Previous Page Table of Contents Next Page


Evolutionary trends of some vegetational formations under the influence of grazing in the Guinean Savannah of the Ivory Coast

J. CESAR *

(*) J. César, Centre de Recherche Zootechnique de Minankro, B.P. 1152, Bouaké, Côte-d'Ivoire.


Outline of methods used
Selective effect of grazing and its role in the evolution of organic matter in two types of soil
Evolution of the floristic composition of 3 types of Savannah under the impact of grazing
Bibliography


SUMMARY

The "spot" method of survey made it possible to compare zones of vegetation at different levels of grazing intensity and to assess the effect of grazing on floristic composition and the soil. It was possible to demonstrate two types of evolution of the rangeland, one slow and progressive on sand/clay soil, the other abrupt with no gradual change, on the sandy substratum, which resulted in the exhaustion of the humic horizon.

The Abokoumékro ranch, located 25 km east of Yamoussokro in the Guinean climatic zone, which receives 1,200 mm rainfall, covers for the most part tropical ferruginous soil derived from a granitic substratum.

The vegetation, which is made up of savannah covered with Ronier palms (Borassus aethiopum mart.) and fairly dense bush savannah, borders on the Loudetia arundinacea sub-association described by Adjanohoun (1964). Other palatable graminaceae are frequently dominant, namely: Elymandra androphila, Schizachyrium sanguineum, Andropogon schirensis, Hyparrhenia smithiana and Hyparrhenia diplandra.

Exploitation of the southern section of the ranch began in 1963. To it was subsequently added the Central and Gofabo sections (1964), the North Yamba section (1966), and the North Morobi section (1968).

The livestock herd is exclusively composed of cattle of the N'dama race. Following the regular increase from 800 head in 1964 to 4,400 head in 1971, which was maintained until 1973, the total stock in December 1974 was 3,900 head.

The observations contained in this first note fall within the framework of a broader study which should lead to a survey of the various natural pasturelands and the state of deterioration of each. Here we shall restrict ourselves to an examination of a few floristic surveys in the various sections of the ranch.

Outline of methods used

In this first approach, only the floristic aspect was dealt with: production will be measured during a subsequent stage. The vegetation is analysed in a linear manner by the spot method (Levi and Madden, 1933; Daget and Poissonet, 1971), which had already been tested in the Ivory Coast (Poissonet, César, 1972). Since the surveys were carried out at different dates and in pasture utilised in different ways, only a vertically stratified measurement of the vegetation will permit comparison of the data obtained at various stages of growth.

While the specific contribution of each cespitose graminacea remains approximately constant throughout the annual cycle in natural savannah, the same is not true in commercial pasture where the growth and earing of the best ingested graminaceae may be halted by grazing. The floristic composition of a given formation at the peak of its development thus varies substantially, depending on whether all the vegetation or only the first two strata are considered.

Table I allows comparison of the specific contributions of two stations 20 metres apart. In the first case, a slight difference may be observed between the full (1a) and partial (1b) surveys, but the order of the principal species is not changed. However, in the second case, Loudetia arundinacea, which appears dominant in the full analysis (2a), is replaced by Schizachyrium sanguineum when the survey is limited to the first two strata, which are the only strata of interest to the grazing animals. In that case, the partial survey better reflects the true grazing potential of the formation.

Table I - Changes in the specific contribution according to the strata studied

1. Savannah with Schizachyrium sanguineum dominant.
2. Savannah with Loudetia arundinacea apparently dominant.

a) All vegetation surveyed.
b) First two strata only surveyed.

Frequency of sp. as percentage

1

2

a

b

a

b

Schizachyrium sanguineum

53.8

65.3

30.0

37.5

Loudetia arundinacea

21.5

17.4

45.0

29.2

Hyparrhenia diplandra

4.6

-

5.0

4.2

Hyperthelia dissoluta

3.1

8.7

-

-

Monocymbium ceresiiforme


-

6.7

12.5

Ctenium newtonii

1.5

-

-

-

Andropogon schirensis

7.7

-

-

-

A. ascinodis

4.6

-

-

-

Digitaria horizontalis

-

-

1.7

-

D. delicatula

-

-

1.7

-

Borreria cf. stachydea

-

-

10.0

16.7

Selective effect of grazing and its role in the evolution of organic matter in two types of soil

Study of mixed pasture with Loudetia arundinacea, Schizachyrium sanguineum and various andropogons makes it possible to assess the choice of animals and grazing. Table II shows the floristic composition of heavily grazed areas on the one hand and of clusters of rejects on the other hand. Preference for Andropogon and Hyparrhenia species over Loudetia arundinacea is clear.

Table II - Comparison of the floristic composition of grazed areas and rejects

 

Centre Zone 1

Gofabo Zone XI

Grazed areas

Rejects

Grazed areas

Rejects

Hyparrhenia diplandra

41

14

48

7

Andropogon schirensis

13

4

20

0

Schizachyrium sanguineum

36

43

12

21

Loudetia arundinacea

3

18

4

72

In Zone I (centre section) on sandy-clayey soil, no edaphic difference appears between the grazed and rejected areas. On the other hand, on sandy soil (Zone XI, Gofabo section) the gritty structure under the rejects becomes particulate and loose for the grazed species. Here grazing gives rise to local weakening of the structure. This phenomenon does not occur if the soil is sufficiently rich in clayey elements.

Moreover, Zone I (centre section) displays locally, in similar ecological conditions, adjoining exclusive populations of Loudetia arundinacea and Schizachyrium sanguineum. The sandy texture of the soil is identical in both cases, but the structure of the first horizons is markedly more resistant under Schizachyrium sanguineum. Local replacement of Schizachyrium sanguineum by Loudetia arundinacea, probably due to overgrazing, is accompanied by a weakening of the structure, and gives grounds for thinking that the higher organic matter content is essential for the retention of Schizachyrium sanguineum and unimportant for Loudetia arundinacea.

Evolution of the floristic composition of 3 types of Savannah under the impact of grazing

a) Ronier palm savannah on tropical ferruginous soil

Three surveys were carried out at the limits of the ranch on the land of Gofabo village in formations that had never been grazed. The topographical locations were plateau, mid-slope, and slope bottom. The beige-to-ochre tropical ferruginous soil displays in its first horizons a sandy texture at slope bottom, and is sandy-clayey on the remainder.

The specific contributions indicate that Loudetia arundinacea is dominant on the plateau while Schizachyrium sanguineum is dominant on the slope (Table III). The optimum for Andropogon schirensis is to be found at mid-slope, while Elymandra androphyla grows equally well on a plateau and at slope bottom.

The sequence of the main species along this natural chain is, therefore, as follows: Elymandra androphila, Loudetia arundinacea, Andropogon schirensis, Hyparrhenia diplandra, Schizachyrium sanguineum and Elymandra androphila. These results are very close to those obtained by Montard in 1964 in similar ecological conditions, although the survey methods were substantially different.

The three surveys were repeated within the ranch at about 20 metres from the previous stations, in formations enjoying exactly the same ecological conditions (topography, soil, fires) but which had for ten years been grazed by animals from the ranch. The evolution of the floristic composition is indicated in Table III.

Table III - Specific contribution of the principal graminaceae on sandy-clayey soil:

a) ungrazed savannah
b) savannah grazed over the past ten years

 

Plateau

Mid-slope

Slope bottom

a

b

a

b

a

b

Elymandra androphila

19.1

5.1

-

-

23.3

12.5

Loudetia arundinacea

36.6

39.0

8.9

12.0

11.6

20.8

Andropogon schirensis

12.2

11.9

22.2

34.0

9.3

-

Hyparrhenia diplandra

8.7

6.8

15.6

14.0

20.9

12.5

Schizachyrium sanguineum

8.7

18.6

24.4

20.0

25.6

20.8

The following may be observed:

1. The retreat of Elymandra androphila over the whole transect.

2. The disappearance of Andropogon schirensis at slope bottom.

3. The advance of Loudetia arundinacea.

4. The advance of Schizachyrium sanguineum on the plateau, while this same species seems to be retreating on the slope.

Ten years of regular grazing have produced an indisputable change in the floristic composition and, in particular, a reduction in the best ingested species.

However, the palatable species as a whole have declined in the mid-slope position. The value of the pasture has not markedly fallen, and in such conditions the evolution is slow and progressive.

Table IV - Floristic composition of pasture on red soil with concretions in 1963 (after G. Boudet) and in 1974

 

1963

1974

Plateau

Plateau

Edge of plateau

Slope bottom

Zone XII

Zone XI

Zone XII

Zone V

1

2

3

4

Plateau

High slope

Mid-slope

Plateau slope

Elymandra androphila

64.2

49.6

42.1

0

23

16

0

0

Schizachyrium sanguineum

15.4

33.3

24.4

24.0

21

8

39

12

Hyparrhenia subplamosa

0.4

1.0

2.0

15.9

4

18

11

15

Andropogon schirensis

14.1

9.3

26.1

33.9

2

8

13

26

Loudetia arundinacea

0.6

0.4

2.4

1.5

42

27

24

29

Total

94.7

93.6

97.0

75.3

92

77

87

82

b) Ronier palm savannah on red soil with ferralitic concretions

This type of soil is to be found on the Southern and Gofabo sections of the ranch. Three surveys in the plateau and slope positions were carried out in normally grazed areas (Zones XI and XII). A fourth survey on the plateau in an area transited by the cattle represents the maximum possible level of exploitation on the ranch (Zone V). The floristic composition of the 1974 coverage may be compared with that calculated from the basic coverage data of G. Boudet (1963).

Although the stations studied in 1974 are not completely identical to those studied in 1963, the decline of Elymandra androphila in favour of Loudetia arundinacea is clear. On the other hand, the other species do not seem to have advanced substantially and, in particular, Andropogon schirensis, a highly palatable species, is still present in abundance even in the most highly exploited areas. The total of the main species remains high (82 percent in the transit area), and the decline in the specific contribution of the palatable species does not exceed 15 percent.

As in the previous case, the conclusion is that the vegetation has relatively high stability.

c) Bush savannah on sandy soil

Soils with sandy surface horizons may be found in each section of the ranch, but they occupy a greater area in the Northern sections.

The station studied is at the edge of the ranch in Zone IV, near the village of Oufouédièkro. The soil displays the following characteristics:

0- 5 cm: Sandy, humiferous, somewhat mid-grey
5-10 cm: Sandy - grey-pink
2540 cm: Sandy - ochre-pink.

The natural vegetation is made up of bush savannah, with Afrormosia laxiflora and Hymenocardia acida comprising a homogeneous graminacean stratum dominated by Schizachyrium sanguineum (Table V).

Table V - Specific contribution of the principal species on sandy soil

 

Natural Savannah

Lightly grazed Savannah

Over grazed Savannah

a

b

c

Schizachyrium sanguineum

74.4

73.9

18.7

Andropogon schirensis

20.5

2.4

-

Hyparrhenia diplandra

2.6

4.8

9.9

Imperata cylindrica

2.6

7.1

-

Monocymbium ceresiiforme



-

(Necs) Staph.

-

-

5.5

Cyperus cf. Rotundas

-

-

48.3

Eragrostis turgida

-

-

12.1

Within the ranch, the herbaceous formation appears as a mosaic of lightly grazed zones and over-grazed areas (Table V, b and c). Comparison of surveys a) and b) brings out the impact of limited grazing over eight years: there is a reduction in the specific contribution of Andropogon schirensis, while Schizachyrium sanguineum remains constant.

The total of palatable graminaceae still represents 81 percent of the stock. On the other hand, the over-grazed areas now contain only 30 percent of palatable species; the disappearance of Andropogon schirensis is complete; and the whole of the station has been invaded by Cyperus rotundus Only Hyparrhenia diplandra has been able to make slight progress. Correlatively, the humiferous horizon, lighter in colour, has declined in thickness by 1 cm.

In the most heavily grazed zones, where the disappearance of savanicultural graminaceae is total and the value of the pasture is virtually nil, the humiferous horizon never exceeds 2 cm in thickness.

In contrast to what occurs on sandy-clayey soils, the transformation of the vegetation is rapid. The fall in the organic matter content of the surface horizon leads abruptly to the replacement of savanicultural graminaceae by unexploitable ruderals with no intermediate stages.

These methods of surveys by points made it possible to specify the selection of species by the animals and to demonstrate the decline in the humic horizon of sandy soils under the impact of overgrazing. The result is two different types of evolution of the pasture depending on the nature of the substratum.

The first, slow and progressive, on sandy-clayey soils, probably produces a new vegetation balance.

The second, abrupt, without intermediate stages, is connected with the rapid exhaustion of sandy soil.

The role of the soil in the ecologically closely related formations is therefore fundamental, and quantitative analysis of the vegetation can be seen to be totally inadequate to anticipate evolution under the influence of the livestock, if not accompanied by a study of the soil.

In practical terms, it would appear necessary to ensure a lower grazing level on sandy soil than on sandy-clayey soil, even if the initial biomass of the vegetation is identical in the two cases.

Furthermore, it would be advisable on the sandy areas of the Abokouamékro ranch, to monitor the humiferous horizon even when the vegetation shows no signs of deterioration.

The reconstruction of the humiferous horizon requires comprehensive protection from fire and from other disturbances from outside.

Bibliography

1. ANDJANOHOUN, E. - Végétation des savanes et des rochers découverts en Côte-d'Ivoire. 1964. (mém. ORSTOM N°7) 178 p.

2. AUDRU J. - Etude de factibilité des Ranches d'Abokouamékro et de Sipilou en République de Côte d'Ivoire. Fasc. V. Ranch d'Abokouamékro et extension. I.E.M.V.T., 1972 (Et. Agrost. n° 35), 154 p.

3. BOUDET G. - Pâturages et plantes fourragères en République de Côte-d'Ivoire. I.E.M.V.T. (Ronéot.), 1963. 102 p.

4. DAGET, Ph. et POISSONET, J. - Une méthode d'analyse phytologique des prairies. Critère d'application. Am. agron., 1971. 22 (1), 5-41.

5. LEVY, E. et MADDEN, E. - The point method of pasture analysis. New Zeaf. Jr. Agr., 1933, 46, 267-279.

6. MONTARD, F.X. de. - Rapport de Stage - année 1964. ORSTOM - Centre d'Adiopodoumé. 1964 (Ronéot.) 62 p.

7. POISSONET, J. et CESAR, Y. - Structure spécifique de la strate herbacée dans la savane a palmier ronier de Lambo (Côte-d'Ivoire). Ann. Univ. Abidjan, E, 1, 577, 601.

(La nomenclature des espèces botaniques est celle de la Flora of West Tropical Africa, éd. 2., par J. Huchinson et J.M. Dalziel, 1954 a 1972.)


Previous Page Top of Page Next Page