Oribatid mite (Acari) community structure in steppic habitats of Burgos Province, central northern Spain

This work is a study of the communities of oribatid mites in steppic areas of Burgos Province, central northern Spain, in relation to different habitat types belonging to natural or disturbed ecosystems. The primary objective was to find the relationship between anthropogenic use of land and several diversity indices. Parameters such as abundance, species richness, real diversity (H′) and equitability (J′) of the mite communities were calculated in 20 soil plots, sampled in spring and autumn 2000, belonging to the predominant habitat types of this area: (1) cultivated lands, (2) abandoned crop lands, (3) grasslands/pastures, and (4) natural oak forests or a pine plantation. The most widespread species in the area as well as the most abundant taxa overall are documented, together with the faunistic checklist of oribatid mites identified in the study. The results show a general decline in the composition of the oribatid community, in terms of species diversity and abundance, from non‐disturbed soils to cultivated lands. The number of species (richness) was usually higher in forest soils than in disturbed ecosystems, in which the most degraded soils showed the lowest number of species. The community analysis shows the differences due to habitat type and ubiquitous bioindicator species.


Introduction
Anthropogenic activity changes the original equilibrium of ecosystems by the introduction of cultivated systems as well as through other human-induced modifications. As ecosystems change from their native states to less natural ones, soil animal communities adjust their quantitative and qualitative characteristics. Some of them display the degree of this ecological alteration in terms of changes to their structure and species composition with respect to the original zoocoenoses (Migliorini et al. 2003).
Real diversity or species heterogeneity was studied in order to investigate the community composition and the relative species abundance of all biotopes (Cancela da Fonseca 1969a, 1969bBower and Zar 1977). This parameter was calculated using the Shannon-Wiener index (H9, Shannon and Weaver 1963): In which P i is the ratio between the number of individuals of species i and the total number of individuals. Relative diversity or equitability was calculated with Shannon's relative diversity index (J9) as the relation between the real diversity (H9) and the maximum diversity for each sample (Hm). Species richness (S) was also obtained.
A relative abundance index (Ab) was calculated as follows: where N i is the total number of individuals of each species and N t is the total number of all individuals in all of the sampling plots, in both spring and autumn. All the biological parameters were calculated using the program BioDiversity Pro V.2. Comparison between communities or species was performed using classification methods, which included the use of the Horn's community overlap index (Bower and Zar 1977) followed by the application of the Flexible algorithm of classification with beta520.25 (Daget 1979). NTSYSpc, version 2.01d, program was used to build the dendrograms. Absolute species abundances in spring and autumn, used in calculations, appear in Appendices 1 and 2, respectively.

Results and discussion
A total of 2543 oribatid mites (1313 in spring and 1230 in autumn) belonging to 111 species were extracted from the 40 soil samples studied. The checklist of oribatid species in the study area and their distribution according to the four studied habitat types are given in Table II. Taxa extracted from the samples belong to 43 families. There were some unidentified species belonging to the genera Topobates, Liebstadia, and Pergalumna. Of the 111 species identified in this study, 55 were new records for this area and Pergalumna formicaria was recorded for the first time in Spain. One taxon was new to the oribatid world fauna (unpublished data).
The results of the biological parameters obtained for each plot are given in Tables III and  IV for spring and autumn, respectively.
Crop management alters the abundance of mite populations in comparison with lessdisturbed systems (Edwards and Lofty 1969). The abundance values in sites belonging to habitat type 1, the cultivated soils (CC, HC, VC, AA), generally showed a lower number of specimens captured in comparison to other habitat types. These soils also showed the lowest values of diversity and species richness as would be expected in soils where the agricultural practices meant that natural plant species were replaced by unnatural ones, and the impact of tillage or the addition of agrochemicals, such as fertilizers or pesticides, strongly influenced the structure of oribatid mite communities (Lagerlö f and Andrén 1988). Mahunka and Paoletti (1984) and Tomlin and Miller (1987) also stated that oribatid mite communities have lower species richness in agro-systems than in nearby natural forest soils.
High diversity values have been found in forest soils of high-altitude plains (sites VR, HR, AR), confirming the values obtained by Iturrondobeitia and Saloñ a (1990) in mountain forest soils in the Basque Country, in places close to Burgos Province but with an Atlantic climate.
However, the crop lands heavily fertilized with an inorganic fertilizer such as AL and AT showed high values of abundance and diversity in spring, probably due to the increase in soil organic matter provided by the addition of sewage sludge. These practices are usually carried out before sowing and provide a source of energy to the microbial and microarthropod communities. This exogenous organic matter enhanced the growth of invertebrate populations in spring, the season in which a significant amount of nutrients is liberated by the mineralization of the most labile fractions.
In spring, the older abandoned crop land (E2-abandoned for about 30 years) showed a higher diversity and abundance of oribatid mites than the more recently abandoned crop land (E1) because of the successional processes. However, this result could not be evaluated in autumn as the E1 land was ploughed by the local farmers.
Parameters such as Shannon's diversity index indicate the biological organization at community level and could be considered as an expression of community structure, yielding the highest values in undisturbed soils.  In general, the species diversity of oribatid communities decreased from natural to cultivated lands, a fact that has been widely corroborated in scientific literature in recent years (e.g. Vu and Nguyen 2000).
The other parameter, equitability or relative diversity (J Shannon's), displays values from 0 to 1 and indicates the population balance in the various sites. Relative diversity values >0.8 are indicative of an optimum population balance (Daget 1979), and were mainly observed during spring in most of the biotopes. Relative diversity values establish   uniformity in the species abundance, giving the relationship between the observed diversity and the probable maximum diversity (Magurran 1988).
The equitability values obtained by J Shannon's for plots VP and AP indicated good population equilibrium. Other communities such as E1 were dominated by specialized taxa living mainly in microhabitats showing low equitability values. In this former crop land, two oribatid mite species, Zygoribatula connexa and Zygoribatula hispanica, dominated the oribatid community.
The variation in species richness for all sites and seasonal sampling periods is given in Figure 1. The loss of species in crop lands is a result of environmental degradation due to human impact, as is widely corroborated by scientific literature concerning oribatid communities (e.g. Tomlin and Miller 1987). The taxa richness (species+subspecies) of the oribatid communities decreased in the following order: from natural or plantation forests (100) to grassland and pastures (47), abandoned crop lands (38) and cultivated lands (30).
The classifications of affinities between mite communities are shown in the dendrogram of Figure 2.
The number of formed groups in the dendrogram was more than the number of habitat types. This clearly showed that it was an important overlap between communities. Some high-affinity associations were between communities of spring and autumn of the same plot. Crop and forest plots seemed to have their own identity because they formed exclusive groups, such as A (crops of barley of habitat type 1), D (natural plots of habitat type 4) or F (all forests of Quercus pyrenaica oak). The other formed groups had mixed oribatid communities: B (crops, recent abandoned crop and pasture), C (old abandoned crop, Holm oak forest and some pasture), E (wheat crop, nitrophilous grassland and marker garden) and the G group is formed by only two plots, wheat crop with inorganic fertilization and pasture with thyme and brush used for livestock farming. All these observations revealed that there was an important pool of species present in many communities. But, let us see if there were species indicating tendencies of habitat type or others. For this purpose affinity between species, based on niche overlap, was investigated. The classifications of affinities between species are shown in the dendrogram of Figure 3. There were clearly eight groups formed, arranged in three other larger ones. The most relevant explanations were the following.
The species forming group H belong to plot AA of Quercus pyrenaica, a forest whose identity differed from other forests.