During 2006, significant progress was made against all project objectives, involving the collaborative efforts of the researchers from the institutes of CCS Haryana Agricultural University (HAU), Punjab Agricultural University (PAU), Rice-Wheat Consortium (RWC) and the University of Adelaide (UA), to investigate zero-tillage rice establishment and crop-weed dynamics in rice and wheat cropping systems of India and Australia.

Excellent progress has been made in the development of baseline information required for the successful establishment of direct seeded rice (DSR) as a productive and sustainable alternative to conventional hand-transplanted system. Several field sites (n=35) were established across states of Haryana, Punjab and Bihar. These states geographically provide an excellent cross section of the Indo-Gangetic plains in terms of climatic and productivity status. Consultation amongst principal scientists, researchers and growers from each region, identified local issues to be investigated within the proposed objectives of LWR/2004/033. More specifically, key experimental questions addressed in the first year of the project for DSR systems included (a) direct validation of rice productivity in comparison to hand-transplanted puddle system, (b) identification of optimal rice sowing densities and times, (c) evaluation and suitability of different rice genotypes, (d) identification of weed recruitment/dynamics and potential shifts in weed flora, and (e) development of robust weed management strategies.

Comparative evaluation of DSR against the traditional hand-transplanted system showed a yield depression was associated with zero-tillage establishment in Punjab (20 to 73%), and Bihar (23%) sites, respectively. However, optimising sowing time and seeding rates (40-60 kg/ha) of rice significantly reduced the yield penalty associated with zero-tillage (7%). Data from sites in Punjab and Bihar showed that rice yields were statistically non-significant between DSR and transplanted systems when sowing dates were similar for the systems. The maturity profile (short versus long) of rice cultivars was also shown to be an important characteristic to final yield outcomes and requires further investigation.

In addition, grower field sites (n=17) were established in Haryana to evaluate the performance of basmati and coarse grain rice established under zero-till and puddle transplant systems. Preliminary data is extremely encouraging showing similar yields for Zero-till (2275 to 6800 kg/ha) and traditional puddle transplant systems (2325 to 7625 kg/ha). It is expected that ongoing interaction between the researchers and growers on alternative rice production systems will greatly facilitate the on-farm adoption of DSR.

Emphasis was also placed on identifying suitable rice genotypes for DSR, with rice breeding lines and cultivars sourced from local breeding programs. Several genotypes (n=33) differing in maturity (short, medium & long) and vigour were evaluated under DSR and transplanted systems at sites in Punjab and Bihar. It was encouraging to note that some of the advanced breeding lines produced similar yields under DSR and transplanted systems. However, it is important that these data be revalidated in 2007, to show the effects of seasonal variation on these genotype sowing system responses. In 2007, It is anticipated that additional breeding lines will be sourced from the International Rice Research Institute (IRRI) and Indian breeding programs and evaluated for suitability under DSR system.

An experiment undertaken in Bihar investigating weed establishment under DSR and transplanted rice showed that final weed densities (grassy, broadleaf and sedges) were similar for the different crop establishment systems (248 to 457 plants/m2). However, a significant shift in the weed spectrum resulted with fewer grassy weeds (48 to 83 plants/m2) establishing in zero-till DSR as compared to the transplanted system (220 plants/m2), respectively. Significant shifts in weed flora under DSR will have important implications for the management and the potential development of herbicide resistance. History has shown that herbicide resistance development (i.e. Phalaris minor) can be a major constraint to the productivity and sustainability of rice-wheat systems in India. Research undertaken as part of project LWR/2004/033 will closely monitor the resistance status of weeds under DSR.

Experiments evaluating different herbicide options for DSR identified useful mixtures (i.e. azimsulfuron + metsulfuron) for the control of broadleaf and sedge weeds (see attached reports). Useful options for controlling grass weeds in DSR were limited; however, herbicide Bispyribac provided excellent control (90%) of Echinochloa crus-galli (Barnyard grass) at sites in Haryana. High levels of weed control obtained with Bispyribac resulted in a 3-7 fold increase in rice yield compared to the weedy controls. Identification of useful herbicides for DSR will continue in 2007.

Data from Australian experimental work was extensive in 2006, with collection of important information on water and nutrient use efficiency in wide-row (WR) cropping under well below average growing season rainfall. In addition, replicated experiments were established investigating the performance and suitability of different crop species and their cultivars to WR cropping. These experiments have identified cultivars showing greater suitability to WR cropping and have consequently been repeated this year. Field experiments were also established investigating weed behaviour and management opportunities in WR cropping. Preliminary data has shown that WR systems appear to be less conducive to the establishment of problematic weeds such as annual ryegrass (ARG). Furthermore, strategic intra- and inter-row herbicide applications were shown to provide high levels of ARG control, and significant crop yield increases. Future research will be undertaken in farmer fields to investigate weed population dynamics under wide-row zero-till cropping systems.

During 2007, significant progress was made against project objectives involving collaborative efforts of researchers from CCS Haryana Agricultural University (HAU), Punjab Agricultural University (PAU), Rice-Wheat Consortium (RWC) and the University of Adelaide (UA), to investigate zero-tillage rice establishment and crop-weed dynamics in rice-wheat cropping systems of India and Australia.
During the reporting period excellent progress was made against project milestones in the identification and development of suitable establishment systems for direct seeded rice (DRS) as an alternative to the traditional hand transplanting method. Several field sites (n=35) were established across states of Haryana, Punjab and Bihar. Research activities focused on a) direct validation of DSR establishment systems against traditional hand-transplant technique, b) identification of optimal sowing times and seeding rates, c) evaluation and suitability of different rice genotypes, d) identification of weed recruitment/dynamics and potential shifts in weed flora and e) development of robust weed management systems.
Yield responses for coarse grain and basmati rice types were similar under comparative evaluations of DSR against traditional hand-transplant system at sites across states of Haryana, Punjab and Bihar. At 2 of the 3 sites established across Punjab there was no significant difference in yield between rice establishment systems, only at Kapurthala which has a sodic soil, rice yields were significantly higher (44-85%) in the traditional hand-transplant system. In contrast, rice established under zero-tillage at PUSA, Bihar, yielded significantly more grain (9-39%) than the traditional hand-transplant system. Similarly, higher yields were obtained with zero-tillage machine transplanted (8.3 t/ha) rice than the conventional puddled transplanted (7.5 t/ha) treatment at Kurukshetra research station, Haryana.
In addition, grower field sites (n=6) were established in the states of Haryana and Punjab to evaluate the performance of basmati and coarse grain rice established under DSR and puddle transplant systems. The results from grower fields in Haryana have been extremely encouraging with similar yields under DSR Basmati (2.6-2.9 t/ha) and the traditional puddle transplant system (2.4-3.1 t/ha). Preliminary results from sites in Punjab were also encouraging with yields in excess of 7.9 t/ha under DSR. Grower feedback from these results has been extremely positive, with some growers planning to investigate DSR on a larger scale in 2008. Furthermore, continual interaction between researchers and growers on alternative rice production systems is expected to greatly enhance on-farm adoption of DSR.
Experiments undertaken in Punjab and Bihar investigating optimal rice sowing densities and time clearly showed that yields were maximised for DSR when it was sown on the same day as nursery sowing for the transplant system, and when seeding rates were maintained at between 45-60 kg/ha (150 plants/m2). Furthermore, increased seeding rates were also shown to greatly improve the rice crops competitiveness with weeds such as Echinochloa and Panicum spp., resulting in significant reductions in weed growth (33-35%) and improved crop yields (27-62%).
For the second year, experiments were undertaken in Punjab and Bihar to assess the performance of several rice genotypes (n=44) under DSR and comparative puddle transplant systems. The evaluation consisted of rice breeding lines, hybrids and cultivars, sourced from International Rice Research Institute (IRRI) and local Indian breeding programs. It was encouraging to see that several of the advanced lines, produced similar yields under DSR and transplant system. Breeding line 17A/R10, sourced from IRRI performed well under DSR (5.7 t/ha) in comparison to transplant system (5.2 t/ha) at PAU. Furthermore, hybrid genotypes evaluated in Bihar performed particularly well under DSR producing some of the highest grain yields (4.6-5.6 t/ha) in the study. The performance of basmati genotypes under DSR in Bihar is also noteworthy, with 5 of the 7 genotypes evaluated under this system producing equivalent yields to the puddle transplant system. It is anticipated that further genotype screening will occur in 2008, with additional breeding lines sourced from IRRI and local Indian breeding programs.
An experiment undertaken in Haryana, investigating weed establishment under puddled and unpuddled DSR and transplanted rice showed significant shifts in weed spectrum (grassy, broadleaf & sedges) under different crop establishment systems. Clear associations were found with Echinchloa, Leptochloa and Fimbristylis spp. showing greater establishment under DSR in comparison to the transplant system. Furthermore, there was greater establishment of Echinochloa under puddled conditions and Leptochloa under unpuddled conditions. In contrast, Cyperus difformis which is a sedge showed clear preference for the unpuddled transplant system. These clear differences in weed spectrum with the different rice establishment systems require further investigation. Studies are needed to investigate the effect of crop establishment system on the behaviour of weed seedbank. In June 2008, it is anticipated that studies on weed seedbank behaviour will be undertaken following a visit from the Australian project research officer (PRO) based in Adelaide.
Useful herbicide options identified in 2006 for controlling weeds in DSR (i.e. azimsulfuron and bispyribac) were evaluated further at sites in Haryana, Punjab and Bihar. Research focused on optimising herbicide application rates, timing and mixtures. Herbicide bispyribac continued to provide excellent control of grass (Echinochloa spp.) and sedge weeds (Cyperus spp.) at sites across all 3 states. This herbicide was particularly effective when applied at higher rates (25-30 g/ha) and early application timings (15-20 DAS). However, in an on-farm study at Kolhar bispyribac provided effective control of E. crusgalli even at the panicle emergence stage of the weed. Azimsulfuron provided effective control of broadleaf weeds, particularly when used in combination with metsulfuron. Importantly, there appears to be good selectivity with these herbicides in rice, with high levels of weed control resulting in increased grain yield. Further evaluation of these herbicides is planned for 2008.
Experiments undertaken in Australia provided second year of comprehensive data on water and nutrient use-efficiency in wide-row (WR) cropping. In addition, experiments evaluating the suitability of different crop species and their cultivars to WR were repeated. In 2007, additional crop types canola and chickpeas and their cultivars (n=6) were included for evaluation under WR systems. These experiments have been extremely useful in identifying cultivars with greater suitability to WR systems. Furthermore, positive results from these trials have prompted significant interest amongst the farming community with some growers adopting WR for the first time. Field experiments were also undertaken to investigate weed behaviour and management under WR. Results have again shown that WR systems significantly reduce the establishment of problematic weeds such as annual ryegrass (ARG). Furthermore, experiments established that ARG can be successfully controlled in WR systems with strategic intra- and inter-row herbicide applications. Research activities planned for 2008 will focus on investigating weed dynamics and management under WR cropping systems.

During 2008, significant progress was made against all the objectives in the project "zero-tillage rice establishment and crop-weed dynamics in rice-wheat cropping systems of India and Australia."
Excellent progress has been made against project milestones in the identification and development of suitable crop establishment systems for direct seeded rice (DSR) as an alternative to the traditional hand-transplant system. Several field sites (n=84) have been established across states of Haryana, Punjab and Bihar, with more than 200 acres demonstrating DSR technology. Research activities in the third year focused on a) comparative assessment of DSR establishment systems against hand-transplant, b) identification of optimal sowing times and seed rates, c) evaluation of suitability of different rice genotypes for DSR, d) identification of weed recruitment/dynamics and potential shifts in weed flora and e) development of effective weed management strategies.
Results summarising the yield responses of coarse grain and basmati rice types to DSR and traditional hand-transplant system across sites in Haryana, Punjab and Bihar are presented in the attached reports (appendix 1). Rice established under zero-tillage at Kushmahot, Bihar, in the presence or absence of residue yielded significantly more grain (9-13%) than the traditional hand-transplant system. Similarly, higher yields were obtained with ZT machine transplanted rice (5-8 t/ha) than the conventional transplant approach (4.2-7.3 t/ha) at several grower sites (n=196) across Haryana. DSR yields in Punjab were also similar to the traditional practice of puddled hand transplanted rice.
During the reporting period significant efforts were made to demonstrate DSR technology with the more than 320 grower assisted sites established in Haryana and Bihar. Major findings from years 1 and 2 of the project relating to optimum sowing time, seed rates, varieties, nutrition and weed management have been used to develop agronomic packages for DSR. The results from grower fields have been extremely encouraging with significant yield gains for coarse grain (200-400 kg/ha) and basmati rice types (100-200 kg/ha) under DSR in comparison to the traditional hand-transplant system. Furthermore, grower feedback about these results has been extremely positive and farmer awareness of the potential benefits of DSR is now very high. For example, grower interest in the ZT machine transplant system has been so strong that government funds have been made available for the purchase of 30 additional transplanters for use in Haryana in the coming season. This system allows growers to plant directly into unpuddled conditions and has clearly demonstrated the benefits of DSR with significant savings in labour, fuel and water.
Experiments undertaken in Punjab for the third year clearly showed that yields for rice were maximised for DSR when it was sown on the same day as nursery sowing for the transplant system, prior to onset of monsoon and when seeding rates were maintained at between 45-60 kg/ha for coarse grain rice (150-200 plants/m2). Based on project research, recommendations on the agronomy of DSR (optimal sowing rates, time of sowing, weed control) can now be given with confidence. Furthermore, high seeding rates were shown to improve the competitiveness of the rice crop with grass weeds Echinochloa and Panicum spp., resulting in significant reductions in weed growth (41-48%) and improved yield (48-75%). In addition, experiments undertaken in Punjab and Bihar clearly showed that fertiliser rates of between 120-180 kg N/ha are required to achieve maximum yield potential under DSR.
Field experiments were undertaken in Punjab and Bihar to assess the performance of rice genotypes under DSR relative to the puddled transplant system. The evaluation consisted of rice breeding lines, hybrids and cultivars sourced from International Rice Research Institute (IRRI) and local Indian breeding programs. In Bihar, genotypes showed similar yield responses irrespective of establishment system, however, cultivar Rajendra Mahsuri and hybrids ProAgro 6444 and PHB 71 produced higher grain yields (6.4-6.6 t/ha) under DSR than the transplant system. Research at PAU showed that hybrid RH 257 was the most weed competitive variety under DSR, with only 12.3% yield loss as compared to 28-59% for other varieties. Rice hybrids have consistently performed well under DSR and appear to be well suited to this establishment system. It is anticipated that further genotype screening will take place during 2009.
Experiments undertaken in Bihar and Haryana, investigating weed recruitment behaviour under different rice establishment systems clearly showed shifts in weed spectrum (grassy, broadleaf and sedges) under DSR. These studies have consistently shown that Echinochloa has the ability to infest both DSR and puddled transplanted crops with slightly higher densities under DSR. Leptochloa on the other hand, only becomes a weed of significance under DSR. Consequently, herbicide evaluation has been undertaken and promising new herbicides identified for the control of this weed species. In contrast, sedge Cyperus difformis was more prevalent under the traditional puddled hand-transplant system. Broadleaf weed spp. Ammania and Fimbristylis showed no association with either establishment system, however, Bracheria, was more prevalent under DSR in Bihar. A recent visit from the Australian project research officer (PRO) in June, provided opportunity to demonstrate seedbank sampling methodologies, which is anticipated to assist future research and understanding of the behaviour of these important weed spp.
Several experiments were undertaken in Haryana, Punjab and Bihar to evaluate herbicide options for DSR (i.e. bispyribac and azimsulfuron) with research focusing on optimising herbicide application rates, timing and mixtures. Herbicide bispyribac continued to provide consistent control of grass (Echinochloa spp.) and sedge (Cyperus spp.) weeds under DSR across all three states. In addition, several grower sites (n=32; 111 acres) were established across 6 districts in Haryana to verify performance of bispyribac as a recommended (i.e. 25 g/ha applied 15-25 DAS) weed control option for DSR. These demonstrations clearly showed growers the ability of this herbicide to provide high levels of grass (ave. 96%) and sedge weed control (ave. 76%) and gives them confidence to pursue DSR with the knowledge that an effective weed management tool is available. In addition, azimsulfuron provided effective control of broadleaf and sedge weeds (>70%) when applied at higher rates (30-35 g/ha), early application timings (15-25 DAS) and in tank mixtures with metsulfuron. Importantly, bispyribac and azimsulfuron have consistently been shown to provide effective and safe weed control in DSR, resulting in significant grain yield increases. As a result of project research and other associated work, bispyribac has been registered for rice production systems (DSR & transplant) and is to be released to growers in 2009.
As in the previous two seasons, comprehensive data was collected in 2008 on crop water use under wide-row (WR) cropping systems in Australia (see attached report). In addition, experiments were undertaken to evaluate the performance of canola and chickpea cultivars (n=6) to WR. These experiments successfully identified cultivars with greater suitability to WR and have provided growers with confidence to pursue these systems for yield and weed control benefits. Field experiments were also established to investigate weed behaviour and management opportunities under WR systems. Results have clearly shown that WR systems reduce establishment of problematic weeds such as annual ryegrass (Lolium rigidum) and provide an excellent opportunity for weed control with shielded herbicide applications in the intra- and inter-row zones.

During 2009, excellent progress was made against all the objectives in the project "zero-tillage rice establishment and crop-weed dynamics in rice-wheat cropping systems of India and Australia".
Excellent progress has been made against project milestones in the identification and development of suitable establishment systems for direct seeded rice (DSR) as an alternative to the traditional hand-transplant system. In 2009 several field research sites (n=31) were established across states of Haryana, Punjab and Bihar. In addition to the trial sites, more than 581 farmer sites demonstrating DSR technology were also established. In Haryana alone, more than 350 acres was sown under DSR in 2009, with the area expected to rise substantially next year.
In fourth year of the project, research activities focused on a) direct validation of DSR establishment systems against hand-transplant, b) identification of optimal sowing times and seeding rates, c) evaluation and suitability of different rice genotypes, d) identification of weed recruitment/dynamics and potential shifts in weed flora, e) development of effective weed management strategies, and f) optimising nitrogen management under DSR.
Yield responses for coarse and basmati rice types under DSR and the traditional hand-transplant system at sites across states of Haryana, Punjab and Bihar are summarised in the attached reports (appendix 1). At the two field sites comparing rice establishment systems in Punjab, there was no significant difference between DSR and hand-transplant system, with grain yields as high as 6.4 t/ha for rice sown using modified seed drills. Rice established under zero-tillage and residue at Samastipur, Bihar, showed higher yields (5.4 t/ha) than the puddle-transplant system (5.1 t/ha) which is an exceptional performance given the dry conditions experienced over the growing season. Similarly, DSR evaluated across several farmer sites (n=354) in Haryana provided equivalent rice yields to conventional transplant system, with ZT machine transplanted rice yielding on average 5.6% more grain than the traditional system. As this work was undertaken in close association with the farmers, their awareness of the benefits of this technology is high which is expected to lead to its rapid on-farm adoption.
The project continues to raise awareness and interest amongst researchers, extension workers, government officials and importantly growers of the potential of alternate rice establishment systems. This was clearly evident from attendance by more than 200 farmers from all over Punjab at a national workshop on DSR at Punjab Agricultural University in September 2009. This workshop was organised by the PAU team of this ACIAR project. Furthermore, efforts to demonstrate DSR and its benefits were significantly increased during the reporting period with more than 581 grower assisted sites established in Haryana and Bihar. These field sites, established using agronomic packages developed within this project, have been extremely encouraging showing similar yields for both coarse and basmati rice types under DSR (2.8-6.4 t/ha) in comparison to the traditional puddle transplant system (3.2-6.4 t/ha). Grower feedback from these results has been extremely positive, with many growers reporting water savings of as much as 30% under DSR as well as savings in labour required for hand transplanting. In addition, considerable savings (3500 to 5000 rupees/ha) were reported for machine transplanted rice in comparison to the labour and puddling intensive hand-transplant system, further encouraging many growers to shift to DSR in 2010.
Experiments undertaken in Punjab last year clearly showed that yields for rice were maximised for DSR when it was sown on the same day as nursery sowing for the transplant system. Delays of sowing DSR beyond 15 days of nursery establishment for PTR resulted in significant losses in yield (26-55%). These studies have been critical in developing the management package for DSR. Research has also shown conclusively that DSR needs to be sown prior to the onset of monsoon, as rice is very sensitive to water-logging during germination, a key finding from Bihar. Improvements in the seed delivery mechanisms of seed-drills by the manufacturers during last two years has resulted in a significant reduction in the seed rate required for coarse grain (20-30 kg/ha) and basmati rice (17-20 kg/ha) without compromising crop establishment (150 plants/m2) and yield. In addition, experiments undertaken showed that fertiliser rates of between 120-180 kg N/ha split into three equal applications (i.e. 33% sowing, tillering and panicle initiation) are required to achieve maximum yield potential under DSR. Further research on drum seeder for DSR establishment has been discontinued as this practice is incompatible with the use of pre-emergence herbicides such as pendimethalin.
Identification of suitable rice genotypes (coarse grain and basmati) for DSR has been a major objective of the project and was undertaken at Punjab and Bihar. These evaluations consisted of rice breeding lines, hybrids and cultivars, sourced from International Rice Research Institute (IRRI) and local breeding programs. At Bihar under drought conditions, it was extremely encouraging to see that many of the rice genotypes were able to produce similar yields under DSR (3.8 t/ha) in comparison to the transplant system (4 t/ha). Furthermore, hybrid genotypes (i.e. Arize dhan and Arize 6444) were able to produce yields in excess of 5.8 t/ha in the study and continued to show excellent potential under DSR. In addition, yield responses for seven different basmati cultivars grown on farmer assisted sites (n=354) in Haryana were consistently higher (4-7%) under DSR than the transplant system and indicates the suitability of already released cultivars for this alternative establishment system.
Research undertaken in Haryana and Bihar has identified significant shifts in weed spectrum (grass, broadleaf and sedge) under the different rice establishment systems. As in previous years, clear associations were found with grass spp. Leptochloa showing greater preference for DSR and sedge Cyperus difformis for the transplant system. However, grass spp. Eragrostis was identified as a new emerging weed threat with significantly greater emergence under DSR. In contrast, Echinochloa spp. showed no association with either establishment system and was found infesting both DSR and puddle-transplanted crops at similar densities.
As a consequence of the significant threat posed by aerobic grass spp. Leptochloa and Eragrostis to success of DSR, herbicide options targeting these and other prominent species (i.e. Echinochloa) were evaluated further at sites in Haryana, Punjab and Bihar. Previous research clearly showed potential of new herbicides bispyribac and azimsulfuron to provide effective and consistent control of Echinochloa (>90%) and sedge spp. (>70%) under DSR, and resulted in registration and release of bispyribac (Nominee Gold) for rice production systems in 2009. Consequently, bispyribac was widely used by growers at many of the 581 grower assisted sites last year. In addition to providing excellent control of Echinochloa spp., bisypribac (25 g/ha applied 25 DAS) used in combination with pre-emergence pendimethalin appears to provide effective broad spectrum control of weeds including Leptochloa and Eragrostis (>90%) under DSR and is a significant project finding. Importantly, these herbicides provide excellent and safe weed control in DSR, and have resulted in significant grain yield increases. Effective herbicide recommendations and strategies for weed control under DSR have now been developed as a consequence of project research, providing growers with greater confidence to pursue DSR.
Herbicide application technology was identified as a weakness in weed management during the mid-term review of this project by Dr Christian Roth. In response to that finding, Sam Kleemann (Australian project research officer - PRO) undertook manufacture of an herbicide spray unit for applying herbicides uniformly and safely in the field (see appendix for images). Evaluation of the spray unit at PAU during 2009 has shown promising results. In response to that success, CIMMYT-India is building 10 replica units for use across Bihar this year and machinery manufactures (i.e. National Agro Industries) in Ludhiana, Punjab, have begun construction of similar units using locally made parts.
Experiments undertaken in Australia have continued to focus on the development of effective weed management strategies and improvements in water use efficiency for farming systems of southern Australia (see attached report). More specifically, experiments were established (n=3) evaluating seeding system, time of sowing and crop-topping effects on weed management and crop safety. These experiments clearly showed that performance of new pre-emergence herbicides (i.e. Boxer-Gold and Sakura) was not compromised under low soil disturbance disc systems or by early sowing. However, some of the weed population was able to escape these new herbicides and set large amount of seed (~5000 seeds/m2). Use of crop-top applications of glyphosate in wheat treated with these herbicides was able to reduce annual ryegrass (Lolium rigidum) seed production by 95%. However, crop safety from crop-topping is very sensitive to timing. Spraying even 2-3 days too early can lead to large yield losses (>50%) in wheat. Follow up work will be undertaken to identify suitable timing for crop safety during crop-topping wheat for seed set control in ryegrass.