Larger bioavailability of soil phosphorus for irrigated rice compared with rainfed rice in Madagascar: results from a soil and plant survey

Irrigated rice is less prone generally to phosphorus (P) deficiency than rainfed rice because redox reactions release P upon soil flooding. It is not known whether that is also true in highly weathered soils of Madagascar where the combination of high soil Fe and low P input may impede significant release of P. Soils and flag leaf samples were collected in 2010 in 38 irrigated rice and 46 rainfed rice fields belonging to private farmers. A critical flag leaf P content was derived from a P‐dosed pot trial study with three soils, and the results suggested 2.4 g P/kg as the critical value. Average flag leaf P was significantly larger in irrigated than in rainfed rice (2.2 compared with 1.7 g P/kg), and flag leaf P was below the critical value in 76% of irrigated rice fields while this fraction was 100% in rainfed rice. Nitrogen and K deficiencies were less prevalent. Flag leaf P increased with increasing soil pH and soil pH explained partially differences in leaf P between irrigated and rainfed rice. Flag leaf P was unrelated to soil organic matter, but increased with oxalate‐extractable soil P (P_o). Multiple regression analysis revealed greater leaf P at equal soil P_o and equal pH in irrigated compared with rainfed rice. Grain yield estimates (1‐m² squares) increased with flag leaf P but not with leaf N and K. In a regression model, about 42 % of the yield variance was explained with soil P_o and a rice‐growing system. The survey suggests that P is the main limiting nutrient for rice, and that soil P bioavailability is larger for irrigated than for rainfed rice in weathered soils of Madagascar.