Due to increased population pressure and limited availability of fertile land, farmers on desert fringes increasingly rely on marginal land for agricultural production, which they have learned to rehabilitate with different technologies for soils and water conservation. One such method is the indigenous zai technique used in the Sahel. It combines water harvesting and targeted application of organic amendments by the use of small pits dug into the hardened soil. To study the resource use efficiency of this technique, experiments were conducted 1999-2000, on-station at ICRISAT in Niger, and on-farm at two locations on degraded lands. On-station, the effect of application rate of millet straw and cattle manure on millet dry matter production was studied. On-farm, the effects of organic amendment type (millet straw and cattle manure, at the rate of 300 g per plant) and water harvesting (with and without water harvesting) on millet grain yield, dry matter production, and water use were studied. First, the comparison of zai vs. flat planting, both unamended, resulted in a 3-to 4-fold (in one case, even 19-fold) increase in grain yield on-farm in both years, which points to the yield effects of improved water harvesting in the zai alone. Zai improved the water use efficiency by a factor of about 2. The yields increased further with the application of organic amendments. Manure resulted in 2-68 times better grain yields than no amendment and 2-7 times better grain yields than millet straw (higher on the more degraded soils). Millet dry matter produced per unit of manure N or K was higher than that of millet straw, a tendency that was similar for all rates of application. Zai improved nutrient uptake in the range of 43-64% for N, 50-87% for P and 58-66% for K. Zai increased grain yield produced per unit N (8 vs. 5 kg kg À1 ) and K (10 vs. 6 kg kg À1 ) compared to flat; so is the effect of cattle manure compared to millet straw (9 vs. 4 kg kg À1 , and 14 vs. 3 kg kg À1 ), respectively, Therefore zai shows a good potential for increasing agronomic efficiency and nutrient use efficiency. Increasing the rate of cattle manure application from 1 to 3 t ha À1 increased the yield by 115% TDM, but increasing the manure application rate further from 3 to 5 t ha À1 only gave an additional 12% yield increase, which shows that optimum application rates are around 3t ha À1 .

In the West African Sahel, farmers use the zai technique to reclaim degraded cropland. Although the nutrients released by the decomposition of the amendments are central to the success of the technique, little is known regarding the impact of the zai pits on the decomposition process and whether the nutrient release is synchronized with plant requirements. The decomposition of millet stalks and cattle manure applied in zai pits or at the soil surface was studied in Niger using litterbags, under controlled irrigation on-station in 1999 and on-farm in 1999 and 2000 at two locations (Damari and Kakassi) with contrasting soils. In addition, a satellite trial was conducted in 2000 on-farm at the same locations to study the relative contribution of termites to manure decomposition. Only at Damari did termite presence enhance manure decomposition, by a factor three for surface placement compared to the zai pits. At Damari, zai pits enhanced the decomposition when termite activity was suppressed. Whereas manure decomposition proceeded two to three times faster than that of millet stalks at Damari, the type of amendment had no effect on decomposition rate at Kakassi. Nutrient release followed the trend of organic amendment decomposition except for K. When applied prior to the rainy season, nutrient release rate of organic amendments strongly exceeded plant nutrient uptake, which could lead to important leaching losses during the first 4-6 weeks after sowing, especially for N and to a lesser extent for K. However, at harvest, total nutrient absorption by plants was generally higher than the total amount released. The results indicate a highly site-specific response of amendment decomposition to zai and the need for a better timing of amendment application to reduce potential leaching losses, possibly through a split application.

a b s t r a c tInadequate nutrient supply and insufficient rainfall are the most important limiting factors for crop production in the Sahelian agro-ecological zones. Targeted technology application may help to improve the efficient use of limited nutrient and water resources. The objective of this study was to determine the optimal combination of fertilizer micro-dosing and manure application rates for improved millet yield and enhanced water use efficiency in low input millet-based cropping system. A two-year field experiment was conducted at a research station in Niger using a randomized complete block design with three replications. The treatments consisted of the factorial combination of: (i) two fertilizer microdosing options (20 kg ha −1 of diammonium phosphate (DAP) and 60 kg ha −1 of NPK corresponding to 2 g hill −1 of DAP and 6 g hill −1 of NPK, respectively), (ii) cattle manure at four application rates (0 kg ha −1 , 1000 kg ha −1 , 2000 kg ha −1 , 3000 kg ha −1 ) and (iii) two methods of manure application (broadcasting and hill placement). Millet grain yields under fertilizer micro-dosing combined with manure was increased on average by 59%, 83% and 113% for 1000 kg ha −1 , 2000 kg ha −1 and 3000 kg ha −1 , respectively compared with fertilizer micro-dosing alone. Combined applications of manure and fertilizer micro-dosing increased water use efficiency significantly. Hill placement of manure increased total dry matter on average by 23% and water use efficiency by 35% relative to manure broadcasting. The total root length density was increased by 66% and 42% in hill placement of manure at 25 cm and 50 cm, respectively, from the hill centre compared with manure broadcast. These results indicate that millet production with the fertilizer micro-dosing technology can be improved further by hill-placement of manure. The combination of 2000 kg ha −1 of manure and 20 kg DAP ha −1 hill-placed were most promising for increasing millet yield and the efficient use of limited nutrients and water in Sahelian millet based systems. There is need for testing this technology further together with farmers to valuate its effectiveness.

The fertilizer microdosing technology deals with the application of small quantities of fertilizers in the planting hole, thereby increasing fertilizer use efficiency and yields while minimizing input costs. In drought years, microdosing also performs well, because larger root systems are more efficient at finding water, and it hastens crop maturity, avoiding lateseason drought. Recent research found that solving the soil fertility problem unleashes the yield potential of improved millet varieties, generating an additional grain yield of nearly the same quantity. Recognizing that liquidity constraints often prevent farmers from intensifying their production system, the warrantage or inventory credit system helps to remove barriers to the adoption of soil fertility restoration. Using a participatory approach through a network of partners from the National Agricultural Research and Extension Systems (NARES), non-governmental organizations (NGOs), farmers and farmer groups and other international agricultural research centres, the microdosing technology and the warrantage system have been demonstrated and promoted in Burkina Faso, Mali, and Niger during the past few years with encouraging results. Sorghum and millet yields increased by up to 120%, and farmers' incomes went up by 130% when microdosing was combined with the warrantage system. This chapter highlights the outstanding past results and the ongoing efforts to further scale up the technology using Farmer field schools (FFS) and demonstrations, capacity and

A study was carried out in the rainy seasons of 2008 and 2009 in Niger to investigate the effects of fertilizer microdosing on root development, yield and soil nutrient exploitation of pearl millet. Different rates of diammonium phosphate (DAP) were applied to the soil at different depths and it was found that although micro-dosing with DAP increased grain yield over the unfertilized control to a similar level as broadcast DAP, doubling the micro-dosage did not increase it further. Increasing the depth of fertilizer application from 5 to 10 cm resulted in significant increases in root length density, and deep application of fertilizer resulted in higher yields, although the increases were generally not significant. It was postulated that the positive effect of micro-dosing resulted from better exploitation of soil nutrients because of the higher root volume. Levels of nutrients exported from the soil were at least as high in plants receiving micro-dosing as the unfertilized control, and plants receiving microdosing exported 5-10 times more phosphorus from the soil than the amount added through fertilization.

Aflatoxin contamination of peanut is a major threat to consumers in West Africa. High levels of aflatoxin have been reported in West and Central Africa, particularly in Niger. Field trials were conducted from 1991 to 1994 at ICRISAT Sahelian Center, Sadore Research Station near Niamey, Niger. Various production practices were compared to examine their effects on water stress and Aspergillus flavus infection and aflatoxin contamination. Different levels of water stress were achieved by varying planting date and frequency of irrigation in two resistant and two susceptible cultivars. Contamination of seed with A. flavus and aflatoxin was determined. The susceptible cultivars 28–206 and JL 24 had much higher levels of seed infection following 3 wk or more of water stress than did the resistant cultivars. Susceptible cultivars showed up to 81% seed infection. Cultivar 28–206 had low levels of contamination when there was low water stress but became very susceptible when the period of water stress increased (3 wk of drought). Seed infection by A. flavus and contamination by aflatoxin were highly correlated across years and cultivars. Although seed infection by A. flavus was very responsive to water stress in the field, aflatoxin contamination did not increase proportionally. This may have been influenced by high soil temperatures in Niger, which may exceed 40 C. Most reports indicate that a minimum of 20 to 30 d of drought is needed for significant aflatoxin contamination, but contamination was observed after 14 d of water stress in this study.

Limited availability of soil organic amendments and unpredictable rainfall, decrease crop yields drastically in the Sahel. There is, therefore, a need to develop an improved technology for conserving soil moisture and enhancing crop yields in the Sahelian semi-arid environment. A 2-year field experiment was conducted to investigate the mulching effects of Acacia tumida pruning relative to commonly applied organic materials in Niger on millet growth, yields and water use efficiency (WUE) under fertilizer micro-dosing technology. We hypothesized that (1) A. tumida pruning is a suitable mulching alternative for crop residues in the biomass-scarce areas of Niger and (2) combined application of A. tumida mulch and fertilizer micro-dosing increases millet yield and water use efficiency. Two fertilizer micro-dosing options (20 kg DAP ha -1 , 60 kg NPK ha -1 ) and three types of organic mulches (millet straw, A. tumida mulch, and manure) and the relevant control treatments were arranged in factorial experiment organized in a randomized complete block design with four replications. Fertilizer micro-dosing increased millet grain yield on average by 28 %. This millet grain yield increased further by 37 % with combined application of fertilizer micro-dosing and organic mulch. Grain yield increases relative to the unmulched control were 51 % for manure, 46 % for A. tumida mulch and 36 % for millet mulch. Leaf area index and root length density were also greater under mulched plots. Fertilizer micro-dosing increased WUE of millet on average by 24 %, while the addition of A. tumida pruning, manure and millet increased WUE on average 55, 49 and 25 %, respectively. We conclude that combined application of micro-dosing and organic mulch is an effective fertilization strategy to enhance millet yield and water use efficiency in low-input cropping systems and that A. tumida pruning could serve as an appropriate mulching alternative for further increasing crop yields and water use efficiency in the biomass-scarce and drought prone environment such as the Sahel. However, the economic and social implications and the long-term agronomic effects of this agroforestry tree in Sahelian millet based system have to be explored further.Keywords Organic mulch Á Fertilizer micro-dosing Á Acacia tumida Á Millet yield Á Water use efficiency

Direct determination of soil hydraulic properties is often costly and laborious hence the use of indirect methods such as pedotransfer functions (PTFs). Despite progress made in PTF development in general, little evaluation of PTFs has been done for the sandy soils of Niger. We tested the ability of three PTFs, (Campbell, van Genuchten, and Vauclin) to determine soil water retention and unsaturated hydraulic conductivity (K) for sandy soils at two villages (Banizoumbou and Bagoua) in Niger. Modeled K was compared to K estimated from neutron probe readings at 1.4 m; and modeled moisture retention was compared to lab measurements derived from the hanging water column method and pressure plate apparatus for the following depth intervals: 0-30, 30-60, and N 60 cm at Banizoumbou; and 0-30, 30-120, and N120 cm at Bagoua. The Campbell PTF resulted in lower root mean square error (RMSE) (0.05-0.06 m 3 m − 3 ) for soil moisture retention for the three depth intervals at the two sites and performed better than the van Genuchten function (RMSE 0.06-0.07 m 3 m − 3 ) for Bagoua soils which had higher sand content. The van Genuchten PTF consistently overestimated dry regime moisture retention for the three depth intervals especially at Bagoua but performed well for the wet regime. The Campbell and Vauclin PTFs underestimated K (RMSE 0.61-1.01 mm d − 1 ) at both sites whereas the van Genuchten model was close to field measurements (RMSE 0.26-0.47 mm d − 1 ). These results show that the Campbell model is a cheaper alternative to direct measurement of moisture retention and the van Genuchten function can be used to estimate K for Niger's sandy soils with modest accuracy.