Fate of applied fertilizer phosphorus in a highly weathered sandy soil under lowland rice cropping, and its residual effect

Phosphate fertilizer use is expected to increase with the intensification of rainfed lowland rice production on highly weathered sandy soils in Indochina, but little is known about the residual effects of P fertilizer to be expected under such conditions. We here measure residual effects of P over four consecutive cropping seasons in which rice straw was removed from plots on a sandy Plinthustalf near Phnom Penh, Cambodia. Six combinations of P fertilizer applications at 16.5 kg P ha(-1) were made: either to no crops, to all crops, or to the first, second, third or fourth crop only, and the responses of the crops and soil P fractions quantified. Freshly applied P increased rice grain yield by 95%. In the first and second crops using residual P fertilizer, yields increased by 62 and 33% relative to the nil-P plot. Grain yields in the third crop using residual P dropped to levels obtained in the nil-P soils. After one crop, there were substantial increases in levels of inorganic labile P (NaOH-Pi), labile organic P (NaOH-P.) and occluded P (residual P) from the fresh P fertilizer application. By contrast, the resin-extractable P and H2SO4-P Pools were relatively small and after harvesting the first crop relatively unresponsive to P fertilizer application. The decline in the value of residual P fertilizer value to succeeding crops was reflected in decreasing levels recovered from all the soil P pools.,P Cumulative removal of P in four successive rice crops accounted for 30 and 55% of the 16.5 kg ha(-1) in the form of harvested grain and whole plants. One-third of the applied P was recovered in the first crop, a further 15 and 6% were recovered in the first (R-1) and second (R-2) crops using the residual P fertilizer, whereas, in the third crop (R-3) the recovery did not differ from the nil-P soil. These results suggest that P needs to be re-applied at about 17 kg P ha(-1) every two crops on acid sandy lowland soils to maintain the grain yield at about 2.5-3.0 t ha(-1) when rice straw is removed from fields. The mechanisms underlying the decline in residual value of P, including P turnover from rice straw and long-term reactions of soil P forms under rainfed conditions in highly weathered sandy soils warrants further investigation. Further studies should examine the consequences of returning all rice straw to the fields for residual P and P turnover. (C) 2002 Elsevier Science B.V. All rights reserved.