O Lyasse

Alley cropping in the moist savanna of West-Africa. III. Soil organic matter fractionation and soil productivity

In cropping systems with limited amounts of external inputs, the soil organic matter pool (SOM) may contribute significantly to plant nutrition. The impact of organic inputs on total SOM and particulate organic matter (POM) N contents as affected by soil type and the relationships between sources of N and maize N uptake were assessed for a set of alley cropping trials in the West- African moist savanna. The trials were established in Niaouli (Bénin Republic), in Glidji, Amoutchou, and Sarakawa (Togo), and in Bouaké and Ferkessédougou (Côte d‘ Ivoire). The total soil N content, averaged over all treatments and years, varied between 324 and 1140 mg N kg−1 soil. The POM-N content varied between 50 and 160 mg N kg−1 soil. The average proportion of soil N belonging to the POM pool ranged between 9% and 29%. This was significantly related to the annual N inputs from maize stover and prunings, when averaged over the different alley cropping treatments. The trial ‘age‘ also appeared to be related to the impact of the different treatments on the POM-N content. The Ferkessédougou soil contained a relatively higher proportion of total soil N in the POM pool because of its relatively high silt and clay content, compared to the other sites. The relative change in POM-N content between 1996 and the initial sampling was about twice the relative change in total soil N content. This suggests that N incorporated in the POM is relatively labile, compared to N incorporated in the other SOM fractions. Maize N uptake was related to the amount of add pruning-N (partial r2 of 27%), the rainfall during the growing season (partial r2 of 17%), the POM-N content (partial r2 of 14%), and to a lesser degree to the POM N concentration (partial r2 of 5%), the fertilizer N addition rate (partial r2 of 3%), and the silt and clay content of the soil (partial r2 of 3%). The POM-N content was shown to be influenced by organic matter additions and soil characteristics and to contribute significantly to maize N supply. This pool may be an important indicator for the soil fertility status of savanna soils.

Alley cropping in the moist savanna of West-Africa: I. Restoration and maintenance of soil fertility on ’terre de barre‘ soils in Bénin Republic

The potential of alley cropping systems supplied with a limited amount of fertilizer to restore crop productivity on a degraded site and to maintain crop productivity on a recently cleared, non-degraded site on ‘terre de barre‘ soils in Southern Bénin was investigated from 1994 to 1996. Leucaena leucocephala, Senna siamea and Gliricidia sepium were used as hedgerow species. Maize yields of the no-tree control plots dropped from the initial (1990) 401 kg ha−1 and 2181 kg ha−1 on the degraded and non-degraded sites, respectively, to 109 kg ha−1 and 1346 kg ha−1 in 1996, even with application of a minimal amount of mineral fertilizer. The alley cropping systems produced on average (mean of three treatments and three years) 107% more grain than the initial 1990 values on the degraded site and 11% less grain than the initial 1990 values on the non-degraded site. Especially the Senna and to a lesser degree the Leucaena treatment yielded consistently more grain than the control. The Senna trees contained a larger amount of N and produced more wood during the first pruning on the degraded site (155 kg N ha−1 and 14.0 ton fresh wood ha−1) than on the non-degraded site (49 kg N ha−1 and 6.6 ton fresh wood ha−1) most likely because of differences in subsoil fertility, as indicated by the higher clay, exchangeable bases, and N content between 60 and 125 cm cm. N accumulation and wood production by the Leucaena and Gliricidia trees was similar in both sites (82 and 36 kg N ha−1 and 4.6 and 9.3 ton fresh wood ha−1, respectively). When a limited amount of fertilizer is available, Senna appears to be the best choice as hedgerow species on sites with a relatively fertile subsoil. For other soils, a N2-fixing species may be a better choice.

Alley cropping in the moist savanna of West-Africa: II. Impact on soil productivity in a North-to-South transect in Togo

The moist savanna of West-Africa is characterized by a wide range of climates and soil types. The impact of the biophysical environment on hedgerow N uptake, wood production and maize grain yield was assessed for three years in three alley cropping trials with a selected number of hedgerow species in Glidji (Southern Togo), Amoutchou (Central Togo), and Sarakawa (Northern Togo). Senna siamea hedgerows accumulated significantly more N in the first pruning in Glidji (129−138 kg N ha−1) and Sarakawa (102−185 kg N ha−1) than in Amoutchou (17–26 kg N ha−1). This difference in N uptake was attributed to the infertile subsoil in Amoutchou, which was sandy up to 1 m and had a shallow groundwater-table. The amount of N accumulated in the Gliricidia sepium biomass varied between 38 kg N ha−1 in Glidji and 142 kg N ha−1 in Amoutchou. Averaged over all species and sites, 9 to 29% and 9 to 39% of the annual N accumulation in the hedgerow biomass is incorporated in the second, respectively third pruning. The Gliricidia trees produced between 12 and 26 ton fresh matter ha−1 of wood and the Senna trees between 4 and 38 ton fresh matter ha−1. Maize grain yield in Glidji was not affected by treatments (3196 kg ha−1, on average). In Amoutchou, the highest grain production was observed in the Gliricidia treatment (2774 kg ha−1 vs 1007 kg ha−1 in the control), while in Sarakawa, the Gliricidia (3786 kg ha−1) and Senna (3842 kg ha−1) plots produced a greater grain yield than the control plots (2123 kg ha−1). Maize yield increase in the alley cropping systems relative to the control plots was related to the soil total N content. Top and sub-soil characteristics were shown to be an important modifier of the functioning of alley cropping systems and should be taken into account when deciding on whether to use alley cropping and when selecting the hedgerow species.

Root distribution of Senna siamea grown on a series of derived-savanna-zone soils in Togo, West Africa

Although crucial for assessing the functioning of alley cropping systems, quantitative information related to the hedgerow tree root distribution remains scarce. Soil mapping and destructive soil sampling was used to assess the impact of soil profile features on selected root characteristics of Senna siamea hedgerows, growing in alley cropping systems in three sites (Glidji, Amoutchou, and Sarakawa) representative for the derived savanna of Togo, West Africa. While the soil profiles in Glidji and Sarakawa contained a clay accumulation horizon, the Amoutchou profile was sandy up to 1 m. The number of small roots (diameter < 2 mm), quantified on a soil profile wall, decreased with depth in all sites. For most soil depths, the abundance of small roots tended to be higher near the tree base, e.g., ranging from 5.3 dm−2 in Amoutchou to 21.4 dm−2 in Glidji for the 0–20 cm layer, than in the middle of the alley, e.g., ranging from 3.1 dm−2 in Amoutchou to 13.8 dm−2 in Glidji for the 0–20 cm layer. Root length density (RLD) of the 0–10 cm and 10–20 cm layers was significantly higher in Glidji than in Amoutchou (P < 0.05) and in Sarakawa (P = 0.08). Differences in RLD between sites were not significant for layers below 30 cm. For each layer, root weight densities (RWD) were similar in all sites, e.g., ranging from 0.44 mg cm−3 in Amoutchou to 0.64 mg cm−3 in Glidji in the 0–10 cm layer, indicating that the roots in the Glidji topsoil had a smaller overall diameter than in Amoutchou. In Amoutchou, the relative RLD was lower than in Glidji or Sarakawa for the top 40 cm of soil, while the inverse was observed for the layers between 50 and 100 cm deep and this was related to the sandy soil profile in Amoutchou. Another consequence of the sandy profile was the larger tap root diameter below 50 cm in Amoutchou compared to Sarakawa. For all sites, significant (P < 0.001) linear regressions were observedbetween RLDs, RWDs, and the abundance of small roots, although the variation explained by the regression equations was highest for the relationship between RLD and RWD. The potential of the hedgerows to recover nutrients leached beyond the reach of food crops or the safety-net efficiency was evaluated for the tree sites.