Are different biodiversity metrics related to the same factors? A case study from Mediterranean wetlands

Biological Conservation 142 (2009) 2602–2612

Contents lists available at ScienceDirect

Biological Conservation journal homepage: www.elsevier.com/locate/biocon

Are different biodiversity metrics related to the same factors? A case study from Mediterranean wetlands S. Gascón *, D. Boix, J. Sala Institute of Aquatic Ecology, University of Girona, Faculty of Sciences, Campus de Montilivi, E-17071 Girona, Spain

a r t i c l e

i n f o

Article history: Received 14 October 2008 Received in revised form 28 April 2009 Accepted 8 June 2009 Available online 2 July 2009 Keywords: Regression tree models Habitat condition Protection policies Crustaceans Insects

a b s t r a c t Conservation biology is mainly interested in prioritizing sites on the basis of their high biodiversity. Although species richness is a commonly used criterion, it does not take other crucial aspects of identifying conservation priority sites into account, such as rarity or taxonomic distinctness. Additionally, management efforts are usually focused on the conservation of a small number of species, mainly vertebrates. However, the biodiversity patterns of these faunal groups and the main factors which determine them cannot be generalized to other faunal groups (e.g. aquatic invertebrates). Therefore, the objectives of the present study are: (1) to compare the response of 11 biodiversity metrics in order to know which ones are redundant, (2) to identify key environmental factors for biodiversity, and (3) to find out whether sites with high biodiversity values also have a good habitat condition and high protection status. The study was done at assemblage level (crustaceans and insects) in 91 wetlands in the NE Iberian Peninsula. Regression tree models were used to identify the key factors influencing biodiversity, including water, wetland and landscape characteristics as explanatory variables. Generalized Linear models were used to establish the relationship between biodiversity metrics and protection status and habitat condition. The results obtained by the two sampled seasons were compared. Conductivity was the main factor influencing biodiversity metrics. Positive significant relationships were found between some biodiversity metrics and wetland habitat condition, whereas there were none for protection status, indicating the inadequacy of conservation policies to protect wetland aquatic invertebrate biodiversity. Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction One of the main interests of conservation biology is to prioritize sites on the basis of their biodiversity values, selecting those that have the highest biodiversity (Gaston and Spicer, 2004). However, the measurement of biodiversity presents a challenge to biologists, mainly due to the effort needed to conduct extensive surveys but also because there is still no agreement on the method to be used to evaluate biodiversity. Although species richness is largely used for this purpose (e.g. Gotelli and Colwell, 2001; Boix et al., 2008; Céréghino et al., 2008b), it does not take into account other crucial aspects for identifying conservation priority sites such as, for example, rarity or taxonomic distinctness (Vanewright et al., 1991; Gaston, 1994). Therefore, other biodiversity measures, which account for these deficiencies, have been increasingly used (e.g. Clarke and Warwick, 1998; Abellán et al., 2005b; Heino et al., 2005). In this sense, some of the metrics that have appeared as promising tools to measure complementary aspects of biodiversity and

* Corresponding author. Tel.: +34 972 418 466; fax: +34 972 418 150. E-mail addresses: [email protected] (S. Gascón), [email protected] (D. Boix), [email protected] (J. Sala). 0006-3207/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2009.06.008

their response to environmental variability are taxonomic distinctness measures. These metrics are based on measuring the taxonomic relatedness of species in a community rather than on measuring the number of species, and overcome most of the problems of the traditional measures of diversity (Clarke and Warwick, 1998). The rationale of these metrics is that samples with species belonging to a single genus or family (and so on up the taxonomic hierarchy) are considered as less diverse than those with species in different genera or families. Although recent studies have been developed in freshwater systems (e.g. Heino et al., 2007; Marchant, 2007; Campbell et al., 2008), they were first applied in marine studies mainly assessing anthropogenic disturbance effects (e.g. Rogers et al., 1999; Warwick et al., 2002; Walters and Coen, 2006). The importance of assessing rarity when biodiversity is measured has also been suggested by other authors (e.g. Collinson et al., 1995; Abellán et al., 2005b; Davies et al., 2008a). Rarity values can be assessed using several criteria (Gaston, 1994), but one of the more accepted methodologies consists on establishing different levels of rarity (e.g. rarity of occupancy, rarity of individuals within an area, and habitat specificity). Examples of the use of different levels of rarity are found in studies dealing on organisms such as flora, water beetles or zooplankton (Rabinowitz et al., 1986; Abellán et al., 2005a; Hessen and Walseng, 2008). Other

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

proposals based on the originality of the assemblage structure within a dataset could also be used to measure rarity (Puchalski, 1987; Rodrigo et al., 2003). The use of biodiversity metrics based on the assessment of different aspects of communities is important since it can provide more complete information. Following the idea suggested by Heino et al. (2005), if two metrics are correlated and show a similar relationship with the same environmental variable, it would be interpreted as they are responding similarly to this change in the environment. Consequently, to optimize the available resources (i.e. economic, material, time, etc.), monitoring could be performed focussing the efforts in only one of these metrics. By contrast, if they respond to different environmental factors, they may describe biodiversity differently, being valuable independent measures. Thus, the identification of biodiversity metrics to supply this kind of complementary information may be of great interest for conservation assessment and management. During the last century many wetlands have been lost, and those that remain face increasing pressure due to agriculture, land drainage, pollution, trampling and urban development. A common approach to prevent the loss of biodiversity is the establishment of reserve areas (areas under a range of in situ protection measures; Pressey et al., 1993). However, conservation biology is usually directed towards large-scale ecosystems leading to a general neglect of small-scale landscape elements (De Meester et al., 2005). Additionally, most of the management efforts are focused on the conservation of a small number of species, mainly vertebrates (e.g. Ramsar, 2006). This process is likely to be sub-optimal since the biodiversity patterns of a faunal group, and the main factors which determine them, cannot be generalized to other faunal groups (e.g.

2603

Bonn and Gaston, 2005). In fact, the existing methods for conservation planning could be inadequate because they are mainly based on assessing biodiversity from well-known taxa assuming that these would be a good surrogate for other taxa (e.g. Rodrigues and Brooks, 2007). Recent studies have highlighted the importance of wetlands as biodiversity ‘‘hot spots‘‘ (Gopal and Junk, 2000; Scheffer et al., 2006; Céréghino et al., 2008b and references therein), but the patterns driving biodiversity in these habitats are still under study. For example, wetland invertebrate biodiversity (measured as species richness) has been related to conductivity, nutrient availability, water permanence, and some landscape factors (e.g. Boix et al., 2008). However, as far as we know, nothing is known about the behaviour of other biodiversity metrics (e.g. taxonomic distinctness) on these variables in these habitats. Therefore, as has recently been noticed (Céréghino et al., 2008a), the knowledge of aquatic invertebrate biodiversity in ponds still needs to be improved. The objectives of the present study are: (1) to compare the response of the different biodiversity metrics in order to know which ones are complementary, (2) to identify key environmental factors for different biodiversity metrics and (3) to study whether high biodiversity sites coincide with protected and less impacted sites.

2. Materials and methods 2.1. Studied sites The study was carried out in 91 representative wetlands (ponds, lagoons or marshes) located throughout Catalonia (NE Iberian

Fig. 1. Map of Catalonia (NE Iberian Peninsula) showing the 91 wetlands analysed, which are coded according to their habitat type. The shaded part of the map represents the area above 800 m a.s.l.

2604

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

Peninsula). The selected wetlands are part of the official Wetland Inventory of Catalonia. All sampled wetlands were below 800 m a.s.l. to ensure that they were influenced by Mediterranean climatic conditions and had a maximum depth of less than 6 m (Fig. 1). Two sampling surveys were conducted during 2003, the first in February (hereafter winter) and the second in June (hereafter spring), in order to encompass seasonal variability. Almost all the wetlands were sampled twice except for some temporary ponds that dried out before the second sampling survey. 2.2. Organism data and water, wetland and landscape characteristics Invertebrates were sampled using a 20 cm diameter dip-net (250 lm mesh size). At each wetland, three sets of 20 sweeps each were carried out per visit, covering all microhabitats visually detected in the littoral zone of the wetland. Samples (60 sweeps) were preserved in situ with 4% formalin. Crustaceans and insects (except dipterans) were sorted under a stereomicroscope and identified to species level. The influence of different variables related to water, wetland and landscape characteristics on aquatic biodiversity has been proved (Oertli et al., 2002; Declerck et al., 2005; Jeffries, 2005). Therefore, in our study we have included a selection of variables informing about these environmental characteristics. Water temperature, conductivity, pH and dissolved oxygen were measured in situ, whereas concentration of dissolved inorganic nutrients (ammonium, nitrite, nitrate and soluble reactive phosphorus), total nutrients (nitrogen and phosphorus) and chlorophyll-a were analysed in the laboratory. Analyses of dissolved inorganic nutrients were carried out from filtered samples, and total nutrients from unfiltered samples, according to Grasshoff et al. (1983). Dissolved inorganic nitrogen (DIN) was obtained as the sum of ammonium, nitrite and nitrate. Chlorophyll-a was extracted using 80% methanol, after filtering water samples (Whatman GF/C filters), and measured according to Talling and Driver (1963). Water permanence was considered as a categorical variable (temporary vs. permanent; 38 and 53 water bodies, respectively). To obtain water body size and landscape variables, freely available aerial photographs were used (DPTOP, 2005; MAPA, 2006). Water body size was calculated as the maximum flooded area, without considering extreme flooding situations. Wetland isolation was calculated as the distance to the nearest water body, and water body density as the number of water bodies within a radius of 500 m from the studied site. Water characteristics were measured simultaneously with organism sampling, whereas wetland and landscape characteris-

tics had a unique value per wetland. See Table 1 for more detailed information of water, wetland and landscape characteristics. 2.3. Protection and habitat condition Each site studied was classified according to its protection status and habitat condition. The protection status indicates if the site is located in a protected and managed area (Managed: 31 sites), in a protected area without management (Protected: 17 sites) or in an unprotected area (Unprotected: 43 sites). Managed protected areas are natural parks, which develop strategies for the conservation of the protected area. In contrast, protected areas without management do not have any specific institution to carry out conservation strategies. To assess the habitat condition, a rapid assessment method, developed for Mediterranean shallow lentic ecosystems, was used (ECELS; Sala et al., 2004). This index is based on the evaluation of five independent components that take several features into account to obtain the habitat condition value of each site (basin littoral morphology, human activity, water appearance, emergent vegetation and hydrophytic vegetation). The values obtained with this index range from 0 (worst habitat condition) to 100 (best habitat condition). 2.4. Biodiversity metrics Eleven biodiversity metrics related to assemblage structure, rarity and taxonomic distinctness were calculated for each sample. Three metrics were related to assemblage structure: (i) number of species per sample (S), (ii) species diversity obtained using the Shannon–Wiener diversity (H0 ), and (iii) Pielou’s evenness based on Shannon’s index (J0 ). Shannon–Wiener diversity and Pielou’s evenness were obtained using a base-two logarithm (Pielou, 1969). Four metrics were measured to asses sample rarity: the index of faunal originality (IFO), rarity of individuals, rarity of occupancy, and habitat specificity. Although IFO was firstly applied on flora data, it has been also used to asses fauna originality in zooplankton studies (e.g. Rodrigo et al., 2003). IFO was calculated according to Puchalski (1987):

P IFO ¼

1=Mi S

where M is the total number of samples in which species i occurs (from i = 1 to S), and S is the number of species in the corresponding sample. The rarity of individuals, rarity of occupancy, and habitat

Table 1 Environmental characteristics of the wetlands under study. For each variable the median and the range (minimum and maximum) are shown. Median

Minimum

Maximum

Water variables Chlorophyll-a (lg L1) pH Temperature (°C) Conductivity (mS cm1) Dissolved oxygen (%) Dissolved inorganic nitrogen (lg N L1) Soluble reactive phosphorus (lg P L1) Total nitrogen (lg N L1) Total phosphorus (lg P L1)

8.62 7.92 14.40 1.19 80.00 242.76 9.30 1587.91 135.16

0.21 6.11 2.34 0.07 4.70 1.10 0.62 146.30 3.21

126.36 10.02 34.70 79.20 300.00 30,207.80 2494.57 34,365.80 26,981.47

Pond variable Size (m2)

7300

26

4,600,000

Landscape variables Water body density (WBD) (n° of water bodies in 500 m radius) Wetland isolation (WI) (distance to the nearest pond, m)

4 80

1 5

32 3200

Habitat condition ECELS values

65

28

100

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

2605

specificity were calculated similarly to the proposal made by Hessen and Walseng (2008) based on the scheme described by Rabinowitz et al. (1986), which considers these three different types of rarity. In our study, the rarity of individuals was established using the maximum observed abundance for each species. However, microcrustaceans (cladocerans, copepods and ostracods) and macroinvertebrates (large branchiopods, malacostracans and insects) were considered separately due to the different total abundances of these two faunal groups (Appendices A and B). A species with a maximum abundance in a sample lower than 10% of the total abundance of their faunal group (either a microcrustacean or a macroinvertebrate) was coded as rare (i.e. rarity of individuals = 1). To obtain the average species rarity of individuals per sample (RI), we summed the rarity of individuals value obtained by each species present in that sample and divided by the total number of species of the corresponding sample. The rarity of occupancy within the studied area was assessed using the proportion of the sites where the species was present, and so species with an occurrence lower than 5% were considered rare (i.e. rarity of occurrence = 1). To obtain the average species rarity of occupancy per sample (RO), we summed the rarity of occurrence value obtained by each species present in that sample and divided by the total number of species of the corresponding sample. Finally, the habitat specificity was determined by the presence of the species in different habitat types. Three habitat types were distinguished according to Boix et al. (2005): freshwater temporary (fwt: 29 water bodies), freshwater permanent (fwp: 35 water bodies) and saline water bodies (sw: 27 water bodies). The habitat specificity scores ranged from 0 to 1: 0 for species found in the three habitat types, 0.5 for species found in two habitat types, and 1 for species only found in one habitat type. To obtain the average species habitat specificity per sample (RH), we summed the habitat specificity scores of each species present in that sample and divided by the total number of species of the corresponding sample. A complete list of species studied with their occurrences, abundances and habitat specificity can be consulted in Appendices A and B. Following Clarke and Warwick (1998), sample taxonomic distinctness, which measures the taxonomic relatedness of the organisms in a sample, was analysed by means of four metrics: (i) taxonomic diversity (D), (ii) taxonomic distinctness (D*), (iii) average taxonomic distinctness (D+), and (iv) variation in taxonomic distinctness (K+). The first two are metrics based on abundance data, and the last two are based on presence–absence data. Taxonomic diversity (D) is a natural extension of the Simpson diversity index since it is the expected path length of the classification tree between any two individuals chosen at random. Taxonomic distinctness (D*) was obtained after dividing D by the Simpson diversity index to remove the dominating effects of species abundance. Average taxonomic distinctness (D+) is an intuitive definition of biodiversity since it is the average taxonomic breadth of the sample. Variation in taxonomic distinctness (K+) measures variance in pairwise path lengths between each pair of species, reflecting the unevenness of the taxonomic tree (Clarke and Warwick, 2001). PRIMER version 6 was used to obtain all taxonomic distinctness metrics, setting at 100 the longest path length in taxonomy. The path lengths between different taxonomic levels of the classification tree (based on standard Linnaean hierarchical classification) were considered equal. There were eight taxonomic levels in the aggregation file for the analyses: species, genus, subfamily, family, order, class, superclass and phylum. The taxonomic groups included in this study were: Branchiopoda, Copepoda, Ostracoda, Malacostraca, Ephemeroptera, Odonata, Heteroptera, Coleoptera and Trichoptera (Appendices A and B).

tree models were used to asses the relationship between biodiversity metrics (response variables) and water, wetland and landscape characteristics (explanatory variables). Samples with more than one species were included in the analysis (161 out of 163 samples). This type of regression displays a binary tree, built by a process known as binary recursive partitioning, which gives a very clear picture of the structure of the data, and provides a highly intuitive insight into the kinds of interactions between variables (Crawley, 2002). However, the models obtained can be too elaborate and over-fit the training data, and so the best tree size needs to be established. In order to solve this problem, the regression tree models were developed using the ‘‘party” package for R, which automatically selects the best tree size (Hothorn et al., 2006). The dataset was split by season and all regression trees were calculated for winter and spring separately. Eleven models were run, one for each biodiversity metric. The explanatory variables used were: dissolved inorganic nitrogen (DIN), soluble reactive phosphorus (SRP), total nitrogen (TN), total phosphorus (TP), chlorophyll-a (Chla), conductivity (Cond), water permanence (temporary vs. permanent), water body size (WBS), wetland isolation (WI) and water body density (WBD). Generalised linear models (GLM) were used to study whether biodiversity values were related to protection status and to habitat condition. These kinds of models provide a flexible and powerful tool to analyse data with non normal distributed errors (e.g. count data). The biodiversity metrics were used, one in each model, as response variables. Twenty-two models were run (11 models using protection status, and the other 11 models using habitat condition as explanatory variables). Season (winter vs. spring) was also included as explanatory variable to study the consistence of the relationships. A significant interaction implied that the relationship between the explanatory variable and the biodiversity metric is not the same for both seasons. Following the principle of parsimony, a backward model simplification procedure was developed to ascertain that the models contained only significant variables. Significance of explanatory variables and their interaction was tested using the deletion test procedure detailed in Crawley (2007). Only terms that caused a significant increase in the deviance were retained in the final model. Non-significant explanatory variables which were included in significant interaction terms were not deleted (Crawley, 2002). Moreover, the significance of the final model respect to the null model (model without any explanatory variable) was also tested, and only significant models (p < 0.05) were further interpreted. All models were run using Poisson and Gaussian distributions, and we checked model error distributions to choose the one that better fitted the assumptions. When overdispersion was detected, quasipoisson error distribution instead of Poisson was used. Similarly, F-test instead of Chi-square test was performed in order to obtain the significance of the deviance explained by each explanatory term (Crawley, 2002). To assess the model goodness of fit, we employed the squared Pearson correlation coefficient between observed and predicted values. To perform GLM analyses we used the ‘‘glm” function written in Rlanguage. All statistical analyses were run using R 2.7.0 (R-DevelopmentCore-Team, 2007).

2.5. Statistical analyses

Spearman correlations showed that several biodiversity metrics were related (Table 2). Although half of the relationships tested were significant (27 out of 55 possible), only seven had a correlation coefficient higher than 0.5. Thus, highly related biodiversity

Spearman correlation was used to test for the congruence between the 11 biodiversity metrics. On the other hand, regression

3. Results 3.1. Relationships between biodiversity metrics

2606

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

Table 2 Spearman correlation matrix between the studied biodiversity metrics (N = 161). In bold: significant correlations. Significance level is indicated as follows.

Species richness (S) Shannon diversity (H0 ) Evenness (J0 ) Index of faunal originality (IFO) Rarity of individuals (RI) Rarity of occupancy (RO) Rarity of habitat specificity (RH) Taxonomic diversity (D) Taxonomic distinctness (D*) Average taxonomic distinctness (D+) Variation in taxonomic distinctness (K+) * **

S

H0

J0

IFO

RI

RO

RH

D

D*

D+

0.461** 0.062 0.278** 0.512** 0.272** 0.012 0.330** 0.035 0.385** 0.175*

0.801** 0.105 0.201* 0.039 0.094 0.827** 0.199* 0.038 0.202*

0.052 0.092 0.137 0.125 0.685** 0.219* 0.273** 0.128

0.477** 0.721** 0.593** 0.010 0.038 0.175* 0.045

0.518** 0.222** 0.144 0.034 0.081 0.046

0.279** 0.075 0.055 0.193* 0.013

0.085 0.023 0.000 0.309**

0.019 0.077 0.054

0.266* 0.242*

0.268**

p < 0.05. p < 0.001.

metrics (rs > 0.5) were: (1) species richness with RI, (2) Shannon diversity with evenness, (3) Shannon diversity with taxonomic diversity, (4) evenness with taxonomic diversity, (5) IFO with RO, (6) IFO with RH, and (7) RI with RO. 3.2. Identifying key environmental factors Analysing the key factors determining wetland aquatic invertebrate biodiversity, our results showed that five of the eleven biodiversity metrics used in this study were significantly related to some of the explanatory variables. Two of them (species richness and RH) showed the same response in both studied seasons (Fig. 2), whereas the rest (all of them metrics informing about the taxonomic relatedness of samples) only had a significant response against environmental variables in one season (Fig. 3). Thus, several key factors were identified for biodiversity metrics related to assemblage structure, rarity and taxonomic distinctness. Water conductivity was the main factor for the two metrics with the same response in both seasons (Fig. 2). Thus, and despite environmental differences between seasons, always two groups

with significantly different richness values and RH specificity were distinguished. Significantly higher values of species richness were found with low conductivity values (less than 15.4 mS cm1 in winter and 11.4 mS cm1 in spring). In contrast, RH showed their higher values at high conductivity values (15.1 mS cm1 in winter, and 27.3 mS cm1 in spring). The rest of biodiversity metrics were only significantly related to environmental variables in one of the seasons (Fig. 3). In winter, conductivity also appeared to be significantly important for variation in taxonomic distinctness. This metric had higher scores in lower conductivity levels (lower than 3.07 mS cm1) and lower values in higher conductivities (higher than 15.1 mS cm1). In contrast, during spring, only one environmental variable (soluble reactive phosphorus) was significantly related to the biodiversity metrics (taxonomic distinctness and average taxonomic distinctness). Although the thresholds were different, higher values of the nutrient implied significant lower values of biodiversity metrics in both cases (2.79 and 273.7 lg P L1 for taxonomic distinctness and average taxonomic distinctness, respectively).

SPRING

WINTER 1

1

COND

COND

p = 0.023

p = 0.044

≤15.41

Species Richness

≤11.44 30

>11.44 (n = 14) 30

25

25

20

20

15

15

10

10

5

5

>15.41 (n = 9)

(n = 80) 15

15

10

10

5

5

(n = 58)

1

1

COND

COND

p < 0.001

p < 0.001

≤15.1

>15.1 (n = 10)

RH

(n = 79)

≤27.3

>27.3 (n = 10)

(n = 62)

1

1

1

1

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0

0

0

0

Fig. 2. Significant regression tree results for species richness, and average species habitat specificity (RH).

2607

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

SPRING 1

SRP

COND p < 0.001

≤3.07

>3.07

3

COND

Variation in taxonomic distinctness

p = 0.034

≤15.1 (n = 17)

(n = 62)

>15.1 (n = 10)

600

600

600

400

400

400

200

200

200

0

0

0

≤2.79 (n = 27)

90

90

80

80

70

70

60

60

50

50

40

40

30

30

>2.79 (n = 45)

1

SRP p < 0.001

≤273.7

Average taxonomic distinctness

WINTER 1

Taxonomic distinctness

p = 0.03

>273.7 (n = 9)

(n = 63) 80

80

70

70

60

60

50

50

40

40

30

30

Fig. 3. Significant regression tree results for variation in taxonomic distinctness, taxonomic distinctness and average taxonomic distinctness.

4. Discussion 4.1. Are the different biodiversity metrics complementary? Our results point out that many of the biodiversity metrics used in this study were significantly related. However, even when a high correlation was observed (for example taxonomic diversity with

2.0 1.0 0.0

5

3.0

r2 = 0.070 Shannon Diversity

10 15 20 25 30

r2 = 0.073

S

S

100

150

r2 = 0.068

50

0.0 0.2 0.4 0.6 0.8 1.0

Taxonomic diversity

r2 = 0.035

W

W

S

0

W

Evenness

None of the biodiversity metrics showed a significant response among factor levels of protection status. Therefore, the sites with the highest protection status (Managed sites) did not have the highest biodiversity values for wetland aquatic invertebrates. The three metrics related to assemblage structure (species richness, Shannon diversity and evenness) were not significantly related to protection status, although they showed the same response to season, reporting higher values in spring. Additionally, taxonomic diversity showed also a significant result, but was due to the interaction between season and protection status. Protection status did not change between seasons, and so the significant interaction was due to seasonal changes in this metric that were more evident in protected and managed sites, but not in sites without management (Fig. 4). On the other hand, species richness, taxonomic distinctness and average taxonomic distinctness showed a significant relationship with habitat condition, that was the same in both seasons (Fig. 5). The remaining biodiversity metrics did not have any significant relationship with the habitat condition or showed seasonal differences, because of the significant interaction between season and habitat condition (Shannon diversity and evenness). Again, these significant interactions arose due to changes on biodiversity metric but not on habitat condition, since habitat condition was the same in both seasons. An important point to note is that the relationships found to be significant and consistent in both seasons were always positive, indicating that in well preserved sites the significant biodiversity metrics were expected to be higher.

Species Richness

3.3. Relationships between biodiversity metrics and protection status and habitat condition

M

P W

U

M

P

U

S

Fig. 4. Notched box-plots showing significant relationships found between biodiversity metrics (Y axis), season and protection status (X axis). Legend in X axis indicates factor levels as follows: M: managed area; P: protected area; U: unprotected area; S: spring; W: winter. Above each box-plot squared Pearson correlation between observed and predicted values (r2). Only significant models are shown.

Shannon diversity; rs > 0.8) the information supplied by the metrics could not be considered redundant. The results reported by the regression tree analyses helped to understand biodiversity metric behaviours and complementarities. For example, taxonomic distinctness metrics revealed interesting responses to environmental variability. Thus, whereas taxonomic distinctness and average taxonomic distinctness were sensitive to nutrient concentration changes (e.g. decrease of the metrics values with increases of water reactive phosphorus concentration), variation in taxonomic distinctness responded to variables more linked

2608

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

r2 = 0.104

r2 = 0.066

= 0.039

30

50

70

90

1.0 0.8 0.6 0.4

Evenness

0.2 0.0

Shannon Diversity 90

30 40 50 60 70 80

Taxonomic distinctness

70

30

50

r2

70

90

30

50

70

90

= 0.030

50 55 60 65 70 75 80

50

r2

Average taxonomic distinctness

30

0.0 0.5 1.0 1.5 2.0 2.5 3.0

25 20 15 10 5

Species Richness

30

r2 = 0.115

30

50

70

90

Fig. 5. Scatter plots showing significant relationships found between biodiversity metrics (Y axis) and habitat condition (X axis). Open symbols and dashed lines indicate spring values, and grey symbols and solid lines winter values. Above each scatter plot squared Pearson correlation between observed and predicted values (r2).

to other water characteristics not directly related to nutrient enrichment, such as conductivity. Similar results linking variation in taxonomic distinctness to conductivity differences, and average taxonomic distinctness to nutrient changes have been obtained in other studies (e.g. Mouillot et al., 2005). Although biodiversity metrics could be significantly correlated, they had different responses to environmental factors and therefore contributed with additional information. In this sense, and in agreement with our findings, Heino et al. (2005) already noticed that taxonomic distinctness metrics provided complementary information to the one supplied by the most commonly used metric (i.e. species richness). Thus, in biodiversity monitoring studies it could be interesting to use several metrics that describe independent aspects of biodiversity (e.g. rarity vs. species richness), as well as metrics that are related to different environmental changes (e.g. different responses of taxonomic distinctness metrics). This idea has already been suggested by other authors, working on stream macroinvertebrate assemblages (Heino et al., 2008). Our results showed that at least four biodiversity metrics supplied complementary information (species richness, habitat specificity, average taxonomic distinctness and variation in taxonomic distinctness), and so may help to understand better the environmental effects on assemblages biodiversity. 4.2. Key environmental factors for wetland aquatic invertebrate biodiversity Several environmental factors related to water, wetland and landscape characteristics revealed themselves to be important for aquatic invertebrate biodiversity. Traditionally, water permanence (e.g. Tarr et al., 2005), water body size (e.g. Oertli et al., 2002) and salinity (e.g. Lyons et al., 2007) have been identified as important factors influencing biodiversity in wetlands. In our case study, neither water permanence nor water body size appeared to be significant factors determining differences in aquatic invertebrate biodiversity. The lack of significance of water permanence may indicate that biodiversity values are similar in both temporary and permanent water bodies. Some authors, who found that biodiversity (using species richness as a surrogate) of temporary ponds was comparable to that found in permanent ones, have called for a

revision of the idea of a low biodiversity in temporary environments (Biggs et al., 1994; Boix et al., 2001; Williams, 2006). In addition, recent works have pointed out that small water bodies make a significant contribution to regional biodiversity and call into question the assumption that aquatic biodiversity is concentrated in larger systems (Davies et al., 2008b). The lack of significance of water body size for any of the biodiversity metrics tested also supports this idea. In contrast, salinity (measured as water conductivity) was the most important factor, since it affected biodiversity at different levels: (i) assemblage structure parameters (species richness), and (ii) rarity (RH). Supporting our results, previous studies have also documented a negative effect of conductivity on aquatic invertebrate species richness (e.g. Piscart et al., 2005), and have classified some high conductivity habitats (saline streams and coastal salt pans) as rarity hotspots for water beetles (Abellán et al., 2007). Thus, the use of rarity values in the assessment of biodiversity in habitats with high conductivity turns out to be important; otherwise, these kinds of systems would be undervalued due to their negative relationship with one of the most used surrogates of biodiversity (i.e. species richness). The effect of water quality on biodiversity metrics was only detected in spring. During late spring (June) the scarcity of water inputs characteristic of the Mediterranean climate (Mariotti et al., 2002), coupled to evaporation processes due to temperature increases, led to a decrease of water volume in wetlands. Thus, phosphorus could increase by a simple concentration process, or because it was released from sediments due to the combined effect of low oxygen concentrations and high temperatures (Hansson et al., 2005). Nevertheless, sites with higher phosphorus concentrations had significant lower invertebrate biodiversity, although only noticeable using taxonomic relatedness metrics (average taxonomic and taxonomic distinctness). Other studies agree with our findings, since they showed similar significant responses of taxonomic distinctness metrics (assessed using average taxonomic distinctness) to water quality (e.g. Heino et al., 2007; Marchant, 2007). 4.3. Implications of seasonal variation Not all biodiversity metrics showed seasonal variations. Average taxonomic distinctness (which is less affected by species richness changes; Clarke and Warwick, 2001) and RH had similar values in both studied seasons (see Figs. 2 and 5). In contrast, biodiversity metrics extracted from assemblage characteristics (as species richness, Shannon diversity and evenness) are highly sensitive to season and always reported higher values in spring (see Fig. 4). Taxonomic distinctness metrics based on abundance data also appeared to be sensitive to seasonal differences, and again showed higher values in spring. These results were not surprising because abundance and richness differences are expected due to population dynamics, life-histories, and dispersion activity, which are seasonally related (Williams and Feltmate, 1992; Fairchild et al., 2003). These findings have implications for further studies on biodiversity based on field surveys, but also for those that use existing information. Thus, field surveys may be more efficient if they are planned to be performed during spring, because it will capture higher amount of biodiversity at all levels. On the other hand, if the study is based on existing information on surveys performed at different seasons, it would be better to use biodiversity metrics not seasonally biased (as, for example, variation in taxonomic distinctness and rarity values) to assure robust results. 4.4. Wetland aquatic invertebrate fauna and conservation policies Many of the international conservation policies neglect aquatic invertebrate fauna. For example, only one of the nine criteria used

2609

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

for the Ramsar Convention is exclusively restricted to fauna other than fish or birds (Ramsar, 2006). Other policies, such as the European Habitats Directive (92/43/CEE), are focused on the protection of habitats because of their uniqueness. The uniqueness of sites is mainly determined by a list of endangered species (Annex II of the Directive). This list overlooks invertebrate fauna as is exemplified by their scarce representation (only 61 out of 224 taxa), in particular the aquatic invertebrates (only 14 out of 61 invertebrate taxa). Moreover, both the Ramsar Convention and the EU Habitats Directive do not consider any kind of biodiversity measure. Therefore, the neglect of aquatic invertebrate fauna and their biodiversity to establish conservation priority sites could be a possible explanation for the lack of protection status effect on invertebrate biodiversity values. Other studies also conclude that focusing on any single biodiversity component alone is insufficient to protect other components (Cabeza and Moilanen, 2001; Bonn and Gaston, 2005). In fact, a recent review on the performance of protected areas already manifested that several studies did not find a positive response of biodiversity against protection for various reasons, and one of the reasons arose because of ‘‘insufficient consideration of biological criteria” (Gaston et al., 2008 and references therein). Moreover, Gaston et al. (2008) emphasize in their concluding remarks the existence of several gaps when trying to evaluate protected areas performance. In this sense, our study has dealt with two of them: (1) supplying information on success of protected areas using different biodiversity metrics, and (2) evaluating different levels of protection to asses their ecological performance. Nevertheless, it is interesting to highlight that maybe protection status was not established through consideration of invertebrate assemblages, but these assemblages are also sensitive to habitat degradation, because they showed higher biodiversity values in well preserved wetlands (although only weak relationships were found). This kind of positive effect of habitat condition on biodiversity is not surprising since it has been documented using other target faunal groups such as mammals, butterflies, birds or amphibians (e.g. Beja and Alcazar, 2003; Stapanian et al., 2004; Rouget et al., 2006). 5. Conclusions Our results showed that conductivity turned out to be the main factor controlling Mediterranean wetland aquatic invertebrate biodiversity. In addition, they showed how the use of several biodiversity metrics that supply additional information would improve the evaluation of wetland biodiversity values (e.g. rarity in brackish and saline systems). It is interesting to note that taxonomic distinctness metrics were the only ones that were significantly related to water nutrient concentrations, although these relationships, at least in Mediterranean wetlands, may be taken with caution because they are seasonally affected. In contrast, more traditionally known metrics, such as species richness and rarity based on habitat specificity, showed similar responses in both seasons, but on the other hand were less sensitive to nutrient variability. Moreover, the fact that managed and protected sites did not have a significantly higher value for any of the biodiversity metrics tested suggests that wetland aquatic invertebrate fauna is neglected by state conservation strategies. Acknowledgments This work was supported by the Ministerio de Educación y Ciencia, Programa de Investigación Fundamental (Ref. GL200805778/BOS). We thank Mònica Martinoy and Jaume Gifre for field and laboratory assistance, and Xavier Quintana and two anonymous reviewers for their useful comments on the manuscript.

Appendix A Microcrustaceans found in this study. The table shows the water body typology in which the species were recorded (WBT; freshwater temporary (fwt), freshwater permanent (fwp), and saline water bodies (sw)), occurrence of the species (O), and the highest abundance in a sample (HA).

CLADOCERA Alonella excisa Alonella exigua Alona guttata Alona rectangula Bosmina longirostris Camptocercus rectirostris Ceriodaphnia dubia Ceriodaphnia laticaudata Ceriodaphnia pulchella Ceriodaphnia quadrangula Ceriodaphnia reticulata Chydorus sphaericus Daphnia curvirostris Daphnia galeata Daphnia longispina Daphnia magna Daphnia obtusa Daphnia pulicaria Ilyocryptus sordidus Leydigia leydigii Macrothrix sp. Megafenestra aurita Moina brachiata Moina micrura Oxyurella tenuicaudis Disparalona leei Pleuroxus aduncus Pleuroxus denticulatus Pleuroxus laevis Scapholeberis rammneri Simocephalus exspinosus Simocephalus vetulus Sida crystallina Tretocephala ambigua COPEPODA Acanthocyclops gr. robustus Calanipeda aquaedulcis Canthocamptus microstaphylinus Canthocamptus staphylinus Canuella perplexa Cletocamptus confluens Copidodiaptomus numidicus Cyclops sp. Diacyclops bicuspidatus Diacyclops bisetosus Diaptomus cyaneus Ectocyclops phaleratus Ergasilus sp. Eucyclops macruroides Eucyclops serrulatus Eurytemora velox Halicyclops magniceps Halicyclops rotundipes

WBT

O (%)

HA (%)

fwt fwp fwp fwt, fwp sw, fwt, fwp fwp fwt, fwp fwp fwp sw, fwt, fwp sw, fwt, fwp sw, fwt, fwp sw, fwt, fwp fwp fwp sw, fwt, fwp fwt, fwp sw, fwt, fwp fwp fwp fwt fwp fwt, fwp sw, fwp fwp fwp sw, fwt, fwp fwp sw, fwp sw, fwp sw, fwt, fwp sw, fwt, fwp fwp fwt, fwp

1.08 1.08 1.08 10.75 11.83 2.15 2.15 1.08 2.15 8.60 9.68 47.31 8.60 1.08 1.08 16.13 8.60 20.43 2.15 4.30 1.08 1.08 6.45 5.38 5.38 1.08 19.35 5.38 9.68 9.68 8.60 29.03 1.08 3.23

27.36 50.00 0.99 9.96 28.20 8.24 78.07 0.40 40.33 42.14 83.89 78.63 95.66 1.15 0.14 93.77 94.97 93.31 17.53 57.01 1.20 3.89 64.28 78.98 17.65 5.51 35.13 15.63 58.39 6.19 27.55 53.42 0.93 10.19

sw, fwt, fwp 52.69 100.00 sw, fwp 20.43 97.38 fwt 1.08 0.91 sw, fwt, fwp 17.20 97.96 sw 7.53 63.68 sw 5.38 72.82 fwp 1.08 98.10 sw, fwt, fwp 37.63 92.52 sw, fwt, fwp 35.48 99.06 sw, fwt 8.60 58.10 fwt 3.23 24.64 fwp 4.30 36.59 sw 1.08 4.26 fwp 2.15 29.41 sw, fwt, fwp 24.73 100.00 sw 7.53 96.89 sw 1.08 0.58 sw 7.53 10.25 (continued on next page)

2610

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

Appendix B (continued)

Appendix A (continued) WBT

O (%)

HA (%)

Harpacticus littoralis Lernaea sp. Macrocyclops albidus Macrocyclops fuscus Megacyclops viridis Mesochra heldti Metacyclops minutus Microcyclops rubellus Mixodiaptomus incrassatus Mixodiaptomus kupelwieseri Nitokra lacustris Nitokra spinipes Onychocamptus mohammed Paracyclops affinis Paracyclops fimbriatus Paracyclops imminutus Phyllognathopus viguieri Schizopera sp. Thermocyclops dybowskii Tisbe longicornis Tropocyclops prasinus

sw, fwp fwp sw, fwp fwt fwt, fwp sw, fwp fwt fwp fwt fwt, fwp sw sw fwp sw, fwp fwp fwp sw fwp sw, fwp sw sw, fwt, fwp

3.23 1.08 22.58 1.08 22.58 5.38 1.08 2.15 1.08 10.75 5.38 2.15 1.08 3.23 2.15 1.08 1.08 2.15 2.15 3.23 18.28

50.95 0.12 57.78 4.71 85.46 47.58 38.73 12.50 0.26 89.14 35.86 2.76 5.88 20.00 10.58 57.00 1.07 50.00 39.39 46.68 94.92

OSTRACODA Bradleycypris obliqua Bradleystrandesia fuscata Bradleystrandesia reticulata Candona neglecta Candona sp. Candonocypris sp. Cyclocypris ovum Cypria ophtalmica Cyprideis torosa Cypridopsis vidua Cypris bispinosa Cypris subglobosa Eucypris virens Fabaeformiscandona fabaeformis Herpetocypris brevicaudata Herpetocypris chevreuxi Heterocypris incongruens Heterocypris salinus Ilyocypris gibba Loxoconcha elliptica Paralimnocythere psammophila Plesiocypridopsis newtoni Sarscypridopsis aculeata Tonnacypris lutaria

fwt fwt fwt fwt sw, fwt fwp fwt, fwp fwp sw sw, fwt, fwp fwt fwp sw, fwt, fwp fwp sw, fwp fwt, fwp sw, fwt, fwp sw, fwp fwt, fwp sw, fwp fwt sw, fwt, fwp sw fwt

1.08 2.15 1.08 1.08 2.15 2.15 4.30 3.23 9.68 24.73 2.15 1.08 26.88 2.15 2.15 5.38 10.75 11.83 2.15 11.83 2.15 3.23 6.45 2.15

1.36 9.78 0.90 2.64 0.68 7.48 95.33 7.39 71.64 80.42 10.10 0.03 50.36 2.82 1.51 21.27 12.12 100.00 10.84 100.00 8.65 75.58 80.49 2.40

Appendix B Macroinvertebrates found in this study. The table shows the water body typology in which the species were recorded (WBT; freshwater temporary (fwt), freshwater permanent (fwp), and saline water bodies (sw)), occurrence of the species (O), and the highest abundance in a sample (HA). WBT LARGE BRANCHIOPODA Chirocephalus diaphanus Tanymastix stagnalis

fwt fwt

O (%)

HA (%)

3.23 3.23

96.93 100.00

WBT

O (%)

HA (%)

1.08

66.22

Triops cancriformis

fwt

MALACOSTRACA Atyaephyra desmarestii Carcinus aestuari Corophium orientale Echinogammarus pacaudi Gammarus aequicauda Heterotanais oerstedi Idotea chelipes Lekanesphaera hookeri Leptocheirus pilosus Mesopodopsis slabberi Palaemon longirostris Palaemonetes zariquieyi Proasellus coxalis Procambarus clarkii Protracheoniscus occidentalis

fwp sw sw, fwp fwp sw, fwp sw, fwp sw sw, fwp sw sw sw fwp sw, fwt, fwp sw, fwt, fwp sw, fwp

4.30 1.08 8.60 3.23 20.43 2.15 1.08 13.98 2.15 4.30 2.15 3.23 5.38 5.38 2.15

56.82 0.55 100.00 92.58 100.00 85.71 9.38 97.41 7.59 66.67 100.00 13.64 100.00 100.00 11.11

ODONATA Aeshna sp. Anax imperator Anax parthenope Ischnura elegans Lestes barbarus Lestes sp. Lestes virens Lestes viridis Libellula depressa Sympetrum fonscolombii Sympetrum striolatum

fwt, fwp sw, fwp fwp sw, fwt, fwp fwt sw fwp fwp fwp sw, fwt, fwp fwp

3.23 2.15 3.23 38.71 5.38 1.08 2.15 1.08 1.08 16.13 1.08

28.57 20.00 20.00 100.00 23.08 6.67 1.37 0.15 0.05 100.00 0.12

EPHEMEROPTERA Caenis luctuosa Cloeon inscriptum

fwt, fwp sw, fwt, fwp

2.15 44.09

34.82 100.00

HETEROPTERA Anisops sardeus Corixa affinis Corixa panzeri Corixa punctata Cymatia rogenhoferi Gerris argentatus Gerris thoracicus Hebrus pusillus Hesperocorixa linnaei Hesperocorixa moesta Mesovelia vittigera Micronecta scholtzi Naucoris maculatus Notonecta maculata Notonecta meridionalis Notonecta viridis Plea minutissima Sigara dorsalis Sigara lateralis Sigara scripta Sigara stagnalis

sw, fwt, fwp fwt sw, fwt, fwp fwt, fwp fwp sw, fwp fwp fwp fwt fwt fwp fwp sw, fwp fwp fwp sw sw, fwt, fwp fwp sw, fwt, fwp fwt sw, fwp

15.05 1.08 5.38 3.23 1.08 8.60 1.08 1.08 2.15 7.53 1.08 11.83 3.23 1.08 1.08 1.08 13.98 3.23 15.05 2.15 4.30

84.21 5.02 37.88 0.33 0.01 6.76 0.01 4.55 26.85 78.53 1.51 100.00 18.18 3.07 6.67 0.06 100.00 37.50 100.00 14.63 72.73

COLEOPTERA Agabus bipustulatus Agabus conspersus Agabus didymus

fwt fwt fwt, fwp

1.08 1.08 2.15

0.37 100.00 2.13

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

Appendix B (continued) WBT

O (%)

HA (%)

Agabus montanus Agabus nebulosus Anacaena bipustulata Anacaena limbata Berosus affinis Berosus hispanicus Berosus signaticollis Cybister lateralimarginalis Cyphon sp. Dryops algiricus Enochrus bicolor Enochrus cf. isotae Graptodytes bilineatus Graptodytes flavipes Graptodytes ignotus Haliplus lineatocollis Haliplus variegatus Helochares lividus Helophorus alternans Helophorus fulgidicollis Helophorus sp. Hydraena sp. Hydrochus sp. Hydroglyphus geminus Hydrophilus piceus Hydroporus sp. Hydrovatus cuspidatus Hygrotus inaequalis Ilybius fuliginosus Laccophilus minutus Laccophilus poecilus Noterus clavicornis Noterus laevis Ochthebius dilatatus Ochthebius punctatus Ochthebius viridescens Oulimnius rivularis Paracymus sp. Peltodytes rotundatus Rhantus suturalis Stictonectes sp.

fwt fwt fwp fwt fwt sw, fwt fwt fwt sw fwt sw, fwp fwt sw, fwt, fwp fwt fwt fwt, fwp sw, fwt fwt, fwp fwt sw, fwt, fwp sw, fwt, fwp fwp fwp fwp fwt fwt, fwp fwp fwp fwp fwp fwp fwt, fwp fwp fwt, fwp fwp fwp fwt, fwp sw fwp fwt, fwp fwt

1.08 1.08 1.08 1.08 1.08 2.15 5.38 1.08 1.08 3.23 12.90 1.08 29.03 5.38 1.08 3.23 3.23 6.45 1.08 5.38 6.45 1.08 1.08 2.15 1.08 2.15 1.08 1.08 1.08 1.08 1.08 4.30 1.08 4.30 1.08 1.08 2.15 1.08 1.08 2.15 3.23

0.11 26.27 0.66 0.11 7.14 3.78 33.33 0.36 2.22 9.76 100.00 0.84 100.00 32.98 0.11 0.30 16.67 52.11 8.21 16.67 20.00 0.22 0.22 28.17 0.11 16.67 6.67 46.67 0.19 0.08 1.41 23.08 14.29 100.00 100.00 5.63 0.11 20.00 0.44 46.67 9.48

TRICHOPTERA Ecnomus sp. Limnephilus sp.

fwp fwt, fwp

2.15 8.60

100.00 50.00

References Abellán, P., Sánchez-Fernández, D., Velasco, J., Millán, A., 2005a. Assessing conservation priorities for insects: status of water beetles in southeast Spain. Biological Conservation 121, 79–90. Abellán, P., Sánchez-Fernández, D., Velasco, J., Millán, A., 2005b. Conservation of freshwater biodiversity: a comparison of different area selection methods. Biodiversity and Conservation 14, 3457–3474. Abellán, P., Sánchez-Fernández, D., Velasco, J., Millán, A., 2007. Effectiveness of protected area networks in representing freshwater biodiversity: the case of a Mediterranean river basin (south-eastern Spain). Aquatic Conservation – Marine and Freshwater Ecosystems 17, 361–374. Beja, P., Alcazar, R., 2003. Conservation of Mediterranean temporary ponds under agricultural intensification: an evaluation using amphibians. Biological Conservation 114, 317–326. Biggs, J., Corfield, A., Walker, D., Whitfield, M., Williams, P., 1994. New approaches to the management of ponds. British Wildlife 5, 273–287.

2611

Boix, D., Sala, J., Moreno-Amich, R., 2001. The faunal composition of Espolla pond (NE Iberian peninsula): the neglected biodiversity of temporary waters. Wetlands 21, 577–592. Boix, D., Gascón, S., Sala, J., Martinoy, M., Gifre, J., Quintana, X.D., 2005. A new index of water quality assessment in Mediterranean wetlands based on crustacean and insect assemblages: the case of Catalunya (NE Iberian peninsula). Aquatic Conservation – Marine and Freshwater Ecosystems 15, 635–651. Boix, D., Gascón, S., Sala, J., Badosa, A., Brucet, S., López-Flores, R., Martinoy, M., Gifre, J., Quintana, X.D., 2008. Patterns of composition and species richness of crustaceans and aquatic insects along environmental gradients in Mediterranean water bodies. Hydrobiologia 597, 53–69. Bonn, A., Gaston, K.J., 2005. Capturing biodiversity: selecting priority areas for conservation using different criteria. Biodiversity and Conservation 14, 1083– 1100. Cabeza, M., Moilanen, A., 2001. Design of reserve networks and the persistence of biodiversity. Trends in Ecology & Evolution 16, 242–248. Campbell, W.B., Arce-Pérez, R., Gómez-Anaya, J.A., 2008. Taxonomic distinctness and aquatic Coleoptera: comparing a perennial and intermittent stream with differing geomorphologies in Hidalgo, Mexico. Aquatic Ecology 42, 103–113. Céréghino, R., Biggs, J., Oertli, B., Declerck, S., 2008a. The ecology of European ponds: defining the characteristics of a neglected freshwater habitat. Hydrobiologia 597, 1–6. Céréghino, R., Ruggiero, A., Marty, P., Angélibert, S., 2008b. Biodiversity and distribution patterns of freshwater invertebrates in farm ponds of a southwestern French agricultural landscape. Hydrobiologia 597, 43–51. Clarke, K.R., Warwick, R.M., 1998. A taxonomic distinctness index and its statistical properties. Journal of Applied Ecology 35, 523–531. Clarke, K.R., Warwick, R.M., 2001. A further biodiversity index applicable to species lists: variation in taxonomic distinctness. Marine Ecology – Progress Series 216, 265–278. Collinson, N.H., Biggs, J., Corfield, A., Hodson, M.J., Walker, D., Whitfield, M., Williams, P.J., 1995. Temporary and permanent ponds: an assessment of the effects of drying out on the conservation value of aquatic macroinvertebrate communities. Biological Conservation 74, 125–133. Crawley, M.J., 2002. Statistical Computing. An Introduction to Data Analysis using SPlus. John Wiley & Sons, Chichester. Crawley, M.J., 2007. The R Book. John Wiley & Sons, Chichester. Davies, B., Biggs, J., Williams, P., Lee, J., Thompson, S., 2008a. A comparison of the catchment sizes of rivers, streams, ponds, ditches and lakes: implications for protecting aquatic biodiversity in an agricultural landscape. Hydrobiologia 597, 7–17. Davies, B., Biggs, J., Williams, P., Whitfield, M., Nicolet, P., Sear, D., Bray, S., Maund, S., 2008b. Comparative biodiversity of aquatic habitats in the European agricultural landscape. Agriculture Ecosystems & Environment 125, 1–8. De Meester, L., Declerck, S., Stoks, R., Louette, G., Van de Meutter, F., De Bie, T., Michels, E., Brendonck, L., 2005. Ponds and pools as model systems in conservation biology, ecology and evolutionary biology. Aquatic Conservation – Marine and Freshwater Ecosystems 15, 715–725. Declerck, S., Vandekerkhove, J., Johansson, L., Muylaert, K., Conde-Porcuna, J.M., Van der Gucht, K., Perez-Martinez, C., Lauridsen, T., Schwenk, K., Zwart, G., Rommens, W., Lopez-Ramos, J., Jeppesen, E., Vyverman, W., Brendonck, L., De Meester, L., 2005. Multi-group biodiversity in shallow lakes along gradients of phosphorus and water plant cover. Ecology 86, 1905–1915. DPTOP, 2005. Hipermapa. Atles electrònic de Catalunya. Departament de Política Territorial i Obres Públiques. . Fairchild, G.W., Cruz, J., Faulds, A.M., Short, A.E.Z., Matta, J.F., 2003. Microhabitat and Landscape Influences on Aquatic Beetle Assemblages in a Cluster of Temporary and Permanent Ponds. Journal of the North American Benthological Society 22, 224–240. Gaston, K.J., 1994. Rarity. Chapman & Hall, London. Gaston, K.J., Spicer, J.I., 2004. Biodiversity: An Introduction. Blackwell Publishing, Oxford. Gaston, K.J., Jackson, S.F., Cantú-Salazar, L., Cruz-Piñón, G., 2008. The ecological performance of protected areas. Annual Review of Ecology, Evolution, and Systematics 39, 93–113. Gopal, B., Junk, W.J., 2000. Biodiversity in wetland: an introduction. In: Gopal, B., Junk, W.J., Davis, J.A. (Eds.), Biodiversity in Wetlands: Assessment, Function and Conservation. Backhuys Publishers, Leiden, pp. 1–10. Gotelli, N.J., Colwell, R.K., 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters 4, 379– 391. Grasshoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis. Verlag Chemie, Weinheim. Hansson, L.A., Bronmark, C., Nilsson, P.A., Abjornsson, K., 2005. Conflicting demands on wetland ecosystem services: nutrient retention, biodiversity or both? Freshwater Biology 50, 705–714. Heino, J., Soininen, J., Lappalainen, J., Virtanen, R., 2005. The relationship between species richness and taxonomic distinctness in freshwater organisms. Limnology and Oceanography 50, 978–986. Heino, J., Mykrä, H., Hämäläinen, H., Aroviita, J., Muotka, T., 2007. Responses of taxonomic distinctness and species diversity indices to anthropogenic impacts and natural environmental gradients in stream macroinvertebrates. Freshwater Biology 52, 1846–1861.

2612

S. Gascón et al. / Biological Conservation 142 (2009) 2602–2612

Heino, J., Mykrä, H., Kotanen, J., 2008. Weak relationships between landscape characteristics and multiple facets of stream macroinvertebrate biodiversity in a boreal drainage basin. Landscape Ecology 23, 417–426. Hessen, D.O., Walseng, B., 2008. The rarity concept and the commonness of rarity in freshwater zooplankton. Freshwater Biology 53, 2026–2035. Hothorn, T., Hornik, K., Zeileis, A., 2006. Unbiased recursive partitioning: a conditional inference framework. Journal of Computational and Graphical Statistics 15, 651–674. Jeffries, M., 2005. Small ponds and big landscapes: the challenge of invertebrate spatial and temporal dynamics for European pond conservation. Aquatic Conservation – Marine and Freshwater Ecosystems 15, 541–547. Lyons, M.N., Halse, S.A., Gibson, N., Cale, D.J., Lane, J.A.K., Walker, C.D., Mickle, D.A., Froend, R.H., 2007. Monitoring wetlands in a salinizing landscape: case studies from the wheatbelt region of Western Australia. Hydrobiologia 591, 147–164. MAPA, 2006. Sistema de identificación de parcelas agrícolas. Ministerio de Agricultura Pesca y Alimentación. . Marchant, R., 2007. The use of taxonomic distinctness to assess environmental disturbance of insect communities from running water. Freshwater Biology 52, 1634–1645. Mariotti, A., Struglia, M.V., Zeng, N., Lau, K.M., 2002. The hydrological cycle in the Mediterranean region and implications for the water budget of the Mediterranean sea. Journal of Climate 15, 1674–1690. Mouillot, D., Gaillard, S., Aliaume, C., Verlaque, M., Belsher, T., Troussellier, M., Chi, T.D., 2005. Ability of taxonomic diversity indices to discriminate coastal lagoon environments based on macrophyte communities. Ecological Indicators 5, 1–17. Oertli, B., Auderset Joye, D., Castella, E., Juge, R., Cambin, D., Lachavanne, J.B., 2002. Does size matter? The relationship between pond area and biodiversity. Biological Conservation 104, 59–70. Pielou, E.C., 1969. An Introduction to Mathematical Ecology. Wiley-Interscience, New York. Piscart, C., Moreteau, J.C., Beisel, J.N., 2005. Biodiversity and structure of macroinvertebrate communities along a small permanent salinity gradient (Meurthe River, France). Hydrobiologia 551, 227–236. Pressey, R.L., Humphries, C.J., Margules, C.R., Vanewright, R.I., Williams, P.H., 1993. Beyond opportunism – key principles for systematic reserve selection. Trends in Ecology & Evolution 8, 124–128. Puchalski, W., 1987. Phytoplankton Assemblages in After Exploitation Reservoirs. Inst. Ecology Polish Academy of Sciences, Dziekanow Lesny. R-Development-Core-Team. 2007. R: A Language and Environment for Statistical Computing, Vienna. Rabinowitz, D., Cairns, S., Dillon, T., 1986. Seven forms of rarity and their frequency in the flora of the British Isles. In: Soulé, M.E. (Ed.), Conservation Biology: The Science Scarcity and Diversity. Sinauer Associates, Sunderland, Massachusetts, pp. 182–204.

Ramsar, 2006. The Ramsar Convention Manual: A Guide to the Convention on Wetlands (Ramsar, Iran, 1971). Ramsar Convention Secretariat, Gland, Switzerland. Rodrigo, M.A., Rojo, C., Armengol, X., 2003. Plankton biodiversity in a landscape of shallow water bodies (Mediterranean coast, Spain). Hydrobiologia 506, 317– 326. Rodrigues, A.S.L., Brooks, T.M., 2007. Shortcuts for biodiversity conservation planning: the effectiveness of surrogates. Annual Review of Ecology Evolution and Systematics 38, 713–737. Rogers, S.I., Clarke, K.R., Reynolds, J.D., 1999. The taxonomic distinctness of coastal bottom-dwelling fish communities of the North-east Atlantic. Journal of Animal Ecology 68, 769–782. Rouget, M., Cowling, R.M., Vlok, J., Thompson, M., Balmford, A., 2006. Getting the biodiversity intactness index right: the importance of habitat degradation data. Global Change Biology 12, 2032–2036. Sala, J., Gascón, S., Boix, D., Gesti, J., Quintana, X.D., 2004. Proposal of a rapid methodology to assess the conservation status of Mediterranean wetlands and its application in Catalunya (NE Iberian Peninsula). Archives Des Sciences 57, 141–151. Scheffer, M., van Geest, G.J., Zimmer, K., Jeppesen, E., Søndergaard, M., Butler, M.G., Hanson, M.A., Declerck, S., De Meester, L., 2006. Small habitat size and isolation can promote species richness: second-order effects on biodiversity in shallow lakes and ponds. Oikos 112, 227–231. Stapanian, M.A., Waite, T.A., Krzys, G., Mack, J.J., Micacchion, M., 2004. Rapid assessment indicator of wetland integrity as an unintended predictor of avian diversity. Hydrobiologia 520, 119–126. Talling, J.F., Driver, D., 1963. Some problems in the estimation of chlorophyll a in phytoplankton, In: Doty, M.S. (Ed.), Proceedings of the Conference on Primary Productivity Measurement, Marine and Freshwater. University of USA Atomic Energy Commission, Division of Technical Information TID 7633, Hawaii, pp. 142–146. Tarr, T.L., Baber, M.J., Babbitt, K.J., 2005. Macroinvertebrate community structure across a wetland hydroperiod gradient in Southern New Hampshire, USA. Wetlands Ecology and Management 13, 321–334. Vanewright, R.I., Humphries, C.J., Williams, P.H., 1991. What to protect – Systematics and the agony of choice. Biological Conservation 55, 235–254. Walters, K., Coen, L.D., 2006. A comparison of statistical approaches to analyzing community convergence between natural and constructed oyster reefs. Journal of Experimental Marine Biology and Ecology 330, 81–95. Warwick, R.M., Dexter, D.M., Kuperman, B., 2002. Freeliving nematodes from the Salton sea. Hydrobiologia 473, 121–128. Williams, D.D., Feltmate, B.W., 1992. Aquatic Insects. C.A.B International, Wallingford. Williams, D.D., 2006. The Biology of Temporary Waters. Oxford University Press, Oxford.