Eco-efficient weed and nutrient management strategies, and modeling rice-weed interactions in mechanized dry-seeded rice

High rice planting density has the potential to suppress weeds, and crop-weed interactions can be exploited by adjusting fertilizer rates and water regimes. The researcher hypothesized that: i) planting density can be manipulated to suppress weed growth, ii) rice competitiveness against certain weeds can be enhanced by increasing nitrogen (N), and iii) water regime can affect rice-weed competition. To test these hypotheses, four screenhouse (2012) and two greenhouse (2012-2014) studies were conducted at the International Rice Research Institute (IRRI), Los Baños, [Laguna] Philippines. To find out the threshold level and critical period of weed competition with rice, two field studies (2013) were conducted. From integrated weed management perspective, the role of herbicides cannot be ruled out. Hence, one greenhouse study (2013-2014) was conducted to determine the effect of irrigation scheduling on efficacy of preemergence herbicides and their phytotoxicity on rice seedlings, and one field study (2012-2013) to evaluate the efficacy and economics of different herbicides. To control weeds through the use of all these cultural and chemical weed control measures, and to find out the most appropriate seed rate and N rate which will enhance the N and solar radiation efficiency and productivity of DSR system, a field study was also conducted at IRRI for three consecutive seasons (two wet and one dry seasons) during 2012-2013. Ischaemum rugosum and Rottboellia cochinchinensis were taller than rice in competition with rice densities of 100, 200, and 400 plants per m sup 2, regardless of N application rates. Both weeds are phenotypically plastic and can alter their morphology and physiology to improve their resource acquisition and to avoid shade. R. cochinchinensis was more responsive to N than rice. Added N favored rice biomass production more than I. rugosum. Under limited N conditions, the plant height of Echinochloa colona and Cyperus iria were completely suppressed by 400 plants per m sup 2 rice density. Application of N improved the phenotypic plasticity in E. colona and C. iria, therefore, these weeds were more responsive to applied N than rice plants. Leptochloa chinensis plants were taller than rice and had better growth in aerobic than saturated conditions. While rice plants were taller than Fimbristylis miliacea in both aerobic and saturated conditions, both rice and F. miliacea were taller in saturated than aerobic conditions. Weed growth and seed production was not affected by N rates, while otherwise for rice growth. Irrespective of N rate and water regime, the biomass of L. chinensis and F. miliacea decreased as rice density was increased from 0 to 640 plants m sup -2. With rice interference under aerobic conditions, L. chinensis increased its leaf weight ratio, inflorescence weight ratio, and specific stem length more than several times than F. miliacea. Rice growth parameters (height, tiller number m sup -2, stem, leaf, panicle, and shoot biomass, and leaf area) and grain yield attributes (panicle density, filled spikelets panicle-1, and average grain weight) were significantly reduced with increased E. crus-galli density, having early emergence time. All these rice growth and production parameters increased with the delay in weed emergence from 2-45 days after rice emergence (DARE). Simulated results showed that weed emergence of varying weed populations after 45 DARE had no effect on rice yield. All preemergence (PRE) herbicides (oxadiazon, pendimethalin, and pretilachlor + safener) provided 100% control of E. colona, E. crus-galli, L. chinensis, C. iria, and Amaranthus spinosus. Murdannia nudiflora was not controlled by pendimethalin. Percent reductions in rice leaf, stem, root, and shoot biomass by different herbicides was in the order of pendimethalin 2 oxadiazon 1 pendimethalin 1 oxadiazon 0.5 pretilachlor + safener 0.6 kg ai ha-1. Herbicide treatments reduced rice growth parameters in the order of soil saturated at 1 3 5 7 days after spray. All herbicide treatments with PRE (oxadiazon and pendimethalin) or early post-emergence (POST) herbicides (thiobencarb plus 2,4-D and butachlor plus propanil) reduced total weed density by 85-100%, and biomass by 80-100%. Late POST (bispyribac-sodium and fenoxaprop plus ethoxysulfuron) treatments, on the other hand, reduced weed density by 32-50%, and biomass by 40-62%. Highest grain yield was achieved in weed-free plots (5.9 to 6.1 t/ha), while the lowest in weedy plots (0.2 t/ha). Among the herbicide treatments, rice treated with oxadiazon, thiobencarb plus 2,4-D, and butachlor plus propanil, followed by the late POST herbicides had grain yield increments of 23-25, 20-26, and 18-23 times than that of the yield in weedy plots, respectively. Sole application of oxadiazon resulted to the highest net profit and benefit-cost ratio in both trials, which was similar to the treatments involving oxadiazon or early POST herbicides, and followed by the sequential application of late POST herbicides. High seeding rate (100 kg/ha) produced higher grain yield (5.99 t/ha) than low seeding rate (50 kg/ha) (5.13 t/ha) in the dry season. In the wet season, high seeding rate produced lower grain yield (4.5 t/ha) than low seeding rate (4.8 t/ha). N rate of 150 kg/ha produced the highest yield component values and grain yield (5.2-6.6 t/ha) compared to 100 kg N/ha (4.7-5.6 t/ha) and 200 kg N/ha (4.9-6.5 t/ha). Higher internal N use efficiency (40-53 kg grain kg/N uptake by plants) was achieved at 150 kg N/ha. Partial factor productivity (kg grain kg-1 N applied) was also highest (41-47 in DS and 32-37 in WS) in plots applied with 150 kg N/ha. However, the best degree of synchrony between crop N demand and supply was also achieved at 150 kg N/ha. Hence, appropriate combinations of rice planting density, N rates, soil moisture status, and weed species-specific herbicides are needed to economically control weeds and reduce the herbicide load on agroecosystems, and enhance the productivity and profitability of DSR systems under varying seasonal conditions.