Evaluation, by coupling sorption and isotopic labelling and isotopic dilution methods, of the bioavailability of phosphorus in 'tanety' soils of Madagascar

Increasing food production in Madagascar requires to find new surfaces for cultivation. 'Tanety' soils cover about 30 % of the arable land of Madagascar (3,250,000 millions hectares) and they are the main potential surfaces available to increase crop production. The 'tanety' soils of Madagascar are frequently Ferralsols characterized by high content of Al- Fe- oxyhydroxydes, low nutrient reserves and low pH. It results in a low fertility of these soils due to Al toxicity and P deficiency leading to their degradation. Phosphorus is often the first limiting factor for crops in the 'tanety' soils, even after liming. Also, if food security has to be reached during the next decades, the question of P, as a limiting factor for crop production, has to be addressed. As a non-renewable and expensive resource for farmers, P fertilizers have to be thoroughly applied and managed. To overcome soil degradation through enhancing crop productivity, it was necessary to identify the origin of soil P deficiency for crops. Recently, substantial progresses took place in the characterization of the plant-available soil P, especially by understanding and describing major mechanisms driving P release from soil reserves and controlling plant P uptake. The soil diffusive P is generally considered as the main source of phosphorus for plants. This fraction of the total soil P is the amount of phosphate ions of the solid phase that can move, with time, towards the soil solution as a consequence of a P depletion in the soil solution due to plant root uptake. It is the amount of sorbed P than can replenish the soil solution with P. The dynamics of soil diffusive P was determined in batch experiments (soil - solution ratio = 1/10) by coupling sorption experiences with subsequent isotopic labelling of phosphate ions in solution of soil suspension at steady state and applying isotopic dilution principle. Such approach allows to determine the fraction of the P sorbed that can leave, with time, the sorbed phase to reach the soil solution, in fact the P that can come back in the soil solution, under a given gradient of P concentration. The characteristics of the dynamics of the soil diffusive P were determined in two cultivated Ferralsols of Madagascar differing in their substrates and compared to other cropped soils sampled in other countries. The concentration of P, determined as phosphate ions, in the soil solution for cultivated and fertilized soils reached only 0.005 mg P l-1 for the Ferralsol on granite substrate and 0.049 mgP l-1 for the alluvial fine Ferralsol. When applying P, as dissolved phosphate ions, the increase of P concentration in soil solution was very limited. Almost 99.5% of applied P quickly leave the solution and was sorbed on the soil solid phase even with massive addition of P, such as 100 mg P kg-1 soil. corresponding to about 400 kg P ha-1 when considering 4000 t.soil ha-1. The weak increase of P concentration was due to the high number of reaction sites onto high-fixing Al- and Fe- oxyhydroxydes. However, experimental results showed, in ferrasols sampled in Madagascar, that the ability of sorbed P to replenish solution is very high and rapid. The P deficiency of 'tanety' soils is therefore primarily attributed to extremely low P ions concentration in soil solution, which is lower to the threshold required for roots to absorb P (estimated to 0.005 mg P L-1 for maize. To improve plant-availability of P in 'tanety' soils and to maximize efficiency of P fertilizer it is necessary first to reduce soil P fixing capacity through soil management practices