YIELD TRENDS OF A LONG-TERM NPK EXPERIMENT FOR INTENSIVE RICE MONOCULTURE IN THE MEKONG RIVER DELTA OF VIETNAM

A long-term experiment to study the effects of nitrogen (N), phosphorus (P) and potassium (K) in intensive rice monoculture commenced at the Cuu Long Delta Rice Research Institute (CLRRI), Omen, Vietnam, in 1986. Two lowland rice crops, transplanted and irrigated, of cultivar IR64 were grown each year, one in the wet season (WS) from May to August, and the other in the dry season (DS) from December to March. The soil is an Fluvaquentic Humaquept representative of much of the double-cropped rice area in the Mekong Delta. The eight treatments were N-P-K fertiliser at the following rates in kg element ha(-1): 0-0-0 (check), 80-0-0, 0-17.5-0, 0-0-25, 80-17.5-0, 80-0-25, 0-17.5-25, and 80-17.5-25 in a randomised complete block with 4 replications. Mean grain yields 1986-1994 ranged from 2.1 t ha(-1) for nil fertiliser in the WS to 5.9 t ha(-1) with N + P in the DS. Wet season yields showed significant decline with time. The most rapid decline was 376 kg ha(-1) year(-1) from an initial level of 3.9 t ha(-1) for the treatment of N without P. Dry season yields of the N + P treatments increased at a rate of 293 kg ha(-1) year(-1) to over 5 t ha(-1). Nitrogen fertiliser-use efficiency declined rapidly in WS, even in combination with P. There was no significant benefit from K throughout. The main yield component reduced in WS was the number of grain per panicle. Phosphorus in combination with N returned up to 180 kg grain per kg P for WS and around 80 kg ha(-1) in DS. Summed over all 16 crops, the application of N + P returned an additional 26 t grain ha(-1) over the check or treatments with N alone. This was after deduction of the cost of fertiliser as grain equivalent. Phosphorus is a key factor for productivity in this environment, and if adequately supplied can prevent WS yield decline.