Version classiqueVersion mobile

Dynamique et gestion des pelouses calcaires de Haute-Normandie

 | 
Thierry Dutoit

Troisième partie. Biodiversité des communautés (phytocénose, banque de graines, pédofaune)/Biodiversity of communities (phytocenosis, seed hanks, pedofauna)

Chapitre troisième. Managing chalk grassland for biological conservation: the case of soil macrofauna communities

Gérer les pelouses calcicoles pour la conservation biologique: le cas des communautés de macrofaune du sol.

Résumé

L'étude de la macrofaune du sol au cours d'une succession secondaire potentielle a été réalisée dans différentes parcelles de pelouses calcicoles après abandon du pâturage. Les résultats montrent une évolution contrastée des différentes populations en rapport avec l'apparition d'une couverture arborée et des changements dans la qualité des litières. Dans les parcelles à végétation herbacée, les espèces épigées dominent contrairement aux parcelles boisées. Si la richesse taxonomique est maximum dans la pelouse en exclos, chaque parcelle demeure cependant caractérisée par la présence de certains groupes taxonomiques. En conséquence, les systèmes de gestion conservatoire mis en place pour la conservation biologique de pelouses calcicoles riches en plantes et en invertébrés doivent également préserver la diversité des communautés de macrofaune du sol par la réalisation de systèmes de gestion tournants.

Texte intégral

INTRODUCTION

1Chalk grasslands have been studied for a long time because of their diversity and species richness. In north western Europe, these ecosystem result from human forest clearing and have been used as sheep walks for many centuries (Tansley, 1939). Since the abandonment of agricultural utilisation in the 1950's, natural succession leads to scrub and woodland extension (Smith, 1980) and/or species-poor coarse grasslands (Willems & Bobbink, 1990). Species diversity and richness of ancient calcicolous grassland are very vulnerable to a lack of grazing (Bobbink & Willems, 1987), either by domestic or wild animals (Thomas, 1962). In consequence, the aims of management plans for chalk grassland ecosystem are generally to remove vegetation in order to promote the diversified flora of grazed stands (Sutherland & Hill, 1995). The management systems which can achieve these objectives have been discussed in general terms by Wells (1965) and are now applied in chalk grasslands of north western Europe (Maubert & Dutoit, 1995).

2Impacts of various management regimes (grazing, mowing, abandonment, etc.) especially on plants communities (Wells, 1969a) and epigeous invertebrates (Morris, 1969) have been studied to orientate management plans. Nevertheless, in soil calcareous ecosystems, despite the studies of Hodda and Wanless (1994a,b) and Wasilewska (1994) about nematodes, very little is known about the richness and diversity of soil invertebrate communities and their evolution along secondary plant successions after grazing abandonment (Decaëns et al., 1995).

3Dynamic of soil macro fauna communities under different management regimes is as important as the knowledge of the evolution of plants or epigeous invertebrates (Solbrig, 1991). Further more than their contribution to taxonomic richness of chalk grasslands, soil macrofauna communities could be considered as key species (System Processes) as regards to they role in the soil system (Bond, 1994). They take part on the formation of a highly active mull by improving microbial activity (Scheu, 1990) and on the functioning of nutrient cycling (Stork & Eggletton, 1992). In ecosystem functioning, some taxonomic groups (earthworms) could play an important role on vegetation dynamics by transport selectively some seeds of the seed bank (Willems & Huijsman, 1994) or enhance seedlings in their worm casts (Thompson et al, 1994).

4In this study, attention was paid to the evolution of the biomass and density of soil macrofauna communities along a potential secondary succession which occurs in some chalk grasslands of Uppcr-Normandy (France). Evenness, diversity and species richness have been assessed as soil macrofauna community descriptors.

STUDY SITES

5Three sites have been chosen to form a pattern of samples of ecological factors and management which characterized chalk grassland ecosystems in this region. These sites represent a set of 6 plots with different management as described below. Nomenclature follows Tutin et al (1964-1980).

Site 1. Val de Lescure, Amfreville-la-Mivoie (1°7'42"E, 49°25'5"N)

6Plot 1 (Grazed Grassland) corresponds to an enclosed pasture (6.5 ha) which is heavily grazed by sheep and horses (1.5 UGB/ha). The vegetation is a very short sward dominated by small perennial herbs e.g. Festuca cf. lemanii, Carex flacca and Lotus corniculatus. Bare grounds, non palatable species and weeds indicative of overgrazing are locally abundant (Eryngium campestre, Verbascum nigrum, Cirsium acaule, Origanum vulgare).

Site 2. Roches de Saint-Adrien, Belbeuf (1°7'30"E, 49°22'22"N)

7Plot 2 (Ungrazed Grassland) corresponds to a fenced area (0.7 ha) where sheep grazing (4 sheep/ha/year) were abandoned in January 1992. The vegetation of this plot is dominated by Sesleria albicans, Brachypodium pinnatum and some annuals weeds which occurred in bare grounds created by sheep trampling (Medicago lupulina, Linum catharticum, etc.).

8Plot 3 (Abandoned Grassland) is immediately adjacent to the fenced enclosure. It's a grassland (0.5 ha) where sheep grazing was abandoned since 1950. The vegetation consists to a tall grassland dominated by tussock species (Brachypodium pinnatum, Sesleria albicans) with encroachment of some scrub species (Rosa canina, Cornus sanguinea, Viburnum lantana, etc.).

9Plot 4 (Scrubs) is adjacent to the tall grassland and corresponds to an uniform stand of Crataegus monogyna and Cornus sanguinea (0.5 ha) and could be assigned an age of 40 years. Viola hirta and Origanum vulgare were by far the commonest species of the herbaceous vegetation.

10Plot 5 (Maple Wood) corresponds to a woody vegetation (0.5 ha) where the tree canopy is virtually complete with dominant Acer pseudoplatanus and Fraxinus exelsior. The scrub layer is composed by Corylus avellana, Sambucus nigra and Crataegus monogyna. The herbaceous layer is dominated by Mercurialis perennis.

Site 3. Belles-Roches, Brosville (1°7’12"E, 49°9"N)

11Plot 6 (Pine Wood) consists on a woody vegetation (14.5 ha) exclusively composed with Pinus sylvestris which has invaded the grassland since the end of grazing in 1947 (Dutoit et al., 1994). Shade is not so important as in scrub community and maple wood. In consequence, an herbaceous vegetation dominated by two grass (Brachypodium pinnatum, Sesleria albicans) and one sedge (Carex flacca) can occur.

12The aim of the study was to assess changes in density and biomass of soil macrofauna under grazing management and during post-pastoral secondary successions. In order to achieve this purpose, the sites must be arranged in succession order to represent an approximate (potential) succession sequence (Fig. 1). To avoid misunderstanding, it must be emphasized that any particular site has been, or is likely to go, through all the described stages. Clearly some of the stages are succession alternatives, e.g. pine wood in site 3. In the same way, scrub communities dominated by hawthorn and dogwood is not absolutely a necessary stage in the development of maple wood in site 2. The sites dominated by woody species have been ordered solely on the basis of the time elapsed since the abandonment of sheep grazing in 1950. These ages being estimated from old aerial photographs (Dutoit et al., 1994). Similar considerations can be applied to the plots with only herbaceous vegetation. The enclosure plot in site 2 is the hypothetical starting point of post pastoral succession after one year of grazing abandonment which continued on the grassland of site 1. The tall grassland in site 2 corresponds to an abandoned grassland since 1950 (Liger, 1952).

MATERIALS AND METHODS

Sampling of soil macrofauna

13Every two months, from March 1994 to January 1995, 5 soil samples of 25x25x10 cm were taken in each plot along a transect which origin and direction were randomly chosen. Two methods were used to sampling soil macrofauna:

  1. Hand sorting, following the method recommended by the Tropical Soil Biology and Fertility program (Anderson & Ingram, 1993). Soil monoliths were dung with a spade and divided in 3 strata (i.e. litter, 0-5 and 5-10 cm). Each stratum was carefully hand sorted in a tray.

  2. Earthworms from deeper soil layers were extracted with formalin. 2x2.5 1 of 0.5% formalin were applied to the soil (Baker & Lee, 1993). After 10 minutes, earthworms were collected and the soil was rapidly dung out up to 10 cm and hand sorted.

Fig. 1: Schematic representation of potential secondary successions occuring in chalk grasslands in Upper-Normandy

14All invertebrates were killed and kept in 75° alcohol except earthworms which were previously fixed in 5% formalin during 3 days. Herbaceous and moss cover were sampled on 50x50 cm square, litter on 25x25 cm and herbaceous roots with a 10x12 cm cylinder. Samples were dried at 75°C during 24 hours before counting and weigthing.

Data processing

15Bouché (1972) was used as identification aid for Lumbricidae, Perrier (1927-1940), Jones (1990) and Paulian (1990) for others groups (Appendix). Richness, diversity and evenness of each sampled were calculated using Shannon index. The difference in mean density, biomass and diversity index were statistically tested by the parametric Fisher test. Two multivariate analysis were performed on Mac Mul and Graph Mu software (Thioulouse, 1990):

  1. A Principal Component Analysis, in order to identify the main gradients in vegetation and soil characteristics. In this view, 22 variables were measured but only 11 have been kept in the PCA as mean variables which can play a role on the distribution of soil macrofauna communities (Tab. 1).

  2. A Correspondence analysis (CA), in order to describe the structure of invertebrates communities under the different management regimes. Density and biomass of the 13 taxonomic units identified were grouped in three ecological groups (i.e. epigeic, endogeics and anecics) (Tab. 2).

Tab. 1: Vegetation and soil characteristics of the six plots (C: Carbon; O.M. : Organic Matter; CEC: Cation Exchange Capacity)

16Key of codes used in PCA and CA, for the different plots, variables and taxonomic groups are given in table 3.

Tab. 2: Macrofauna density and biomass of the six plots (standart error in brackets)

Tab. 3: List of abreviations used in the PCA and CA figures

RESULTS

1778 taxonomic units (species for earthworms, family for other groups) of soil invertebrates have been sampled during the study along the secondary succession (Appendix). Multivariate analysis on the mean environmentally factors (Fig. 2) and macrofauna communities (Fig. 3) reveal two factors which are responsible for 91.7% (PCA) and 80.7% (CA) of the total variance observed.

18The first environmental factor is identified as the effect of the vegetation structure and particularly the opposition between herbaceous cover (coor: 19.8) and tree cover (-45.5). It accounts for 75.8% of the environmental variance (PCA), for 46.7% of the variance in invertebrates communities (CA) and clearly separates herbaceous plots from woody plots (Fig. 2). In the grasslands, the absence of a tree cover which can reduced adverse micro climatic conditions occurring in the superficial soil (i.e. summer drought and extreme temperature) induce important adaptations of macrofauna communities like predominance of endogeic groups i.e. Coleoptera (Melolonthidae, Nilidulidae) and one earthworm (Apporectodea caliginosus). In contrast, the tree cover of woods and soil protection maintained both the moss and the litter layer (Tab. 1), lead to buffered humidity and temperature and permit the development of important epigeic populations i.e. Isopoda (Oniscidae) and Formicidae.

Fig. 2: PCA carried out on the vegetation and soil variables (Tab. 1) which characterize six plots of a potential succession in chalk grasslands (Upper-Normandy. France). Keys for plots and variables are given in Tab. 3. See text for the signification of groups

Fig. 3: CA carried out on soil macrofauna taxonomic groups (Tab. 2) which caracterize six plots of a potential secondary succession in chalk grasslands (Upper-Normandy, France). Keys for plots and toxanomic groups are given in Tab. 3. See text for the signification of groups.

19The second factor represents the effect of herbaceous layer and root layer which influence the quality of litter brought to the soil. It explains 15.9 % of the environmental variance, 34 % of the community variance and opposes the grazed grassland and deciduous woody plots, with high herbaceous biomass (-14.7), roots biomass (-13.9) and low C/N litter value (-9.3), to the pine wood and other non-grazed grasslands (Tab. 1). In the grazed plot, the herbaceous layer decreases under the effect of grazing (Hutchinson & King, 1980). In the scrub and maple wood this decrease is more due to the decline of light under the tree canopy. Furthermore, dead leaves or animal faeces provide a high nutrient value food substrate with low C/N (Tab. 1) which allow the growth of epigeic earthworms population (Lumbriscus castaneus).

20In the pine wood, the high C/N litter layer (Tab. 1), mostly composed of non palatable pine needles, seems to favour the development of other epigeic groups i.e. Gasteropoda (Pomatiidae) and Diplopoda (Glomeridae, Iulidae).

21Along the succession, relative contribution of endogeic invertebrates is important in the first plot (78 % of total biomass) and decreased during reforestation which is marked by the appearance of epigeic groups such as epigeic earthworms (25 to 28 % of total biomass for the woody plots).

22In the grazed plot, the quality of the litter brought to the soil allows the maintenance of macrofauna population with a medium biomass (100.5 g/m2) and high density (1259 in/m2). Contribution to the biomass of anecic ant populations (70%) and endogeic larva of Coleoptera (6%) are particularly important in this plot. The highest biomass (140 g/m2) and density (1365 ind/m2) is observed in the ungrazed grassland while the lower biomass (114.1 g/m2) and density (839 ind/m2) is observed in the abandoned grassland. In the woody plots (scrubs and maple wood) the epigeic fauna of the litter is mostly composed of little arthropods. This could explain the low biomass (84.9 & 81.7 g/m2) and the high density (968 & 1159 ind/m2) respectively observed in scrubs and maple wood. The macrofauna of the pine wood has the lowest density (421 ind/m2) and biomass (70.2 g/m2).

DISCUSSION

23The evolution of macrofauna taxonomic diversity and richness between the different plots is under the influence of environmental factors which influence the composition and structure of soil macrofauna communities along secondary plant successions of chalk grasslands.

24The lowest value calculated for the diversity and evenness of the grazed plot (Fig. 4) could be linked with the spatial and temporal heterogeneity of the litter which is brought to the soil in the form of animals dung and faeces (Hirschberger & Bauer, 1994). Abandonment of grazing leads to a rapid development of the herbaceous vegetation and to the accumulation of a litter layer which protect the soil from drought and extreme temperature (Scheu, 1990). Colonisation by tussock grasses (B. pinnatum) is heterogeneous and leads to a mosaic of different micro habitats (i.e. open and tall vegetation areas). The spatial variability of plant cover is assumed to increase the variability of others ecosystem members and can be responsible for the high taxonomic richness (47 T.U) observed in this stage (Babel et al., 1992). The decrease in taxonomic richness (41 T.U) and the increase of species diversity in the abandoned plot should result from the domination of a dense herbaceous vegetation with a high C/N litter layer (Scheu, 1992; Bobbink & Willems, 1993). This vegetation reduced micro habitats and brought spatially a more homogeneous litter.

25Reforestation of the grasslands by deciduous species leads to a deep modification of the macrofauna communities composition and structure. Woody environment presents characteristics that favour the development of epigeic macrofauna, low light density, buffered micro-climatic conditions and a thick layer providing both a new habitat with a low C/N food supply. A general increase in evenness, species richness and species diversity can be observed from the scrub to the maple wood plots (Fig. 4). At the opposite, in the pine wood, the very low nutrient value of the litter brought to the soil in the form of pine needle lets down species richness (34 T.U.). However, the spatial and temporal homogeneity of the litter enhances the species diversity which shows here its highest value (2.78).

26Certain taxonomic units are confined to one or two plots i.e. Dendrobaena mammalis is only find in the pine wood. Symphita in the grazed grassland; Allelulidae and Cantharidae in the ungrazed plot; Seegestriidae in the abandoned grassland (appendix). Some of them like the spider Atypus affinis, which where found in the scrubs plot, could be considered of nature conservation interest as regards to its habitat strictly confined to chalk grasslands in north western Europe (Jones, 1983). The presence of the calcicolous earthworm Allobophora muldali restricted to scrub and maple wood can also be noted for its autecological significance i.e. neutrophilous species (Bouché, 1972).

27The structure of soil macrofauna communities depends more of the quality of food supply than processes of predation or competition (Watkin & Wheeler, 1966). This is due to the extreme specialisation for feeding developed by macrofauna populations which permits the coexistence of numerous populations (Zinclair, 1971). Furthermore, the heterogeneity shown by soil and litter composition along the succession should lead to surface migrations between plots when there is a competition for the same source of food (Mather & Christensen, 1992).

28In chalk grasslands, along secondary successions, soil macrofauna species are greatly influenced by the form and quality of litter brought to the soil, which is at first dependent of the nature and structure of plant communities. More than a source of food, this litter constitutes an habitat for a lot of soil macrofauna species. In these conditions, functional groups (i.e. anecic, epigeic, endogeic) or even specific groups (taxonomic) cannot survive when changes are important in food supply and habitat conditions along the succession (i.e. change in litter quality between grazed and abandoned plot, apparition of a thin moss layer under woody plots, etc.). These results must be opposed to those obtained for plant or other animals communities. In this case, changes caused by the cessation of grazing or the elimination of a predator lead out to differences in animal or plant community structure (diversity) which is at least dominated by competitive species for plant (Grime, 1979) or by a population of one specie in the case of animals communities (Barbault, 1992).

Fig. 4: Evolution of soil macrofauna richness, diversity (Shannon) and evenness, calculated from macrofauna population densities in the six plots

29Nevertheless, despite these differences, some similarities exist for biological conservation. For example, some functional units or some specific taxonomic units could only be present under one stage of the secondary succession for plants as for invertebrates (Duffey et al., 1974). Moreover, in our study, stages which highest diversity and plant richness roughly correspond to grassland where grazing have been abandoned since a few months or one year (Dutoit & Alard, 1995c). This results can be compared to those of Bara and Vanderheyden (1985) which have reached the same observations for the distribution of vagiles arthropods communities along a secondary succession sere on a chalk grassland in Belgium.

CONCLUSION

30Rotational management systems which have been used for the maintenance of species-rich plant communities (Wells, 1980) or epigeous invertebrate communities (Morris, 1971b) could also be advised for the conservation of species-rich soil macrofauna communities. Moreover than their contribution to the general taxonomic biodiversity of chalk grassland ecosystems, soil macrofauna communities must be also considered as functional groups to understand their key-role in soil and vegetation processes (Briones et al., 1995).

Appendix: List of the taxonomic units collected in the soils of the six plots during the sampling period

Appendix: List of the taxonomic units collected in the soils of the six plots during the sampling period

Table des illustrations

Légende Fig. 1: Schematic representation of potential secondary successions occuring in chalk grasslands in Upper-Normandy
URL http://books.openedition.org/purh/docannexe/image/8098/img-1.jpg
Fichier image/jpeg, 40k
Légende Tab. 1: Vegetation and soil characteristics of the six plots (C: Carbon; O.M. : Organic Matter; CEC: Cation Exchange Capacity)
URL http://books.openedition.org/purh/docannexe/image/8098/img-2.jpg
Fichier image/jpeg, 52k
Légende Tab. 2: Macrofauna density and biomass of the six plots (standart error in brackets)
URL http://books.openedition.org/purh/docannexe/image/8098/img-3.jpg
Fichier image/jpeg, 92k
Légende Tab. 3: List of abreviations used in the PCA and CA figures
URL http://books.openedition.org/purh/docannexe/image/8098/img-4.jpg
Fichier image/jpeg, 76k
Légende Fig. 2: PCA carried out on the vegetation and soil variables (Tab. 1) which characterize six plots of a potential succession in chalk grasslands (Upper-Normandy. France). Keys for plots and variables are given in Tab. 3. See text for the signification of groups
URL http://books.openedition.org/purh/docannexe/image/8098/img-5.jpg
Fichier image/jpeg, 40k
Légende Fig. 3: CA carried out on soil macrofauna taxonomic groups (Tab. 2) which caracterize six plots of a potential secondary succession in chalk grasslands (Upper-Normandy, France). Keys for plots and toxanomic groups are given in Tab. 3. See text for the signification of groups.
URL http://books.openedition.org/purh/docannexe/image/8098/img-6.jpg
Fichier image/jpeg, 36k
Légende Fig. 4: Evolution of soil macrofauna richness, diversity (Shannon) and evenness, calculated from macrofauna population densities in the six plots
URL http://books.openedition.org/purh/docannexe/image/8098/img-7.jpg
Fichier image/jpeg, 48k
Titre Appendix: List of the taxonomic units collected in the soils of the six plots during the sampling period
URL http://books.openedition.org/purh/docannexe/image/8098/img-8.jpg
Fichier image/jpeg, 64k
URL http://books.openedition.org/purh/docannexe/image/8098/img-9.jpg
Fichier image/jpeg, 37k

© Presses universitaires de Rouen et du Havre, 1996

Conditions d’utilisation : http://www.openedition.org/6540

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search