Moreno Et Al. SUPPLEMENTARY MATERIAL: Extrapolation and Rarefaction in Estimates 9

Moreno et al. SUPPLEMENTARY MATERIAL: Extrapolation and Rarefaction in EstimateS 9.

Diversity is scale-dependent: the larger the sampled area the higher number of species will be found. This makes good intuitive sense in that the more quadrats one samples in a community, the more likely rare species will be picked up. In practice, monitoring species richness for large areas is quite difficult and imprecise as not all individuals can be observed. If species richness of a sampled area is taken to assess and compare the diversity and composition of species assemblages of the underlying habitat, values are only comparable if sampling efforts are standardized in an appropriated way. A classical method toaccount for the inevitable confounding effects of sample size (or sampling effort) is the estimation of expected richness based on asymptoteof fitted mathematical functions to the species accumulation curves, which are built by successive random-addition of samples.

Recently rarefaction (interpolation) and extrapolation (prediction) approaches have been proposed (Chao 2005; Colwell et al 2012). EstimateS 9, and open source software (Colwell, 2013), computes these parameters from counts of individuals for each species in single samples (individual-based abundance data), or in each of a set of samples (sample-based abundance data). Whereas the extrapolation approach estimates the total species richness, including species not present in any sample, rarefaction curves estimate species richness for a sub-sample of the pooled total species richness, based on an empirical reference sample (all samples pooled).

Rarefaction is used to bring samples of different sizes to a common footing. It works resampling at random, 1, 2, ..., n individuals or i= 1, 2, ..., t sampling units until all individuals or sampling units in the reference sample have been accumulated. EstimateS re-samplesseveral times at each level (e.g. 100 times), and the means (and standard deviations) among resamples for each level of accumulation are reported. The effect of differences in sample size for two or more habitats is substantially reduced by comparing at the same level of samples accumulation (usually the lowest number of samples among the units compared). In this work we selected Coleman-rarefied index of species richness (Coleman et al 1982).

Extrapolation is based on the number of species only represented in one or a few plots (or by one or a few individuals). This allows us to estimate the species richness for the whole population from which the sampling was drawn. EstimateS 9introduces extrapolation from a reference sample to the expected richness (and its standard deviation) for an augmented number of individuals or sampling units (recommended up to by 2-3 the number of samples). In this work we selected Chao2 for sample-based incidence data as a conservative estimator of species richness (Chao 2005).

As discussed by Colwell & Coddington (1994), the problem of estimating the true number of species shared by two (or more) sites based on sample data (necessarily biased for incomplete sampling of rich communities) presents a difficult but important challenge. EstimateS 9 also compute the compositional similarity among samples, either by the classical estimator of shared species (Sorensen index), but also taking into account the number of unobserved species by statistical methods (Chao-Sorensen Estimator; Chao et. al. 2005). This latter method requires species abundance data and is based on the probability that two randomly chosen individuals, one from each of two samples, both belong to species shared by both samples (but not necessarily to the same shared species).

Moreno et al. SUPPLEMENTARY MATERIAL

TABLE S1. Relation and details of general habitat categories found in 10 dehesa farms studied. For each GHC the total surface or length mapped and the number of plots mapped for the 10 farms are showed. Mean size of the plots, with the range (max. and min.) is also given, with the number of plots sampled for each type of GHC.

Area-based Habitats / Surface, Ha
(10 farms pooled) / Number of Plots (10 farms pooled) / Mean Plot Size,
Ha / Number of
Sampled Plots
Wood pastures / 3134 / 204 / 23.0 (0.1 – 106.8) / 41
Annual-plant open pastures / 228 / 26 / 10.4 (0.2 – 42.7) / 11
Perennial-plants open pastures / 246 / 54 / 8.3 (0.4 – 27.0) / 15
Mix open pastures / 838 / 174 / 12.3 (0.5 – 39.5) / 16
Shrubs / 207 / 38 / 12.6 (0.2 – 184.0) / 19
Agricultural crops / 128 / 40 / 4.5 (0.1 – 23.6) / 12
LinearFeatures / Length, Km
(10 farms pooled) / Number of Features (10 farms pooled) / Mean Feature Length,m / Number of
Sampled Plots
Woody lines / 21 / 50 / 583 (63 – 1263) / 11
Herbaceous lines / 17 / 63 / 334 (88 – 1589) / 10
Water bodies / 55 / 73 / 1114 (156 – 3212) / 10

GHC Codes: Wood pastures (evergreen and/or deciduous trees of 5-30m height); Annual-plant open pastures(open pastures dominated by annual species); Perennial-plants open pastures(open pastures dominated by perennial species); Mix open pastures(open pastures evenly formed by both annual and perennial species); Shrubs (soil mostly cover by shrub species, that include low phanerophytes (0.3-0.6m), mid phanerophytes (0.6-2m) and tall phanerophytes (2-5 m), with or without presence of the tree cover); Agricultural crops (cultivated herbaceous and woody crops and other artificial areas (e.g., artificial grass) or areas with bare soil (e.g., boulders, fallow lands)); Wood lines that include lines of trees and/or scrubs; Herbaceous Lines; Water bodies (stream waters and artificial ponds).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURE S1

Figure S1. Map of the area, with the 10 studied dehesa farms.Coordinates in UTM (ETRS 1989 – Huse 30). Size of the farms ranges from 150 ha (dark green) to 835 ha (Light blue).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURES2

Figure S2. Curves of relative accumulation of species with the number of sampling plots, including extrapolation till twice the number of sampled plots (145).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURE S3A

Figure S3a. Species accumulation curves for Plants extrapolated till n= 50 plots (for reference sample (number of measure plots) for each GHC see Table 2).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURE S3B

Figure S3a. Species accumulation curves for Bees extrapolated till n= 50 plots (for reference sample (number of measure plots) for each GHC see Table 2).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURE S3C

Figure S3c. Species accumulation curves for Spiders extrapolated till n= 50 plots (for reference sample (number of measure plots) for each GHC see Table 2).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURE S3D

Figure S3d. Species accumulation curves for EWs extrapolated till n= 50 plots (for reference sample (number of measure plots) for each GHC see Table 2).

Moreno et al. SUPPLEMENTARY MATERIAL. FIGURE S4

Figure S4. Curves of accumulation of species with different habitats types compared with the accumulation of surface occupied by those habitats: Wood pastures (WoodPast;), Annual-plant open pastures (AnnPast), Perennial-plants open pastures (PerenPast), Mix open pastures (MixPast), Shrubs, Agricultural crops (AgriCrop), Woody lines (WoodLine), Herbaceous lines (HerbLine) and Water bodies (Water).