SEARCH

SEARCH BY CITATION

Introduction

  1. Top of page
  2. Introduction
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

In undisturbed environments, Sahelian rodent population dynamics are determined by climatic (Hubert & Adam, 1985) and edaphic factors (Hubert, Leprun & Poulet, 1977). Demographic change is therefore relatively predictable (Hubert, 1982; Sicard, Diarra & Cooper, 1999). Such situations are rare in the Sahel, notably, because of the high incidence of natural or anthropic disturbances. The impact of such disturbances on rodent population dynamics has been confirmed by various workers (Duplantier, 1998; Granjon et al., 2005; Papillon, Godron & Delattre, 2006). However, to develop rodent control strategies, allowance must be made for the effects of disturbances on the spatial distribution of rodent communities in different habitats and these effects must be measured in the short and medium term.

This study investigates spatial distribution variations of a rodent community over four seasons (wet and dry) following burning and correlates these with population variations (analysed and described elsewhere; Papillon et al., 2006).

Materials and methods

  1. Top of page
  2. Introduction
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

The study site (Fig. 1), climate and rodent population dynamics have been described in detail elsewhere (Papillon et al., 2006). Nine landscape elements (LE) were identified and classified on the basis of three descriptors: (i) vegetation, defined by the dominant forest species, whether mixed or not with crops; (ii) soil typology and rooting conditions: pedo 1 – with mottled ferruginous, more or less hydromorphic concretions and good rooting conditions; pedo 2 – with mottled, lessivated of tropical ferruginous soil and moderate rooting conditions; pedo 3 – calco-alkaline granite and poor rooting conditions and (iii) the presence of termite mounds.

image

Figure 1.  Study site experimental tracts (rodent community studies): ANA = Acacia nilotica variety adamsonii; PJM = Prosopis juliflora; PBI = Parkia biglobosa; PAS = grass covered buffer zone; ARA = Acacia radiana; PAC = Parkinsonia aculeata; PBC = Parkia biglobosa and crops; ANT = Acacia nilotica variety tomentosa; AAC = Acacia albida

Download figure to PowerPoint

About half of the area was pedo 1 type and was used for crops [Sorghum bicolour and Penissetum glaucum on Parkia biglobosa and crops (PBC) an Acacia albida (AAC) facies] or was unburnt [Acacia nilotica (ANA) or PBC].

The occurrence of termitaries for each LE was evaluated with a chi-squared test. Overall, termitaries were distributed evenly over type 1 and 2 soils (chi-squared nonsignificant) but were under-represented on type 3 soils (chi-squared highly significant: 8.33 < 0.005).

The affinity between a rodent species and each LE was evaluated by the corrected frequency of each species in each LE and estimated by the ratio (expressed as a percentage) of relative frequency of capture per LE to the relative frequency of each LE (Godron, 1968, 2005; Daget & Godron, 1982).

Results

  1. Top of page
  2. Introduction
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Figure 2 illustrates the seasonal affinities of each species: Taterillus gracilis was regularly present on the site whatever the season. It was the first species to colonize the site after burning (Wl) and displayed an immediate positive affinity for ANT and grass covered buffer zone (PAS) that lasted through all seasons. The species also regularly frequented Acacia radiana (ARA) and ANA in Wl and Parkinsonia aculeata (PAC) in D2. During colonization (Wl), the preferred LE were associated with habitats featuring Acacia: ARA, ANT and ANA. At this period, the mostly adult T. gracilis population clustered on the soils that were best for digging. Over the following seasons (Dl, W2 and D2), their distribution became more even.

image

Figure 2.  Positive and negative (regular or episodic) seasonal affinities of rodents. The strongest affinities occurring in several seasons are in white boxes and joined by lines

Download figure to PowerPoint

Nannomys sp. colonized the site from Wl and concentrated mostly in a grass-and shrub-rich corridor (PAS). Nannomys sp. deserted the site during Dl but recolonized it rapidly and massively during W2. This demographic explosion involved colonization of all LE and a very even distribution.

Tatera guineae colonized remained chronically low during the two wet seasons and fell even further during the two dry seasons. Spatial distribution was strongly clustered with Prosopis, Parkinsonia and ANA.

Mastomys erythroleucus did not colonize the plot until D1. Thereafter it spread rapidly throughout the LE (except for PBC), with a very strong affinity for PAS. It largely abandoned the plot during W2, but recolonized it during D2 preferentially repopulating PAS and ARA.

A single LE (PAS) was particularly attractive to rodents. By contrast, three LE were little frequented: AAC, PBC and PAC.

Figure 3 illustrates the apparently antagonistic relations identified among species. There are nine of them. Four concern T. gracilis, two of which are versus T. guineae, one versus M. erythroleucus and one versus Nannomys sp. Three ‘antagonistic’ relations concern T. guineae; all three versus T. gracilis. Lastly, two antagonistic relations concern M. erythroleucus versus Nannomys sp. and Nannomys sp. versus T. guineae.

image

Figure 3.  Apparently antagonistic relations among species. Solid lines join positive affinities of one species with affinities of another

Download figure to PowerPoint

Figure 4 shows the three clearest examples of potentially competitive relations for a given LE. They involve PAS, PAC and ANA. Potential competition is also expressed for ARA and PJM, respectively between T. gracilis (Wl) and M. erythroleucus (D2) and between Nannomys sp. (D2) and T. guineae (W1-W2).

image

Figure 4.  Potentially competitive relations for grass covered buffer zone, Parkinsonia aculeata and Acacia nilotica. Positive and negative affinities of one species for landscape element are joined by solid and dashed lines respectively

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Introduction
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

After burning (Wl), spatial behaviour seemed selective – the presence of one species in one LE seemed to exclude others – and the number of apparently antagonistic relations was highest during this season. The succession of species over time favoured the progressive installation of populations and allowed mechanisms leading to highly selective spatial occupation, to be put in place.

With Dl, T. gracilis and M. erythroleucus developed a loose spatial behaviour with no antagonistic relations. This can be explained by M. erythroleucus arriving late on parcels occupied for several months by T. gracilis. Despite its greater reproductive capacities, the M. erythroleucus population was less competitive. Mechanisms dissuading the occupation of its preferred LE could not therefore have been fully in place.

During W2 and in the absence of M. erythroleucus, there was a resumption of selective behaviour among T. gracilis and T. guineae (as in Wl), associated with antagonistic relations between both species. Nannomys sp. alone occupied all LE with no antagonistic relations. Its behavioural and morphological characteristics probably facilitate its being maintained in the LE where it comes least into confrontation with M. erythroleucus.

With D2, M. erythroleucus was already present and actively breeding. It rapidly became dominant and developed selective spatial behaviour on the most favourable LE.

This process is accompanied by antagonistic relations. The prolific character of M. erythroleucus supposedly explains why its population increases earlier and more swiftly. It becomes dominant only when colonization coincides with its dispersion phases.

Fox (1982) and Fox, Taylor & Thompson (2003) show how the succession of rodent species after burning matches the plant stages favourable to them. They emphasize, although, that this is not a sufficient cause to explain colonization by these species. Other criteria such as the pattern of disturbance which must be analysed with the methods developed in landscape ecology, the mobility phase of each species and selective and aggregative occupation of LE are also decisive. By taking all these criteria into account we can better understand the colonizing mechanism affecting both outbreak risks and local biodiversity.

References

  1. Top of page
  2. Introduction
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  • Daget, P. & Godron, M. (1982) Analyse de 1’écologie des espéces dans les communautes. Masson Ed, Paris.
  • Duplantier, J.M. (1998) Les petits rongeurs indicateurs des modifications du climat, des milieux et des pratiques agricoles dans la vallée du fleuve Sénégal Actes ducolloques « Eau et sante » (1994). Editions de 1’Orstom, Paris, pp. 5365.
  • Fox, B.J. (1982) Fire and mammalian secondary succession in an Australian coastal heath. Ecology 63, 13321341.
  • Fox, B.J., Taylor, J.E. & Thompson, P.T. (2003) Experimental manipulation of habitat structure: a retrogression of the small mammal succession. J. Anim. Ecol. 7, 927940.
  • Godron, M. (1968) Quelques applications de la notion de frequence en écologie végétale (recouvrement), information mutuelle entre espéce et facteurs ecologiques (échantillonnage). Oecologica Plant 3, 185212.
  • Godron, M. (2005) Ecologie et evolution du monde vivant. Ed. du Conseil international de la langue francaise, Paris.
  • Granjon, L., Cosson, J.F, Quesseveur, E. & Sicard, B. (2005) Population dynamics of Mastomys huberti (Rodentia; Muridae) in an annually flooded agricultural region of central Mali. J. Mammal. 86, 9971008.
  • Hubert, B. (1982) Dynamique de populations de Mastomys erythroleucus et de Taterillus gracilis au Sénégal. Mammalia 46, 137166.
  • Hubert, B. & Adam, F. (1985) Outbreaks of Mastomys erythroleucus and Taterillus gracilis in the Sahelo-Sudanian zone in Senegal. Acta Zool. Fennica 173, 113117.
  • Hubert, B., Leprun, J.C. & Poulet, A.R. (1977) Importance ecologique des facteurs édaphiques dans la répartition spatiale de quelques Rongeurs du Sénégal. Mammalia 45, 3559.
  • Papillon, Y., Godron, M. & Delattre, P. (2006) Changes in a Sudano-Sahelian rodent community after slash-and-burn farming (Gonsé Forest, Burkina Faso). Afr. J. Ecol. 44, 379387.
  • Sicard, B., Diarra, W. & Cooper, H.M. (1999) Ecophysiology and chronobiology applied to rodent pest management in semi-arid agricultural areas in Sub-Saharan West Africa. In: Ecologically-Based Rodent Management (Eds G.Singleton, L.Hinds, H.Leirs and Z.Zhang). ACIAR editions, Camberra, pp. 409440.