Positive associations involving the tussock grass Panicum turgidum Forssk. in the Aïr-Ténéré Reserve, Niger

https://doi.org/10.1016/j.jaridenv.2006.07.007Get rights and content

Abstract

In the Saharan mountain range of the Aïr, Niger, the tussock-grass Panicum turgidum Forssk. was hypothesized to act as a nurse plant for a number of species. Multiple spatial positive associations involving P. turgidum were detected along an aridity gradient with an index comparing species richness inside vs. outside tussocks. Associated plants were annuals and young and/or small perennials. Comparative abiotic measurements inside tussocks and in openings showed that P. turgidum diminished seriously the stress induced by vapor pressure deficit (VPD) during the harshest periods of the day (p<0.001, randomization test). This trait identified it as a potential nurse plant. The preference for hosted plants to develop inside dead tussocks rather than in living tussocks at the driest site may indicate that the nurse plant potential of P. turgidum tussocks declined in the presence of extreme water stress. Further experimental work, especially removal experiments, should improve the understanding of the underlying mechanisms of facilitation processes in this particularly harsh environment.

Introduction

The structure of the vegetation in arid ecosystems, although variable in scale, shape, and composition, is characterized by its spatial discontinuity, with an organization in either banded, spotted, or scattered patches. The term two-phase mosaic (or ‘two-phase patterns’, Whittaker et al. (1979)) has been used by Aguiar and Sala (1999) to describe these differences between high-cover patches of vegetation and a low-cover matrix. Vegetation is concentrated in ‘sink’ patches where the resources are collected either by abiotic heterogeneity or biotic heterogeneity (due to the plants themselves). In opposition, ‘source’ patches are essentially characterized by bare soils (Aguiar and Sala, 1999).

Once a patch is colonized by a pioneer species, complex interactions may develop among plants of different species. The frequency of positive interactions has been shown to be high in harsh environments, in particular in arid environments, where it is has been identified as a driving mechanism of vegetation dynamics (Callaway, 1995; Bruno et al., 2003). These processes commonly involve nurse plants sensu Niering et al. (1963) that facilitate the establishment of other plants species beneath their canopies (Franco and Nobel, 1988). Although cited as a succession mechanism in earlier works on vegetation dynamics (Clements, 1936), positive interactions have recently been incorporated into ecological theory (Bruno et al., 2003). They are now considered to be one of the main causal factors in the organization of terrestrial and aquatic communities (Bruno et al., 2003), on the same level as negative interactions (Lortie et al., 2004).

One of the principal opinions on positive interactions is that their importance is positively correlated with the level of abiotic stress (Bertness and Callaway, 1994; Bertness, 1998; Menge, 2000; Bruno et al., 2003). However, recent studies diverge from this proposal. Brooker and Callaghan (1998) focused on the importance of disturbances in driving plant interactions to facilitation. Choler et al. (2001) also insisted that the increase in positive interactions was related to a high level of both water stress and disturbances in alpine plant communities. In contrast, some authors have shown that in conditions characterized by severe constraints, in particular in arid communities, positive interactions may become insignificant, either because of the very low biomass of potential nurse plant species (Kitzberger et al., 2000), or because of an increase in negative interactions (Maestre and Cortina, 2004), due to higher competition for water (Davis et al., 1998). This debate underlined the relevance of comparing ecological processes from different points of view, particularly in extreme environments.

On the American continents, both the Mexican and South American (semi-)arid ecosystems have been documented, such as for example, the study of the interactions between a giant Cactaceae (Carnegia gigantea (Engelm.) Brit. and Rose) and a Fabaceae (Cercidium microphyllum Rose and I.M. Johnst., Niering et al., 1963), or tussock–woody plant interactions in the Patagonian steppes (Soriano et al., 1994; Kitzberger et al., 2000). Australian arid lands have been well studied too, particularly banded vegetation in central and eastern Australia (Tongway and Ludwig, 1990; Ludwig and Tongway, 1995). In Africa, studies led to the characterization of tiger bush, i.e. banded vegetation in the Sahelian zone of Niger (e.g. Valentin and d’herbes, 1999). In contrast, the Saharan zone has not been taken into account when considering positive associations and the underlying mechanisms of plant interactions in two-phase mosaic ecosystems.

Our purpose here is to contribute to the understanding of the ‘two-phase mosaic’ spatial patterns and their effects on vegetation diversity by studying spatial associations in a plant community from a Saharan mountainous region. We hypothesized that a structuring perennial Poacae, Panicum turgidum, Forssk. may act as a potential nurse plant for a number of other plant species in the inter-mountain ecosystems of the Aïr (Republic of Niger, West Africa). In this region, this plant is the main component of vegetation patches surrounded by a bare soil matrix, in association with Acacia spp. and a restricted number of other woody species (Saadou, 2001). Our objective was to characterize the positive associations of P. turgidum with other species and their variations along a water stress gradient in a Saharan environment. Because positive associations are not necessarily due to positive interactions (but see Kikvidze et al., 2005), we discussed to what extent our results indicate the existence of facilitation processes, especially considering abiotic data on temperature and relative humidity. The contribution of this study is to improve understanding of the structure and dynamics of arid plant communities and to provide an introductory insight into facilitation processes in Saharan ecosystems.

Section snippets

Study area

The National Aïr-Ténéré Nature Reserve (RNNAT), located in the Saharan zone of northern Niger (Fig. 1), was created in 1988 (Saibou, 1988). Its arid ecosystems, provide a number of resources, traditionally used by approximately 18 000 local inhabitants of the Tuareg ethnic group (Newby, 2000). It is bordered on the West by the Aïr mountain range (1988 m a.s.l.) and on the East by the Ténéré desert. The mean annual rainfall at the station of Iferouane (see location on Fig. 1) was 58.2 mm between

Community structure

A total of 66 plant species were identified. A large proportion was found at sites T and M (44 and 45, respectively, Table 2). In contrast, only 12 species were identified at the driest site I. As a beta-diversity index, the ratio of species shared among sites is low, especially with site I (I and T: 13.6%, I and M: 16.7%). It remained relatively low between T and M (34.8%) in spite of a high species richness at both sites.

Site I also displayed the lowest number of P. turgidum tussocks, and by

Multiple positive associations in stressful conditions

As a component of the Saharan zone, the study area is constrained by high levels of abiotic stress sensu Grime et al. (1988). Ozenda (2004) described Saharan aridity as the highest in the world due to the lack of soil water availability and the important water loss due to the high temperatures and low relative humidity of the air. Accordingly, the abiotic data measured at Moungoul during the rainy season detected a peak of stress at 13:00; with a relative humidity less than 10% and a

Acknowledgments

We wish to thank Antoine Cornet and Jean-Louis Devineau for their helpful comments on earlier drafts, Ahmed Tcholli and Mahamane Saidou for their contribution with the data collecting, Deborah Taylor for linguistic revisions, and Yves Vigouroux for statistical advises. We are grateful towards the government of the Republic of Niger who gave us the possibility to work in the Natural National Reserve of the Aïr and Ténéré.

References (42)

  • R.M. Callaway et al.

    Positive interactions among alpine plants increase with stress

    Nature

    (2002)
  • F.E. Clements

    Nature and structure of the climax

    Journal of Ecology

    (1936)
  • P. Choler et al.

    Facilitation and competition on gradients in alpine plant communities

    Ecology

    (2001)
  • M.A. Davis et al.

    Competition between tree seedlings and herbaceous vegetation: support for a theory of resource supply and demand

    Journal of Ecology

    (1998)
  • A.C. Franco et al.

    Interactions between seedlings of Agave desertii and the nurse plant Hilaria rigida

    Ecology

    (1988)
  • S.J. Francks

    Facilitation in multiple life-history stages: evidence for nucleated succession in coastal dunes

    Plant Ecology

    (2003)
  • F. Giazzi

    Étude initiale. La Réserve Naturelle Nationale de l’Aïr et du Ténéré (Niger). La connaissance des éléments du milieu naturel et humain dans le cadre d’orientations pour un aménagement et une conservation durables. Analyse Descriptive

    (1996)
  • J.P. Grime et al.

    Comparative Plant Ecology—A Functional Approach to Common British Species

    (1988)
  • Z. Kikvidze et al.

    Linking patterns and processes in alpine plant communities: a global study

    Ecology

    (2005)
  • T. Kitzberger et al.

    Effects of climatic variability on facilitation of tree establishment in northern Patagonia

    Ecology

    (2000)
  • Ch.J. Lortie et al.

    Rethinking plant community theory

    Oikos

    (2004)
  • Cited by (0)

    View full text