Wet rice organic weed control trials

Project Overview

Project Type: Farmer
Funds awarded in 2015: $13,128.00
Projected End Date: 02/28/2018
Grant Recipient: Boundbrook Farm
Region: Northeast
State: Vermont
Project Leader:
Erik Andrus
Boundbrook Farm

Annual Reports

Information Products

NERICE powerpoint (Conference/Presentation Material)


  • Agronomic: rice
  • Animals: poultry


  • Crop Production: crop improvement and selection, cropping systems, water management
  • Education and Training: extension, farmer to farmer, workshop
  • Farm Business Management: agricultural finance, budgets/cost and returns, business planning, new enterprise development
  • Natural Resources/Environment: biodiversity, drift/runoff buffers, habitat enhancement, soil stabilization
  • Pest Management: cultural control, physical control, weeder geese/poultry
  • Production Systems: agroecosystems, integrated crop and livestock systems

    Proposal summary:

     Japonica rice grown in flooded paddies is crop being considered as an alternative crop by an increasing number of growers. Takeshi and Linda Akaogi released their Rice Growing Manual for the Northeast in 2009 and have been hosting the Rice in the Northeast conference since then. A large number of growers have begun to experiment with rice as a result and a handful are seeking to advance this crop to the commercial production stage.  For rice is to be advanced to the commercial level, significant new questions of management techniques must be addressed. Boundbrook Farm has been working with wet rice since the completion of the Akaogi's SARE-funded research and we have made many strides towards a viable commercial model for operations ranging from .5 to 10 acres of paddies.  We have identified weed control as being an issue of particular significance. In the aquatic paddy environment, aggressive wetland sedges and grasses easily outcompete slow-growing rice. We have ruled out herbicide as a solution, and seek to develop replicable practices for managing weeds in rice paddies on a small commercial scale. We will trial the effectiveness of weed control with ducklings (a common Japanese organic method), with floating cover (azolla filiculoides) and through hand control both in a continuously flooded paddy and in a pulse-flooded paddy, using a minimal-irrigation and mechanical weeding strategy known as SRI (System of Rice Intensification).  Establishing an ecologically-sound solution to weed control that uses modest amounts of labor would raise the profile of rice and give the new community of interested rice growers (about 100 attended the aforementioned conference in 2014) the confidence necessary to increase their rice-growing activities. We will reach this community with our findings through presentations at NOFA and the Rice in the Northeast Conferences and through online publication.  

    Project objectives from proposal:

    Our objective is to perform trials of the most promising techniques of ecological weed control. Over the course of two years we aim to evaluate the comparative costs and benefits of weed control using multiple strategies, measured against a control. We aim to measure the labor and other associated costs of weed control methods against gains in productivity.

    These trials must take place over two consecutive growing seasons to gauge the impact of fallowing prior to planting. Fallowing has an entailed cost in that a season of potential production is forfeited, and we aim to assess whether the benefits outweigh that cost.

    Additionally we will trial the effectiveness of weed control with ducklings (a common Japanese organic method), with floating cover (azolla filiculoides) and through hand control both in a continuously flooded paddy and in a pulse-flooded paddy, using a minimal-irrigation and mechanical weeding strategy known as SRI (System of Rice Intensification).

    Following transplanting, all but the pulse-flooded trial plot will all be kept flooded with 4 inches of water until the ripening phase begins (August) but will be managed for weed control using seven different methods.

    Plots 1 and 8 (Control #1) will receive no weeding. Reasonable effort will be made to prevent azolla from entering the plot, using a plastic barrier.

    Plots 2 and 9(Control#2) will also not receive azolla but will be hand weeded until canopy is established (mid-July).

    Plots 3 and 10 (Azolla#1) will be seeded with azolla immediately after transplanting but will not be hand weeded.

    Plots 4 and 11 (Azolla#2) will be seeded with azolla immediately after transplanting and will be hand weeded until canopy is established.

    Plots 5 and 12(Ducks#1) will receive 4 ducklings (approximately the stocking density recommended by Takao Furuno) one week after transplanting but will not be seeded with azolla or be given manual weed control

    Our array of rice paddies at Boundbrook farm consists of 5.5 acres of growing area leveled for irrigation and surrounded by berms. The trial plots will be set up in three separate but adjacent paddies known to be largely uniform in their soil properties. All have clay soils with about 10-12% organic matter and have been flooded for rice with heavy weed pressure since 2012. We will label them A, B, and C. An experiment site plan is attached.

    Since the integration of fallowing or rotation seems so significant in that it can deplete weed seed reserves and/or disrupt weed life cycles, we have designed this as a two year trial. In 2015, rice will be grown in our normal way (flooded, with ducks and azolla) in Paddy A, while Paddy B and Paddy C will be subjected to a wet and a dry fallow respectively.

    Throughout the 2015 growing season, Paddy B will be maintained in a wet bare fallow state. No crops will be planted, but the paddy will be flooded with over 4 inches of water and puddled at regular intervals throughout the season to destroy emerging weeds. Concurrent with this activity, this paddy can also be used to propogate azolla and to house surplus ducks.

    Also in 2015, Paddy C will be maintained in a dry state. We will plant consecutive crops of buckwheat as a green manure and till them in at 6-week intervals beginning in May 2015 with the last planting to take place around September 15th.

    In 2016, a series of 14 trial plots each measuring 500 square feet will be laid out in each of Paddies A, B, and C, plus one extra 1000 sf trial plot in Paddy C for a total of 43 trial plots. Using our standard practices, we will prepare all plots with spring wet tillage (known to rice farmers as "puddling") and 1000 lbs of dehydrated chicken manure to the acre. The trial plots will be laid out in the prepared area measuring 25 feet by 20 feet each, with seven weed control regimens repeated twice in each paddy in order to control for soil variation. All will be planted with healthy seedlings of Matsumae, providing sufficient seed can be secured, Akitkomachi if not. Transplanting takes place late May or early June.

    Plots 6 and 13 (Ducks #2) will receive 4 ducklings one week after transplanting and will also be seeded with Azolla but will have no manual weed control

    Plots 7 and 14 (Ducks#3) will receive 4 ducklings as in plots 5 and 6 and azolla as in plot 6 and will also have manual weed control as necessary until canopy formation.

    This series of 14 plots will be repeated with randomized layout in paddies A, B, and C to assess the impact of prior-year dryland green manuring and wet fallowing relative to simply following wet rice.

    Additionally, in Paddy C, we will arrange a minimal-water trial using pulse irrigation in a separate 1000 sf plot. Pulse irrigation is not compatible with the use of ducks or azolla so those variables are excluded. SRI (System of Rice Intensification) is a model using only minimal pulse irrigation and hand/mechanical weeding. It seems most suitable to site an SRI-method plot in a zone that follows dry crop land use, so we are locating this only in Paddy C, which will be subdivided with an internal earth berm to house the SRI pulse-irrigation trial as well as the fully-flooded plots 1-14 as described above.

    The SRI trial plot will be pulse irrigated only from transplantation onward. Each irrigation pulse will include just enough water to saturate soil. Weeds will be controled manually and mechanically until canopy formation.

    Our chief concerns are the cost of our inputs, chiefly labor hours but also including the cost of buying or raising ducklings, equipment hours, and fencing supplies and food for weeder ducks. We will carefully tally the respective costs of each method with particular attention to labor input requirements. The plots are sized at 500 square feet to have a larger sample size crop area and a more accurate projection of extended costs than possible in, for instance, a 10 x 10 plot. Also, methods like inter-row cultivation and duck weed control do not work as well in very tiny plots and would likely produce skewed results without adequate economy of scale.

    Naturally we are also concerned with productivity. We will thresh grain from each plot separately and weigh the resulting total yield. Note that certain of the plot methods, like the ones containing ducks, will add fertility during the growing season. Since this bonus fertility is simply “part of the package” with ducks no attempt will be made to quantify the bonus fertility itself, but any resulting effects may affect the final crop weight.

    The final report will assess the costs and benefits of these approaches in order to craft a method that combines reasonably low input costs and satisfactory crop performance.

    The first steps in the project will take place in the beginning of the 2015 growing season. In Paddy A, wet rice will be planted and grown as normally done on our farm, in a flooded setting, probably with ducks present from early June to mid-July. This work will be done by the farmer and project manager, Erik Andrus, with support from a farm worker. This normal, unbudgeted, day-to-day work sets the stage for trials in Paddy A in 2016.

    Farmer Erik Andrus will wet-fallow Paddy B by tilling with a tractor-mounted rototiller in flooded conditions beginning in May with repeat tillages every two weeks or so, as necessary to ensure that weeds are tilled in before they can flower. This is budgeted work that sets the stage for trials in Paddy B in 2016.

    Farmer Erik Andrus will plant and incorporate green manure in Paddy C by tilling in dry conditions (with a tractor-mounted rototiller) and broadcast seeding buckwheat. This green manure crop will be incorporated and a new planting made on 6-week intervals. The last planting will be left standing over the winter and incorporated in spring 2016 at the onset of trials.

    In April or May 2016 all plots are prepared for planting by Erik Andrus and a farm worker by puddling. Seedling care is concurrent with field preparation. Transplanting in all plots takes place late May or early June, and is done by hand and by machine.

    Following transplanting, the forty-three plots will be cared for by Andrus and the farm worker, according to the regimen described above. Erik Andrus will ensure documentation of all steps performed. Weed control is rendered largely moot by the 3rd or 4th week of July, when either the rice plants form a full canopy to fill empty space or become permanently stunted and unproductive due to weed pressure. Therefore the focus of the entire project is the weeks between the beginning of June and the end of July 2016.

    The weeks between the end of weed control efforts and the onset of harvest season are lighter in labor requirements. Farm personnel remove ducks and begin to dry off paddies. Pest deterrents such as fishing line and mylar tape and balloons are deployed as necessary. In late September and early October harvest is performed by Andrus and the farm worker. Each plot will be harvested separately with a farm-owned mini-combine and the weight of rough rice taken right in the field with a platform scale.

    Following completion of trials the data will be compiled and evaluated, and conclusions formed. Erik Andrus will be responsible for the writing of a final report and for giving presentations, to take place from Winter 2016 until August 2017. We will also ensure that other aspects of our rice farming, such as seedling care and post-harvest processing, are well-enough described for farmers to understand the context of commercial-scale weed control in rice.

    We will host a farm field day during the trial, between June and mid July 2016 and will incorporate this into NOFA-VT's summer on-farm workshop series. We will also announce this event well in advance through the Rice in the Northeast network.

    Following completion of the trial, Erik Andrus will publish a report of approximately 30 pages and will print hard copies in booklet form for distribution at conferences, and will make it available as a SARE information product.

    The Rice in the Northeast conference at Akaogi farm is now an annual event for a regional audience of both current and prospective growers and is the logical forum to address both groups. Because the 2016 fall harvest must be completed before a final assessment can be made this must be scheduled in August 2017.

    Additionally, we will circulate our report through UVM Extension, and will propose to present at the winter conferences of NOFA-VT and MOFGA in early 2017.

    Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.