Agrobiodiversity Indicators and Measurement using R for Description, Monitoring, Comparison, Relatedness, Conservation and Utilization

Agrobiodiversity is the most important part of biodiversity. It can be described, quantified, compared, and related by using different statistical tools called agrobiodiversity statistics (agro-statistics). Six components and 25 groups of agrobiodiversity should be used for agrobiodiversity analysis. Six types and levels of agrobiodiversity can be quantified. Both quantitative and qualitative data are used for estimating scores and indices. The measurement objects for describing agrobiodiversity are community, household, site, crop group, species, landrace


Introduction
Agrobiodiversity (also called agricultural genetic resources, AGRs) is a part of biodiversity and includes all genetic resources that are economically beneficial.In majority of the countries, native agrobiodiversity is neglected and underutilized due to their high priority to monomorphic and high yielding varieties.Many different factors are contributing to losing the AGRs.Among them the major factor is the rapid expansion of single improved homogenous varieties and breeds in the world.Such single improved variety is generally developed through studying a single species or variety or set of genotypes, and there are limited studies on the whole agrobiodiversity at a particular site.The general trend is that, rather than evaluating, describing and improving the native agrobiodiversity, improved variety or breed is easily adopted and expanded due to which many indicators are being affected.Indicators are any values, scores or status which explain about the agrobiodiversity of a particular location.Agrobiodiversity indicators have not been standardized across the world; and even the methodologies to estimate and measure the indicators are not available.Indicators are very important to manage the agrobiodiversity better, to plan programs and activities, and to monitor the trends (Sthapit et al., 2017;PAR, 2018).
For the conservation of forest biodiversity (non-agrobiodiversity), different indicators and approaches have been used, for example red listing of the species.Many types of species have been defined and given due attention.Different types of species include Alien, Charismatic, Dominant, Emblematic, Endangered, Endemic, Exotic, Flagship, Focal, Foundation, Indicator, Indigenous, Invasive, Keystone, Landscape, Priority, Rare, Specialty, Substitute, Surrogate, Target, Threatened, Tourism, Umbrella and Vulnerable species.Similar approaches can be applied at species and landrace level to support AGRs.Quantification of AGRs is another aspect that identifies such species or landraces.
Different types of scores and indices along with coefficients can be estimated and used as indicators (Joshi et al., 2005;Jarvis et al., 2000;Grum and Atieno, 2007)).Several statistical tools can be applied using computer software to quantify agrobiodiversity.Quantifications (measurements) of agrobiodiversity are generally done at different levels e.g., at the agroecosystem, species, varieties, and administrative units.Agrobiodiversity in any area should be estimated properly that leads to choosing the conservation approaches effectively.This paper, therefore, describes different operational agricultural units (OAU) for estimating diversity indices using R packages.Among the various components under agrobiodiversity statistics, this paper focuses on the measurement of agrobiodiversity.With the approaches described in this paper, one can rank any household, community, district, or the country and can locate a center of the diversity.A hotspot of agrobiodiversity and red zone for agrobiodiversity can be identified, in addition to identifying the indicator species and landraces.

Agrobiodiversity Components and Groups
Agrobiodiversity covers all genetic resources that have value for food, nutrition, health, and other economic uses to human beings.It has six components, and they are crops, forages, livestock, insects, microorganisms, and aquatic genetic resources (Joshi et al., 2020c).Insects and microorganisms include only economic and beneficial species.Under aquatic genetic resources, only economically important species are included e.g., fish.Each of these components can further be divided into four sub-components.They are cultivated/ domesticated, semi-domesticated, wild relatives, and wild edible species (Joshi and Shrestha, 2017;Joshi and Shrestha, 2019).
These components, sub-components, and economic groups (Joshi et al., 2020c;Joshi and Shrestha, 2019) are very useful to estimate different types of diversity indexes, indicators, and scores of a particular site, community, or household over a certain period.The AGRs may be of exotic and native types and both types can be considered for agrobiodiversity measurement, but measurement based on only native AGRs would be more valuable and important.There are many other grouping systems of AGRs (Joshi and Shrestha, 2019), and these groups can also be considered to quantify agrobiodiversity.

Agrobiodiversity Levels and Types
Agrobiodiversity can be measured and studied at different levels or hierarchies by using different traits.Based on levels (coverage of objects), there are six types of agrobiodiversity (Figure 1) (Joshi et al., 2020b;Bajracharya et al., 2012).Genetic diversity includes three levels of diversity i.e., varietal diversity, genotypic diversity, and allelic diversity.Agrobiodiversity can also be described under six types of diversity based on traits and use-values.These include functional diversity, morphological diversity, molecular diversity, use-value diversity, nutritional diversity, and food diversity.All these 12 types of diversity should be measured and studied at a particular site in a given period.Based on the data types, objectives, and objects, different measures are used to estimate and compare these different types of agrobiodiversity.Diversity can also be assessed based on cropping patterns, growing season, land type and habitat.at species and varietal levels.Morpho type is very simple indicator to measure the diversity.
With the development of different molecular markers and computing software, genetic parameters are also commonly estimated.Description of these tools has been described by Joshi et al. (2005).Both parametric and non-parametric tests are also commonly used to compare agrobiodiversity.Appropriate test statistics are given in figure 3 based on data types and the number of objects (factors) used.Both temporal and spatial analysis (called trend analysis) can be carried out to see the status and changes in agrobiodiversity.

Agrobiodiversity Measurement (Quantification)
Agrobiodiversity measurement includes the quantification of AGRs at different levels.Based on the quantification, AGRs can be grouped at the level of different strata e.g., red list, endangered, rare, common, etc. (Joshi and Shrestha, 2019).The main measures of agrobiodiversity are richness, evenness, diversity indices (Shannon, Simpson indices), similarity coefficients, dissimilarity coefficients, scores (Joshi et al., 2005;Kindt and Coe, 2005;Joshi et al., 2018;Jarvis et al., 2000;Grum and Atieno, 2007).Another measure is species density, which takes into account the number of species in an area.Similarly, landrace density can also be estimated.These measures should be measured at six different levels and types of agrobiodiversity (Figure 1) e.g., household, community, ward, municipality, district, province, and country.Such estimates are generally calculated based on native agrobiodiversity and are, therefore, useful for identifying the hotspot areas for agrobiodiversity.Quantification helps locate the center of diversity, identify the hotspot and red zone areas for agrobiodiversity.Hotspot areas are those areas that have the higher diversity score and indices, high diversity on wild relatives, endemic species, many rare and unique landraces, and species, and different types of land and cropping patterns.
Measurement (quantification) may be based on phenotypic, genotypic, perception, and survey data.Such data can be collected and measured through community biodiversity register and community seed bank, diversity block, diversity collection, diversity fair, field/transect walk, focus group discussions, food fair, household survey, key informant interviews, online survey, lab experiment, literature review, local market, on-farm, and on-station trials.Diversity changes over time and space are also estimated using different diversity measures, which are important for monitoring and applying appropriate methods for conservation and utilization.
For the index calculation at different levels, one can count the number of species within-group, or several landraces within species as well as group (PAR, 2018;Pudasaini et al., 2016;Borcard, Gillet and Legendre, 2011;Grum and Atieno, 2007;Joshi and Baniya, 2006).Taking the natural logarithms of species richness or landrace richness, an index can be calculated.The proportion of each group, species, or landraces can be calculated by dividing the number of that groups, species, or landraces by the total number of all groups, species, or landraces in a given area.The formula for calculating the Shannon diversity index, Simpson index, evenness, and other indices can be applied on these data.Agrobiodiversity index (ABDI) can be of household (HH), village or community, district, province, agroecozone, and country.A weighted index using either agrobiodiversity components or groups can be estimated as described in the literature 1 .In some cases, microorganisms, insects, ornamental plants, and the medicinal plant may be excluded from the calculation due to data unavailability.
The percentage of species or landraces in each group or species can be calculated considering the total number of species or landraces in the country or studied areas (Pudasaini et al., 2016;Joshi et al., 2018;Joshi et al., 2007).Based on the data obtained, each household or area or district can be ranked.For example, ABDI (based on landraces) for each household is equal to the number of landraces in each species or group divided by the total number of landraces in a community or district.

Agrobiodiversity Indicators (Score and Index)
1 https://news.mongabay.com/2016/05/top-10-biodiverse-countries/Agrobiodiversity indicators are any scores, indices, signs, symptoms, values, drivers, or marks that speak about the status of total diversity, trends on diversity, the status of intra-and inter-level diversity of species, and landraces in a particular area.It indicates that the agrobiodiversity is increasing, remaining constant, or decreasing.There is a wide range of methods of measuring various dimensions of agrobiodiversity, which is often referred to as the agrobiodiversity indicators, scores, and indices (Boversity International, 2017;Sthapit et al., 2017;PAR, 2018;Kindt and Coe, 2005;Joshi et al., 2020b).Diversity indicators, indices, and scores can be used to compare within and between different populations at species, landraces, and genetic levels over locations and years.
Agrobiodiversity indicators can be assessed at three different systems, namely, in consumption and market system, in production system, and in genetic resource management system (Sthapit et al., 2017).Some indicators include the red zone, red list, landraces coverage (based on five cell analysis), cropping pattern, mixture, monocrop vs. multicrops, land type, food items, native products in the market, the richness of species and landraces, population size, etc.A red list is the list of names of genetic resources (at genotype, landrace, variety, strain, and breed levels) under different groups based on the analysis of distribution and population size (also called five cell analysis), and trait distribution.Among these indicators, scores and indices are more commonly estimated and used.
Diversity indices and scores are calculated using both qualitative and quantitative data.In case of quantitative data, it needs to be converted into qualitative groups.The proportion of entries in i th class can be calculated using morphological data considering the different phenotypic classes of traits.Similarly, frequency data on genebank collection can be used to estimate different indices.Many ways can be used to estimate several types of household scores and indices.Household-level diversity can be of household diversity score and index as given below.

A1. Household Agrobiodiversity Score (HHABDS)
1. Number of species (species richness, n) in each of 6 agrobiodiversity components (crops, forages, livestock, economical insects, economically important microorganisms, aquatic agricultural species) over a year 2. Number of landraces (landrace richness, n) per species for each of 6 components in a year 3. Land type, n (marshy/ wetland, pond/aquatic, slopy upland, terrace upland, slopy low land, terrace low land, riverside, agroforestry land, grassland) 4. Functional diversity (number of special functions using special landraces) in a year 5. Unique diversity value (the number of specialty/ unique landraces divided by the total number of landraces) 6. Agrobiodiversity group score (or agrobiodiversity group richness) (based on 25 agrobiodiversity groups i.e., cereals, pseudocereals, millets, sugar and starch crops, grain legumes, oilseed crops, summer vegetables, winter vegetables, roots and tubers, winter fruits, summer fruits, spices, beverages and narcotics, fibers, forage trees, forage grasses, ornamental plants, medicinal plants, supportive plants, economical and beneficial (EB) insects, EB microorganisms, fish and aquatic animals, aquatic plants, poultry, and livestock), at 0 or 1 scale over a year with maximum 25 score 7. Dietary diversity score (based on 15 groups: cereals, pseudocereals, millets, roots and tubers, vegetables, fruits, nuts, meat and poultry, eggs, fish and aquatic animals, pulses and legumes, milk and milk products, oil/fat and ghee2 , sugar and honey, and miscellaneous) at 0 or 1 scale on half-year basis with maximum 15 score 8. Social agrobiodiversity score (number of religious or culturally associated landraces, considering all 6 agrobiodiversity components) 9. Food diversity score (number of food items/recipes eaten per meal, average of morning, day, and evening foods) 10.Food component score (number of species in food per meal, average of morning, day, and evening foods) 11.The average area per species (crops and forages) in square meter 12. HH agrobiodiversity score: sum from above 1 to 10 scores.Similarly, we can estimate agrobiodiversity indices and scores at district, province/ state levels or any defined specific areas.OAUs can be further ranked based on these scores and indices.The followings are additional measures of agrobiodiversity.

A2. Household Agrobiodiversity Index (HHABDI)
• Agrobiodiversity index at HH, community, district, province, ward levels using the number of species or landraces divided by the total number of species or landraces in a country • Analog site index of a particular landrace or species, calculated from climate analog tool based on reference site of a particular landrace or species • Driver index can be estimated for each of different drivers (factors) in a particular area over the particular time frame, using the formula, lost landraces divided by the total number of landraces available before the effect of this driver.

Data Types and Collections
Different types of data are generated and collected for the measurement and other studies of agrobiodiversity.Different data types for agrobiodiversity study are given in figure 4. Data could be agro-morphological, molecular, and perception, which can be generally collected from on-station research, on-farm trial, surveys, and lab research.Several methods and techniques can be used to collect data and information (see Joshi et al., 2005 for detail).
Apps and software are available for collecting data and information electronically both online as well offline.FieldLab is an application for Android tablets that are used for data collection in the field.It is developed by IRRI3 and is available freely.Field Book is a simple app for taking phenotypic notes.It is an open-source application for field data collection on Android4 and is available from Google Play5 .The Fieldbook2020 software developed by CIMMYT6 provides offline capabilities for managing pedigrees, phenotypic data, seed stocks, and field books for a breeding program.It provides integrated management of global information on genetic resources, crop improvement, and evaluation for individual crops.R Package7 included in this software is useful for statistical analyses.Biologer8 is a simple and free software designed for collecting data on biological diversity.online tools are very useful to minimize errors and speed up data processing.Some electronic media-based survey tools are given below.▪ Surveymonkey9 : A cloud-based survey tool that helps users create, share, collect and analyze surveys.▪ Google forms10 : It is used to create online forms and surveys.▪ SoGoSurvey11 : A cloud-based platform that enables creation, distribution, and multilingual analysis of surveys, forms, polls, quizzes, and assessments.▪ mWater Portal12 : Free platform for data collection, data visualizations, and datadriven management of infrastructure in emerging economies.▪ ODK13 : It is an Open Data Kit, open-source software for collecting, managing, and using data in resource-constrained environments.

Measurement Objects
The information for measuring agrobiodiversity comes from different levels.These levels are alleles, genes, genotypes, cultivars (varieties and landraces), crops, species, components and groups, agroecosystems or agroecozones, parcels or plots, households (farmers), villages, communities, ethnicities, wards, municipalities, landscapes, regions, districts, provinces/ states, countries, and continents.These levels are measurement objects, called OAU (operational agricultural unit).
In addition, there are several crop groups that are OAU based on different criteria e.g., use-value base, economic importance base, national list base, habitat base, red list base, growing season base, national priority base, etc. Examples are cereals, vegetable fruits, released variety, registered variety, major, minor, primary, secondary, staple, commodity, high value, commercial, industrial, food crops, feed crops, manuring crops, pesticidal plants, cash crops, cover crops, trap crops, catch crop, cultivated, semidomesticated, wild edible, field crops, garden crops, aquatic plants, common, rare, endangered, extinct, localized, vulnerable, winter crops, summer crops, and off-season (Joshi and Shrestha, 2019).
Object or OAU refers to the things being analyzed, interpreted, evaluated, or described.Variable or character refers to the properties used to describe the objects under study.Variables may be both qualitative and quantitative, and include agromorphological, genotypic, and perception data.These are measured or observed from an individual, representative samples, or population.In some cases, agromorphological markers, traits, and molecular markers can be treated as OAU.

Software for Agrobiodiversity Statistics
Many software are available for agrobiodiversity statistics.The general and molecular software are given below.

I. General Statistical Software
▪ AGROBASE14 : For data management, experiment management, and statistical analysis.
▪ CropStat15 : For data management and basic statistical analysis of experimental data.▪ DIPVEIW: For genebank data management and analysis.▪ DIVA-GIS16 : For mapping and geographic data analysis (a geographic information system (GIS).▪ Genstat 17 : For data analysis, particularly in the field of agriculture.▪ GGEbiplot 18 : For biplot analysis, conventional statistical analysis, and decision making based on univariate and multivariate data.▪ GDA35 : For the analysis of discrete genetic data.▪ GenAlEx36 : Excel Add-In for the analysis of genetic data, particularly useful for dominant data such as RAPD and AFLP data.▪ MEGA 37 : For reconstructing phylogenies using distance matrices and maximum parsimony methods, and includes neighbor-joining, branch-and-bound parsimony methods and bootstrapping.▪ PHYLIP 38 : Extensive package of programs for inferring phylogenies.▪ POPGENE 39 : For the analysis of genetic variation among and within populations using co-dominant and dominant markers, and quantitative data.▪ PowerMarker 40 : A comprehensive set of statistical methods for genetic marker data analysis, designed especially for SSR/SNP data analysis.▪ STRUCTURE 41 : Uses a clustering method to identify population structure and assigns individuals to those populations.

R Packages for Agrobiodiversity Measurement and Study
Most of the software and R packages used in biodiversity analysis can be used for agrobiodiversity analysis.Past is simple and free software that can be used for agrobiodiversity data.It is good for generating a graph, doing multivariate analysis, estimating different diversity indices, and analyzing time-series data.Some of the R packages useful for analysis of agrobiodiversity data are: ▪ adiv 42 : Analysis of Diversity, with functions, data sets, and examples for the calculation of various indices of biodiversity including species, functional and phylogenetic diversity.▪ agricolae 43 : Statistical Procedures for Agricultural Research, offers extensive functionality on experimental design especially for agricultural and plant breeding experiments and other statistical analysis.▪ analogues 44 : To calculate the climatic similarity between a reference site and a prescribed area, helps identifying locations with similar climates.▪ BAT 45 : Biodiversity assessment tools, assess alpha and beta diversity in all their dimensions (taxonomic, phylogenetic and functional).▪ BiodiversityR 46 : For statistical analysis of biodiversity and ecological communities.▪ BioFTF 47 : To study biodiversity with the functional data analysis.▪ BIO-R 48 : Biodiversity analysis using molecular data.▪ GGEBiplotGUI 49 : A graphical user interface for the construction of, interaction with, and manipulation of GGE biplots.
S # to display a richness library(vegan) #loading vegan package H=diversity(hhdata)#estimate Shannon diversity index help(diversity)# look for description of function diversity simp=diversity (hhdata, index="simpson") #estimate simpson index J = diversity (hhdata, index ="simpson")/log(S) #estimate Pielou's evenness (J) diversity(hhdata[-1], index="shannon")#exclude first column in case of data file with first column as row name barplot(simp) #plot simpson index pairs(cbind(H, simp), pch="+", col="blue") #plot all ## Species richness (S) and Pielou's evenness (J): S <-specnumber(hhdata) #estimate richness cor(H,simp) #correlation coefficient between the Shannon and Simpson indices A useful picture of diversity across several units is the function anosim() in the package, vegan.This analysis ranks all the dissimilarities among accessions and produces a boxplot of the ranks of dissimilarities within a given unit e.g., household.As an example, iris data set within this package is given below.data(iris) #loading data in R memory distiris<-dist(iris[,1:4]) #distance matrix computed by using the specified distance measure to compute the distances between the rows of a data matrix anoiris<-anosim(distiris,iris$Species) #analysis of similarities (anosim) provides a way to test statistically whether there is a significant difference between two or more groups of sampling units.plot(anoiris) #produces a boxplot of the ranks of dissimilarities within a given unit.Another useful R package is BiodiversityR, which is a graphical user interface for statistical analysis of biodiversity and ecological communities, including species accumulation curves, diversity indices, Renyi profiles, GLMs for analysis of species abundance and presence-absence, distance matrices, Mantel tests, and cluster, constrained and unconstrained ordination analysis.It is menu-driven built within Rcmdr package.BiodiversityR analyzes two datasets simultaneously as does the vegan community ecology package.These data sets are the community datasets (rows correspond to sample units and columns correspond to species) and the environmental datasets.
It is suggested to install the package in R following the guidelines55 as described in the installation guide.The manual56 can also be accessed.
Followings are the commands and steps for analysis in BiodiversityR.An analysis can be carried out either through menu driven or using commands: library (BiodiversityR) #load BiodiversityR package library (Rcmdr) #load Rcmdr package BiodiversityRGUI() #open graphical interface help("BiodiversityRGUI", help_type="html") #to see details.These are the steps for doing analyses with the menu options of BiodiversitR.To select the species and environmental matrices, follow these menu-driven steps: BiodiversityR > Environmental Matrix > Select environmental matrix Select the dune.envdataset as an example Biodiversity > Community Matrix > Select community matrix Select the dune dataset as an example.

Interpretation
Richness (S) is a number of species, landraces, particular traits in household, community, sites, or landrace.It quantifies types of the dataset.Shannon index (Shannon diversity index or Shannon Weaver index, H') includes both species number and evenness, where a greater number of species increase diversity, as does a more equitable distribution of individuals among species.High H' is representative of a diverse and equally distributed community.H' is strongly influenced by species richness and by rare species.Simpson index (D) is a measure of diversity, which takes into account both richness and evenness.The value of D ranges from 0 to 1, the greater the value the greater the diversity.The Simpson index gives more weight to evenness and common species.Evenness (Pielou's evenness, E) is a measure of the relative abundance of the different species making up the richness of an area.A community dominated by one or two species is considered to be less diverse than one in which several different species have a similar abundance.Its value ranges from 0 to 1 and 1 is complete equitability.

Conclusion
Native agrobiodiversity is generally neglected for conservation, quantification, evaluation, and monitoring.Different statistical tools can be used under agrobiodiversity statistics.Many software and R package are now available for agrobiodiversity study including measurement.Six types and levels of agrobiodiversity need to quantify and study for better management of agrobiodiversity.An operational agricultural unit is like a factor in which variables are generated and analyzed.Multivariate analysis and diversity indices are the major statistical components used in agrobiodiversity measurement.Estimates help generate the agrobiodiversity indicators that ultimately drive the program plans and activities.Many different types of scores and indices can be measured for household, community, any other administrative unit, and other OAUs.Among the many software and R packages, vegan and BiodiversityR are very useful packages for estimating diversity indices and multivariate analysis along with many statistical features.Such estimates should be measured over a certain geo-region and period to monitor the status, plan the program, and rank the geo-regions.

Acknowledgments
The Grassroots Institute organized a Summer Field School on Mountain Ecosystem and Resource Management in September 2021.Based on the presentation in this Summer School, this review article was written.A special thank goes to Dr Hasrat Arjjumend for his initiation and Dr. Lila Khatiwada for valuable suggestions and grammar correction.

Figure 2 :
Figure 2: Different statistical tools for agrobiodiversity study.

Figure 3 :
Figure 3: Statistical testing tools (parametric and non-parametric) for comparing agrobiodiversity based on data types

Figure 4 :
Figure 4: Data types for measuring on-farm agrobiodiversity at ecosystem, species, and cultivar levels

6.3 B.1. Village Agrobiodiversity Score (VABDS)
A. Based on species within agrobiodiversity group ▪ HH agrobiodiversity group richness, n 1. HH Shannon diversity index (based on number of species within a group) 2. HH Simpson index (based on number of species within a group) 3. HH species evenness (specie within a group) B. Based on landraces within the agrobiodiversity group ▪ HH agrobiodiversity group richness, n 4. HH Shannon diversity index (based on number of landraces within a group) 5. HH Simpson index (based on number of landraces within a group) 6. HH landraces evenness (specie within a group) C. Based on landraces within species ▪ HH agrobiodiversity species richness, n 7. HH Shannon diversity index (based on number of landraces within a species) 8. HH Simpson index (based on number of landraces within a species) 9. HH species evenness (specie within a group) HH agrobiodiversity index (HHABDI): sum of above 1 to 9 index values.In the similar way of household scores and indices, one can estimate village or community agrobiodiversity scores and indices as follows.
summer fruits, spices, beverages and narcotics, fibers, forage trees, forage grasses, ornamental plants, medicinal plants, supportive plants, economical and beneficial (EB) insects, EB microorganisms, fish and aquatic animals, aquatic plants, poultry, and livestock) at 0 or 1 scale over a year with maximum 25 score Bal Krishna Joshi ▪ Instat+ 19 : A general statistical package.▪ Minitab 20 : Simple and general statistical package.▪ MS Excel 21 : Spreadsheet software program, a powerful data visualization, and analysis tool.▪ MSTAT-C 22 : For the design, management, and analysis of agronomic research experiments.▪ NTSYSpc 23 : Commonly used package for numerical taxonomy and multivariate analysis system.▪ Past 24 : For scientific data analysis, with functions for data manipulation, plotting, univariate, multivariate statistics, ecological analysis, time series, and spatial analysis.▪ R 25 and RStudio 26 : For statistical computing and graphics.▪ SAS 27 : For data management, advanced analytics, and multivariate analysis.▪ SPSS 28 : A software platform that offers advanced statistical analysis, a vast library of machine learning algorithms, and text analysis.▪ STAR 29 : Statistical tool for agricultural research.▪ Statistica 30 : A data analysis and visualization program.▪ Statistix 31 : Statistical analysis program.▪ PDA 32 : For biodiversity analysis and conservation prioritization problems.▪ BioDiversity Pro 33 : A free statistical package program enabling many measures of diversity to be calculated for a dataset of taxa by samples.