Environmental features influencing Carabid beetle (Coleoptera) assemblages along a recently deglaciated area in the Alpine region

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report
Category:

Economy & Finance

Published:

Views: 14 | Pages: 8

Extension: PDF | Download: 0

Share
Description
Abstract 1. The spatio-temporal approach was used to evaluate the environmental features influencing carabid beetle assemblages along a chronosequence of an Italian Alpine glacier foreland. The influence of environmental variables on species
Transcript
   Ecological Entomology  (2007), 32 , 682–689 DOI: 10.1111/j.1365-2311.2007.00912.x © 2007 The Authors682 Journal compilation © 2007 The Royal Entomological Society  Introduction Predicting the impacts of global change on species richness and species distribution has been studied frequently (e.g. McCarty, 2001; Parmesan & Yohe, 2003; Harte et al.  , 2004 ). Ecologists realise the importance of relating environmental change and community change with current climate warming, with parti-cular reference to the temporal-scale approach. Assemblage changes over time therefore appear to be a good key to interpret global change ( Bardgett et al.  , 2005 ). These kinds of studies are important to try to identify keystone taxa or bioindicators. Soil-dwelling macroinvertebrates can indicate changes in the envi-ronment through their responses as individuals or communities. Hodkinson and Jackson (2005) highlighted the potential role of invertebrates for monitoring environmental change with partic-ular references to the montane ecosystems. They suggested the use of Carabid beetles (Insecta, Coleoptera) as good indicators for studies involving landscape and habitat changes. Correspondence: Mauro Gobbi, Department of Biology, University of Milan, Via Celoria 26, I-20133, Milan, Italy. E-mail: gobbi.mauro@tiscali.it Environmental features influencing Carabid beetle (Coleoptera) assemblages along a recently deglaciated area in the Alpine region MAURO GOBBI 1  , BRUNO ROSSARO 1  , AMBER VATER 3  , FIORENZA DE BERNARDI 1  , MANUELA PELFINI 2  and PIETRO BRANDMAYR 4   1  Department of Biology, University of Milan, Milan, Italy , 2  Department of Earth Science, University of Milan, Milan, Italy , 3  Department of Geography, University of Swansea, Singleton Park, Swansea, U.K. and 4  Department of Ecology, University of Calabria, Arcavacata di Rende (CS), Italy Abstract . 1. The spatio-temporal approach was used to evaluate the environmental features influencing carabid beetle assemblages along a chronosequence of an Italian Alpine glacier foreland. The influence of environmental variables on species richness, morphology (wing and body length), and distribution along the chronosequence was tested. 2. Species richness was found to be a poor indicator of habitat due to weak influences by environmental variables. It seems that the neighbouring habitats of a glacier foreland are not able to determine significant changes in carabid species richness. 3. Instead it appears that history (age since deglaciation) and habitat architecture of a glacier foreland are strongly correlated to species adaptive morphological traits, such as wing morphology and body length. Assemblages characterised by species with reduced wing size are linked to the older stages of the chronosequence, where habitat is more structured. Assemblages characterised by the largest species are linked to the younger sites near the glacier. These morphological differentiations are explained in detail. 4. Habitat age can therefore be considered the main force determining assemblage composition. On the basis of the relationship between morphological traits and en-vironmental variables, it seems likely that age since deglaciation is the main variable influencing habitat structure (primary effect) on the Forni foreland. The strong rela-tionship between carabid assemblages and habitat type indicates that site age has but a secondary effect on carabid assemblages. This may be utilised to interpret potential changes in assemblages linked to future glacier retreat. Key words . Body length , climate change , Forni Glacier , ground beetles , wing morphology .   Carabid assemblages along an alpine glacier foreland 683   © 2007 The AuthorsJournal compilation © 2007 The Royal Entomological Society,  Ecological Entomology , 32 , 682–689  Literature focusing on the response of carabids to the chang-ing environments is diverse. For example, Niemelä et al.  (2000) studied communities living in different lowland agroecosys-tems and described how assemblages react to human activities. Carabids found in European mountains have been studied by several authors. A recent up-to-date review of carabid assem-blages in the Alpine region of the Alps was given by Brandmayr et al.  (2003a,b ). While more is known about changes in carabid composition during the past (e.g. Foddai & Minelli, 1994; Ashworth, 1996; Ponel et al.  , 2003 ) and about the relation be- tween changes in species richness along altitudinal – vegetation gradients (e.g. Brandmayr et al.  , 2003a,b ; Eyre et al.  , 2005; Hodkinson, 2005 ), little is known about responses to climate and ecosystem changes. To understand carabid beetle responses to climate change and thus to environmental change it is important to select an area where the composition of communities is determined by abiotic and biotic factors (and not by anthropogenic activities). The area must be able to react rapidly to changes in climate of vary-ing intensity. Deglaciation since the Little Ice Age maximum (LIA; 16th – 19th centuries) creates an ideal area in which to study reactions to changes in communities. Ecosystem develop-ment on terrain freed by retreating glaciers represents an excel-lent opportunity for the detailed study of primary succession triggered by climatic change. With their well-known chronol-ogy of glacial recession called chronosequence, glacier fore-lands have been used as a unique model system to study the action of climate changes on living soil communities ( Bardgett et al.  , 2005 ). Along a chronosequence direct comparisons can be made between communities established on soils of different ages since deglaciation ( Matthews, 1992 ). It can be assumed, therefore, that plots of different ages represent different stages in the development of the community. Invertebrate assemblages along glacier forelands have previ-ously been studied in the Austrian Alps ( Kaufmann, 2001, 2002 ), Italian Alps ( Gobbi et al.  , 2006a,b ), in Svaldbard and Sweden ( Hodkinson et al.  , 1998, 2004 ) as well as in southern Norway (A. Vater, unpublished data). The aim of this research was to analyse the spatial patterns of carabid beetles assemblages in relation to altitudinal and environ-mental gradients along the Little Ice Age chronosequence of an alpine glacier. Species richness and species assemblages were ana-lysed across an area with a spatial and temporal gradient. The first objective was to analyse if diversity changes in relation to environ-mental features or in relation to the altitudinal gradient. The second objective was to analyse and discuss spatial species replacements and the nestedness  in the patterns of species distribution along the spatio-temporal succession, and the third objective was to deter-mine if habitat architecture determined morphological features (wing morphology and body length) of species. Material and methods Study area The study area was the Forni Valley (46°25   N, 10°34   E) in the Italian Middle-Eastern Alps ‘SRII/A’ ( Marazzi, 2005 ). The Forni Glacier covers a 12 km 2  surface (the biggest valley glacier in the Italian Alps). The valley is characterised by a 2.5 km fore-land with a well preserved 155-year chronosequence distributed along an altitudinal gradient of 500 m ( Pelfini, 1992; Pelfini & Smiraglia, 1992 ). The major late Holocene advances are marked by frontal moraines dated to 1850, 1904, 1926 and 1980. Using various records it was possible to determine the position of the glacier in 1943, 1953, and 2000. Sampling stations Sites A to G are found within the proglacial area of the fore-land covering terrain from 1850 to 2000. Sites H and I are posi-tioned on the glacier surface. H is located in the debris cover of the glacier while I is set into the ice itself. Site L is located be-tween the stony detritus on the hydrographical left margin of the glacier. Sites are therefore distributed to cover altitudinal and environmental gradients both on the glacier and on the foreland [ Fig. 1 ; see also Gobbi et al.  (2006a) for a detailed study site description]. The sampling season was completed during the snow-free season (July – September) of 2004 and 2005. During 2004 the right side of the valley was analysed from A to F (sites identified by number 1 following the letter), together with samples H. During 2005 the left side of the valley was analysed (sites are depicted with a number ‘2’ following the letter), together with samples G, I, and L ( Fig. 1 ). A total of 108 pitfall traps were used, arranged in transects of six traps for each sampling site, with a 10 m distance between each trap. Pitfall traps were plas-tic cups (7 cm diameter) baited with a standard mixture of wine vinegar and salt to preserve the catch ( Greenslade, 1964;  Southwood, 1978; Brandmayr, 2005 ). Pitfalls were collected and re-set every 20 days. A total of 33 traps were found to con-tain no specimens. Analyses were therefore carried out using data from the 75 traps found to contain carabids and considering data from single traps and not total data from sites.  Analyses Species richness . Carabid beetles were determined to species level following the Fauna Europea Web Service for the nomen-clature ( Audisio & Vigna Taglianti, 2004 ). Species richness was computed as number of species found in each trap. Species replacement . Some sites were sampled during two sampling seasons. It could be hypothesised that during 1 year no differences in species diversity for each site would be expected, but sampling sites from A to F in 2004 and 2005 were at differ-ent sites in the valley. Therefore, the reliability of this merged data was assessed by testing species diversity for each site with coupled sampling in 2004 and 2005. A correlation test was used between coupled species diversity for   each site in 2004 and 2005 merging all data from each trap for every site in each year. To analyse species replacement between sites, a meta-com-munity  structural analysis was performed ( Leibold & Mikkelson, 2002 ). The probability of obtaining the observed spatial turno-ver between species and communities was calculated using a   684  Mauro Gobbi  et al. © 2007 The AuthorsJournal compilation © 2007 The Royal Entomological Society,  Ecological Entomology , 32 , 682–689 Monte Carlo simulation proposed by Leibold and Mikkelson (2002) . The significance of differences in the number of species replacements among communities was tested (between the ob-served and the simulated matrices using a two-tailed  Z   -test). Significantly lower rates of species replacements than expected ( =  low spatial turnover) represent nested structures, with hier-archical structures among communities. Higher rates than ex-pected are known as anti-nested structures, and provide evidence of groups of species that are mutually exclusive.  Environmental variables and morphological species  features . Several environmental variables were measured at each sampling station and used in the analysis. Variables in-cluded altitude, distance from the glacier, age since deglacia-tion, soil dryness (presence or absence of a water layer above the ground) as well as herbaceous vegetation and stone cover. Environmental variables, morphological variables (species traits), and species abundances were log 10  (  x    +  1) transformed, and percentage values (  p  ) were transformed to arcsin √  (  p   /100) to normalise the distribution. Wing morphology and body length were measured for each species. Wing morphology was determined observing the pres-ence or the absence of functional wings (metathoracic alae). Species with wings longer than the elitrae were considered macropterous  . Such species are known therefore to have good dispersal abilities. Species with wings shorter than the elitrae were considered brachypterous  , and therefore have low disper-sal ability. These species are known as ‘  per pedes  colonisers’ ( Brandmayr, 2005 ). Species dispersal ability can give informa-tion about the ability of carabid beetles to colonise new environ-ments, the ecological succession of a habitat, and the relation between the elevation gradient and carabid dispersion ( Brandmayr, 1983; Nilsson et al.  , 1993 ). Macropterous species are better dispersers than brachypterous ones that are only ground-dwelling colonisers ( Lövei & Sunderland, 1996 ). Wing morphology therefore influences the pattern of species distribu-tions in the environment ( Gutiérrez & Menéndez, 1997 ). Brandmayr (1983) has extended the concept of ‘power of dis-persion’ from the unit of species to community, quantifying this parameter as number of species able or unable to fly. We are therefore able to identify a stable environment if the community contains a high number of brachypterous species. Body length was determined for each specimen as the dis-tance from the margin of labrum to the apex of the elitrae. Average body length for each assemblage was then calculated. Both male and female body length was measured as carabids show sex dimorphisms. The average body length of the assem-blages is therefore a result of elevation, habitat trophic availabil-ity and habitat disturbance ( Sustek, 1987; Blake et al.  , 1994 ). Climatic harshness at high altitudes affects the length of larval development ( Sota, 1996 ). Assemblages composed by species with overwintering larvae on average produce larger-bodied adults as a result of a length of development period ( Sota, 1996;  Hodkinson, 2005 ). The relation between mean body length in each assemblage and life-cycle strategy of each species gives information about the success of colonisation of alpine habitats. Assemblages with autumn breeders are able to colonise cool habitats more successfully than those with spring breeders ( Sota, 1994 ). Phenology and average body length expressed by the assemblages could therefore provide information about the ability of species to colonise the recently deglaciated areas. The influence of independent environmental variables on species richness and on morphological features was examined by evaluating the correlation coefficients of paired variables ( Table 1 ). Separate canonical correspondence analyses were carried out using six environmental variables and five species traits in the first analysis (CCA-traits) and 22 species abun-dances in the second analysis (CCA-species). Multiple corre-lation between the two sets was calculated in both analyses. The correlation coefficients and CCA were performed using MATLAB ©  7.0 and canoco version 4.0. Fig. 1. Geographical location of the sampling sites.   Carabid assemblages along an alpine glacier foreland 685   © 2007 The AuthorsJournal compilation © 2007 The Royal Entomological Society,  Ecological Entomology , 32 , 682–689  Results Of the 75 pitfall traps, 567 carabid beetles were collected (from 22 species) ( Fig. 2 ). Species diversity in each sampling site was similar in 2004 and 2005 with a highly significant correlation coefficient be-tween years ( r    s   =  0.452, P   =  0.01). The spatial turnover in species and assemblages distribution per site was significantly higher than expected by random chance ( P   species   <  0.0001; P   communities   =  0.022) with an anti-nested pattern found ( Fig. 2 ). Environmental variables were not found to affect species richness along the valley. Highly significant relationships exist between the environmental variables ( P   <  0.0001 for each pairs of variables) ( Table 1 ). The eigenvalues and multiple correlations between traits and environmental variables and between species and environmen-tal variables are given in Table 2 . The test of significance of correlation between environmental variables and traits in CCA-traits analysis gave an F   -value of 5.65 ( P   =  0.005), and in CCA-species analysis gave an F   -value of 6.89 ( P   =  0.005). Species with reduced wing size showed a negative correlation to stony cover and altitude, but were positively related to soil age, vegetation cover, and distance from the glacier ( Table 1 and Fig. 3 ). Environmental variables had different influences on both male and female body length. Male length was positively related to stony cover and negatively linked to vegetation cover, soil age, and distance from the glacier. Female length, however, was correlated only to male length, with environmental varia-bles having a marginal influence ( Table 1 and Fig. 3 ). The influence of environmental variables on species distribu-tion ( Fig. 4 ) show that the species highly influenced by the envi-ronmental features are: Oreonebria castanea  linked to stony – dry sites at high altitude,  Nebria jockischi  and Sinechostictus doderoi  linked to stony – wet sites at high altitude sites. Ocy-dromus incognitus  , Cymindis vaporariorum  ,  Amara quenseli  , Orinocarabus silvestris  , and Carabus depressus  were associ-ated with stony and wet sites on the intermediate part of the valley, while Calathus melanocephalus  ,  Bradycellus caucasi-cus  , Synuchus vivalis  ,  Leistus nitidus  , Calathus micropterus  , and Pterostichus multipunctatus  were associated with dry, old soils. Discussion Many papers report that abiotic variables (e.g. altitude, tempera-ture, precipitation) and habitat architecture are able to affect species richness, species distribution, and community composi-tion (e.g. McCoy, 1990; Brehm et al.  , 2003; Sanders et al.  , 2003; Sinclair et al.  , 2003; Eyre et al.  , 2005; Hodkinson, 2005 ). On the Forni foreland species richness does not seem to be influenced by such environmental variables. Small differences between neighbouring habitats, like in proglacial areas, do not produce significant changes in carabid richness. This result sug-gests that carabid beetle richness is not influenced by habitat architecture but is more sensitive to habitat management ( Dauber et al.  , 2005 ). In the high alpine environment, where few (or no) anthropogenic activities occur, species richness is not thought to be a good indicator to describe differences in assemblages or to describe the relation between assemblages and environmental variables at each site ( Table 1, Fig. 3 ). The absence of a species richness trend, but the presence of species turnover along the chronosequence led us to investigate other species traits with the potential to react to habitat differentiation ( Figs 2 and 4 ). In summary, only terrain age and habitat architecture appear to be related to species morphological traits, as previously reported in Brandmayr (1983) and Sota (1996) . As already hypothesised in Gobbi et al.  (2006a) , wing morphology appears to be related to environmental architecture. In the youngest parts of the Forni Valley, low brachypterous numbers characterise the stony, loose, and unstable recently deglaciated soils. At older, drier sites, with Table 1. Correlation coefficients and probabilities of obtaining these correlation values by chance ( italic  ). Environmental variablesSpecies variables AltitudeDistanceAgeStony coverVegetation coverDrynessMale lengthFemale lengthMacropterousBrachypterousSpecies richness Environmental variables Altitude1.000 –  0.920 –  0.8640.505 –  0.9300.0740.172 –  0.005 –  0.315 –  0.369 –  0.226 Distance  0.000 10.974 –  0.7470.9910.098 –  0.343 –  0.1180.2960.4430.109 Age  0.000 0.000 1 –  0.7270.9390.075 –  0.315 –  0.1080.3370.4490.085 Stony cover  0.000 0.000 0.000 1 –  0.741 –  0.3400.4190.127 –  0.068 –  0.4150.027 Vegetation cover  0.000 0.000 0.000 0.005 10.090 –  0.359 –  0.1210.2920.4530.141 Dryness  0.531 0.404 0.523 0.001 0.011 1 –  0.1340.018 –  0.3180.026 –  0.209 Species variables Male length  0.140 0.003 0.006 0.000 0.002 0.251 10.6356 –  0.2505 –  0.194 –  0.2753 Female length  0.968 0.312   0.356 0.279 0.301 0.879 0.000 1 –  0.2995 –  0.1765 –  0.2871 Macropterous  0.006 0.010 0.003 0.561 0.011 0.005 0.374 0.964 10.20930.6701 Brachypterous  0.001 0.000 0.000 0.000 0.000 0.827 0.510 0.109 0.072 10.8661 Species richness  0.052 0.351 0.471   0.817 0.226 0.073 0.036 0.001 0.034 0.000 1   686  Mauro Gobbi  et al. © 2007 The AuthorsJournal compilation © 2007 The Royal Entomological Society,  Ecological Entomology , 32 , 682–689 a greater percentage of vegetation cover, high numbers of low dispersal species were found. Therefore it can be asserted that brachypterous species (  per pedes  colonisers) are linked to old, dry, and mature soils. Brandmayr (1991) found that brachypter-ous species show minimum values at the first stage of a habitat succession, and medium values occur as open land begins to be colonised by trees and shrubs. Very high values are then associ-ated with the climax environment. Wing morphology analyses provide information about dispersal powers of species. On the basis of data from the present study, it can be concluded that in a recently deglaciated area, carabid colonisation is determined by habitat evolution. Therefore, the ability of carabids to colo-nise newly exposed land is linked to habitat age and architec-ture, not to the presence of functional wings. Habitat structure is the more deterministic factor controlling mean body length of the carabids assemblages. In the Forni foreland, assemblages with the longest specimens are situated at the highest altitude where stony cover is high. Males are more sensitive to habitat structure and changes than females, which are associated only with male body length. The longest species, found at stony sites are Oreonebria castanea  ,  Nebria jockischi  and C arabus silvestris.  According to Sota (1996) , autumn breed- ers with overwintering larvae such as  Nebria  spp. have the greatest colonisation success in alpine habitats. Moreover, it is known that several small-sized or riverside-preferring carabids live mostly on fine soils (silt, clay), whereas middle-sized ground beetle taxa prevail in montane soils, debris and screes that offer larger shelters also. Further research is perhaps needed to obtain more insight in the relationships between carabid size and soil granulometry. Environmental analyses ( Fig. 4 ) show the connections between these species and the stony substrates. Stony cover and substrate moisture appear to be the main varia-bles driving species distribution. Altitude has a strong influence on Oreonebria castanea  and  Nebria jockischi. Nebria  spp. are micro-term  species living near glaciated environments (Brandmayr & Zetto Brandmayr, 1988). It is interesting to note these species do not coexist in the same habitat and therefore it can be supposed that high niche segregation exists ( Gereben, 1995; Kaufmann, 2001;  Kaufmann & Juen, 2002 ). Oreonebria castanea  have been found at    O  r  e  o  n  e   b  r   i  a  c  a  s   t  a  n  e  a   S   i  n  e  c   h  o  s   t   i  c   h  u  s   d  o   d  e  r  o   i   N  e   b  r   i  a   j  o  c   k   i  s  c   h   i   O  r   i  n  o  c  a  r  a   b  u  s  s   i   l  v  e  s   t  r   i  s   A  m  a  r  a  q  u  e  n  s  e   l   i   P  r   i  n  c   i   d   i  u  m   b   i  p  u  n  c   t  a  u  m   O  c  y   d  r  o  m  u  s   i  n  c  o  g  n   i   t  u  s   C  a  r  a   b  u  s   d  e  p  r  e  s  s  u  s   N  o   t   i  o  p   h   i   l  u  s  a  q  u  a   t   i  c  u  s   C  y  m   i  n   d   i  s  v  a  p  o  r  a  r   i  o  r  u  m   N  e   b  r   i  a  p   i  c   i  c  o  r  n   i  s   A  m  a  r  a  p  r  a  e   t  e  r  m   i  s  s  a   A  m  a  r  a  e  r  r  a   t   i  c  a   P   t  e  r  o  s   t   i  c   h  u  s  u  n  c   t  u   l  a   t  u  s   C  a   l  a   t   h  u  s  m  e   l  a  n  o  c  e  p   h  a   l  u  s   N  e   b  r   i  a  r  u   f  e  s  c  e  n  s   P   t  e  r  o  s   t   i  c   h  u  s  m  u   l   t   i  p  u  n  c   t  a   t  u  s   C  a   l  a   t   h  u  s  m   i  c  r  o  p   t  e  r  u  s   L  e   i  s   t  u  s  n   i   t   i   d  u  s   S  y  n  u  c   h  u  s  v   i  v  a   l   i  s   C  y  c   h  r  u  s  c  a  r  a   b  o   i   d  e  s   B  r  a   d  y  c  e   l   l  u  s  c  a  u  c  a  s   i  c  u  s    A  g  e  o   f   d  e  g   l  a  c   i  a   t   i  o  n   (  y  r   )   S  a  m  p   l   i  n  g  s   t  a   t   i  o  n  s noneHnoneILG2455F125F251E152E262D178C261D279C1101B1100B2155A1154A2 wing statusbmmbmmmbbmmmmbbmbbmbmb Species typical of glacial habitats:glacier front with unstable soils,debris-covered glacier andunvegetated soils.Species typical of alpineprairie near the tree line.Species typical of alpine pastured prairie andtypical of forest with sparce trees(  Rhododendro-Vaccinietum cembretosum ).   Fig. 2. Species distribution across the Forni glacier foreland. Sampling sites are ordered following age of deglaciation; species are ordered according to the first axis obtained from a correspondence analysis ( Leibold & Mikkelson, 2002 ). Wing status is indicated for each species: b, brachypterous; m, macropterous. Table 2. Eigenvalues and multiple correlations between environmen-tal variables and traits (CCA-traits, above) and between environmental variables and species (CCA-species, below). Axis 1Axis 2Axis 3Axis 4 Traits  Eigenvalues0.2770.0510.0040 Traits – environment correlations0.6910.3960.2080.219 Species  Eigenvalues0.1510.1040.0610.046 Species – environment correlations0.7990.7530.7570.834
Recommended
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x