FALLOW AND SESBANIA EFFECTS ON SOIL-NITROGEN DYNAMICS IN LOWLAND RICE-BASED CROPPING SYSTEMS

Vast areas of rice-growing (Oryza sativa L.) lowlands in Asia are fallowed or cropped with non-rice crops for part of the year. Nitrate can accumulate during the fallow or non-rice crop, but this nitrate can be lost upon flooding for rice production. To determine fallow and green manure crop effects on soil nitrate and ammonium dynamics in lowland riceland, a 2-yr field study was conducted in the Philippines. Treatments before wet season rice were (i) Sesbania rostrata grown for either 45 or 60 d, (ii) weedy fallow, and (iii) weed-free fallow. Sesbania rostrata was sown with irrigation in late April-early May, rains started in early (1989) or mid-May (1990). Weeds and S. rostrata were incorporated after soil flooding on 23 June. Rains increased soil water-filled pore space to above 0.75 mL mL-1 between mid-May and soil flooding. Weeds and S. rostrata assimilated soil nitrate, as evidenced by lower (P < 0.05) nitrate in those treatments than in the weed-free fallow. The decrease in soil nitrate in the weed-free fallow from 24 April to before soil flooding (15 kg N ha-1) was apparently due to denitrification or leaching; additional nitrate (19 kg N ha-1 in 1990) disappeared after soil flooding. Ammonium-N was rapidly released from incorporated weeds and S. rostrata. It reached a maximum by 36 d after incorporation, which correlated (r = 0.95) with N accumulation by rice at 45 d after transplanting. Results suggest that weeds and crops before rice can reduce soil N loss by assimilating nitrate-N and then cycling this N through incorporated plant residues back to the soil where it is rapidly mineralized and used by rice.