Previous Page Table of Contents Next Page

Measurement of herbaceous production in the Sahelian zone


(*) J.C. Bille, O.R.S.T.O.M., P.O. Box 1386, Dakar, Senegal.

A. Measurement of standing biomass
B. Comparison of net production and biomass


Herbaceous standing biomass in the Sahelian zone is highly dependent on the climate. A correlation appears to exist between production above the ground and the amount of rainfall or the length of the period during which the soil is able to supply enough water to the vegetation. The effects of one year may carry over to the next. Maximum biomass is often erroneously confused with net production, whereas the latter is always higher: an example of how to measure net production is given and discussed. Finally, difficulties encountered in measuring Sahelian productivity are brought into focus.

Between 1969 and 1974, a study was made of the productivity of an area of Sahelian savannah which, during the period of study, received rainfall on the order of 200 mm. The area consisted of shrub savannah in northern Senegal, in which the herbaceous stratum comprised mainly of Aristida mutabilis. A. funiculata, Schoenefeldia gracilis and, less frequently, Cenchrus spp., Blepharis linariifolis, Eragrostis tremula and Polycarpes linearifolia. This formation may be regarded as being very representative of the dry tropical zone.

A. Measurement of standing biomass

The growth period of the lower stratum is limited, beginning with the first rains, generally at the beginning of August. Three periods may be distinguished in the development of the biomass:

1. In August and September, the plant material remaining from the previous year disappears, the new material develops, and the biomass increases;

2. In October, the growth falls off to nothing and all the grass progressively dries up;

3. Subsequently, the dispersion of the diaspores and the falling of small pieces of leaf and stalks lead to a marked reduction in the standing biomass, first rapidly and then more slowly.

The most meaningful level of biomass is clearly the one that corresponds to the maximum attained; and the task of measuring it calls for a fairly wide range of controls because of the transience of this particular stage, the date of which varies, and the considerable heterogeneity of Sahelian formations.

Nevertheless, the following values were recorded for the years 1969 to 1974:


Rainfall (mm)

Maximum biomass






















The graph shows that the biomass (W) may be linked to the rainfall (R) by a linear relation, although the correlation is of little significance at the probability threshold 0.95 (r = 0.84). The relation was written W = 0.3 P-9.

In parallel, the period (N. expressed in days) was determined during which the water status of the soil permitted plant growth. During this period, real evapotranspiration was higher than or equal to 2 mm per day, and the soil reserves permitted continued soil activity on an average of ten days after the last rain. This favorable season lasted 110 days, 70 days, 55 days, 0 day, 65 days and 70 days, respectively, in the years from 1969 to 1974. The correlation with the biomass was better (r = 0.89) and the relation W = 0.9 N - 2 may be regarded as satisfactory (Fig. 1 b).

However, additional investigation showed that the lower stratum had been reduced in 1973, following the total absence of production in 1972, because of excessive depletion of the quantity of diaspores able to germinate in this environment. On the other hand, it appears that 1974 enjoyed an increase in biomass due to the exceptional presence of exigent species, particularly Chloris prieurii.

Graphique 1

Graphique 2

Consequently, it would only have been possible to determine the average biomass of this formation on the basis of a single year's measurements in 1971, or one year in six. In addition, the effects of the very exceptional year of 1971 were still discernible 2 years later, and systematic cross-effects between successive years were demonstrated in the case of production of woody species. In any attempt to make a correlation between a climatic factor and plant production, account should be taken of the years preceding the test year.

B. Comparison of net production and biomass

Even during the period of development of the plant biomass it was possible to observe the partial dying-off of certain plants or parts of plants that were directly subjected to the action of decomposants. For that reason, an analysis was derived from the Wiegert and Evans method (Lomnicki, A. et al., 1968: Modification of the Wiegert, Evans method for estimation of net primary production, Ecology 49: 147-149).

The method consists in taking from a first plot at time t (0) the dead material W (0), leaving the living material so as to allow it to grow and die naturally. At time t (1), the dead material H produced between t (0) and t (1), and the living material B are collected, while the dead material G is taken from a second plot. The term G represents the algebraic sum of the dead material at time t (0), the dead material added during the period and the dead material which has disappeared, or:

G = W (0) + H - [W (0) - W (1)]

and consequently W (1) = G - H. while the biomass is represented by the sum B + G. and the productivity by:

P = B (1) - B (0) - H (1) + W (0) - (1)

Figure 2 expresses the values of the various components of the biomass, expressed in g of dry matter per m² during the 1970 rainy season. It can be seen that the disappearance of dead material (W terms) ceased from October, but the H values were certainly underestimated at the beginning of the season, as decomposition was intense.

The values calculated for primary production during each month may therefore be regarded as minimal: 43.2 g/m² in August and September and 7.6 g/m² in October. Annual net production stands at 94 g/m² and is thus 40 percent higher than the maximum observed biomass (67 g/m²), but it is likely that this proportion is not identical from one year to another and is higher the more humid the year.

At all events, any method for estimating the off-take of plant material by animals during the humid season from biomass measurements alone (for example, comparison of biomass under pasture and biomass protected from grazing) would be quite misleading in the Sahelian zone, unless the experiment was of very short duration. In general it should be admitted that net production is unknown if only biomass measurements are available, and the behaviour of annual tropical graminaceae partially consumed during the process of growth is not known.

Previous Page Top of Page Next Page