Integrated soil fertility management for bean-maize based farming systems in Gitega Province, Burundi: Understanding and enhancing the agronomic and economic benefits of organic and mineral inputs

Securing sustainable food security remains a huge challenge in sub-Saharan Africa. Yet, there is an urgent need to do so as the region faces high population density, continuous cultivation of already depleted soils, limited availability of resources to farmers all of it leading to low crop productivity. In this region, soil acidity and low N and P supply capacities are common problems over large areas of agricultural land. These problems are aggravated by the lack of improved and resilient germplasm of the major crops, including maize, and various legume species. Hence, it provides the ideal testing ground to evaluate the Integrated Soil Fertility Management (ISFM) paradigm with its different options, thereby proving the conceptual 'staircase' of improving yield effects and profitability of fertilizer use. The current research is conducted in three action sites (Buraza, Makebuko and Mutaho districts) of the Consortium for Improving Agriculture-based Livelihoods in Central Africa (CIALCA) in Gitega Province. Diammonium phosphate (DAP, 18-46-0) was used to remediate soil N and P deficiencies but its use is limited in rural areas because of its poor availability, its wrong application rates, high cost, lack of store rooms and good roads. Hence the region needs to find strategies that would enhance yield increments with concomitant increased agricultural net returns. The general objective of this work was to validate the ISFM concept on bean-maize based farming systems in Gitega Province, Central Burundi; thereby proving the underlying hypothesis of stepwise maximization of the profitability of DAP fertilizer use. First, we carried out a field survey to evaluate the local farmers' knowledge on DAP fertilizer use and assess DAP effects on bean and maize production in two districts of Gitega Province: Buraza and Makebuko. We found a huge variability in plot sizes, plant densities, mineral DAP fertilizer rates on beans and maize production. This illustrates how farmers have different attitudes and this should be taken into account as a precondition for the establishment of profitable and sustainable nutrient management systems. Profitability of DAP use on local germoplasm under local farmers' practices was also assessed. Only 3 % and 40 % of farmers' fields provided profitable net returns for beans and maize, respectively. These results justify the need for ISFM. Secondly, we assessed the applicability of the Compositional Nutrient Diagnosis (CND) approach under smallholder farming practices, and the frequency of nutrient deficiencies for maize cropping in Buraza and Makebuko districts of Central Highlands of Burundi in the short rains of 2012 (2012SR). The trials were carried out on 76 fields for Buraza and 66 fields for Makebuko districts, hence totalling 142 fields covered in this study. In each field, two plots were demarcated to accommodate two treatments, a control without fertilizer and a treatment with 1kg of DAP fertilizer. Composite soil samples and maize ear leaves were collected from all plots and shipped to the International Center for Tropical Agriculture (CIAT) in Nairobi for analysis. Indophenol blue method was used for N-analysis and ICP-OES for the determination of other nutrients. A total of 284 samples was used in the CND computations. The maize simplex comprises eight nutrients N, P, K, Ca, Mg, S, Mn and Zn and the filling value R9 which takes care of all other nutrients not included in this analysis. The occurrence of nutrient deficiencies showed that 1 kg of DAP resulted in alleviating some of them, especially in Buraza district. However, N and P remained the main limiting ones for maize productivity in both control and fertilized plots in Makebuko district. The persistent P-deficiency may be due to the strongly acid, P-fixing soils. Zn and Mn-deficiencies occurred mostly in both control and fertilized plots of Buraza district. Zn and Mn-deficiencies are expected to limit yields. We observed in this study that actually in Buraza district, the DAP fertilizer induced a smaller yield increase on average (29 %) compared to the one in Makebuko district (46 %) where these two nutrients are not limiting. Next, we assessed the profitability of DAP fertilizer use on local and improved varieties of climbing bean (LR 2012) and maize (SR 2013) in smallholder farms of Buraza district in Gitega province in Central Burundi. Local varieties used were AMAKUTSA and ISEGA for bean and maize while G13607 and ZM 605-24C were improved varieties for climbing bean and maize. Fertilizer and variety effects were significantly larger when planted on time in a soil with larger total N (TN > 1.5 g kg-1) and total C (TC > 16 g kg-1) contents for beans while characterized by a moderate soil clay content (38 - 40 %) with high total C (TC ≥ 19 g kg-1) content for maize. Large fertilizer and improved variety effects consequently resulted in large value cost ratios (VCR). For improved climbing beans, VCR values on average were 4.49 $ $-1 which is more than the average of 3.19 $ $-1 obtained from local climbing beans. The same trend was also observed for maize, where the value cost ratio for the improved maize on average, was almost twice the one of local maize. The use of DAP fertilizer (100 kg ha-1) for the improved climbing bean and maize varieties resulted in larger grain yields, larger net benefits, larger value cost ratios and larger marginal rates of return than for local climbing bean and maize varieties. Finally, we tested the application of DAP fertilizer with improved bush and climbing bean varieties, in rotation with a subsequent improved maize variety. Rotational effects of a bush and climbing bean variety were compared, and the profitability of DAP fertilizer assessed in bean-maize rotations in 59 smallholder farms of Mutaho district, Gitega Province in Central Burundi in 2 cropping seasons: LR 2012 for beans and SR 2013 for maize. The improved varieties used were MLB 122-94B and G13607 for bush and climbing beans and ZM 605-24C for maize, respectively. The use of DAP fertilizer (50 kg ha-1) increased on average grain yields up to 1000 kg ha-1 for bush beans and 1800 kg ha-1 for climbing beans, respectively. Application of DAP fertilizer to the two bean varieties resulted in large yield differences when beans were planted on time in soils with a total C below 2.5 %. Such response was expected since the initial soil analysis indicated that the soils in which beans were cropped, were of low fertility, acidic, with small amounts of total N, total C and available P. The positive effect of DAP on maize grain yields was realized for about 60 % of the farmers. With the use of the same nutrients, maize following climbing beans out yielded maize following bush beans for about 48 % of the farmers. Large fertilizer and bean variety effects consequently resulted in a large value cost ratio. For climbing beans, VCR values on average were 4.25 $ $-1 , more than double the average of 1.83 $ $-1 obtained from bush beans. The same trend was also observed for maize planted after beans, where the control maize plots on average yielded 4000 kg ha-1. The value cost ratio was on average two times larger when maize followed climbing beans than when it followed the bush beans. Such profitability was realized when maize grain yields ranged between 1067 and 4622 kg ha-1. These yields were obtained from fields with high total C (TC > 23 g kg-1), low total N (TN < 2 g kg-1) and high clay content (more than 37 % clay). Results from this study revealed that an improved climbing bean preceding an improved maize on soils enriched with DAP fertilizer is a promising alternative to the local smallholder farmers' practices. In summary, the study confirmed the insufficient knowledge of smallholder farmers on how to use fertilizer according to their local conditions, which consequently leads to soil nutrient mining. There is a dire need for more trainings and awareness creativity emphasizing the use of fertilizer at their best advantage and development of site specific fertilizer recommendations instead of blanket. We noted that the use of fertilizer to the improved varieties was more profitable than for local varieties. In this way of implementing ISFM, we also found that use of fertilizer on climbing bean-maize rotation generated more rotational effects and was more profitable than that of bush bean-maize rotation. We therefore concluded that use of site specific fertilizer recommendations and high yielding improved grain legume varieties would be the most important factors to consider in initiatives to improve agronomic and economic productivity of legume-cereal rotations in such depleted soils.status: publishe