individuals will not affect the diversity. So, if we have a community where some species are very common and we are only interested about those common dominant species, then Simpson index is better for interpretation of the diversity.

Nice to see good answers.

Thanks

I agree with Pal. Shannon index is the widely used one

I agree with Dr. Arshad Ali, "it depends on your research questions and hypotheses".

Thanks Dr. Fabrizio Maturo for giving a very perspective!

Alena Rendeková I recommend this research paper may it help you

The Shannon diversity index (1949) is the most commonly used. It expresses the relative importance of the number of abundant species in a given environment. This H index is minimal if all the individuals in the stand belong to a single species, it is maximum when all the individuals are distributed equally over all the species (Frontier, 1983). In nature, whatever the taxonomic group studied, the Shannon diversity indices are between 0 and 4.5, rarely more (Frontier and Pichod-Viale, 1998). The value of the Shannon index varies according to the number of species present and their relative proportions. This makes its use less exhaustive for the comparison of two environments with very different specific riches. Thus, Pielou's regularity or fairness index (E) appears interesting to compare different environments (Devineau et al., 1984).

Ple se re d my p per on biodiversity qu lity where I point out th t Sh nnon is limited in use by the influence of species richness nd restricted numeric l r nge; Simpson's index voids these problems without ny need to ch nge the input d t

Sorry no first letter of the lph bet

Species diversity is a complex concept without universally-accepted definition. Species diversity indices and their corresponding Hill's Number (effective number of species) are preferred by some ecologists largely based on statistical arguments, instead of biological relevancy. The uses of these measures in biological monitoring, restoration and conservation are questionable at the best and misleading at the worst. It appears better to treat species richness, evenness, and composition separately than to combine these elements into a single measure. You might like to check this paper: Article Weighting effective number of species measures by abundance ...

Yong Cao

I agree totally with your point that using single indices for recording/registering biodiversity loses the very point of biodiversity (its multiple properties) and that it can only be truly understood by viewing the properties/indices holistically