Cumulative annual dung beetle diversity in Mediterranean seasonal environments

ORIGINAL ARTICLE

Rossana Agoglitta • Claudia E. Moreno

Mario Zunino • Gabriella Bonsignori

Marco Dellacasa

Cumulative annual dung beetle diversity in Mediterraneanseasonal environments

Received: 30 August 2011 / Accepted: 29 November 2011 / Published online: 14 December 2011� The Ecological Society of Japan 2011

Abstract Species diversity assessments should consider thedynamic nature of ecological communities, especially inhighly seasonal ecosystems. Here we provide a compre-hensive framework for analysing seasonal changes inspecies composition, richness and diversity in two localdung beetle (Coleoptera: Scarabaeoidea: Aphodiidae,Geotrupidae, and Scarabaeidae) communities from Wes-tern Tuscany (Italy), in the Mediterranean ecoregion. Wetest whether, in this highly seasonal region, cumulativeannual diversity is an oversimplification of well differen-tiated seasonal communities. Data were obtained throughrepeated standardised samples collected regularly over anentire year. We clearly identify different summer andwinter communities at each site based on species compo-sition and abundance. Seasonal richness and diversityvalues are different from the cumulative annual values, asa consequence of beta diversity between seasons, andsome dung beetle species are identified as idiosyncratic of

each particular season. Both ecological (niche partition-ing) and biogeographical factors are suggested as driversof these temporal variations. Thus, because local inven-tories of fauna that include records over long time periodsactually reflect situations where coexistence and interac-tions are unlikely to occur, highly seasonal sites must beviewed as having temporally differentiated communities inorder to reach feasible and reliable baselines for localdiversity assessments.

Keywords Species diversity Æ Temporal beta diversity ÆTemporal species turnover Æ Scarabaeoidea ÆIdiosyncratic species

Introduction

Species diversity has been studied extensively for manydecades, mainly in relation to spatial variation (e.g.Ricklefs and Schluter 1993; Rosenzweig 1995), and morerecently to possible connections with human disturbance(Hanski 2005), ecosystem functioning (Kinzing et al.2001) and climate change (Sanchez-Rojas et al. 2011).Even though diversity has become a fundamental issuein ecology and conservation biology, its definition andquantitative expression have been under debate recently(Ricotta 2007; Moreno and Rodrıguez 2011; Tuomisto2011; and references therein). In spite of this discussion,we are aware that, to protect local and regional speciesdiversity effectively, we need to know how such speciesdiversity is spatially and temporally distributed. Forexample, does a local community consist of different tem-poral communities, or is it homogeneous? And, if so, howdifferent are the temporal communities? The answers tothese questions would improve our understanding ofdiversity patterns and processes within communities, andcould be useful for developing clear recommendations forproperly sampling and protection strategies.

The term a diversity is applied to the species diversitywithin a community (Whittaker 1960). But this conceptcan be refer to point, mean or cumulative a diversity

R. Agoglitta Æ M. ZuninoDipartimento di Scienze dell’Uomo, dell’Ambiente e della Natura,Universita degli Studi di Urbino ‘‘Carlo Bo’’, Localita Crocicchia,61029 Urbino, ItalyE-mail: [email protected]

M. ZuninoE-mail: [email protected]

C. E. Moreno (&)Centro de Investigaciones Biologicas, Universidad Autonoma delEstado de Hidalgo, Apdo. postal 69, Plaza Juarez, 42001 Pachuca,Hgo, MexicoE-mail: [email protected].: +52-771-7172000Fax: +52-771-7172112

G. BonsignoriCRIM Lab, Polo Sant’Anna Valdera, Scuola Superiore Sant’Anna,Viale Rinaldo Piaggio, 34, Pontedera, 56025 Pisa, ItalyE-mail: [email protected]

M. DellacasaCentro Interdipartimentale Museo di Storia Naturale e delTerritorio, Universita di Pisa, via Roma, 79, 56011 Calci, ItalyE-mail: [email protected]

Ecol Res (2012) 27: 387–395DOI 10.1007/s11284-011-0910-8

(Halffter and Moreno 2005): the number of species ateach elementary sampling unit was called point diversityby Whittaker (1977), average species richness for a set ofsampling units within a community is mean a diversity,and the accumulated richness for the complete set ofsampling units in a community is the cumulative adiversity (Halffter and Moreno 2005). So, cumulative adiversity describes the sum of the species that can befound between two temporal limits. It includes everynew local record, and does not exclude species that werepreviously recorded but are no longer present in thecommunity (Halffter and Moreno 2005).

In a spatial context, b diversity is the extent of changein species composition among different communities;thus regional diversity results from both a and b diver-sity (Whittaker 1960). Similarly, the cumulative diversitywithin a community may result from both the diversitywithin relatively homogenous time periods, and tempo-ral b diversity, which may be defined as the change inspecies composition between different time periods(Moreno and Halffter 2001). Spatial b diversity is ana-lysed as the ratio between regional and mean a diversity(Jost 2007), and, correspondingly, the ratio betweencumulative and mean diversity within a community hasbeen useful for studying temporal b diversity (Morenoand Halffter 2001; Romanuk and Kolasa 2001; Errouissiet al. 2009).

Ecologists have long realised that local communitieschange through time, and that this may be due to severalfactors (Fjeldsa and Lovett 1997). Consequently, itseems clear that local inventories of fauna, includingdata that are sometimes distributed over many months,years or even decades, actually result from the inclusionof data from situations that are not necessarily homo-geneous; such inventories are therefore not a feasiblebaseline from which to evaluate local diversity. More-over, while species richness might remain stable, com-munity composition can be dynamic due to temporal bdiversity (Werner et al. 2007). This is particularly evidentfor the cumulative a diversity obtained over a whole yearin seasonal environments (Agoglitta 2008). Thus, whenmean a is considerably lower than cumulative diversity,a high b diversity over time occurs (Moreno and Halffter2001). If a community has high temporal b diversity,single month or seasonal samples will not be represen-tative of the complete annual species diversity, andcumulative annual samples will not be representative oflocally coexisting assemblages.

Dung beetle communities are well known for theirrole in ecosystem function and are considered an excel-lent focal group for studying species diversity patterns(Halffter and Favila 1993; Halffter 1994; Davis et al.2004; Spector 2006; Nichols et al. 2007, 2008).

Seasonal changes in temperature and rainfall, likethose typical of the Mediterranean climate, are criticalfor dung beetles (Hanski and Cambefort 1991). Thisregion is characterized by a clear seasonality, which isparticularly evident in the contrasting trends in tem-perature and precipitation (Fig. 1); i.e. the winter is cold

and rainy, while the summer is hot and quite dry. Thecombination of high temperatures and lack of rainfallduring summer usually reduces Mediterraneancoprophagous communities drastically, in both theirspecies number and population abundance. Wintertemperatures are usually less harsh and do not cause ananalogous effect (Lumaret and Kirk 1991). Overall, inthis type of climate, dung beetles may be more activeduring spring and autumn (Hanski and Cambefort1991).

Several studies have shown seasonal changes in dungbeetle community structure worldwide, and in the Cen-tral Mediterranean ecoregion (sensu Bulgarini et al.2006) there is also recent evidence of this. In south-western Germany, Wassmer (1994) compared thestructure and composition of coprophagous communi-ties every month from April 1992 to May 1993, revealingimportant structural changes that corresponded to twoseasonal cycles: February–April and May–September.In Spanish agro-ecosystems, Zamora et al. (2007) ana-lysed the temporal b diversity of dung beetles within asequence of 15-day-long sampling periods from May2001 to June 2002. They reached the conclusion that thehigh temporal b diversity probably did not result fromtemporal variation in species composition, but ratherwas the result of a well-established community thatadapted to the drastic seasonality of Mediterraneanconditions. Jay-Robert et al. (2008) studied the rela-tionship between behavioural models (dung-dwellers vssoil-diggers) and seasonality in south-western France,and stressed that seasonal changes in activity ‘‘constitutea phylogenetically inherited character’’. Errouissi et al.(2009) monitored dung beetle assemblages from Januaryto December 2006 in Northern Tunisia, and found a

Fig. 1 Monthly average temperature (solid line) and precipitation(dashed line) in the Maremma Regional Park (a), and the SanRossore Presidential Estate (b) in central Italy, during the periodsof dung beetle sampling. Source: http://www.ilmeteo.it/portale/archivio-meteo

388

significant seasonal variation in dung beetle assemblagediversity and composition, with the highest temporal bdiversity in October and in February. In Italy, Carpa-neto et al. (2005) referred to seasonal differences withinthe coprophagous communities, although they used adifferent approach based on aspects of the beetles’feeding preferences.

Here we provide a conceptual and methodologicalcontribution to the analysis of biodiversity in the CentralMediterranean ecoregion, and therefore a baseline formonitoring it, taking seasonality into account and usingdung beetle communities (Scarabaeidae, Aphodiidaeand Geotrupidae families) as the model. We focussedonly on adult individuals that are active each month;thus, our analysis and conclusions do not apply to larvalstages. We discuss the implications of this partial view ofthe community. Our specific research questions were: (1)over an annual cycle in highly seasonal localities ofcentral Italy, are there seasons of the year clearly dif-ferentiated by temporal b diversity between them, basedon the variation of dung beetle species composition andabundance? (2) If so, is seasonal b diversity related moreto a nested pattern of richness than to species turnover,at each site? (3) How do species richness and diversitychange between seasons, and between seasons and thedata set for the entire year? And (4), which dung beetlescan be considered as idiosyncratic species of particularseasons of the year?

Methods

Study areas and dung beetle sampling

We used data from annual samples taken at two sites ca.200 km from Tuscany (central-western Italy), separatedby approximately 90 km: the Maremma Regional Park(Grosseto province), located at the south of the Tuscanycoastal strip, and the San Rossore Presidential Estate(Pisa province), located at the Northeast of Pisa. Atboth sites, the main dung resources during field samplingwere from cattle and horses.

Dung beetles were sampled using standard method-ologies (pitfall traps baited with cow and horse dung) atboth sites. The Maremma site was sampled from June2003 to May 2004 (Bonsignori 2006). During this period,the maximum temperature was recorded in August 2003and the minimum in January 2004, while precipitationwas high in November 2003 and February 2004 (Fig. 1a).Dung beetles were collected every 15 days from 22 trapsdistributed among five sampling stations. Traps werelocated at a minimum distance of 10 m, in natural hal-ophilous grasslands with few or no trees, in Pinus pineapine forests on sand, and in degraded olive tree fields oncalcareous soil with apparent rocks. At San Rossore, 20traps were placed with a minimum distance of 10 m intwo artificial pastures as sampling stations (10 traps persampling station; Dellacasa 1995), on a sandy-limestonesoil, and on a medium-texture soil (essentially clayey),

where sampling was repeated weekly over 1 year (1992).During this period the maximum temperature was re-corded in August, and the maximum precipitation was inOctober (Fig. 1b).

In this way, sampling sites cover a wide spectrum oftypical habitat conditions in the Central Mediterraneanecoregion, although spatial variation is not consideredas a factor design in this study. In order to verify thecompleteness of species inventories before data analysis,we estimated the expected total richness for each siteusing the Chao2 estimator (Chao and Shen 2010), andfound that species inventories were 92.66 and 92.90%complete at Maremma and San Rossore, respectively.Thus we consider that our databases are reliable repre-sentations of dung beetle communities at both sites. Forthe systematics of Geotrupidae, we referred to Zunino(1984), and Baraud (1992); for Scarabaeidae, Baraud(1992), and for Aphodiidae, Dellacasa and Dellacasa(2006).

Data analysis

Data from each site were pooled by month for theanalysis. In order to assess the variation in speciescomposition between months in the two study areas, wefirst used non-metric multidimensional scaling (NMDS)to graph the relative position of the months (samples)according to their similarity in species composition(based on species number of individuals per sample,without transformations). NMDS ranks the values ofdistance among all pairs of samples, in this case usingthe Horn index, which is one of the most easily inter-preted similarity measures because it is a measure of trueoverlap of order 1, and is related to the concept of truebeta diversity (Jost 2007). The ordination is then used inan iterative procedure to locate the samples in a two-dimensional space that maintains the distances amongsamples. For each site, the months grouped in thisordination space were considered as a ‘‘season’’. Then,in order to partition the variability of species responsesinto two factors (site and season), we carried out apermutational multivariate analysis of variance usingdistance matrices (PERMANOVA), as a non-paramet-ric alternative to the multivariate analysis of variance(Anderson 2001).

For each site, we compared species richness anddiversity between seasons and the data set for the entireyear. For species richness we used rarefaction curveswith 95% confidence intervals, using the minimum valueof abundance at each site (7,500 individuals in Mare-mma and 16,500 in San Rossore) as standard samplingsize for comparisons. For species diversity we calculatedthe number of effective species using the measure of truediversity of order 1 (Jost 2006), which weights eachspecies exactly by its frequency in the community (i.e.favouring neither rare nor common species). Given thepresence of rare species (singletons) in all communities,we calculated the estimated diversity using the Chao and

389

Shen (2003) method, which is a non-parametric ap-proach that allows for an accurate estimate when thereare unseen species in a community. This method has alsothe advantage of providing 95% confidence intervalswith satisfactory sample coverage probability (Chao andShen 2003). Then, in order to separate the contributionof temporal turnover and nestedness to the total amountof b diversity between seasons, we used the methodproposed by Baselga (2010). This partitioning of bdiversity was performed for each site using presence–absence data. We also plotted the rank–abundancecurves for seasonal and cumulative beetle communitiesat each site.

Finally, we identified the idiosyncratic species ofeach seasonal community using the indicator value in-dex (IndVal) proposed by Dufrene and Legendre(1997), which is a method that has been used success-fully for dung beetles by McGeoch et al. (2002). Thehighest value of the index (100%) is obtained when allthe individuals of a species are found in one seasononly, and when that particular species occurs in allsamples from that specific season. The statistical sig-nificance of IndVal was evaluated using a randomisa-tion procedure. The IndVal analysis was performedwith the software PC-ORD v. 4.0 (McCune and Mef-ford 1999). The effective species number for truediversity of order 1, with its estimated 95% confidenceintervals based on sample coverage, was calculatedusing the program SPADE (Chao and Shen 2010). Allthe other analyses were done with the R-statisticalcomputing language (R Development Core Team 2009)using the vegan package.

Results

Seasonality from the point of view of dung beetles

The complete data set included 81,171 dung beetlespecimens belonging to 63 species. The NMDS ordina-tion of this data (two-dimensional stress: 0.159) showedgradual changes among samples based on species com-position and the relative abundance of dung beetles(Fig. 2). Axis 2 (R2 = 0.147) separates the two studysites, while Axis 1 (R2 = 0.589) orders the samples bytime in cyclic continuous gradients through the year foreach site. This is a foreseeable pattern given that com-munities do not usually exhibit sharp compositionalchanges. However, in order to establish main seasons ofthe year based on dung beetle temporal b diversity, wearbitrarily delimited summer and winter groups oftemporal samples. Some samples were considered astransition months between seasons, according to theirintermediate species composition between the summerand winter assemblages. This decision was consistentwith discernible changing times in temperature andprecipitation at each site (Fig. 1). For the MaremmaRegional Park, the group of summer months runs from

April to September, while winter is from November toFebruary. For this site March is a transition month fromwinter to summer, and October is the transition fromsummer to winter. At San Rossore, summer is from Mayto September, and winter is November–March. As inMaremma, in San Rossore October is also the transitionfrom summer to winter, but April is a transition monthfrom winter to summer. In San Rossore, this seasonalcategorisation is clear. However, at the Maremma sitewe found that the September sample is as far fromOctober (the transitional month) as from August(Fig. 2), the former and latter being considered withinthe same summer group.

Both site and season, as well as the interaction be-tween them, are significant sources of variation for thedistance matrices between months (Table 1). Takinginto account these results, we decided to consider thesummer and winter communities from each site as dif-ferent, and performed separate analyses of species rich-ness and ecological diversity, and also for between-seasons nestedness and turnover components of bdiversity. Leaving aside the four transitional months, the

Fig. 2 Non-metric multidimensional scaling of monthly samplesbased on dung beetle species composition and abundance in theMaremma Regional Park (black circles) and San Rossore Presi-dential Estate (gray circles) in central Italy. Dotted lines groupsummer months, and continuous lines group winter months of eachsite. Samples outside of the seasonal groupings are consideredtransitional months between seasons. Stress value 0.1593

Table 1 Permutational multivariate analysis of variance on com-positional similarity for assemblages of dung beetle species duringtwo seasons (summer and winter) at two sites in central Italy (99permutations)

Source df SS MS Pseudo-F P

Sites 1 1.1095 1.1095 6.6435 0.01Season 1 2.0314 2.0314 12.1637 0.01Site · season 1 0.7951 0.7951 4.7607 0.01Residual 16 2.6721 0.1670Total 19 6.6081

390

data set was then restricted to 10 months at each site.This way, the data includes a total of 63,287 dung beetlespecimens belonging to 62 species: 19,720 individualsbelonging to 52 species for Maremma Regional Park,and 43,567 individuals in 46 species for San Rossore.

Species richness and diversity in seasonsand the entire year

At the Maremma Regional Park, when sampling size isstandardised at 7,500 individuals, the summer commu-nity species richness is almost the same as the cumulativeannual richness (47.5 and 49.05 species, respectively),while richness was clearly lower for the winter commu-nity (25 species), as seen in the rarefaction curves(Fig. 3). We found the same pattern when we comparedtrue diversity taking into account species frequencies. Inthis case, both the summer and the annual communitieshave ca. 17 effective species, while the winter communityhas a significantly lower diversity, with only four effec-tive species (confidence intervals shown in Fig. 3), whichmeans that, in Maremma, the summer dung beetlecommunity is at least four times more diverse than thewinter community.

In contrast, species richness of the winter and sum-mer communities at San Rossore is very similar (37 and37.44 species, respectively), but the cumulative annualrichness is higher (41.37 species) when samples with thesame size (16,500 individuals) are used (Fig. 3). Speciesdiversity of the winter, summer and annual communitiesis different (3.85, 6.61 and 9.05 effective species, respec-tively; Fig. 3), indicating that the summer community is1.72 times more diverse than the winter community, andthe cumulative annual community is 1.37 times morediverse than the summer community. However, thesample coverage method used for estimating confidenceintervals did not show significant differences in diversityamong communities (Fig. 3).

Accordingly, b diversity between seasons is driven bycompositional turnover and nestedness in different waysat the two sites. At the Maremma Regional Park,70.40% of total beta diversity is related to a nestednesspattern, because the winter community is a subsample ofthe summer community. Thus, in this site species turn-over explains only 29.60% of b diversity. On the otherhand, at San Rossore, nestedness accounts for only16.75% of total beta diversity, while species turnoverexplains 83.25% of it.

The low diversity found for the winter communities isrelated to the presence of highly dominant species, whichare drawn in the rank–abundance graphs (Fig. 4). In theMaremma Regional Park, Nimbus johnsoni accounts for55.77% of the total number of individuals of the wintercommunity, followed by Melinopterus consputus with19.10%; while Sericotrupes niger accounts for 16.13% ofthe total abundance during summer. For San Rossore,Melinopterus consputus represents 68.28% of the totalnumber of individuals of the winter community, followed

Fig. 3 Rarefaction curves of dung beetle species richness forsummer (open triangles), winter (black triangles) and cumulativeannual communities (solid line) in the Maremma Regional Parkand San Rossore Presidential Estate, central Italy. The inner graphsshow the species diversity of order 1 (effective number of species)for these communities, with 95% confidence intervals

Fig. 4 Rank–abundance graphs of cumulative annual dung beetlecommunities in summer and winter in the Maremma Regional Parkand San Rossore Presidential Estate

391

by Sericotrupes niger with 10.50%. Onthophagus (Pala-eonthophagus) grossepunctatus accounts for a noteworthy54.27%of the total abundance of this site during summer.

Dung beetle idiosyncratic species in each season

The high relative abundance of some species, togetherwith their frequency in the samples, allows for theidentification of species idiosyncratic for the summer

and winter communities of both sites. Seven of the 52species analysed from Maremma were statistically sig-nificant idiosyncratic species of the winter season, while11 species were significant idiosyncratics of the summerseason (Table 2). From the 46 species of San Rossore, 6and 11 species were statistically significant idiosyncraticspecies of the winter and the summer seasons, respec-tively. In order to show their temporal variation, weselected four of these idiosyncratic species that hadsimilar total abundance, each representing one commu-

Table 2 Dung beetle specieswith high, and significantlydifferent from random, valuesas idiosyncratics for thesummer and wintercommunities of the MaremmaRegional Park and San RossorePresidential Estate in centralItaly

Site Season Species IndVala P

Maremma Summer Otophorus haemorrhoidalis 100 <0.01Acanthobodilus immundus 100 <0.01Onthophagus (Furconthophagus) furcatus 100 <0.01Onthophagus (Onthophagus) illyricus 100 <0.01Subrinus sturmi 99.9 <0.05Euoniticellus fulvus 99.8 <0.01Labarrus lividus 99.7 <0.05Onthophagus (Onthophagus) taurus 99.6 <0.01Sisyphus schaefferi 99.4 <0.01Onthophagus (Palaeonthophagus) grossepunctatus 83.3 <0.05Onthophagus (Palaeonthophagus) vacca 83.3 <0.05

Maremma Winter Nimbus johnsoni 100 <0.01Melinopterus stolzi 100 <0.01Melinopterus consputus 99.8 <0.01Melinopterus prodromus 97.9 <0.01Sigorus porcus 95.9 <0.05Agrilinus constans 93.4 <0.05Bubas bison 92.5 <0.05

San Rossore Summer Onthophagus (Furconthophagus) furcatus 99.8 <0.01Onthophagus (Onthophagus) taurus 99.8 <0.01Labarrus lividus 99.7 <0.01Euoniticellus fulvus 99.7 <0.01Onthophagus (Palaeonthophagus) grossepunctatus 99.2 <0.01Planolinoides borealis 98.5 <0.01Volinus sticticus 98.4 <0.05Onthophagus (Palaeonthophagus) vacca 95.5 <0.05Onthophagus (Palaeonthophagus) ruficapillus 92.7 <0.05Bodiloides ictericus ghardimaouensis 80 <0.05

San Rossore Winter Ceratophyus rossii 100 <0.01Melinopterus consputus 99.9 <0.01Bubas bison 98.1 <0.01Typhaeus typhoeus 95.3 <0.01Geotrupes spiniger 89.2 <0.01Planolinus fasciatus 75.8 0.05

aIndVal values were obtained with the method proposed by Dufrene and Legendre (1997)

Fig. 5 Phenology of four idiosyncratic species with similar total abundance for the summer and winter communities at each site

392

nity (summer and winter at the two sites). Their abun-dance clearly exemplifies the phenology of the obviouslythermophilous species that characterize the summercommunities, and the typical psychrophilous andhygrophilous species of the winter communities (Fig. 5).

Discussion

Our framework for the analysis of changes in speciesrichness, diversity and composition allow us to disclosedifferent drivers of the cumulative annual dung beetlediversity in two localities of the Euromediterranean areathat experience strong seasonal differences in climate.High temperatures and reduced rainfall characterise thesummer season, while precipitation is limited mostly tothe few months of the cold winter season.

Under such climatically variable conditions, hightemporal b diversity is expected to occur (Moreno andHalffter 2001). Seasonal fluctuations in environmentalconditions offer opportunities for temporal niche parti-tioning, i.e. habitat conditions favour different species atdifferent times, depending on their niche requirements(Chesson and Huntly 1997; Chase and Leibold 2003).Species may partition resources on a temporal scale,likely influenced by thermal constraints (Albrecht andGotelli 2001). Given that patterns in local speciesdiversity are driven by coexistence rules (Weiher andKeddy 1999; Chase and Leibold 2003), we can expect ahigh diversity when species coexist by partitioning theirniches. Although further research is required to under-stand the processes underlying seasonal b diversity in thehighly diverse dung beetle communities in this part ofItaly, temporal niche partitioning determined by thermalrequirements seems to be the most feasible mechanism.

All of our results are based only on adult dung beetlecommunities, as they are most usually sampled in studiesof diversity patterns (Halffter 1994; Davis et al. 2004;Spector 2006; Nichols et al. 2007, 2008). However, thelack of data on larval stages is not likely to have a majorinfluence on our interpretation of community interac-tions. Dung beetle larvae are largely immobile and de-velop in underground chambers with a predeterminedamount of resource provided by their parents, as abrood ball mass (Halffter and Edmonds 1982; Halffter1998). Once this dung is finished, larvae do not searchfor additional food sources. Thus, only adults activelylocate food resources and thus may have strong intraand interspecific interactions (Halffter and Edmonds1982; Hanski and Cambefort 1991).

Differences in species richness between seasons, andtheir relationship to the cumulative annual richness(Moreno and Halffter 2001; Romanuk and Kolasa2001), as well as separating the influence of nestednessand turnover components (Baselga 2010), are the basisfor understanding temporal b diversity at each site. Asshown in the partitioning of total b diversity, at theMaremma Regional Park, seasonal b is driven mainly by

nestedness, which results from differences in richness(Baselga 2010). Thus, the winter community could rep-resent an impoverished assemblage of the summercommunity; though there may also be a biogeographiceffect (see below). At this site, ca. 94% of the totalnumber of species found throughout the year is presentin the summer community (of this, 52% are exclusive tothe summer and 42% are shared between the summerand winter communities), while only three species(5.77% of the total) are found exclusively during thewinter. Conversely, at San Rossore the cumulative an-nual richness is higher than the richness of both thesummer and winter communities, which coincides withthe fact that species turnover plays a major role indetermining b diversity between seasons. At this site,although ca. 67% of the total number of species foundthroughout the year are shared between both seasons,ca. 20% were found exclusively during summer, while13% were only recorded during the winter months.Consequently, under the circumstances of this study, theoversimplification of the data obtained over an entireyear leads to values that may be conceptually homolo-gous to those of the cumulative a diversity. Thus,interpretation of these values from an ecological per-spective is problematic because cumulative a diversityincludes all new species records, but does not exclude therecords of species that are no longer active in the com-munity because of dormancy, emigration, death or evenlocal extirpation. Thus, the cumulative a diversity doesnot reflect the real coexistence and interaction of indi-viduals in a community, and strays further from realityas the time interval of sampling increases (Halffter andMoreno 2005).

From a biogeographical standpoint, our results agreewith the conclusions of Jay-Robert et al. (2008; see alsoZunino 1985), who proposed that the Euromediterraneancoprophagous Scarabaeoidea fauna as a whole reflects abiogeographical transition process; however, evaluationof the importance of both current and historical biogeo-graphical constraints on the taxonomical structure(Zunino 2005; Bellucci et al. 2007) of the winter andsummer communities requires further study. Indeed,while the Aphodiinae and Geotrupinae are groups with aclearly Holarctic affinity, the Scarabaeidae include com-ponents with Afrotropical or, perhaps secondarily, Cen-tral Asiatic affinity. Considering both ecological andbiogeographical factors, it seems that there are two dis-tinct assemblages present at each site. On one hand, thewinter community is well represented by idiosyncraticspecies with North centred distribution areas, such asNimbus johnsoni, Melinopterus stolzi, Melinopterus con-sputus,Melinopterus prodromus, Sigorus porcus,Agrilinusconstans, Typhaeus typhoeus, Geotrupes spiniger andPlanolinus fasciatus; however, Bubas bison is a Mediter-ranean species, while Ceratophyus rossii is endemic to asmall area of Tuscany but belongs to a genus with scatterdistribution along the ancient Northern margins of Te-thys and that has an isolated species in California. On theother hand, the summer community is represented by

393

some idiosyncratic species with South centered (mainlyMediterranean) distribution areas, including Bodiloidesictericus ghardimaouensis, Euoniticellus fulvus, Onthoph-agus (Palaeonthophagus) grossepunctatus, Onthophagus(Palaeonthophagus) ruficapillus, Sisyphus schaefferi andSubrinus sturmi; or extending to Central Asiatic areas,such as Acanthobodilus immundus, Labarrus lividus, On-thophagus (Onthophagus) illyricus, Onthophagus (Onth-ophagus) taurus,Onthophagus (Furconthophagus) furcatusand Onthophagus (Palaeonthophagus) vacca. However,three species (Otophorus haemorrhoidalis, Planolinoidesborealis andVolinus sticticus) are distributed largely alongthe Palaearctic.

Variations in the summer and winter communitiescould be a useful instrument for predicting and moni-toring the effects of climate change on biodiversity. Forexample, in a scenario of climate change where droughtincreases, we would expect a reduction in the temporaldifferentiation of dung beetles in our study area, leadingto the homogenisation of communities throughout theyear. Also, the phenology of the species identified here asrepresentatives of summer and winter communitiescould be monitored to detect changes throughout theyear. However, the categorical seasons used in this workmay not accurately reflect fine temporal changes, asthose related to the transitional months, which we havedeliberately removed from our analysis of species rich-ness and diversity. Further research may be focused oncyclic patterns, analysing time in a continuous scale.Time series analysis may be useful to explore fine phe-nological changes.

From a realistic point of view, we should think aboutwhether cumulative a diversity should be taken as areliable measure of community diversity or not. For acomprehensive evaluation of coexistence in seasonalenvironments, either communities are sampled over ashort period of time (which may be difficult if a numberof communities must be sampled to assess spatial vari-ation), or temporal b diversity is also analysed byincluding time as a factor in sampling design. If the goalis to inventory the species diversity of any site in thisregion, time will be required to sample a high proportionof the species, and surveys carried out during only oneseason would be inadequate. But if we are looking forthe natural delimitations of ecological communitiesbased on species coexistence and on population inter-actions, highly seasonal sites must be viewed as havingtemporally differentiated communities. Consequently,for a conceptually coherent and a practical alternative, adiversity values should refer to species that coexist in aparticular ecological time.

Conclusion

The strong seasonality of the Central Mediterraneanecoregion influences dung beetle richness, diversity andspecies composition. At both sites, it seems that surveys

conducted during the summer only could be appropriateto characterise at least 90% of local species diversity andrichness. However, species composition varies in timeand space: at the Maremma site, the winter communityseems to be a nested subsample of the summer com-munity, while at San Rossore species turnover betweenseasons plays a more relevant role in structuring com-munities. Further research is required to disentangle theprocesses involved in the temporal dynamics of diver-sity, and to develop better methods for the analysis oftemporal b diversity, taking into account the influence ofniche partitioning and the biogeographical drivers ofthese temporal variations. Overall, our results suggestthat, when species diversity is evaluated in highly sea-sonal environments, we must take into account the dy-namic nature of ecological communities. This propertymay be disclosed with the combined information ofspecies richness, diversity, abundance and compositionin ecological communities.

Acknowledgements C.E.M. is grateful to Jorge Montero for fruitfuldiscussions about temporal b diversity and advice on some statis-tical methods, and to Anne Chao for her suggestions aboutdiversity methods. Data analysis and manuscript writing weresupported by the SEP-CONACYT Basic Science project 84127(Mexico).

References

Agoglitta R (2008) Il popolamento a coleotteri Scarabaeoidea de-gradatori del settore siciliano. Biodiversita e biogeografia. PhDthesis dissertation, University of Urbino, Italy

Albrecht M, Gotelli NJ (2001) Spatial and temporal niche parti-tioning in grassland ants. Oecologia 126:134–141

Anderson MJ (2001) A new method for non-parametric multivar-iate analysis of variance. Austr Ecol 26:32–46

Baraud J (1992) Coleopteres Scarabaeoidea d’Europe. Faune deFrance. France et Regions Limitrophes, vol 78. Societe Linne-enne de Lyon, Lyon

Baselga A (2010) Partitioning the turnover and nestedness com-ponents of beta diversity. Global Ecol Biogeogr 19:134–143

Bellucci S, Agoglitta R, Zunino M (2007) Gli scarabeidi degrada-tori dell’area marchigiana: appunti corologici e biogeografici.Biogeographia 28:631–638

Bonsignori G (2006) Analisi del popolamento a Scarabaeoideacoprofagi del Parco Regionale della Maremma. BSc thesis,University of Pisa, Italy

Bulgarini F, Putrella S, Teofili C (2006) Biodiversity vision: laconservazione della biodiversita nell’Ecoregione MediterraneoCentrale, contributi al Piano Nazionale per la Biodiversita.WWF Italia, Ministero dell’Istruzione, dell’Universita e dellaRicerca, Roma

Carpaneto GM, Mazziotta A, Piattella E (2005) Changes in foodresources and conservation of scarab beetles: from sheep to dogdung in a green urban area of Rome (Coleoptera, Scarabaeoi-dea). Biol Conserv 123:547–556

Chao A, Shen T-J (2003) Nonparametric estimation of Shannon’sindex of diversity when there are unseen species in sample.Environ Ecol Stat 10:429–443

Chao A, Shen T-J (2010) Program SPADE (Species prediction anddiversity estimation). Program and User’s Guide. http://chao.stat.nthu.edu.tw

Chase JM, Leibold MA (2003) Ecological niches: linking classicaland contemporary approaches. The University of Chicago,Chicago

394

Chesson P, Huntly N (1997) The roles of harsh and fluctuatingconditions in the dynamics of ecological communities. Am Nat150:519–553

Davis ALV, Scholtz CH, Dooley PW, Bham N, Kryger U (2004)Scarabaeine dung beetles as indicators of biodiversity, habitattransformation and pest control chemicals in agro-ecosystems.S Afr J Sci 100:415–424

Dellacasa M (1995) Gli scarabeidi coprofagi (Coleoptera Scarab-aeoidea) e il controllo biologico dello sterco nei pascoli di SanRossore (Pisa). BSc thesis, University of Pisa, Italy

Dellacasa G, Dellacasa M (2006) Coleoptera Aphodiidae, Apho-diinae. Fauna d’Italia, vol XLI. Calderini edizioni, Bologna,Italy

Dufrene M, Legendre P (1997) Species assemblages and indicatorspecies: the need for a flexible asymmetrical approach. EcolMonogr 67:345–366

Errouissi F, Labidi I, Nouira S (2009) Seasonal occurrence andlocal coexistence within scarabaeid dung beetle guilds (Cole-optera: Scarabaeoidea) in Tunisian pasture. Eur J Entomol106:85–94

Fjeldsa J, Lovett JC (1997) Biodiversity and environmental sta-bility. Biodivers Conserv 6:315–323

Halffter G (1994) Que es la biodiversidad? Butlletı de la InstitucioCatalana d’Historia Natural 62:5–14

Halffter G (1998) A strategy for measuring landscape biodiversity.Biol Int 36:3–17

Halffter G, Edmonds WD (1982) The nesting behavior of dungbeetles (Scarabaeinae). An ecological and evolutive approachInstituto de Ecologıa. Galache, Mexico

Halffter G, Favila ME (1993) The Scarabaeinae (Insecta: Coleop-tera) an animal group for analyzing, inventorying and moni-toring biodiversity in tropical rainforest and modifiedlandscapes. Biol Intl 27:15–21

Halffter G, Moreno CE (2005) Significado biologico de las diver-sidades alfa, beta y gamma. In: Halffter G, Soberon J, Koleff P,Melic A (eds) Sobre diversidad biologica: el significado de lasdiversidades alfa, beta y gamma. Monografıas Tercer Milenio,vol 4. Sociedad Entomologica Aragonesa, Zaragoza, Espana,pp 5–18

Hanski I (2005) The shrinking world: ecological consequences ofhabitat loss. Excellence in ecology, vol 14. International Ecol-ogy Institute, Oldendorf

Hanski I, Cambefort Y (1991) Resource partitioning. In: Hanski I,Cambefort Y (eds) Dung beetle ecology. Princeton UniversityPress, Princeton, pp 330–349

Jay-Robert P, Errouissi F, Lumaret JP (2008) Temporal coexis-tence of dung-dweller and soil-digger dung beetles (Coleoptera,Scarabaeoidea) in contrasting Mediterranean habitats. BullEntomol Res 98:303–316

Jost L (2006) Entropy and diversity. Oikos 113:363–375Jost L (2007) Partitioning diversity into independent alpha and

beta components. Ecology 88:2427–2439Kinzing AP, Pacala SW, Tilman D (2001) The functional conse-

quences of biodiversity: empirical progress and theoreticalextensions. Princeton University Press, New Jersey

Lumaret JP, Kirk AA (1991) South temperate dung beetles. In:Haski I, Cambefort Y (eds) Dung beetle ecology. PrincetonUniversity Press, Princeton, pp 97–115

McCune B, Mefford MJ (1999) PC-ORD for windows multivariateanalysis of ecological data V.4.0. MjM Software, GlenedenBeach, Oregon

McGeoch MA, Van Rensburg BJ, Botes A (2002) The verificationand application of bioindicators: a case study of dung beetles ina savanna ecosystem. J Appl Ecol 39:661–672

Moreno CE, Halffter G (2001) Spatial and temporal analysis of thealpha, beta and gamma diversities of bats in a fragmentedlandscape. Biodivers Conserv 10:367–382

Moreno CE, Rodrıguez P (2011) Commentary: do we have aconsistent terminology for quantifying species diversity? Backto basics and toward a unifying framework. Oecologia 167:889–892

Nichols E, Larsen T, Spector S, Davis ALV, Escobar F, FavilaME, Vulinec K (2007) Global dung beetle response to tropicalforest modification and fragmentation: a quantitative literaturereview and meta-analysis. Biol Conserv 137:1–19

Nichols E, Spector S, Louzada J, Larsen T, Amezquita S, FavilaME (2008) Ecological functions and ecosystem services pro-vided by Scarabaeinae dung beetles. Biol Conserv 141:1461–1474

R Development Core Team (2009) R: a language and environmentfor statistical computing. R Foundation for Statistical Com-puting, Vienna, Austria. http://www.R-project.org. ISBN3-900051-07-0

Ricklefs RE, Schluter D (1993) Species diversity in ecologicalcommunities: historical and geographical perspectives. TheUniversity of Chicago Press, Chicago

Ricotta C (2007) A semantic taxonomy for diversity measures.Acta Biotheor 55:23–33

Romanuk TN, Kolasa J (2001) Simplifying the complexity oftemporal diversity dynamics: a differentiation approach. Eco-science 8:259–263

Rosenzweig ML (1995) Species diversity in space and time. Cam-bridge University Press, Cambridge

Sanchez-Rojas G, Ballesteros-Barrera C, Pavon NP (2011) Cambioclimatico: aproximaciones para el estudio de su efecto en labiodiversidad. Universidad Autonoma del Estado de Hidalgo,Pachuca, Mexico

Spector S (2006) Scarabaeine dung beetles (Coleoptera: Scara-baeidae: Scarabaeinae): an invertebrate focal taxon for biodi-versity research and conservation. Coleopt Bull 60:71–83

Tuomisto H (2011) Commentary: do we have a consistent termi-nology for species diversity? Yes, if we choose to use it. Oeco-logia 167:903–911

Wassmer T (1994) Seasonality of coprophagous beetles in theKaiserstuhl area near Freiburg (SW-Germany) including thewinter months. Acta Oecol 15:607–631

Weiher E, Keddy P (1999) Ecological assembly rules. CambridgeUniversity Press, Cambridge

Werner EE, Yurewicz KL, Skelly DK, Relyea RA (2007) Turnoverin an amphibian metacommunity: the role of local and regionalfactors. Oikos 116:1713–1725

Whittaker RH (1960) Vegetation of the Siskiyou Mountains,Oregon and California. Ecol Monogr 30:279–338

Whittaker RH (1977) Evolution of species diversity in land com-munities. In: Hecht MK, Steere WC, Wallace B (eds) Evolu-tionary biology, vol 10. Plenum, New York, pp 1–67

Zamora J, Verdu JR, Galante E (2007) Species richness in Medi-terranean agroecosystems: spatial and temporal analysis forbiodiversity conservation. Biol Conserv 134:113–121

Zunino M (1984) Sistematica generica dei Geotrupinae (Coleop-tera, Scarabaeoidea: Geotrupidae), filogenesi della famiglia econsiderazioni biogeografiche. Boll Mus Reg Sci Nat Torino2:9–162

Zunino M (1985) Gli Scarabaeoidea coprofagi dell’area euromed-iterranea: relazioni filetiche e biogeografiche. In: Atti XIVCongresso Nazionale Italiano di Entomologia, Palermo,pp 321–325

Zunino M (2005) Corotipos y biogeografıa sistematica en el Eu-romediterraneo. In: Llorente J, Morrone JJ (eds) Regional-izacion biogeografica en Iberoamerica y topicos afines. UNAM,Mexico DF, pp 181–187

395