Version classiqueVersion mobile

Dynamique et gestion des pelouses calcaires de Haute-Normandie

 | 
Thierry Dutoit

Première partie. Contexte scientifique et cadre de l'étude/Scientific context and geographic setting

Chapitre second. The conservation of species-rich chalk grasslands in Central-Normandy: an agro-ecological and multi-level approach

Approche agro-écologique et multi-niveaux de la conservation de pelouses calcicoles diversifiées en Normandie centrale (France).

Résumé

Les facteurs écologiques qui influencent la composition botanique et la structure de la végétation des pelouses calcaires de Normandie centrale sont appréhendés grâce à une étude régionale. Les résultats des analyses multivariées montrent que: (1) Deux types d’habitats peuvent être définis à partir des données climatiques et pédologiques (mésophile et mésoxérophile). (2) Les facteurs d’exploitation ont également une influence importante. Les processus d’intensification (accroissement de la pression de pâturage) sont ainsi responsables de la perte de richesse spécifique des pelouses calcicoles la plus rapide, tandis que l’abandon du pâturage conduit à des changements plus lents. En considérant les pelouses calcicoles dans leurs contextes agricoles, nous insistons alors sur la nécessité de procéder à une approche multi-niveaux pour répondre aux objectifs de la conservation biologique de ces milieux.

Texte intégral

INTRODUCTION

1Chalk grasslands, a plagioclimax species-rich vegetation in western Europe, have been studied for several decades in many fields of research: conservation biology (Jermy & Stott, 1973; Rorison & Hunt, 1980), plant communities or population dynamics (Hillieretfl/., 1990), phytosociology (Willems, 1982b; Royer, 1985), plant strategies (Grubb, 1977; Grime, 1979, Mortimer, 1992). Most of these works have focused on a given issue at a given scale but there is a lack of a global understanding of the agro-ecological functionning of chalk grasslands.

2Indeed, the dynamics of chalk grasslands, over the last decades, can be considered as the result of many changes which have occured simultaneously or not, at different scales in time or space. For example, land use and landscape changes, agricultural intensification or abandonment (Green, 1990), air pollution or isolation processes (Bobbink & Willems, 1987) are often mentioned as possible influences. But the hierarchization of ecological factors influencing the plant communities dynamics, the importance of temporal versus spatial changing patterns of land use or management are largely unknown. In the same way, the relationships between the chalk grasslands and the agricultural systems are not studied even in the countries where the ecology of these ecosystems is well known (e.g. U.K. and NL).

3Thus, several questions remain about the most important driving factors on chalk grasslands evolution and, consequently, the priorities of intervention for the conservation or the restoration of these ecosystems. New approaches are needed, which require considerations from conservation biology as well as from landscape ecology and agro-ecology (Baudry et al, 1993). This appears to be an important issue in France and especially in Normandy, where the NGOs have not yet became an alternative driving force for biological conservation as in U.K., but where the divorce between agriculture and nature conservation is not engaged so far as it is in the Netherlands and in Belgium. New questions come into view, from this specific situation: What kind of farms actually manage chalk grasslands? What are the effects of this management? What can be the respective roles of agriculture and NGOs (Non Governmental Organisations) in the protection and management of chalk grasslands in France?

4These questions have framed our research about the conservation and the restoration of these species-rich ecosystems in Normandy. On the basis of a regional survey carried out in central Normandy (Alard, 1990), we have tried to assess the main agro-ecological factors influencing chalk grassland vegetation in this region, and their role in plant richness and diversity. Contrasted situations were studied in order to determine the possible effect of surrounding changes or agricultural context on the plant communities. Finally, a global perspective was roughed out with a view to integrate the future of chalk grasslands within agricultural development as well as nature protection regional plans.

MATERIELS AND METHODS

Study area

5The central Normandy (Fig. 1a&b) is a contrasted agricultural region showing many gradients, as far as the climate, the agriculture, the landscape are concerned. From the north western part which is a grassy bocage managed by small-size farms (52% have less than 20 ha) to the south est, mostly dominated by ploughed openfield and big farms (a mean size of 60 ha), a wide variety of situations makes this region a good one for monitoring the effects of contrasted evolution on the rural landscape and its biodiversity (Alard et al, 1991, 1994). Chalk grasslands occur in central Normandy in the main valley, mostly north-south oriented, and in a wide range of situations (management, agriculture and landscape contexts), which strengthen the test role of this region for nature conservation issues.

Methods

6During 4 successive years (1987-1991), plant “relevés” were made on the widest range of plant communities (grasslands and scrubs) occuring on chalk soils in order to sample the main situations and vegetation types. For each relevé, botanical, environmental and management data were collected. Plant communities were studied by the mean of qualitative and quantitative data: relative cover-abundance for vascular plants occuring in about 20 square meters, plant richness (R), diversity (FF) and evenness (E) were assessed. Field data (slope, exposure, soil depth) were collected for the habitat characterization. Management variables (stocking rate, fertilization rate) were recorded for the same plots when available.

Fig. 1: (1A) Map showing the study area. (1B) Dark area represent grassland contribution in farmland in 1989 (per administrative canton)

7When the chalk grassland was managed, which was always in the context of a farm, the farm structure (size, grassland to ploughed land ratio,...) was recorded. The landscape context of the chalk grassland was described, according to an anthropization gradient (semi-urban, ploughed, grazed or wooded contexts).

Data analysis

8Vegetation cover abundance data were arranged in a general matrix to perform a Factorial Correspondance Analysis (FCA) with a view to identifying the main gradients of species composition and to correlate them with agro-ecological factors (Escofier & Pages, 1990). A complementary classification (HAC) was then carried out using plant and relevés similarities calculated from the correspondence analysis. Multivariate analysis were performed with the ANAPHYTO software, at the University of Paris XI.

9The habitat and ecological requirements of the plant species were assessed following their phytosociological statute (Royer, 1985, 1987) and the autecological data available (Grime et al, 1988). This information was used as an help for the interpretation of both floristical and ecological analysis.

RESULTS

Chalk grassland plant communities: ordination and classification

10A 94 relevés x 130 species matrix was used for the correspondance analysis, after the elimination of species occuring in 5 relevés or less, inorder to reduce the possible interference of statistically rare species in the analysis (Orloci, 1978). The axis 1 and 2 plan of the FCA (Fig. 2) exhibit a parabolic cloud, a particular figure dispersion known as the Guttmann effect (Bachacou, 1973) which is the result of the non indépendance of the axis of the plan. It indicates (Legendre & Legendre, 1979) a dominant gradient in species composition which is the reflect of a determining factor influencing the plant communities. This gradient, materialized by the dispersion, opposes CSR and CR (Poa trivialis, Lolium perenne,...) to stress tolerant (Avenula pratensis, Campanula rotundifolia) plant species and can be interpreted as a general productivity gradient, whether this gradient is caused by management or habitat variations.

11The complementary classification (Fig. 3) determines 5 groups of relevés which are ordered along the gradient of the FCA. One of these groups is clearly isolated from the others by the presence of mesotrophic grassland plant species. This group A is characterized by the combination between typical chalk grassland species and mesotrophic grassland species. The other four groups are separated following a classical distinction between meso-xerophilous and xerophilous habitats. From a phytosociological point of view, the group A belongs to the Cynosurion alliance, while the groups B and C, in the one hand, and the groups D and E, in the other, belong respectively to the sub-alliances Eu-Mesobromenion and Seslerio-Mesobromenion (Royer, 1987).

Tab. 1: Comparison of plant richness and evenness (Shannon index) beween chalk grasslands under different management in central Normandy for mesophilous habitat (Standart error in brackets; difference with high (**), low (*) and no (0) significance). From Bance et al. 1991 & Alard et al., 1994

Fig. 2: Correspondence analysis of the floristic matrix (94x130). Key for community groups and for species (in circle) are given in Fig. 2

Synecological niche of chalk grassland plant communities

12The classification (Fig. 3) emphasizes the role of habitat and management factors in the distinction between plant communities. Floristic differences between plant communities can mostly be interpreted in terms of habitat variations, as far as slope, exposure, soil depth or climate context are concerned, or in term of management practices. Flabitat factors allow us to determine two major types of habitats in which chalk grassland can occur: meso-xerophilous habitat, mostly in the eastern part of the studied region, represented by the chalk slopes of the Seine, the Eure and the Iton rivers valleys; mesophilous habitat, recorded in the whole area, on slight slopes and marl soils.

13Management factors have mainly to do with grazing management because none of the chalk-grassland recorded were fertilized. From the classification, three levels of management can be inferred from the survey: semi-intensive, extensive management and abandonment. Semi-intensively managed chalk grasslands are segregated from the others, while the differences between abandoned and extensively managed chalk grasslands appears not be so discriminant in the definition of plant communities (Fig. 3). From the two components of grazing pressure intensity (Fig. 4), the IGP/AGP ratio seems to be determinant in the segregation between semi-intensively grazed plots (ratio above 2) and the extensive ones. This means clearly that, despite that the range of the mean annual grazing pressures (AGP in livestock unit per ha and per year) is independant of the plant communities recorded (from 0.5 to 1.5 lu.ha-1), the grazing period is more condensed in the year, with higher instantaneous grazing pressure (IGP) for Cynosurion communities. Moreover, the level of grazing intensity is clearly not correlated with the landscape context (Fig. 4) but seems to be more dependant of the farm context (Fig. 5). (See pages 60-61).

Dynamics and structures of plant communities

14Plant richness and evenness have been assessed for the plant communities, in order to record the structure of the phytocenosis. For the mesophilous habitat (Tab. 1), one can see that the highest diversity is observed in the extensively managed chalk grassland, as above defined. These data (ibid.) suggest that a decrease of plant richness takes place after the intensification of management, especially when fertilisation rates increase, while plant evenness changes appear to be not significant in this case. At the opposite, abandonment is followed by a decrease of both plant richness and evenness. Despite these changes, abandoned and extensively managed chalk grassland don’t exhibit major floristic differences because they are not segregated in the plant communities classification. This is not the case in the intensification process.

DISCUSSION

15Data about chalk grassland in central Normandy allow us to consider the respective role of habitat and management factors on the plant communities. These two sets of factors occur at different scale: while habitat gradients are mostly spatial ones in rural landscape (soil catena, micro-climate), management gradients can also occur in time, as in the case of changes in agriculture practices.

16The plant communities classification illustrates the influence of such gradients on the botanical composition in the one hand, and on the vegetation structure in the other (Tab. 1). Clearly, agriculture intensification induces major changes in chalk grassland plant composition while agricultural abandonment influences more the vegetation structure (Alard et al, 1994). If agricultural changes are, as a whole (i.e. intensification and abandonment), responsible of the loss of plant richness and diversity in grassland ecosystems (Green, 1990; Hopkins, 1991; Garcia, 1992) our data suggest that the processes involved in such changes have not the same importance and consequences. Indeed, if in both cases, plant competition processes can explain species dominance, exclusion or extinction in the plant communities (Grime, 1979; Tilman, 1988), the driving factors of these competitive interactions are relevant to different set of perturbation or stress (grazing, trampling, nutrient level,...) (Bobbink, 1991; Plantureux et al., 1989; Tilman & Wedin, 1991; Aerts & Van Der Peijl, 1993).

17Agricultural practices and their evolution are obviously under the dépendance of the farming systems which occur in the rural area (Baudry, 1989; Baudry et al., 1993; Alard et al, 1991; 1994).

18The management of the chalk grasslands in central Normandy seems to be correlated mainly with the farm structures, independently of the landscape context. This means that extensively managed chalk grasslands are recorded mainly in small farms involved in permanent grasslands (Alard et al, 1994), and that this situation can occur, at the regional level, in bocage grassy landscape as well as in ploughed contexts. This illustrates the fact that management at the scale of the plot is much more important than landscape context (e.g. isolation processes,...) for the definition and the ecological interest of chalk grassland plant communities (Gibson, 1986).

19Nevertheless, species-rich chalk grasslands, which are extensively managed, seem to be more or less correlated with some given farm structures. In the context of the evolution of farming systems in central Normandy for the last 50 years (i.e. the increase of the farm size and the decrease of grassland proportion in farmland), this appears to be of consequence on the persistence of such ecosystems. In the bocage context, such farms are still well represented. In this case, conservation objectives have to consider the conservation of both species-rich grassland ecosystems as well as agricultural systems. This kind of support is the aim of some EEC subsidies, delivered with a view of preserving traditional management practices (Baldock, 1991; Buxton, 1991; Dutoit & Alard, 1995d) At the opposite, conservation objectives in ploughed (or semi-urban) landscape has to deal with the scarcity of farming systems compatible with grazing extensification. In these cases, the role of NGOs is considerable, as an alternative of agricultural systems. Such development has recently been initiated in the Seine valley under the activities of the natural sites conservatory of Normandy (Alard & Frileux, 1992).

Fig. 3: Hierarchical classification of grassland plant communities and its agro-ecological interpretation

Fig. 4: Management factors as a discriminant of chalk grassland plant communities in central Normandy. AGP: Annual Grazing Pressure. IGP: Instantaneous Grazing Pressure. (∆) Cynosurion (Group A in Fig. 1). () Eu-Mesobromion in ploughed area () or bocage landscape ().

20Clearly, conservation strategies have to care about two main levels, including both ecological and socio-economical features: internal features (richness, diversity, management,...) and external ones (agricultural context). This leads to the necessity of a multi-scale approach (Fig. 6) for which agro-ecological data are required. In the same way of conservation issues, the restoration objectives lead to contrasted challenges, depending of the processes involved in plant impoverishement (Muller et al, 1995). This issue involves questions about management as well as the persistence of seed banks (Graham & Hutching, 1988a; Milberg, 1992), possibilities of plant reinvasion (Baudry et al, 1993) or soil fertility reduction (Oomes, 1990; Berendse et al., 1992).

Fig. 5: Relationship between the plot management and the farm structure for the chalk grasslands. SAU: farm size. STH/SAU: grassland proportion in farmland. Grazing pressure is synthetized by the IGP/AGP ratio (<1,5 ; 1,5<<2,5 ; >2,5).

CONCLUSION

21Changing agricultural practices represent the main threat for the conservation objectives of species-rich chalk grasslands in central Normandy. In the range of habitat variation of chalk grassland, management intensification induce the major changes in terms of botanical composition and biodiversity. If agriculture can still be an important actor in the conservation objectives, provided that an appropriate financial support is available, the NGOs will have, in the future, a more important role in rural areas dominated by intensive agricultural systems and in semi-urban contexts.

Fig. 6: Definition of protection strategies in respect of the ecological and the socio-economical context. After Alard & Frileux (1992)

Table des illustrations

Légende Fig. 1: (1A) Map showing the study area. (1B) Dark area represent grassland contribution in farmland in 1989 (per administrative canton)
URL http://books.openedition.org/purh/docannexe/image/8091/img-1.jpg
Fichier image/jpeg, 48k
URL http://books.openedition.org/purh/docannexe/image/8091/img-2.jpg
Fichier image/jpeg, 268k
Légende Tab. 1: Comparison of plant richness and evenness (Shannon index) beween chalk grasslands under different management in central Normandy for mesophilous habitat (Standart error in brackets; difference with high (**), low (*) and no (0) significance). From Bance et al. 1991 & Alard et al., 1994
URL http://books.openedition.org/purh/docannexe/image/8091/img-3.jpg
Fichier image/jpeg, 40k
Légende Fig. 2: Correspondence analysis of the floristic matrix (94x130). Key for community groups and for species (in circle) are given in Fig. 2
URL http://books.openedition.org/purh/docannexe/image/8091/img-4.jpg
Fichier image/jpeg, 148k
Légende Fig. 3: Hierarchical classification of grassland plant communities and its agro-ecological interpretation
URL http://books.openedition.org/purh/docannexe/image/8091/img-5.jpg
Fichier image/jpeg, 344k
Légende Fig. 4: Management factors as a discriminant of chalk grassland plant communities in central Normandy. AGP: Annual Grazing Pressure. IGP: Instantaneous Grazing Pressure. (∆) Cynosurion (Group A in Fig. 1). () Eu-Mesobromion in ploughed area () or bocage landscape ().
URL http://books.openedition.org/purh/docannexe/image/8091/img-6.jpg
Fichier image/jpeg, 32k
Légende Fig. 5: Relationship between the plot management and the farm structure for the chalk grasslands. SAU: farm size. STH/SAU: grassland proportion in farmland. Grazing pressure is synthetized by the IGP/AGP ratio (<1,5 ; 1,5<☼<2,5 ; ▲>2,5).
URL http://books.openedition.org/purh/docannexe/image/8091/img-7.jpg
Fichier image/jpeg, 24k
Légende Fig. 6: Definition of protection strategies in respect of the ecological and the socio-economical context. After Alard & Frileux (1992)
URL http://books.openedition.org/purh/docannexe/image/8091/img-8.jpg
Fichier image/jpeg, 298k

© 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