Development and Validation of Nutrient Expert for Maize in Bangladesh

In South Asia, most nutrient management recommendations for maize and rice are derived from static soil test results and/or agroecological zoning recommendations that do not consider the dynamic nature of soil nutrient supply. The current work represents an initial-step towards developing site-specific nutrient management (SSNM) recommendations for maize in Bangladesh, through the Maize Crop Manager and Nutrient Expert^® for Maize decision support tools. The current work also assesses if there are differences in yield and agronomic efficiency between strip tillage (ST) and conventional tillage (CT).

Split-plot experiments were initiated with CT and ST as main plots, and ample (+NPKZn), +NPK-Zn and -N, -P, and –K omission as sub-plots in a rice-maize rotation in central (Jamalpur) and southern (Barisal) Bangladesh. In CT, tilled maize was rotated with puddled transplanted rice. The ST included ST maize followed by un-puddled transplanted rice, both with 30% anchored residue retention. The +NPKZn rates followed Government recommendations by agroecological zone in Bangladesh. In Jamalpur, rates were 250, 100 and 160 kg N, P₂O₅ and K₂O/ha for maize with N applied in three splits. In Barisal, maize nutrient application rates were 200, 80 and 120 kg/ha, with similar splitting of N as in Jamalpur. The N, P₂O₅ and K₂O application was 100, 20 and 75 kg/ha for rice in Jamalpur; in Barisal, rates were 15 kg/ha lower for P only. For rice and maize, 5 kg Zn/ha was also applied to the +NKPZn treatment and each macro-nutrient omission plot. One additional +NKP plot was maintained without Zn to assess the importance of Zn on yields. In the NPK omission plots, the same rates for each nutrient were used as the +NKPZn plot, but with N, P and K sequentially excluded.

No significant differences between ST or CT yields were observed for the winter season maize or rainy season rice in Jamalpur. However, significant differences in yield were consistently found between the nutrient addition and omission plots, though no tillage system vs. nutrient addition or omission interactions were recorded. This pattern also extended to system (maize+rice) yields. In Barisal, significant differences were observed between ST and CT in maize yields. Lower ST yields may have been caused by poor stand establishment, and by cutworm (Agrotis ipsilon) in un-tilled soil. No differences were found between the nutrient addition and omission plots, and no system by nutrient addition, or omission, interactions were observed. In the rice crop, no tillage system differences were found. The lower preceding maize yields resulted in significantly lower ST system yields. No difference in agronomic N use efficiency were observed under the two tillage system in Barisal or Jamalpur.