A Comparison of Organic Weed Management Strategies in Onions

Bryan Brown, Ph.D. Student, and Eric Gallandt, Associate Professor

 What’s your strategy for managing weeds? Cultivate until the crop is large enough to tolerate late-emerging weeds, sometimes returning to harvest from a dense patch of weeds? Cultivate season long and pull any mature weeds as part of a longer-term strategy to prevent weed seed rain and make weeding easier and less costly over time? Intensively mulch to prevent weeds, perhaps improve soil quality, and reduce labor demands for weeding later in the season?

There are successful organic farmers who rely on each of these strategies, some using different strategies for different crops, others with a singular focus. Clearly there is no “best” strategy, but rather, trade-offs and compromises associated with each. Our aim with this field study comparing weed management systems is to quantify multiple dimensions of each system so farmers can evaluate and choose a strategy that is best aligned with their own philosophy, priorities and infrastructure constraints.

which strategy

Using yellow storage onions as our test crop (planted with two onions per hole, spaced 6” apart within rows and 3 rows per bed) on a field with a moderate weed seedbank at the University of Maine Rogers Farm, we implemented several prominent strategies:

1) Critical Period Weed Control (CPWC) – Control weeds only during the crop’s sensitive adolescent stage. This is the minimum amount of weeding you can do and still get a viable crop. However, it allows late-season weeds to go to seed.

2) Zero Seed Rain (ZSR) – Frequent cultivation with the goal of not letting any weeds set seed so that none “rain” to the ground. A strategy expected to be initially costly, but with decreasing cost over time as weed pressure declines.

3) Black Plastic Mulch (BPM) – Suppresses weeds and warms soil. Requires cultivation for the paths.

4) Black Plastic Mulch with Straw-Mulched Paths (BPMSP) – Suppresses weeds in the path as well; added organic matter in the paths.

5) Straw Mulch – Suppresses weeds and adds organic matter to the soil. Applied by hand in June after soil has warmed and onions are approximately the diameter of a pencil.

6) Junk Hay Mulch – Similar to straw but less expensive.

  Aside from primary and secondary tillage and application of plastic mulch, all activities were done by hand. Cultivation was achieved by wheel hoeing the paths, scuffle hoeing the shoulders and between rows, and using short-handled hoes for within rows. Drip irrigation was used to keep soil moisture levels optimum for each plot.

labor by activity

Labor by Activity from our 2014 field season (above) demonstrates that CPWC plots required the least amount of labor. Although they were weeded clean through early July, by the end of the season these plots were a weedy mess, and resultantly, had the longest harvest times. In the spring of 2015 we will collect soil samples to see how much weed seed was added to the seedbank.

In ZSR plots, weeding events took place about every ten days in the early- and mid-season, depending on weather, and less often later in the season as weed germination slowed.

Plastic-mulched plots required three hand-weedings to control the crabgrass coming through the planting holes, suggesting that plastic might be better suited to crops with wider spacing. Also, transplanting by hand took longer in plastic mulch. Soil temperatures under the black plastic were consistently 5-10 degrees Fahrenheit higher than the others. Onions under black plastic matured several weeks earlier, which may have contributed to the decreased yield. The BPMSP required the lowest amount of weeding labor of all strategies.

We used high quality oat straw mulch that didn’t have any weed seed but it did bring in a lot of oat seed (112 seeds/lb) that germinated within the mulch and forced us to hand pull twice. The straw mulch was much easier to apply than the junk hay, which stuck together. The junk hay mulch brought in a lot of weed seed (170 seeds/lb) but few weeds emerged through the mulch.

Not surprisingly, at the end of the season, mulched plots had less compacted soil and better water infiltration than unmulched plots. Plots with organic mulch had more earthworms than the others.

After harvest, the onions were cured in a greenhouse and weighed to measure marketable yield. Insect damage and disease were minimal for all strategies.

Breakdown of Net Income

In the Breakdown of Net Income (above), labor costs were set at $10/hour. Materials costs included fertility, mulches, tractor use, and an estimate of curing, packing, and shipping costs. Sales were calculated by assuming that 90% of the cured marketable yield could be sold at the organic wholesale price of $0.75/lb. Net income was determined as the difference between sales and costs of labor and materials.

It was surprising that the strategies typically used for long-term aims of reducing the weed seedbank (ZSR) or building soil quality (organic mulches), were the most profitable in the first year of implementation. That these more expensive strategies were more profitable than the others demonstrates the importance of high yields.

The CPWC was the lowest yielding strategy, indicating that the weed control period was not long enough. Based on growing-degree-days we expect that the onions should have been weeded through late July to avoid yield loss. This highlights the sensitivity of onions to competition.

In 2015, soil samples will be collected to determine the effects of the contrasting weed management strategies on the weed seedbank and soil organic matter. The onion experiment will also be repeated on a new field to show which results are consistent and which results are subject to yearly variation, so stay tuned for this season’s results!

Managing weeds with crop rotation

“Rotation of crops…is the most effective means yet devised for keeping land free of weeds.  No other method of weed control, mechanical, chemical, or biological, is so economical or so easily practiced as a well-arranged sequence of tillage and cropping.”

— C.E. Leighty.  1938 Yearbook of Agriculture

Diversity is key.  Dissimilar crop species with disparate management practices impose a wide range of stresses and mortality factors, creating an unpredictable environment to which the weed community is continually adjusting (Liebman and Staver, 2001).  Diversity, however, may also establish and/or perpetuate weed problems.  Fall cucurbits, for example, may permit considerable weed growth after vines run, causing abundant seed rain (see, Figure 1, below).  Cover crops, while frequently noted for their ability to reduce weed biomass, often contain weeds going to seed.  Perennial legumes or sod crops favor perennial weeds such as quackgrass, and do not include timely soil disturbance events that promote germination losses of annual weeds.  Thus, while diversity is key, successful weed management requires cropping sequences that feature practices that minimize, or better, eliminate, “credits” to the weed seedbank, while maximizing seed “debits.”

Short-season cash or cover crops, whose growth is terminated before weeds set seed, are the most useful elements in preventing weed seed credits (Figure 1).  The tillage events necessary for these crops are often well-timed to preempt seed rain of winter annual weeds.  Ideally, the crops are then terminated before their associated summer annual weeds set seed.  In considering longer-season crops, good weed control, a competitive canopy, and opportunity for hand roguing surviving weeds are key attributes.

We measured common lambsquarters weed seed rain in a broccoli, winter squash rotation, managed without cover crops (control), with fall cover crops, two consecutive years of red clover (2-Yr. CC), or alternate years of vegetables and cover crops with summer fallowing (e.g., after strategies described by Nordell and Nordell, 2007; Figure 1).  The alternate year cover crop system consistently had the lowest common lambsquarters seed rain (see Alt.-Yr. CC, solid boxes, below).  This, combined with the seedbank depleting fallowing periods during the cover crop years, prevented this species from increasing over the four years of the experiment (data not shown).


Figure 1.  Effect of cover crop systems on common lambsquarters seed rain in 2001 through 2004.  Within a year means labeled with different lowercase letters are significantly different based on Tukey’s HSD (P < 0.05).

Debiting strategies require consideration of weed seedbank ecology (Gallandt, 2006).  Because germination is the most effective way to deplete the seedbank, it may be useful to consider primary tillage practices that maintain seeds at or near the soil surface, in the “active seedbank,” where seeds are most likely to experience environmental conditions that encourage germination.  Seed predation is also an important source of loss from the seedbank, and a further reason to keep seeds at the soil surface.  Avoiding fall tillage and rapid weed seed burial maintains seed at the soil surface where they are more readily consumed by predators.

Initial conditions of the seedbank should be carefully considered in short-term crop sequence planning.  Where the starting weed pressure is very high, a clean fallow period is the best strategy for drawing down the seedbank (Mohler, 2009; Nordell and Nordell, 2007).  Because weed species vary in their seasonal patterns of emergence, the timing of fallow periods should target the most problematic species or group of weeds.  Winter annuals, for example, exhibit peak emergence in the late fall and early spring—summer annuals, in the warmer periods of June and July.  Shallow tillage coincident with this emergence periodicity will stimulate germination of the targeted group of weeds, and subsequent tillage kills these seedlings.

If the seedbank is at a moderate level, cropping options may be expanded to include crops that are both amenable to effective cultivation, and are sufficiently competitive that abundant weed seed rain is unlikely.  These so-called “cleaning crops,” seem to vary from farm to farm.  Onions, for example, are a cleaning crop for some growers.  A long-season and uncompetitive crop, growers know onions must be nearly weed free, so they make frequent cultivation a priority for this crop.  This frequent, shallow cultivation offers benefits similar to fallowing strategies, encouraging successive flushes of weeds that are removed by subsequent cultivation events.  While the need for repeated cultivation may be viewed negatively in the short-term, the long term effect is depletion of the seedbank. Potato and sweet corn are cleaning crops for some growers.  These crops can be aggressively cultivated and weeds kept at a minimum.  Slow to establish, uncompetitive species, e.g., carrot and parsnip, onion and leek, are ideally planted in the cleanest of fields.

Given the importance of the relative size of the weed seedbank to the success of subsequent weed management practices, it seems counterproductive for an otherwise clean rotation sequence to include a crop likely to result in abundant weed seed rain (e.g., my problem with winter cucurbits).  Rotation blocks could consider likelihood of seed rain as a first separating criterion.  “Weed-free” blocks could be managed with a long-term vision for improving weed management conditions.  Elsewhere, the commonly used “critical period” for weed control can continue to guide management, focusing on control of weed seedlings in the early to mid-period of crop growth.  Before the start of this period, weeds are too small to reduce crop yield; after this period crop competition alone will avoid weed-related yield losses.  In other words, “beat the weeds back” early to ensure a good crop and don’t worry about weedy crops late in the season.  There are many successful farmers who rely on this approach to weed management.  They focus on repeated cultivation and hand weeding until crops are judged to be sufficiently weed-free.  However, seed rain from weeds surviving the critical period means that weed pressure is likely to increase over time.  In response, the frequency of cultivation and hours of hand weeding will have to increase to simply maintain a given level of weed control.  The alternative, managing for improving weed conditions, requires careful deployment of diversity, minimizing credits and maximizing debits to the seedbank (Gallandt, 2006).


Gallandt, E. R. (2006). “How can we target the weed seedbank?” Weed Science 54: 588-596.

Leighty, C.E.  1938.  “Crop Rotation.”  in Soils and Men, U.S.D.A Yearbook of Agriculture.  pgs. 406-430

Liebman, M. and C. P. Staver (2001). Crop diversification for weed management. Ecological Management of Agricultural Weeds. M. Liebman, C. L. Mohler and C. P. Staver. Cambridge, UK, Cambridge University Press: 322-374.

Mohler, C. L. (2009). The role of crop rotation in weed management. Crop Rotation on Organic Farms.  A Planning Manual. C. L. Mohler and S. E. Johnson. Ithaca, NY, Natural Resource, Agriculture, an dEngineering Service (NRAES): 44-46.

Nordell, A. and E. Nordell (2007). Weed the Soil, Not the Crop.  A Whole Farm Approach to Weed Management. Trout Run, PA: 42.

Prepared for the 2009 Proceedings of the New England Vegetable & Berry Conference.