What a Cow Does to Veld, and What Veld Does to a Cow

#Most of this country grows grass and nothing else

Rangeland is roughly 80% of South Africa's land surface; only about 15% is arable (Kirkman et al.). On most of that rangeland there is no crop, no orchard and no pasture-establishment plan that pays. There is grass, there is a dry winter, and there is a question about what harvests the grass.

A cow is the answer for a specific reason, and it is not sentiment. She is a bulk roughage harvester: an animal built to take large mouthfuls of mature, stemmy, low-quality material and ferment it into something useful. The South African Large Stock Unit is defined around exactly that appetite — an animal of 450 kg gaining 500 g a day, needing about 75 MJ of metabolisable energy and 9 kg of dry matter a day (Jordaan et al. 2021). Nine kilograms is 2% of her bodyweight, every day, in dry matter.

That appetite is the ecological tool. It is also the liability, and the whole of this module hangs on holding both in view at once.

#The job a browser will not do

Ask what happens to a Highveld camp with no large grazer in it. The grass grows, sets seed and stands. Next season's growth points sit under last season's stems, moribund material accumulates, light does not reach the base of the tuft, and the palatable Decreaser grasses lose ground to whatever tolerates shading.

Different animals attack that problem differently, because they eat differently. The working rule of thumb — indicative, not South African trial data — is that cattle take about 70% of their diet as grass, sheep about 50% grass and 30% forbs, and goats about 60% browse. Those proportions are the mechanism behind multi-species grazing, and they are the reason a goat is not a substitute for a cow on grassland. A goat will work on the Vachellia and the Seriphium; she will walk past the coarse standing grass that is the actual problem.

So the cow's specific contribution to a sward is removal of bulk, plus the trampling and dung deposition that come with a large animal moving in a group. The gains from running species together are real and they are mostly dietary and behavioural: the species use different strata of the sward, browse works against graze, and each species avoids only its own dung patches, so total utilisation of the camp rises (SA Smallholder).

One tension is worth naming. A burn hot enough to check encroaching shrubs needs a heavy load of standing grass to carry it, and cattle eat exactly that fuel. Grazing a camp hard and then hoping to burn the bush out of it are two plans fighting each other. Decide which job that camp is doing this year.

And there is a limit on removal that is not negotiable. The DALRRD capacity map is built on a 40–50% utilisation factor, and the field heuristic Highveld farmers actually use is the beer can: the grass should never be higher than a can standing up, nor lower than the can lying on its side (Voermol). The cow's job is to take the top, not the base. Residual is the exit rule, and the frosted Highveld winter is what makes it non-negotiable — bare soil in June is erosion waiting for the October storms.

#Three numbers, restated for a cattle herd

Module 3 built these. Here they are with cows in them, because this is where they get used.

• Grazing capacity is what the land can carry long-term, in ha/LSU. The department publishes biome means of Grassland 6, Savanna 12, Nama-Karoo 25 ha/LSU, with polygon values nationally ranging from 1.5 to 140 (DALRRD 2016).
• Stocking rate is what you actually run.
• Stocking density is LSU per hectare at a moment, in the camp being grazed. It carries no time term.

The ARC's blunt framing of the first two is that capacity and rate "sometimes differ by as much as 300%, leading to total degradation" (Mokolobate et al. 2017). Nobody sets out to overstock by 300%. They set out to count heads.

The planning currency that actually adds up is stock days per hectare: 365 ÷ (ha/LSU). On 6 ha/LSU veld that is 60.8 SDH per year (regenagsa).

Now do it on your own animals, and do it in LSU rather than heads. The ARC publishes frame-based factors because LSU is not linear in bodyweight: a 450 kg small-frame cow with a calf at foot is 1.32 LSU and eats about 12 kg DM/day; the same 450 kg cow in a large-frame breed is 1.60 LSU and eats about 15 kg DM/day — a 25% difference in intake at identical liveweight. Lesson 2 turns that into hectares. For now, the point is that your ledger is written in LSU and your kraal is counted in heads, and the two are not the same document.

#The physical enabler nobody budgets for

You cannot graze for residual, move a mob daily, or run a leader-follower sequence with cattle that will not handle. Every grazing plan in this course assumes an animal you can put where you want it, calmly, with two people and no shouting.

The working principles come from Bud Williams and they transfer here, though they are international rather than South African in origin. Animals want to move where they are headed, to see what is pressuring them, to exit where they entered, and to follow other animals (Grandin on Williams' methods). Almost every handling failure you have watched is one of those four being violated: pushing a mob at a gate they were not already pointed at, working from a blind spot behind them, asking them to leave by a strange opening, or splitting an animal off alone.

The cost of getting it wrong is not just time. Excited cattle take 20 to 30 minutes for heart rate to return to normal after a few seconds of agitation (Grandin). Lesson 4 shows you what that same roughness looks like stamped on the carcass in brown ink.

What is genuinely South African about this is that herding is three decisions at once: a grazing technique, theft and predation control, and a wage bill. Night-kraaling, the oldest herding practice in the country, maps neatly onto herd-effect nutrient concentration — and just as neatly onto parasites and hoof damage concentrated on one site.

#Leader-follower, stated correctly

Run cattle first and small stock behind them, and you will be told at every farmers' day that the camp is now clean, because cattle and small stock do not share worms.

They do. In South Africa Haemonchus species, Trichostrongylus axei, Cooperia species and the liver flukes Fasciola hepatica and F. gigantica all cross between cattle and small stock. A camp grazed by cattle is not a clean camp for sheep, and believing otherwise kills lambs.

What is true is weaker and still worth building into a grazing plan. Most of the economically dominant nematodes have a strong host preference, so cattle ahead of small stock, with an adequate rest period between, dilutes the larval challenge each species meets rather than concentrating it — which slows the selection pressure driving anthelmintic resistance. Module 7 handles the drenching decision itself. Note what nobody can give you: no South African trial quantifying that dilution benefit could be found for this course. The mechanism is sound; anyone who attaches a percentage to it is guessing.

#What the veld does back

Now turn the relationship around, because this is the half farmers manage worst.

Take one cow on Delta Farms in July. She weighs 450 kg, she needs roughly 9 kg of dry matter a day, and she is standing in a camp with more visible bulk in front of her than there was in December. Sourveld occurs above roughly 650 mm on acidic leached soils with cold winters and supplies adequate nutrition for only about 5 to 7 months of the year (Voermol). By July, dry-season veld runs at 2 to 6% crude protein against a rumen threshold of about 7% (Virbac SA).

Below that threshold the rumen microbes cannot multiply fast enough, fibre digestion slows, the rumen empties slowly, and voluntary intake falls. She does not eat 9 kg. She eats less as the grass gets worse, which is the opposite of what a farmer standing at the fence expects. Module 6 built this out in full; what matters here is where the bill lands.

It lands on her body condition, and body condition lands on reconception. A cow that loses condition through July and August walks into the November mating season in the wrong shape, does not get in calf, and then eats a full year of grazing capacity — at least six hectares of your farm on Grassland-biome veld, more once she is carrying a frame above 1 LSU — and weans nothing. That is the actual cost of the winter protein gap, and it is paid fourteen months later, which is exactly why it does not get attributed to the right cause.

That is the relationship the rest of this module sits on. The cow is the harvester that keeps the sward functioning and the soil covered, and she is a client whose winter intake decides next year's calf crop. Do the conversion in the box above before the next lesson — because if your LSU total and your land-type value are far apart, no breed decision will fix it.