SUMMARYThe contrasted global population growth with the multiplication of the constraints to developing new irrigation systems puts a special challenge upon human crop production systems that needs to be taken up. The populations in many countries in Asia, Middle East and Africa are expected to double in the coming 50 years. The experience of the green revolution in Asia – during which 70% of food production increase was provided by irrigated agriculture – shows that there is not only a need to strive to increase such crop production systems, but also to improve the production efficiency of existing ones. In fact, as a much worrying case, rice production in valley bottom irrigated lands of African Tropical Savannah is far to yield the expected amount of cereals. One of the major constraints to this production is iron toxicity subsequent to poor drainage conditions. According to Africa Rice, at least 60% of the Tropical Savannah swampy valley bottoms are affected by different degree of iron toxicity. The yield in many areas drops to zero, leaving behind millions of disappointed and impoverished farmers. Therefore, it is not surprising that there is strong research dynamic – ranging from agronomy to microbiology – that strives to propose alleviating solutions to rice iron toxicity. Because prevalent anoxic conditions in the soil combined with iron reducing bacteria development were found a basic contributing factor to iron toxicity, this research has chosen to investigate subsurface drainage potential contributions to solving this issue.Two complementary series of operations – designed within two project- components and focused on five basic questions closely related to the contributing factors to iron toxicity development – were performed. In fact, the research project was implemented in two major components: field surveys and designed experiments. The field surveys investigated iron toxicity triggering or aggravating factors such as clay proportions, ferrous ion Fe2+ concentration, dissolved oxygen, soil acidity or water management. Drawing profit from the knowledge gained in survey research and literature review, two parallel experiments were designed using concrete microplots on one hand and buckets on the other hand, to statistically ascertain the impact of subsurface on soil acidity and ferrous iron concentration changes. All the operations performed within the two components of this research project endeavoured to answer the following five research questions: how is ferrous iron formed and distributed in soils invaded by iron toxicity?how is clay spread within the valley?how is soil permeability affected by clay distribution in the valley?how can water management help improve soil conditions?what is the impact of subsurface drainage on iron toxicity? The answers to these research questions – already published or in press – are exposed below, followed by the contribution of this research project in two areas: i) science en engineering, and ii) socio-economy.Clay and ferrous iron may deposit in strataHigh ferrous ion Fe2+ concentration, inserted into dense clay strata, constitutes an important threat to rice production in several tropical Savannah valley irrigation schemes of West Africa. Many actions are currently undertaken to alleviate iron toxicity. In this study, we have investigated the presence of clay and ferrous iron stratifications within a typical flood prone valley bottom called Tiefora in Burkina Faso. Taking into account the multiple slopes of the valley, two randomized soil samplings were implemented at various depths. Samples were collected as deep as 500 cm, but especially at 30, 50 and 100 cm. The clay percentage was determined by grain size analysis. Ferrous iron concentrations were obtained through the reflectometric method. The stratifications of clay and ferrous ion Fe2+ were checked using statistical hypothesis testing (ANOVA and Welch t-Test). Clay percentage within the first 100 cm top soil – 28.9% – was found twice higher than in the layers underneath. Furthermore, ferrous iron was mainly located in the top 30 cm, with a mean concentration of 994 mg/l. This ferrous iron concentration is much higher than found at depths 50 and 100 cm underneath (73 mg/l), while the pH of all the three layers is almost neutral. This striking stratification suggests several means of alleviating iron toxicity. Among these means, we propose maintaining wet conditions during the growing period in the irrigated lands in combination with leaching by subsurface drainage in the fallow periods. Iron toxicity risk is higher in single season irrigation schemes With the aim of finding the geochemical differences and helping to build alleviating strategies against iron toxicity, two hematite dominant valley bottoms irrigated rice soils were investigated in the Tropical Savannah region of Burkina Faso. The first site was Tiefora, a 16 ha modern double-season irrigated rice scheme and moderately affected by iron toxicity (10% of the area with a toxicity score of 4). The second site was Moussodougou, a 35 ha traditional single-season irrigated rice valley-bottom, with 50% facing more severe iron toxicity (score 7). Nine soil extracts were taken from three depths – 30, 50 and 100 cm – i.e. 27 at Tiefora and 27 at Moussodougou. Five techniques were used to measure the data: i) the ferrous iron concentration was determined using a reflectometer, ii) a pH-meter yielded the pH, iii) clay-proportions were obtained by United States Department of Army (USDA) grain size analysis and densitometry, iv) the organic matter was determined by oven drying and v) the dry bulk density was determined by using undisturbed soil samples. Statistical hypothesis testing of One-way ANOVA and Welch t-test were applied to the data to isolate the similarities and the differences between the two sites. A geochemical analysis followed to find the causes of these differences. The results showed that while oxidation of pyrite leads to a simultaneous increase in Fe2+ concentrations and acidity in the soils of coastal floodplains and mangroves, the oxidation of hematite in Tropical savannah valleybottoms decreases Fe2+ but also increases acidity during the dry season. As a consequence, it was found that the single-season irrigation scheme of Moussodougou is significantly (p-value 0.4%) more acidic (pH 5.7) than the double-season system of Tiefora (pH 6.4) with also 750-1800 mg/l higher ferrous ion Fe2+. The ferrous iron reached 3000 mg/l in some layers in Moussodougou. This result is a justification to modernize traditional single-season spate irrigation schemes into double-season irrigated rice schemes. Subsurface drainage type depends on clay distribution Waterlogged valley bottom soils of Tropical Savannah are areas where the richest traditional cropping systems are found, but they also face adverse physical and chemical conditions which can drastically drop rice yield. Subsurface drainage has been used for many areas to alleviate waterlogging. However, this drainage is dependent of claydistribution, type and location. The current research analysed these factors using the case of Tiefora. For this purpose nine boreholes, with depths from 2 to 6 m, were realised. Some 50 samples of soils were extracted at various depths, based on soil changes in texture and colour. These samples underwent grain-size-analysis. A comparative non-linear regression was performed on the clay distribution. Quadratic regression was the most appropriate. Clay proportions were high - 20-30% in the 2 m topsoil - in the upstream and middle areas. A more important - 30-40% - peak was reached in the downstream area at 1 m-depth, with a much smaller thickness (less than 50 cm) and higher permeability. These results suggest the application of mole drainage in the valley, except downstream where the classical Hooghoudt pipe subsurface drainage can be implemented. Subsurface drainage cost can be reduced by taking into account permeability distribution in valley In flood prone Tropical Savannah valley soils very low infiltration rates often result in acidic conditions favourable to high concentrations of metallic ions, toxic for rice. The infiltration rate determination is important in drainage design to reclaim degraded soils. Several studies have addressed the mapping of the infiltration rate. Yet its relationship with the toposequence of the valley is not clarified. This research has investigated such possibility, examining the case of the irrigated rice valley of Tiefora. Nine boreholes – 1 to 5 m deep – were implemented from upstream to downstream. The Lefranc permeability test of under phreatic conditions in waterlogged soils – used when the impervious layer is close to soil surface or absent – was conducted. First, a comparative regression was applied to the data, including all the parameters of the regression curves. In case of dissimilarity of the infiltration processes, the comparison focused on the final permeability. Our results show a permeability increase from upstream (0.10 ± 0.10 cm/hr) to downstream (greater than 20.0 ± 10.0 cm/h in some places). Taking into account such permeability increase in subsurface drainage system design would result in the implementation of more efficient and cost effective systems. Data based water management can help to reduce water losses and solve water inequity frictions between farmers Surface irrigation represents more than 99% of the irrigated area in West Africa and generally includes valley bottoms dedicated to irrigated rice production, which are often denounced as water overusing schemes. Surprisingly, there is neither follow up nor analysis of the irrigation water used in these gravity irrigation systems. Such a work was carried out in the case of the 16 ha Tropical Savannah irrigated rice valley bottom scheme of Tiefora. Using the flow equation of the concrete weir at the headwork, daily water use volumes were calculated as time series covering more than one-year period. The moving average trend analysis reveals that during both the rainy season (1200 mm of rainfall) and the dry season (no rainfall), the main canal gate is almost never closed, keeping a minimum discharge of 200 m3/day for 4 ha (50 mm/day versus. a local evapotranspiration of 7 mm/day). That stresses the necessity of a more rigorous water management. Furthermore, the autocorrelation analysis by using the ARIMA model showed that the irrigation cycle that ensures equity in water distribution among farm plots is 20 days instead of five. The knowledge of this fact can defuse potential conflicts about equity among farmers: the lack of water in day 4 may be compensated later during the 20-day cycle. It appeared that a simple water level measuring device – installed at the headwork of the main irrigation canal – can produce a time series towhich autoregressive moving average model can be applied to yield, at low cost, a thorough assessment of water management in this surface irrigation system. Subsurface drainage alleviates iron toxicity in mean and long run Iron toxicity is one of the most important constraints that hinder rice productivity in Tropical Savannah valley bottom irrigated fields, but fortunately that can be alleviated. A too high ferrous iron level in the soil can nullify rice yield. Several research fields – agronomy, pedology through microbiology – strive to provide a solution to this issue. Up to date, the contribution of hydraulics to tackle iron toxicity remained limited. The current research addressed this aspect through controlled experiments on highly ferrous iron contaminated rice hematite soils. Twelve concrete microplots and eight buckets were used to implement two independent designed experiments during a period of 86 days. Drainage and liming were the two factors whose impacts were investigated. Drainage was used with two treatment conditions: 0 mm/day and = 10 mm/day, and liming also had two treatment conditions: Lime- = 0 kg/m² and Lime+ = 1 kg/m² per unit increment increase of the pH. Four different responses in the soil were measured: ferrous ion concentration Fe2+, pH, oxido reduction potential, and the dissolved oxygen. For the rice, toxicity scores of the International Rice Research Institute were followedup. The results indicate an increase of Fe2+ from 935 mg/l to more than 1106 mg/l (at 95% of confidence level), but, which is interesting, with a significant decrease of soil acidity from pH 5.6 to 7.3 (95% confidence level). Liming had the same effect in alleviating the acidity. Reduction processes were not hindered by subsurface drainage since the oxydo reduction potential dropped from 84.6 to 9.2 mV, and dissolved oxygenmoved from 1 mg/l to less than 0.1 mg/l. Despite of the reduction of the acidity, with such a high ferrous iron level as 1106 mg/l, the iron toxicity score reached 7 in the twelve microplots and the rice died. Still, the reduction of soil acidity provides a new insight on the hematite soils behaviour, opposite to the acidification with subsurface drainage in coastal floodplains and mangrove pyrite. Furthermore, it will lead to less ferrous iron intake by rice roots and in such perspective improve the rice yield. Finally, though liming can achieve the same result, subsurface drainage takes the advantage when this mineral is not available or is expensive. Project outputs for Tiefora farmers From the investigations and their supporting activities, two major benefits were brought to the farmers of Tiefora. First, in order to alleviated iron toxicity – which is much less severe in this place than in Moussodougou – and improve rice yield (less than 4 tons/ha), it would be essential to apply according the norms of the Institute of Environment and Agricultural Research (IN.ERA) the complex fertilizer NPK. However, this application should go along with making well built bunds around the farm plots in order to confine the fertilizer and make the mineral more available for the rice roots. This will invigorate the crop and thus strengthen its resistance to iron toxicity. Secondly, the project handed to the farmers’ association of Tiefora three key documents: i) an aerial photo the environment of the valley of Tiefora, including the reservoir, the village, the roads and the irrigated valley, ii) a topographical map of the valley bottom, intended to help in potential engineering works on the irrigation system, and iii) a detailed map of the farm plot system, accompanied with the complete list of the farmers and their farm sizes, and the location of iron intoxicated plots for their daily activities Project outputs for Moussodougou farmers Based on the investigation results and due to the severe iron toxicity in Moussodougou, the project provided several advices and handed some key documents to the farmers. Ferrous iron concentration in the soil of Moussodougou can reach 3000 mg/l in many farm plots with acidity as severe as pH 4. Since its incorporation into the soil was found to induce the growth of iron reducing bacteria activity, and given the positive conservation impact of organic matter in lightening the soil structure, the project advised the farmers to reduce its use but not to eliminate it completely. In parallel, farmers would have to use the complex NPK as in Tiefora, according to the norms of IN.ERA, but combine it with a careful erection of plot bunds to make the mineral element more available for the rice. Due to the fact that the current single irrigation season during the year in Moussodougou is an aggravating factor of iron toxicity, the project also introduced to the farmers association its ongoing work of developing sprinkler irrigation from groundwater during the dry season. Finally, the project handed to the farmers’ association the same set of documents as in Tiefora, but related to the valley bottom of Moussodougou. Other social impacts In an ultimate effort to share the insights gained about the iron toxicity alleviation process, this research project produced and uploaded onto the social media YouTube several useful videos. The 15 videos uploaded and accessible for everybody, deal with areas as varied as hydrometrics, microbiology, geochemistry and small scale water saving irrigation equipment assembling at village level (without electricity). Many of these videos were very appreciated by the audience. For example, the video of "Innovative irrigation systems in Sub-Saharan Africa (French)" has been viewed/downloaded 500 times/month. Similarly, the video "How to take a sample of disturbed soil or resting in soil immersed at different depths (English)", was viewed/downloaded some 45 times/month. These two videos were classified "creative common" due to their high potential appropriation by third party video productions. Hence, it is expected that the project will have an even higher social impact in the coming months or years.