Session 3 Genetics, Provenance Trials, and Vegetative Propagation

Session Papers

Provenance Trials

Faidherbia albida in Northern Cameroon: Provenance Trials and Crop Associations

J.M. Harmand¹ and C.F. Njiti²

Abstract

At Mouda, Cameroon (850 mm annual rainfall), a 5-year old Faidherbia albida agroforestry trial with dominant 4-m trees showed very heterogeneous growth due to ferruginous and stony soils. At this age the stand did not have a significant impact on yields of associated crops. Intraprovenance variability on this site masked any differences between six provenances from the Sudano-Sahelian zone. The Burundi (eastern Africa) provenance proved to be unadapted to the site. Plantings of F. albida in a mechanically cultivated crop system was shown to be feasible when planted at a wide and fixed spacing.

Introduction

Faidherbia albida parks in northern Cameroon, often associated with agropastoral systems, are principally found on alluvial soils north of Maroua. These naturally regenerating stands exist throughout numerous formerly cultivated areas. However, the species is rarely found in recently cleared farmlands south of the 900 mm isohyet. Studies of the effect of F. albida on associated crop systems have been carried out by the Centre de recherche forestière (CRF) in collaboration with the Institut de la recherche agronomique (IRA) since 1985 (Peltier and Eyog Matig 1988).

The objectives of the Mouda Gazal trial were to:

• determine the age and stage of development at which F. albida has a positive effect on soil fertility and yields of associated crops;

• compare tree-crop interactions in an intensive system (groundnuts, cotton, and fertilized sorghum) and a traditional extensive system (cotton and non fertilized sorghum);

• identify constraints in the tree/crop system in order to develop F. albida systems in farmlands;

• compare performances of different local and introduced provenances of the species.

Materials and Methods

Mouda is situated in the Sudano-Sahelian zone at an altitude of 500 m. Annual rainfall averages 850 mm and the potential evapotranspiration is 1832 mm. The soil, formed over gneiss and quartzite, is ferruginous and stony, with a depth of 1.70 m. The sandy clay soils at the surface become more clayey at greater depth. Soil water-holding capacity is low (35-50 mm 50 cm^-1), as is the organic matter content (1-2.5%) (Brabant 1988). Tree savannas dominated by Anogeissus leiocarpus typify uncultivated sites. Excessive drainage and a high erosivity limit the value of this soil.

The trial was installed in a 7-ha area cleared of tree savanna utilizing a split-plot design with four replications. The treatments involved three factors. Main plots were 4 cropping systems in rotation comparing improved varieties with a local crop rotation. The improved rotation treatments were composed of groundnut (var 28206), cotton, and sorghum (var S35) on 3-year rotations. Treatment 1 started with sorghum in 1985, and then groundnut in 1986, cotton in 1987, back to sorghum in 1988, etc. Treatments 2 and 3 had the same rotation, but Treatment 2 started  with groundnut and Treatment 3 with cotton in 1985. In this manner, each crop was represented every year. The traditional treatment involved local sorghum (djigari) and cotton in a 2-year rotation.

Groundnut was fertilized with 200 kg ha^-1 single super phosphate (SSP), cotton with 200 kg ha-' complete fertilizer (22-10-15), and sorghum with 50 kg ha^-1 complete and 150 kg ha^-1 KC1 at planting and 50 kg ha^-1 urea as a split dose. In the traditional cropping system, djigari sorghum was not fertilized. Subplots were F. albida provenances, which were planted on an 4 x 4 grid within the field and along the borders. Eight provenances were used (4 local and 4 ex-Cameroon, Table 1). Subplots measured 20 x 24 m. At the edge of each block, control plots (38 x 24 m) of each cropping system was included without the trees. From 1985 to 1989, crops were mechanically cultivated between tree rows. In order to avoid seedling damage, 50-cm strips were left unplanted along both sides of row—a 25% loss in arable area.

Crop production was measured annually. Soils were sampled to a depth of 40 cm on all plots and analyzed at the CIRAD soils laboratory in Mont-pellier, France. Tree heights and mortality were observed.

Results and Discussion

Growth of F. albida

In the first year, the Burundi provenance had the best growth (Table 2). Five and a half years after planting (Dec 1990), the six local and introduced provenances of the Sudano-Sahelian zone (in order of performance, Bogo, Mokyo, Mokolo, Mali, Senegal, and Burkina Faso) performed better than the Ngong and Burundi provenances. In 1990, remaining trees of the declining Burundi provenance (29% survival versus 83% from other 7 provenances) exhibited stagnated growth.

Table 1. Descriptions of Faldherbia albida provenances used in a cropping trial, Mouda Gazal, Cameroon, 1985-91.

Table 2. Height of F. albida provenances by year, Mouda Gazal, Cameroon, 1985-90. (Source: Peltier 1988; Harmand 1989).

Intraprovenance variability (CV = 47-60%) was much higher than interprovenance variability (CV = 16%), so distinctions could not be made between provenances. Further, intraprovenance variability of height was high due to soil heterogeneity. Fractures in the lateritic upper horizon resulted in a spatial heterogeneity of tree growth over the entire trial site. The larger trees (3 m in height) of Block 1 formed a relatively distinct group independent of the provenances.

After 5 years of growth, the dominant tree height of the stand (average height of the 100 largest trees) was 4 m, twice the overall average height (Table 3). This showed that F. albida can develop well on fissured, lateritic soil. A height gradient between Block 4 (average = 1.45 m) and Block 1 (2.12 m) was due to a corresponding increase in soil depth across the gradient. Lateritic outcroppings were more evident in Block 4 than in Block 1.

Table 3. Average height (m) of different tree sub-populations in the Faidherbia albida trial, Mouda Gazal, Cameroon, 1987-90. (Source: Peltier 1988; Harmand 1989.)

Trees planted along the borders of vegetated strips exhibited poorer growth (-36%) and had a lower average survival rate than those planted within the plots. Despite annual weedings around seedlings planted along plot borders, root competition with grasses within the strips suppressed tree development. The fact that few F. albida are encountered in tree savannas or even in the vicinity of farmlands constitutes further proof of its requirement for animal husbandry by farmers.

At this stage in development, tree crowns were small and low foliar biomass production had no discernable positive impact on soil fertility or crop yields (Table 4). In fact, the tree density (500 trees ha^-1) hindered plowing operations and diminished arable area.

Table 4. Crop yields (t ha^-1) in a study, Mouda Gazal, Cameroon, 1990.

Conclusion

The Burundi provenance was not adapted to Sudano-Sahelian conditions. Because of great variability within provenances, differences in performance of six F. albida provenances of the Sudano-Sahel zone could not be distinguished. Dominant trees of F. albida planted on ferruginous and lateritic soil lacking a high water table attained a height of 4 m in 5.5 years and exhibited heterogeneous growth. At this time, no positive impact of the species on associated crops was evident.

Further studies on this site should focus on available mineral nutrients and the water budget in the soil water utilization by trees and crops, and quantities of biomass returned to the soil.

Planting of F. albida at a fixed spacing of 10 x 10 m in a mechanically-cultivated crop system would be feasible. Plantings along borders of vegetated strips existing between crop plots, however, resulted in poor survival rates due to competition with weeds.

References

Brabant, P. 1988. Sols ferrugineux et sols apparentés du Nord-Cameroun. Aspects de leur pèdogenése. Paris, France: Institut français de recherche scientifique pour le développement en coopération. 43 pp.

Harmand, J.M. 1989. Résultats des essais sylvicoles de l'Antenne CRF/IRA de Marona. Marona, Cameroon: Centre de recherches forestières. 170 pp. (Limited distribution.)

Peltier, R. 1988. Resultats des essais sylvicoles de l'Antenne CRF/IRA de Marona. Marona, Cameroon: Center for Forestry Research. 240 pp. (Limited distribution.)

Peltier, R., and Eyog Matig, O. 1988. Les essais d'agroforesterie au Nord-Cameroun. Bois et Forêts des Tropiques 217(3):3-31.

Footnote__________

1 Centre technique forestier tropical (CTFT), B.P. 415, Maroua, Cameroon.

2 Centre des recherches forestieres (CRF)/Institut de la recherche agronomique (IRA), B.P. 415, Maroua, Cameroon.

Harmand, J.M., and Njiti, C.F. 1992. Faidherbia albida in northern Cameroon: provenance trials and crop associations. Pages 79-81 in Faidherbia albida in the West African semi-arid tropics: proceedings of a workshop, 22-26 Apr 1991, Niamey, Niger (Vandenbeldt, R.J., ed.). Patancheru, A.P. 502 324, India: International Crops Research Institute for the Semi-Arid Tropics; and Nairobi, Kenya: International Centre for Research in Agroforestry.