🥛 Housing the Dairy Cow

Zero-grazing means the cow does not walk to her feed. The feed walks to her. She lives in a small housed area, and you cut, carry and place every mouthful she eats in front of her, along with her water. It is the standard system for most African smallholder dairy farms, and it is standard for a reason: most smallholdings are simply too small to graze a productive dairy cow or a good cross on their own land.

But it is worth understanding that confinement is not just a way of coping with small land. It buys you three separate things, and each one shows up somewhere else in this course.

The first is energy. A cow walking long distances to poor grazing burns energy she could have put into milk. In early lactation she is already in a deep energy hole - her milk yield climbs toward peak far faster than her appetite can climb to match it, so she is burning her own body fat to make up the difference. That is negative energy balance, and it is a normal phase of every lactation, not automatically a sign that you have done something wrong. What matters is how deep it goes and how long it lasts. Every kilometre of walking for thin grazing makes it deeper. Bringing the feed to her removes that cost entirely.

The second is cleanliness. A cow lying in mud, or in her own dung, has a dirty udder, and a dirty udder is the entry route for the environmental mastitis organisms you will meet in the next module. The primary reservoir for those organisms is named in the veterinary literature as bedding, with contaminated water, mud and teat dips also implicated. Every one of those is a housing problem, not a milking problem. You cannot wash your way out of a filthy floor.

The third is disease exposure between herds, and here there is a real African number worth knowing. A Ugandan study compared cattle in pastoral systems against cattle in zero-grazing systems - 497 pastoral animals and 226 zero-grazing animals, 723 in total. It found brucellosis seroprevalence of 34.0 percent in the pastoral animals against only 3.3 percent in the zero-grazing animals. At herd level, 100 percent of the pastoral herds had at least one positive animal, against 5.5 percent of the zero-grazing herds. Abortion in seropositive cows was 23 percent in the pastoral system and zero in the zero-grazing system.

Read that carefully, because it is easy to over-read. That is one study, in one country, in one period, comparing two systems that differ in many ways at once. It does not prove that zero-grazing prevents brucellosis. What it does show, with real numbers from Africa, is that housing system and disease exposure are linked, and the most likely reason is obvious: a housed cow does not stand nose to nose with cattle from twenty other herds at a communal water point or on shared pasture. Brucellosis moves between animals by that kind of contact, and by contact with birth fluids and aborted material.

So the wall around your cow is doing several jobs at once. It saves her energy, it keeps her udder off the ground, and it cuts her contact with other people's animals.

What confinement takes from you is equally clear, and you must plan for it. A housed cow eats nothing you have not cut and carried. She drinks nothing you have not fetched. She gets no exercise unless you give her a yard. And every kilogram of dung and every litre of urine she produces lands inside a structure you now have to drain and clean, every single day, forever. A zero-grazing unit is a commitment to daily labour, and a unit designed without thinking about drainage and dung removal becomes exactly the wet, dirty, ammonia-smelling place that undoes all three of the gains above.

The rest of this module is the dimensions. They come mainly from the FAO structural design reference for cattle housing, which is the most detailed sourced dimension table available. Be honest about what it is: it was written for tropical climates generally, not specifically for East African smallholder zero-grazing units. Numeric wall-opening and airflow design figures tailored specifically to East African zero-grazing units were not available for this course and must be obtained from your own national extension service if you need them. The FAO figures are the best sourced starting point you have, and they are a great deal better than guessing.

Space is the first decision and the hardest to change later. Build the walls too close together and you will live with it for the life of the structure, so it is worth doing the arithmetic before the foundation goes in.

Here are the sourced FAO floor-space figures for bedded housing with yard access, by class of animal:

• Young stock 1.5 to 3 months, 70 to 100 kg: 1.4 square metres
• Young stock 3 to 6 months, 100 to 175 kg: 1.8 square metres
• Young stock 6 to 12 months, 175 to 250 kg: 2.1 square metres
• Young stock 12 to 18 months, 250 to 350 kg: 2.3 square metres
• Bred heifers and small cows, 400 to 500 kg: 2.5 square metres
• Milking cows, 500 to 600 kg: 3.0 square metres
• Large milking cows, over 600 kg: 3.5 square metres

Notice that the table is driven by liveweight, not by breed name. That matters in African conditions, because breeds differ enormously in size. A mature Boran cow weighs 400 to 550 kg, which puts her in the 2.5 square metre band or just into the 3.0. A Jersey at 380 to 450 kg mature bodyweight sits in the small-cow band too. A large Friesian cross may sit in the 3.5 band. Weigh or estimate your animal, then read the table. Do not read it from a breed label.

Now work an example. Say you keep two milking cows at about 500 kg each, one in-calf heifer at about 420 kg, and one weaned calf of eight months at about 200 kg.

1. Two milking cows at 3.0 square metres each: 6.0 square metres.
2. One bred heifer at 2.5: 2.5 square metres.
3. One 6-to-12-month animal at 2.1: 2.1 square metres.
4. Total resting and standing area: 6.0 plus 2.5 plus 2.1 equals 10.6 square metres.

That 10.6 square metres is the animals' area only. It does not include the feed alley, the manger, the manure gutter, the passage you walk down, the calving pen or the milking area. A very common mistake is to measure the whole shed, divide by the number of cows, and conclude there is plenty of room. Measure the space the animals actually stand and lie in, and check that figure against the table.

Shade structures are a different case, because a shade is not full housing. A shade roof needs a minimum of about 2.5 to 3.0 square metres per animal, and the orientation of the roof changes the requirement. An east-west oriented shade roof, which lets the cows feed in the shade, needs 3 to 4 square metres per cow. A north-south orientation, with the feeding done outside the shaded area, needs only 2.5 to 3 square metres per cow. That is a real design choice: if you want the cows to eat under the roof, you must build more roof.

Exercise or loafing yards are where the numbers get startling. A paved yard needs roughly 4 to 5 square metres per animal. An unpaved earth yard needs 40 to 60 square metres per animal. That is roughly ten times as much land for the same number of cows.

Why the difference? Because earth under cattle turns to mud. A small earth yard is churned into deep, wet, dung-mixed mud within days, and mud is not a cosmetic problem. It coats the udder and the legs, it is a reservoir for the environmental mastitis organisms named in Lesson 1, and a cow that will not lie down in mud spends more hours standing, which costs her rest and costs you milk. A large earth yard survives because the dung and urine are spread over enough ground to dry out between uses. If you cannot give an unpaved yard 40 to 60 square metres per animal, then either pave it, or accept that you are running a mud yard and plan for the consequences.

One honest caution about all of these figures. They come from the FAO structural design reference, which was written for tropical climates in general and not developed specifically for East African smallholder units. Locally tailored figures were not available for this course. Use these as your design basis, and if your national extension service publishes something different for your country, use theirs, because they know your rainfall and your building materials better than a general reference can.

A cow spends a large part of her day lying down, and where she lies decides how clean her udder is when you come to milk her. This lesson is the resting place: the cubicle, or in most small units, the tie-stall.

A cubicle, sometimes called a free stall, is a defined lying space the cow walks into and backs out of freely. The point of it is that it is exactly the right size. Too short and she lies with her hindquarters in the alley, which means she lies in dung. Too long and she dungs inside the cubicle instead of in the alley behind it, which means she lies in dung again. Too narrow and she cannot rise comfortably. Too wide and she turns around in it. There is no forgiving direction, which is why the dimensions matter.

The sourced FAO cubicle table, length by width:

• Young stock 1.5 to 3 months: 1.2 m long, 0.6 m wide
• Young stock 3 to 6 months: 1.5 m by 0.7 m
• Young stock 6 to 12 months: 1.8 m by 0.8 m
• Young stock 12 to 18 months: 1.9 m by 0.9 m
• Bred heifers and small cows: 2.1 m by 1.1 m
• Milking cows: 2.2 m by 1.2 m
• Large cows: 2.3 m by 1.2 m

Notice how tight the range is at the top end. From a bred heifer to a large cow the width moves only 10 cm and the length only 20 cm. This is not a place for rounding off to the nearest convenient timber length.

Most small African zero-grazing units do not use cubicles at all. They use tie-stalls, where the cow is tethered on a platform in front of a manger. The sourced tie-stall figures are:

• Platform length: 1.6 to 1.8 m
• Platform width: 1.1 to 1.3 m
• Manger width: 0.5 to 0.65 m
• Manure gutter: 0.4 to 0.7 m wide, 0.25 to 0.35 m deep
• Platform slope: 2 to 4 percent

That platform slope is the most important number on the list and the one most often left out. A slope of 2 to 4 percent means the platform falls by 2 to 4 cm for every metre of length, running from the manger end down toward the gutter. It exists so that urine runs away from the cow while she is lying down, instead of pooling under her. A flat platform holds urine against her udder and her hocks for hours. That is a direct route to a dirty udder, to environmental mastitis organisms sitting on the teat end, and to the ammonia smell that tells you the air in the shed is bad.

Work the slope out with real numbers. A 1.8 m platform at 3 percent falls by 0.03 times 1.8, which is 0.054 m, or about 5.4 cm from the manger end to the gutter end. That is roughly the thickness of two fingers. It is a small, cheap, easy thing to get right at building time and effectively impossible to correct afterwards without breaking up the floor.

The gutter behind the platform is the other half of the same system. At 0.4 to 0.7 m wide and 0.25 to 0.35 m deep, it catches what runs off the platform and what the cow drops behind her, and it has to run somewhere - to a pit, a soakaway, or a collection point for manure. A gutter that does not drain is just a trough of slurry, and it will overflow onto the platform, which defeats the whole design.

Alleys and passages need width too, or the shed becomes unworkable. The feed alley is 1.6 to 2.0 m wide where the cows feed nose-out, narrowing to 1.2 to 1.4 m where the manger design lets the cow take a smaller step. Service and passage alleys are 1.4 to 2.0 m. If you have ever tried to get a wheelbarrow past a cow in a narrow passage you already understand why this figure exists.

For a larger unit or a cooperative building something bigger, the free-stall barn alley figures are: between a stall row and the feed trough, 2.7 to 3.5 m, widening to 4.0 m for herds over 60 cows; between a stall row and a wall, 2.0 to 2.4 m; between two facing stall rows, 2.4 to 3.0 m. Most one-cow and two-cow smallholders will never need those, but a group planning a shared unit should have them.

All of these are FAO figures for tropical cattle housing generally. They are not African-smallholder-specific, and no East-African-specific version was available for this course. Use them as your design basis and check them against your own extension service if it publishes its own.

You cannot see air. That is why ventilation is the part of a zero-grazing unit that gets built wrong most often, and why the mistake is invisible until the cow's yield drops or her udder gets infected.

The sourced minimum is straightforward: the roof or eave height should be about 3 metres for adequate air movement. That is a floor, not a target. A shed built lower than that traps three things you do not want. It traps heat, because hot air has nowhere to rise to. It traps humidity from the cow's breath and from the wet floor. And it traps ammonia, which comes off urine and dung and which you will smell at cow height long before you smell it standing up.

Each of those three has a cost you have already met. Heat works directly against the cow's energy balance in early lactation. Humidity combined with heat is exactly what the heat-stress research measures. And ammonia and damp are the conditions in which environmental mastitis organisms thrive on the bedding and the floor.

Now the cheap intervention, and it is one of the best value items in this whole course. Putting a locally made insulating layer under a metal roof - the example given in the source is a woven mat of local materials - can lower the internal shed temperature by roughly 10 degrees Celsius or more compared with bare corrugated iron. Ten degrees. For a woven mat.

Bare iron in African sun is an oven lid. It absorbs the sun's heat and radiates it straight down onto the cow's back, which is why a shed can be hotter inside than the shade of a tree outside it. A mat, a layer of thatch, or any insulating material under the sheets breaks that radiant path.

Why does ten degrees matter in milk terms? Because there is a genuinely African, genuinely quantified study on this. A Ugandan study of 53,730 daily milk-yield records from 183 cows compared purebred Ankole against three crosses, using the Temperature-Humidity Index, which combines heat and humidity into one number. It found:

• Purebred Ankole: yield starts falling above a THI mean of 66, and then falls by 0.055 kg/day for each further THI unit.
• Ankole crossed with Friesian: yield starts falling above THI mean 69, and then falls by 0.107 kg/day per THI unit - the most heat-sensitive group of all.
• Ankole crossed with Jersey: threshold THI mean 69, loss 0.067 kg/day per unit.
• Ankole crossed with Sahiwal: threshold THI mean 68, and the yield loss was not statistically significant - no measurable heat penalty at all.

Read that properly. The purebred Ankole not only tolerates more heat before losing anything, she also loses it more slowly once she crosses her threshold. The Ankole-Friesian cross gives the most milk on an ordinary day and pays the steepest penalty on a hot one. And the zebu-by-zebu cross showed no detectable penalty in that dataset. The study's own authors described the overall heat effect in that herd as small compared with studies of high-producing temperate breeds.

So the practical link is this: the more exotic blood your cow carries, the more your roof is worth to you. A mat under the iron is not a luxury item for a Friesian cross. It is protecting the exact animal the study measured as most heat-sensitive.

Beyond the roof, East African smallholder zero-grazing design guidance names the basics: leak-proof roofing, good drainage, adequate lighting, and walls that keep draughts off the animals while still allowing air to move. A shed that leaks makes the bedding wet, and wet bedding is the mastitis reservoir again. A shed with no light is a shed where you cannot see clots in the foremilk or dirt on a teat.

And here is the honest gap. Specific numeric wall-opening sizes or airflow rates designed for East African zero-grazing units were not available for this course, and this module will not invent them. What you have is the 3 m eave height, the insulating-layer effect, and the principles. If you need a wall-opening specification, ask your national extension service or livestock research institute - in Kenya that is KALRO, in Ethiopia EIAR, and each country has its equivalent. They are the correct authority, and they know your rainfall and your prevailing wind.

The calf is not a by-product. She is the next milking animal in your herd or she is the animal you sell, and either way she is worth as much over her lifetime as several months of her mother's milk. Housing her badly in the first weeks throws that away, and it is cheap to house her well.

Start with the floor, because that is where the sourced detail is most specific. An individual calf pen floor is built from sawn timber boards 37 to 50 mm thick by 75 to 100 mm wide, laid with a gap of 25 to 30 mm between the boards.

That gap is the whole design. It lets urine drain straight through the floor and away, so the calf is not standing and lying in it. But 25 to 30 mm is narrow enough that the calf's foot cannot fall through and she gets secure footing. Wider and she will injure a leg. Narrower and it blocks with bedding and dung and stops draining. Cut your gap with a spacer block, not by eye.

The individual pen dimensions, by age:

• Birth to 2 weeks: 1,200 by 800 mm
• 6 to 8 weeks: 1,200 by 1,000 mm
• 6 to 14 weeks: 1,500 by 1,200 mm

So a newborn calf needs 1.2 m by 0.8 m, and by three months she needs 1.5 m by 1.2 m, which is nearly double the area. If you are building calf pens, build for the older size or build so the divisions move, because the calf grows faster than you will get round to rebuilding.

Now the calving pen. The sourced minimum is 3.3 m by 3.3 m, with a recommended provision of one calving pen per 30 cows, or one per 20 cows where calvings are seasonally concentrated. For a smallholder with one or two cows those ratios sound absurd - you obviously do not need a thirtieth of a pen. Read them the other way round: they tell you that a calving pen is standard, expected provision, and that one pen serves a herd far larger than yours. So for a one-cow or two-cow unit, one 3.3 m by 3.3 m space that can be cleaned out and bedded before a calving is the whole requirement. It does not have to be a permanent room. It has to be available, clean and dry on the day.

Why does it matter so much? Because of what happens in the first hours of the calf's life. A calf is born with almost no antibodies of her own - the cow's placenta does not pass meaningful immunity before birth. Everything she has to fight disease in her first weeks must come through colostrum, the antibody-rich first milk, and her gut can only absorb those whole antibody molecules for a short window. With each minute after birth her ability to absorb them falls, and by 24 hours the gut is almost completely closed.

A cow calving in mud, in a passage, or in a dung-covered yard produces a calf that lands in exactly the bacterial load you have spent this whole module trying to design out, at the one moment in her life when her immune system has nothing in it. The calving pen exists to give her a clean landing.

The target for the first feed is as soon as possible. One veterinary source gives within 6 hours of birth as the target for the first feed. Another gives within 1 hour as ideal, and says any calf that has not suckled naturally within 1 to 2 hours should be given colostrum by hand or by tube without further delay. Those two sourced timings genuinely differ and this course does not pretend to reconcile them. Teach yourself the faster one: get colostrum into her as soon as possible, within 1 to 2 hours if you can manage it at all, and no later than 6 hours. Treat every extra hour as a real loss.

The quantity targets converge on roughly 4 litres of good colostrum as the first-day total, with one source specifying that within the first 6 hours and giving 200 to 300 g of IgG at that volume. A minimum of 3 litres of high-quality colostrum within 6 hours is given as a lower bound. A second feeding 8 to 12 hours after birth is recommended on top of the first.

And here is the evidence that makes the pen worth building. Calves reaching a serum IgG level of 30 g per litre, described as excellent passive transfer, experienced over 10 percent less illness than calves whose serum IgG was only around 8 g per litre. That is a directly measured illness reduction from getting one free feed right, in a clean place, at the right time.

Almost everyone who builds a cow shed plans the drinking water. Almost nobody plans the washing water, and then discovers that keeping the milking area and the equipment clean takes more water than they thought and there is nowhere to get it.

The sourced reference figures, per wash, by milking method:

• Hand milking: 10 litres hot and 20 litres cold
• Bucket milking: 20 litres hot and 40 litres cold
• Pipeline milking: 30 litres hot and 60 litres cold

So even a smallholder hand-milking one or two cows should plan on the order of 30 litres of clean water per wash, split between hot and cold. Milking twice a day, that is around 60 litres a day for washing alone - and that is before the cow drinks a drop.

That is on top of her drinking water, and her drinking water rises sharply with her milk yield, because milk is roughly seven-eighths water. Be honest about what this course can tell you here: a specific sourced figure for litres of drinking water per litre of milk in African dairy cows was not available and this module will not invent one. The safe and important teaching point is that a milking cow's water access and water quality must be treated as seriously as her feed, and that any restriction on water shows up first as a drop in milk yield, before it shows up as visible thirst. By the time you can see she is thirsty, you have already lost milk.

Now let us total up what bad housing actually costs, because it is never one bill. It arrives as several separate losses in different parts of the business, which is exactly why it is so easy to ignore.

1. Milk lost to heat. A shed with a low roof and bare iron sheets runs hot. In the Ugandan THI study, an Ankole-Friesian cross lost 0.107 kg per day for every THI unit above her threshold of THI mean 69. An insulating mat under the iron, which lowers internal temperature by roughly 10 degrees C or more, is the cheapest thing standing between your cow and that loss.

2. Milk lost to mastitis. Bedding is named as the primary reservoir for environmental mastitis organisms, along with contaminated water and mud. A wet, dung-covered lying area is not a hygiene inconvenience, it is the infection source itself. And subclinical mastitis is invisible - no clots, no swelling, nothing you can see - while it quietly cuts yield from that quarter. You will meet the test for it in the next module.

3. Feed energy wasted. A cow walking for poor grazing, or standing all day because she will not lie in mud, spends energy that does not become milk. In early lactation, when she is already mobilising her own body fat to cover the gap between her yield and her appetite, that waste comes straight out of her body condition. Loss of body condition between calving and insemination carries an approximate 10 percent decrease in conception rate for every 0.5 point of body condition score lost. That is a review figure from the general dairy-science literature, not measured in African herds, and it is flagged as such - but the direction is not in doubt, and it links your floor directly to your calving interval.

4. Disease exposure. The Ugandan comparison found 34.0 percent brucellosis seroprevalence in pastoral cattle against 3.3 percent in zero-grazing cattle. One study, one place, one time - but housing and between-herd contact are linked.

5. Calves lost or set back. A dirty calving place at the one moment the calf has no immunity, and a calf pen floor that does not drain, both cost you the animal that was going to be your next milker.

So where do you start? Not with the biggest item. Start with the one that is failing now.

Walk your own unit and score four things honestly:

• Is the lying area dry, or does urine pool where she lies? If it pools, the platform slope is your first job.
• Is the roof over 3 m, and is there anything under the iron? If not, a mat is your cheapest gain.
• Does the animal have the floor area the table gives for her weight? If not, you know how much you are short by.
• Is there clean water, in enough quantity, for both drinking and washing? If not, no amount of milking discipline in the next modules will save your milk quality.

One last thing to keep in proportion. All the dimension figures in this module come from the FAO structural design reference, which was written for tropical climates generally, not developed for East African smallholder zero-grazing units specifically. Locally tailored numeric design figures were not available for this course. These are the best sourced starting point you have. If your national extension service or livestock research institute publishes figures for your country, theirs is the number to build to.