🌿 Harvesting: Timing, Lifting and What the Yield Really Is

Cassava does something no cereal does. A maize crop tells you when to harvest and you obey, or you lose it. Cassava gives you a window months wide and lets you choose. That flexibility is one of the crop's real commercial advantages, and learning to use it deliberately rather than by accident is worth money.

The sourced window. Early or short-cycle varieties can be harvested from around 9 months after planting. Reported optimum root quality falls at 12 to 15 months after planting. Some varieties and some uses extend to 18 to 24 months. So the crop can sit in the ground, alive and growing, for a year or more after it first becomes usable.

That is what people mean when they call cassava a standing food-security reserve. Your roots are stored in the safest warehouse available, which is the soil, and they keep bulking while they wait. You harvest when the household needs food, or when the price is right, or when the processing equipment is free. A cereal farmer cannot do that. Once maize is dry it must come off, be dried properly and be stored somewhere that rats and weevils will attack.

But flexible is not unlimited, and this is where growers lose value. Roots left too long become fibrous and more woody. The texture changes, the eating quality falls, and a buyer paying for table roots will notice. For starch and flour markets it may matter less, because those roots are processed anyway, but there is still a point past which you are feeding the plant rather than filling the store.

Now the part almost nobody teaches, and it is a safety point. Age at harvest changes cyanogenic potential. The Ugandan study covered in Module 3 found cyanogenic content peaked at 8 to 10 months of plant maturity and then declined by month 13. Read that carefully. It means the toxin level in the root is not fixed at planting and it does not simply rise with age. It moves. In that study, the earliest harvests sat near the peak.

Be honest about the limits of that finding. It is one study, of one set of varieties, in one country. It does not license you to say that all cassava is safest at 13 months. What it does establish is the principle: harvest age is a cyanide-relevant variable, alongside variety, soil and stress. Your processing route still has to do its job at any age. If you are lifting young roots for a fermented or dried product, that is not a reason to shorten fermentation.

How do you decide, then? Work through four questions in order.

1. What cycle length does your variety have? This comes from the seed source or the extension office that supplied the material, not from guessing. A 9-month variety and an 18-month variety planted on the same day are two different businesses.

2. What is the end use? Fresh table roots reward lifting near the good-eating window rather than as late as possible. Starch and flour reward dry matter, which keeps accumulating.

3. What does your cash and food calendar need? A household that runs short in the hungry season may lift early and accept a smaller root. That is a legitimate commercial decision, not a mistake, as long as you make it knowingly.

4. What did your test dig show? This is the only question that looks at your actual field, and Lesson 2 is entirely about it.

One last practical point about the window. Every extra month in the ground is another month of exposure to whatever is attacking your crop. Brown streak disease damages the storage root itself and gets worse the longer the root sits. Rodents, pigs and thieves all have more time. Weighing the gain from more bulking against the risk of more loss is the actual decision, and it is different on every farm.

There is no sourced, quantified field-maturity index for cassava. A single validated measure, such as a percentage of leaves senesced tied to starch content, was not retrievable and this course will not invent one. The commonly taught visual signs, yellowing and dropping of the lower leaves, slowing production of new leaves, and toughening of the root skin, are reported practice and worth watching, but none of them proves maturity on their own.

So what do you actually rely on? Two things together. The known cycle length of your variety, and a test dig. The test dig is the more useful of the two because it looks at your field, in your soil, in this season.

Here is how to do one properly.

Choose three plants, from three different parts of the block, not three next to each other and not the three biggest. Take one from a good area, one from an average area and one from the poorest corner. If you only sample your best plants you will build your whole harvest plan on a lie.

Lift each plant carefully and keep its roots separate. Weigh the roots from each plant on their own. Any scale will do, including a market trader's. Write the three weights down with the date.

Now the arithmetic that turns those three weights into a yield figure. Suppose your three plants gave 3.5 kg, 2.5 kg and 2.0 kg of root. Add them: 3.5 plus 2.5 plus 2.0 equals 8.0 kg. Divide by 3 to get an average of about 2.67 kg per plant.

Next you need your plant population. At the common monocrop spacing of 1 metre by 1 metre, planting density is approximately 10,000 plants per hectare. That figure comes straight from the spacing: one plant per square metre, and a hectare is 10,000 square metres.

Multiply: 2.67 kg per plant times 10,000 plants equals 26,700 kg per hectare, which is 26.7 tonnes per hectare.

Now be honest about that number, because a raw multiplication always flatters you. Your field almost certainly does not have a full stand of 10,000 living plants. Gaps from cuttings that never sprouted, plants lost to disease, plants eaten, all reduce it. So count. Walk one row of known length, count living plants, and work out your real population. If a hectare's worth of rows holds 7,500 living plants rather than 10,000, your estimate becomes 2.67 times 7,500 equals about 20 tonnes per hectare, not 26.7. That is a quarter of your expected income gone, and you found it before harvest day instead of after.

Now the second dig, which is where this becomes a decision tool rather than a measurement. Repeat the same exercise two weeks later, on three different plants chosen the same way. If the average per-plant weight has risen meaningfully, the crop is still bulking and waiting is still earning you something. If it has barely moved, the plant has largely stopped filling roots and every further week is risk without reward.

That comparison is the closest thing you have to a real maturity index, and you built it yourself from two afternoons of work.

While the roots are out, use them. Cut two roots crossways and look inside for the brown corky necrosis of brown streak disease, which cannot be seen from outside the plant. Feel the texture: fibrous, stringy, woody flesh says you are late. Note the skin. Write all of it on the same sheet as the weights.

One warning. Test-dug roots are harvested roots. They are now on the deterioration clock covered in the next module, so cook them, sell them or process them the same day. Do not dig ten plants to be thorough and then leave the roots in a heap.

Somebody will quote you a cassava yield figure. An input seller, a cooperative recruiter, a government pamphlet, a man at the market. The number will be large and it will be true somewhere. The skill this lesson teaches is asking what it assumes before you build a plan on it.

Here is the sourced spread, and it is enormous.

Traditional smallholder methods are cited at 5 to 12 tonnes per hectare in one synthesis and 5 to 8 tonnes per hectare in another. Improved varieties, well managed, are cited at 40 to 60 tonnes per hectare, with another source giving above 30 tonnes per hectare as an attainable ceiling with improved management.

National and regional figures sit in between and are mostly low. South Africa is cited at about 50 tonnes per hectare. Nigeria averaged 14.7 tonnes per hectare over 1995 to 1999, Ghana 13.1 and Cote d'Ivoire 10.8 in the same period. The African continental average for 2014 is cited at 8.38 tonnes per hectare, with a range across leading African producer countries of 7.72 to 23.36. A world average cited in an FAO post-harvest compendium is under 10 tonnes per hectare, with Thailand at about 15. Trained farmers under one Nigerian weed-management programme achieved more than 20 tonnes per hectare against a cited Nigerian national average of 9.

So the same crop is described at 5 tonnes and at 60 tonnes per hectare, a difference of twelve times, in one honest reference document. Nobody is lying. They are measuring different things.

Start sorting them by asking what kind of figure it is.

A national average includes every neglected plot, every crop hit by disease, every field of old local varieties on exhausted soil. It is a floor description of a whole country, and it is why national averages sit near 8 to 15 tonnes.

A well-managed improved-variety figure describes a field with good planting material, correct spacing, fertiliser matched to a soil test, weeds controlled through the critical period, and pest and disease pressure handled. That is where 40 to 60 comes from. It is achievable but it is a bundle, not a purchase.

A trained-cohort figure, like the more than 20 tonnes per hectare cited for trained Nigerian farmers, sits between the two, and it is honest about what training plus support achieves in real farmer conditions rather than on a research station.

Now the number that matters most. The gap between traditional practice at roughly 5 to 12 tonnes per hectare and improved varieties with improved management at 40 to 60 is commonly a three to six fold difference. That is not a margin, it is a different business.

But here is the trap, and it is where money is lost. That gap is not attributable to any single lever. It comes from variety choice, planting material quality, spacing, fertiliser, weed control, which is itself worth 30 to 50 percent of root yield on its own, and pest and disease pressure, all together. Nobody sells you the gap. An improved variety planted into unweeded, unfertilised ground with infected cuttings does not give 40 tonnes per hectare, and the person who sold you the variety is not obliged to tell you that.

So when a figure is quoted, ask five questions. Which country and which year? Is it a national average, a research trial, a demonstration plot or a trained-farmer cohort? Which variety? What management is assumed for weeding, fertiliser and planting material? And crucially, is it fresh root weight, which is what you sell, or something else?

Then do the only thing that settles it. Compare it to the number you generated yourself in Lesson 2 from your own test dig. Your own field's figure is the only one that is definitely about you.

Cassava roots are awkward to harvest and the reason is physical. Reported root dimensions are 15 to 100 centimetres in length and 0.5 to 2.5 kilograms per root. Read that range again. One root can be the length of your forearm or longer than your leg, and a single root can weigh half a kilo or five times that. They grow downward and outward from the base of the stem, often deep, often at awkward angles, and they snap.

That variability is not a curiosity. It determines your labour planning, and it determines what size of peeling and grating equipment you need if you are processing, because a machine sized for small roots will not swallow a metre-long one.

Labour for combined harvesting and processing has been cited at approximately 721 man-hours per hectare in one FAO compendium figure. Treat that as a single cited figure, not a constant. It depends heavily on whether the harvesting is manual or mechanised and on how many tonnes of root are actually in the ground. But it is a useful order of magnitude: harvesting and processing a hectare of cassava is hundreds of hours of human work, not tens.

Now the technique, and the reason it matters.

Start by cutting back the stem. Cut the plant off about half a metre to a metre above the ground, leaving a stub you can grip with both hands. A long stem gives you leverage but whips about and tires the crew; no stem at all leaves you nothing to pull.

The stub is a handle, and it is also planting material. Set good stems aside as you go, in the shade, bundled, rather than treating them as rubbish. Module 4's whole argument about the multiplication problem is decided at this moment.

Loosen before you pull. This is the single instruction that separates a clean harvest from a broken one. Work a hoe, fork or digging stick around the plant, at a distance from the stem, and lever the soil upward on two or three sides. Do not chop straight down beside the stem; that is how roots are cut in half. You are trying to break the grip of the soil, not to dig the roots out one by one.

Then lift with the stub, straight up, with a steady pull rather than a jerk. The whole root cluster should come with the plant. If it resists, loosen more soil. Do not fight it, because the plant does not break, the roots do.

Soil condition decides how hard this is. Moist soil releases roots. Baked, dry, hard soil grips them and every plant becomes a wrestling match with a broken root at the end of it. This is one of the practical arguments for ridges and mounds in heavier soils, which are recommended for drainage and ease of harvest. Loose ridge soil lifts far more easily than compacted flat land.

Detach roots from the stub with a clean cut rather than by twisting and tearing. A torn root has a large ragged wound. A cut root has a small clean one.

Which brings us to why any of this matters commercially. Every break, bruise, cut and scrape is a wound, and the deterioration process covered in the next module is a wound response. It begins the moment the root is detached and it starts at the damaged surfaces. A field lifted roughly and thrown into a heap arrives at the processing point already deteriorating faster than a field lifted carefully.

There is a second, immediate cost too. A broken root left in the ground is a root you paid a full season to grow and will never sell. Walk the lifted rows and pick up what snapped off. On a heavy crop that is real tonnage.

Finally, handle roots as food, not as firewood. Do not drop them from height into a truck. Do not walk on the heap. Do not leave the pile in full sun while you finish the next row; stage it in shade. None of that costs money. All of it buys you hours on the clock that starts the moment the root leaves the soil.

One technology comparison gives sourced figures for manual and mechanised cassava planting and harvesting. Before quoting them, note the caveat that comes with them: this is a single technology-catalogue comparison, and the source does not state which country, which year, or what scale the base figures come from. The dollar amounts are unverified as current, country-specific numbers. The relative pattern is the finding. The exact figures are not.

With that stated, here is what the comparison reports.

Planting labour: 8 to 10 people per hectare per day manually, against a two-row mechanical planter covering 7 to 10 hectares per day.

Harvesting labour: 40 to 60 people per hectare per day manually, against a two-row mechanical harvester covering 3 to 5 hectares per day.

Stop on that line, because it is the whole lesson. Manually, one hectare consumes 40 to 60 people for a day. Mechanically, one outfit covers 3 to 5 hectares in a day. Working from the source's own figures, mechanised harvesting requires roughly an eighth to a fifteenth of the labour-days per hectare of manual harvesting.

Costs in the same comparison: planting at US$29 per hectare manual against US$13 mechanised; harvesting at US$61 per hectare manual against US$25 mechanised.

And now the trap that this course wants you to spot before anyone shows you a table like this again. Total production cost is given as US$328 per hectare manual and US$367 per hectare mechanised. The mechanised system costs more per hectare. But per tonne of roots it is US$20.50 manual against US$16.68 mechanised. The mechanised system costs less per tonne.

How can both be true? Because the mechanised farm in that same comparison is credited with a reported 38 percent higher yield. More cost spread over much more output gives a lower cost per unit.

This matters because the two figures point in opposite directions and you must know which one your decision hangs on. If you are asking whether you can afford this season's field operations from the cash in hand, the per-hectare figure is what your pocket feels. If you are asking whether you can sell profitably at the price your buyer offers, the per-tonne figure is the only one that answers it, and it is the figure that goes into the break-even calculation in Module 12. A salesman who shows you only the per-hectare number, or only the per-tonne number, has chosen which story to tell.

Be careful with the 38 percent yield increase too. It is attributed to the mechanised system in this comparison, but a farm that has a mechanical harvester almost certainly also has capital, better planting material and larger consolidated fields. This is a single comparison, not a controlled experiment isolating the machine.

Now the harder question: is mechanisation actually available to you? Mechanisation requires capital access, machinery servicing and land consolidated enough to manoeuvre equipment. The source quantifies none of those. Most smallholders have none of them. Land in fragmented small parcels, without a service network for a broken harvester and without credit, does not become mechanised because a table says it should.

There is a middle path worth naming. Between one person with a hoe and a two-row tractor-drawn harvester lie shared or hired equipment, cooperative ownership, and contractor services where an owner harvests for neighbours by the hectare. If you want the labour compression without the capital, that is the door to try, and your cooperative or extension office is who to ask about whether such services operate in your district.

One last reason this decision belongs in the harvest module rather than an equipment module. Labour compression is not only a cost question. Cassava starts spoiling within days of lifting. A harvest that takes ten days with a hand crew has roots that sat for ten days at different stages of deterioration. Compressing that to one or two days directly reduces the risk that the next module is about.

Everything in this module comes together on one day, and a cassava harvest that is not planned turns into roots sitting in the sun while somebody looks for a vehicle. This lesson is the plan.

Start with the sourced loss picture, because it tells you where to spend your attention. One FAO compendium breakdown attributes cassava post-harvest losses as harvesting 13.6 percent, handling 8.5 percent and processing 23.2 percent. Processing is the largest single category, which surprises most people, and Modules 11 and 12 deal with it. But harvesting plus handling together is 22.1 percent, and that is the part decided by how today goes.

The same source reports that improved technology reduced total losses from 22.3 percent to 10.1 percent in one cited intervention, roughly a 55 percent relative reduction. The excerpt does not detail what that technology was, so this course cannot teach it as a method. What it does establish is that these losses are reducible, not fixed.

Now the plan itself, in the order you build it.

First, the destination. Decide before you lift a single root where the roots are going and when they must arrive. Cassava is not a crop you harvest and then look for a buyer. If you are selling fresh, the buyer and the vehicle are arranged first. If you are processing, the grater, the press and the labour must be ready and free on that day. Harvest to the capacity of what happens next, not to the capacity of your field. If your processing point can handle two tonnes a day, do not lift six.

Second, the crew. Use the labour figures from Lesson 5 as a starting estimate and adjust to your own experience. Decide who cuts stems, who loosens soil, who lifts, who detaches and sorts roots, who bundles the planting-material stems, and who carries. Splitting the crew into these roles is faster than everyone doing everything, and it means the stems get saved instead of trampled.

Third, the sequence. Start at the end of the field furthest from your loading point and work toward it, so that carrying distance shrinks as the crew tires and the pile grows. Harvest block by block and finish a block before opening the next, so that no roots sit half-lifted overnight.

Fourth, staging. Choose a shaded staging point before you start, not when the first heap is already cooking. Roots go from soil to shade, sorted, in a single movement. Damaged and broken roots go into a separate pile, because those are the ones that spoil fastest and they should be processed or eaten first, not mixed into a load that has to travel.

Fifth, the clock. This is the record that matters more than any other in the whole course. For every load, write down the time it was lifted and the time it reached the processing point or the buyer. Harvest-to-process time, not just harvest-to-sale time, is the metric a processing business should track, and every deterioration timeline in the next module supports treating hours since harvest as a number worth writing on every batch.

Make the sheet simple enough that a tired person fills it in: date, block, load number, time lifted, time delivered, weight, and a note on condition.

Sixth, what leaves the field with you. Planting-material stems, bundled and shaded. Broken roots picked up from the loosened rows. Your record sheet. And the test-dig comparison from Lesson 2, so that at the end of the day you can compare the yield you predicted against the yield you actually got. That comparison is how your estimating improves, and after two or three seasons of doing it you will be able to predict your own field better than any quoted figure can.

One last discipline. Write down what went wrong. The vehicle that came late, the crew that was short, the block that ran into darkness. Next season's plan is this season's problems, written down while you still remember them.