Weeds are one of the main problems for organic market gardeners like us. Over the years, we’ve battled against them and know many other folks who deal with the same issues. Weeds compete for nutrients, water and light in vegetable growing beds. Being wild and unselected they tend to grow faster than cultivated plants, especially in fertile, irrigated conditions, and will often set and disperse seed before the crop plant is ready to be harvested and cleared. An unweeded plot will produce much less crop and the weed problem will persist or, likely, get even worse as the weeds drop their seeds and “seed bank” increases. We’re not talking about weeds that spread by root growth like twitch or couch grass here. That’s a different problem that needs a whole different discussion!
Farmers and gardeners know that even when they do manage their annual weed pressure, they still get weeds emerging every season. The mix of weed species may change over time but weed-free farming seems to be constantly beyond reach. The weed seed bank seems to provide an infinite supply of new weeds!
In order to manage weeds in an optimised way, which is with the least labour and resources and interference with vegetable production, we’ve been thinking and reading about all the factors, including our farm practises, that contribute to the dynamics of weed germination. Happily, there are things we can do to deplete the weed seed bank over time.
One day while weeding, Bri asked “why don’t the weed seeds deeper in moist soil germinate, try and grow and then die when they can’t get to the surface?” It was a great question which came up again when we read about the Beal seed experiment. For a great description of this experiment have a look at https://www.atlasobscura.com/articles/the-worlds-longestrunning-experiment-is-buried-in-a-secret-spot-in-michigan. In 1879, Dr William Beal buried a bunch of different local weed seeds mixed with wet sand at locations around the University of Michigan. Every 5 years, initially, and now every 20 years, a bottle is dug up and the seeds germinated to try and get an idea of how long weed seeds can persist in the seed bank. Even farmers in the 1800’s were battling the persistence of weeds! This experiment is set to run until the last bottles are dug up in 2100. Once again, we wondered, why don’t those seeds germinate in the damp sand in those bottles. We wrote to Prof. Frank Telewski asking this very question and he and his colleagues started sending us articles about seed germination triggers. As farmers or gardeners we simplistically think that we just need to get our seeds wet to germinate. There’s a lot more to it!
Moisture is one requirement but light, temperature, gas exchange, nutrient levels and fire-smoke are some of the others.
Moisture:
A seed is a partially developed plant embryo held in “freeze-frame” along with a store of nutrients in the endosperm that are slowly used to keep this embryo alive while it’s in its dormant state. Once activated, the enzymes within the seed cause the embryo to start developing again. They also metabolise nutrients from the endosperm to fuel that process. Seeds stored dry, obviously, will not germinate. They must imbibe - soak up - water before germination can be triggered because the enzymes involved in germination require water. Depending on the history of the site, weeds seeds can be found anywhere from the surface to a few hundred millimetres below. So, why don't deep weed seeds germinate when they get wet and then die because they cant reach the surface - wouldn’t that be great if they did? It turns out that, at depth, water is held tightly by the soil. Plants can actively take up that water but seeds cannot. Their passive water absorption requires freely available water. Closer to the surface, water is less tightly bound by the soil and it’s here that seeds can imbibe (for a much more detailed trip through the physics of this look up matric potential - don’t say I didn’t warn you!). Regardless of this, water alone isn’t enough to trigger germination…..
Light:
Although we are taught that seeds don’t need light to germinate, in fact, many seeds do require light - sometimes even a “flash” exposure - to trigger the process . All the seeds that we deliberately plant have had their light exposure as we handle them. Buried weed seeds are exposed to light when soil is inverted bringing them to the surface. Also, when weeds or crops are pulled out, the soil stuck to the roots carries weed seeds from deeper in the soil profile which can then be deposited close to or at the surface where they can access light. Seeds contain light-sensitive protein pigments known as phytochromes. Light can penetrate the seed coat and be detected by phytochromes which then triggers germination. Interestingly, different types of phytochrome can trigger germination in response to different wavelengths and intensities of light. Individual seeds within a single species will have genetic differences in their phytochromes. This allows different seeds to be triggered to germinate by different light conditions. This maximises the chance that at least some of that species will germinate and grow and ensures the reproductive success of the whole population.
Temperature:
There are several ways that temperature triggers germination. Some weeds will germinate with a simple temperature increase, for example as the soil warms in spring. Others require a certain amount of chilling (stratification) before the temperature increase. A change in the diurnal fluctuation of temperature can also trigger germination. Seeds deeper in the soil profile sit at a more constant temperature because soil is a good insulator. When seeds are brought closer to the surface, again by inversion tillage or roots being pulled, the diurnal temperature fluctuation increases. The seed is exposed to higher temperatures during the day and lower temperatures at night. This fluctuation triggers germination in many species.
Gas exchange:
During germination, a range of enzymatic processes occur including the breakdown of the energy reserves in the endosperm so that the embryo can grow and develop. These processes are mostly aerobic which means that they need oxygen. Oxygen permeates through the seed coat and provides this. Seeds closer to the surface have more access to oxygen than seeds deeper in the soil profile and this is another trigger for germination.
Nutrient:
Some seeds can detect nitrate levels and this is also a cue for germination. Nitrate is taken up by plants as an essential nutrient and so it makes sense for seeds to germinate when this resource is present. “Conventional” farmers add nitrate directly and in organic farms, biological sources of nitrogen are eventually broken down and nitrified to produce nitrate. This process is microbially driven by bacteria, archaea or fungi, is aerobic and so happens closer to the soil surface where there is more oxygen. Seeds closer to the surface, therefore, are more exposed to nitrate than seeds deeper in the soil profile.
Fire-smoke:
Perhaps not so relevant to market gardeners is that some seeds, particularly in Australia with it’s fire-based ecosystems, require fire or, at least, smoke to germinate. If you happen to have a newly established farm that was once bush and you burn or use a gas-flame weeder, it’s possible that you’ll get a whole range of natives plants emerge. This is a reminder that we and the plants we’re trying to grow don’t belong here!
The problem of variable dormancy
With all the triggers for germination present, why don’t all the weed seeds germinate? Dormancy is the ability of a seed to hold in “freeze-frame” until the conditions are right and the triggers for germination are present. The plants that humans have selected over many generations to be our food crops tend to have uniform dormancy. When we provide them with germination triggers - we plant them, water them, and make sure the temperatures are correct - the majority of them will germinate. We’ve selected for that because it ensures maximum yield. As market gardeners we’ve come to expect that when we seed a tray or a bed, pretty much all the seeds will emerge. Weeds are different. As wild plants, they look after themselves and have evolved strategies to maximise their reproductive success rather than maximum yield. Imagine a situation in the wild where all the triggers for germination are present. It’s spring, the soil is warming, there is increased daylight, there have just been rains, microbiological activity is ramping up and soil nitrates are increasing. All the seeds of plant X germinate and start to grow beautifully but then a few weeks later there’s a late cold snap. Temperatures plummet, there’s a heavy frost and all of those young plants die. Sounds like a typical Tasmanian November! That would lead to the local extinction of that species. Variable dormancy has been selected for so that this doesn’t happen. Instead of all those seeds germinating in the seemingly perfect conditions, only a fraction of them will. Due to genetic variability, the germination triggers that each seed requires are slightly different. Perhaps it’s slightly different temperature fluctuations, light wavelength or nitrate concentrations that trigger germination. Perhaps some seeds need multiple cycles of cold-heat before they’re ready to germinate. This feature of wild plants and also the ability of many seeds to be able to survive dormant for even up to hundreds of years explains why things keep germinating year after year.
Observant market gardeners will have noticed that they may have pulled every single chickweed plant that germinated this season before they dropped their seeds but then more grow later in the season and then again in subsequent seasons. You might be tempted to think that it’s because you missed a few and they set seed which then distributed in your patch. That could be the case but it could also be because of variable dormancy. Not all the seeds have germinated. Yet!
The problem with (inversion) cultivation
The other reasons weeds keep coming, even if you are disciplined enough to remove all weeds before they seed is that some of our farming practices bring new weed seeds up to the surface where the triggers for germination are present. Over many years, and depending on the history of the site, there may be weed seeds sitting deeper in the soil profile. Inversion tilling such as rototilling, ploughing and discing will bring these seeds to the surface (as well as bury fresher ones making deposits into the seed bank for the future!). Tilling also aerates the soil providing oxygen to the seed and the microbes which make nutrients like nitrate available - double whammy!
What can we do?
No-till
The surest way to subvert weed germination is to practise no-till market gardening as espoused by Charles Dowding and others. Here, the soil to become market garden is not disturbed at all. Thick layers of compost (>100mm), prepared properly so that any seeds within are sterilised, are placed directly or with an intervening layer of cardboard onto the ground. Seeds or seedlings are planted directly in the compost. Deeper-rooted plants break through into the natural ground. This system works because the weed seeds on the surface of the soil are smothered and don’t have access to the germination triggers - particularly light, oxygen and temperature fluctuations. This system can also work with weeds that spread by root like twitch or couch grass. When harvesting plants or pulling the occasional weed, it is important to not bring soil carrying weed seeds from below the compost to the surface. This is usually done by cutting the plants below the crown and leaving the roots in the ground. Except for root crops of course! The only weeds that need to be managed, in theory, are from seeds that drift onto the farm from nearby. From our own trials on our “top block” of approximately 400m2, using commercial green-waste compost, it’s remarkably effective. Weed pressure drops to negligible levels and is mostly due to blown-in seeds either locally or onto the compost pile at the depot. It’s not without drawbacks, however. On a small scale, it’s feasible, but on larger scales, the compost quantities become cost and labour prohibitive. At 100-150mm thick, every 20m x 0.75 bed requires 1.5-2.25 cubic metres of compost in the establishment phase. A further 20-30mm is required to top up the compost each year as it inevitably breaks down due to microbial action. Purchased compost can cost around $75/m3 (here in Tasmania) and can be of questionable quality. Poor compost - unfinished or nutritionally imbalanced - will lead to poor crops. There’s also the, possibly aesthetic, consideration that you are not “farming” your soil, you’re buying in your growing medium. Making that quantity of compost on-farm would require a significant amount of raw materials that must be imported or harvested on the farm as well as large scale equipment (loaders, compost turners). You might also end up with substandard compost. Then there’s the labour involved in laying down the compost as well as other materials, usually wood products (which in Tasmania, can come from old-growth forest), in the pathways. Most of this is done by barrow and hand tools although small-scale compost spreaders are now becoming available. YouTube stars can rely on volunteers or interns but this labour cost can go uncounted.
If you can access excellent quality compost at a reasonable price and your garden or farm is small, this method is worth doing. You can forget about weeding, mostly, and spend your time making YouTube videos.
Non-inversion
If you are committed to farming your soil, minimising the import of inputs and reducing weed pressure, the best thing to do, for many reasons, is to stop inversion tillage. Although soil types, conditions and techniques used can minimise the damage to soil structure, inevitably, damage occurs during tillage. Not to mention the concomitant loss of soil carbon and biotic diversity. And, to complete the trifecta, weed seeds are brought up into the germination zone! Prior to a market garden, our land was an unimproved cow pasture for over 35 years. We prepared beds using a rotary plough and then a rototiller and each season would re-prepare beds using shallow cultivation with the rototiller. We’re good but not perfect with our weed control and for years we have been battling with weeds - mostly the grasses that were in the original pasture. We have subsequently converted one block on the farm to the no-till system as mentioned above and for the remainder the plough and tiller have been retired. The tiller has actually been decommissioned and is being converted into a compost spreader but that’s another story!
We have changed our bed preparation system a lot over the last few years. At the start of the season, beds are cultivated once to a depth of 50mm with a power harrow. This tool mixes the soil on the horizontal plane (non-inversion), incorporates amendments and rolls the bed to a perfect 750mm wide seed bed. Where we previously used a paperpot transplanter, we are converting more to planting seedling plugs into dibbled holes. This is more laborious but the paperpot transplanter itself inverts soil with its trailing soil closers. Harvest or bed clearing is done by cutting plants just below the surface and roots are left in the ground so as not to invert the soil when we pull them out. The root mass also contributes to persisting soil carbon and soil structure. Weeding is done with tools that cut rather than dig. The inevitable missed weeds that get bigger are now cut just below the surface rather than being pulled out.
It’s too early to tell but it seems that this kind of non-inversion management is reducing the weed load. We’re also noticing other benefits. The soil seems firmer but still beautifully friable and we are able to get away no cultivation at all. Our power harrow is getting less use. Where we would use it at each “bed flip” now we are doing multiple successions without it.
Even with this kind of bed preparation and management system, we still need other systems to assist with weed control.
Germination control systems
stale-seed bed technique
This method is practiced more commonly now on larger scale farms. Essentially, soil is cultivated, and irrigated, if required. Weeds are triggered to germinate due to their new conditions and then a subsequent shallower cultivation or flame treatment kills those weeds just before planting the crop. This method is effective but not perfect because of variable dormancy. Weeds that didn’t germinate originally may germinate after the crop has been planted. Hopefully, these weeds will be lower in number and impact but they may still seed before the crop has matured. In the market garden, stale seed bed techniques can be used with subsequent, vigilant removal of weeds using tools. If plantings are well-spaced and straight, this is not too labour-intensive. We incorporate stale seed bed technique into our non-inversion system described above
occultation/tarping
We use large plastic silage tarps as an addendum to stale-seed bed techniques. After bed preparation and irrigation, we allow the first signs of weed germination to show and then cover with a tarp to occlude light. This kills the newly germinating weeds. After a period of time ranging from 1 - 4 week, depending on the air temperature, the tarp can be removed and the crops sown or another cycle of weed germination and tarping can be done. Tarping has other benefits such as warming the soil and keeping it from water-logging. Tarping for too long might result in soil microbial diversity reduction due to the absence of of the photosynthesis from plants that the microbes depend upon.
hybrid compost system
On our farm, we have found that an effective system hybridises the no-till compost system mentioned above with the non-inversion tillage system. After bed preparation with a power harrow, we either direct seed before covering the bed with ~20mm of commercial compost, or we add the compost first prior to transplanting. Plantings of seedlings are then done as plugs into dibbled holes to prevent exposing the soil. We have found that this thinner layer of compost will inhibit weed germination to some extent meaning that tool-based cultivation is a lot quicker. It also saves a lot of time and expense as the amount of compost used is much less.
cover cropping
Cover cropping has numerous soil-building benefits but, on it’s own, is a poor weed management technique, contrary to what we have been taught. Densely planted cover crops do not inhibit weeds and, in our experience, many of them will mature and seed within the cover crop adding to the weed seed bank. Crimping the cover crop to produce a dense mat on the soil surface will inhibit germination until the mat breaks down or is removed for next season’s bed preparation. We cover crop significant sections of our market garden and, now, combine this with our non-inversion hybrid system. Cover crops are planted using an Earthway seeder in straight lines. We terminate the cover crops by slashing then tarping over the stubble.The tarp helps terminate the cover crop, accelerate its decomposition and kill freshly germinated weeds. After the tarp is lifted, plants are directly transplanted into the beds, between the rows of cover crop residue, using the plug method. This is done directly through any residual cover crop “straw” with or without added compost (depending on how much cover the straw provides).
Conclusions
Weeds, and wild plants in general, have a range of strategies to maximise their abundance and survival.
Traditional farm management causes weed germination but modifications to cultivation techniques with a few added management systems can minimise germination. Several seasons of this kind of management can diminish the near-surface weed population.
Don't let weeds seed and leave existing seeds buried!
