Build Your Farm File

#Three rainfall figures and no rain gauge

This module has already left Delta Farms with an honest gap: gridded portals put Meyerton near 732 mm a year, this project's own frame assumed 600–700 mm, and nobody has a station record. There is a third figure, and it makes the gap worse. A widely used climate portal gives Meyerton about 520 mm a year — NASA reanalysis model output, not a rain gauge. Its rainfall figure is modelled; only its temperature estimate blends nearby station data.

Three numbers, none of them a gauge on that farm, spanning 40%. And they land on opposite sides of the line that decides what kind of farm it is: sourveld is associated with rainfall above roughly 650 mm on leached, cold-winter ground, and mixed veld with roughly 500–600 mm. At 520 mm this is a mixed-veld property carrying a drought reserve as well as a fodder flow. At 732 mm it is a sour property where the winter fodder flow dominates everything else. Same coordinates, two different management years.

Somebody has to read a gauge on that ground, and pull the Weather Service record for the nearest long-run station. Until then every number this course computes for Delta Farms carries an asterisk — and so does every number you compute for your own farm.

#The six facts

There are exactly six things to go and find. Everything else in this course is arithmetic performed on them.

1. Your grazing-capacity polygon value, in hectares per large stock unit, under the Conservation of Agricultural Resources Act 43 of 1983 (CARA).
2. Your veld type — biome, vegetation unit, and sour, sweet or mixed.
3. Your soil forms, at crest, midslope and valley bottom at minimum.
4. Your water sources and their legal status — where the water is, what it yields, what is in it, and whether you are lawfully entitled to it.
5. A camp map with measured hectares, not paced, and not estimated off a title deed that includes homestead, roads and rock.
6. First and last frost dates, recorded on your own ground.

#Counting what you actually have

You now have a polygon value and a hectare figure. The step where most farm files go wrong is the third: converting your animals into the currency the map is written in.

Large Stock Units are not head. CARA Regulation 11(2) and its Table 2 set the legally applicable equivalents, and they are not the round numbers people carry in their heads:

• medium-frame beef cow or heifer, about 525 kg — 1.21 LSU
• large-frame cow, about 550 kg — 1.32 LSU
• weaner from seven months, about 200 kg — 0.53 LSU
• mutton-breed ewe, about 53 kg — 0.15 LSU; woolled ewe, about 47 kg — 0.14 LSU

A herd of forty medium-frame cows is not forty units. It is a little over forty-eight, before you count the bull, the weaners, the horses and the resident game — all of which eat, and all of which count.

Elsenburg's worked example is small enough to feel safe, and is not. A farm runs 100 wool ewes, 110 lambs, 2 rams and 4 cattle on 200 ha of veld at 30 ha/LSU. By head count it looks like a modest operation. Converted on the Meissner class factors Elsenburg uses — ewe 0.15, lamb 0.10, ram 0.19, cow 1.0 — it is 30.38 LSU.

Against the full-year capacity of 200 ÷ 30 = 6.7 LSU, that is a 4.5-fold overstock. Against the working figure where the veld is only used for eight months of the year — 200 ÷ (245/365 × 30) = 10 LSU — it is a threefold overstock. Both denominators are correct; they answer different questions, so say which one you are using. Either way the farmer counting sheep would never have seen it.

That is the whole argument for measured hectares and stated conversion factors sitting in a file rather than in your memory. And note that those Meissner factors are close to CARA's Table 2 but not identical — a woolled ewe is 0.14 there — so the file has to say which set produced the number. Elsenburg's own recommendation follows from it and is worth taking literally: keep fewer animals than the recommended grazing capacity allows. The reserve you leave standing is what makes a dry year survivable without buying fodder.

#Soil, water and the frost line

A soil form name is a water-behaviour forecast. On a Highveld catena you typically walk shallow Mispah or Glenrosa on the crest, Hutton or Avalon on the footslope, and Katspruit in the valley bottom — three different grazing, trampling and infrastructure decisions inside one 20 ha camp. The plinthic forms mottle because a water table perches seasonally, and roots stop at the plinthite, which caps how much dry weather your grass can ride out. The dark swelling clays crack when dry and seal when wet, and are the soils to keep a heavy mob off when saturated. Dig to 30 cm at three marked points, photograph them, and write down the depth to the first restricting layer.

Water needs two pages, and most farmers only keep one. The physical page is yield, recovery rate at the trough when a whole mob arrives at once, and a laboratory analysis. Salinity, sulphate and nitrate are the common South African offenders — the guideline target range for nitrate in livestock water is 0–100 mg/L — and a borehole that yields well can still sit outside the range.

Frost is where a portal figure fails quietly. The reanalysis portal that gives Meyerton 520 mm also gives it a frost-free window of about 284 days, from mid-August to late May. Another gives roughly five days a year below zero at the town station. Both are screen-temperature figures measured in a town. Ground frost on open Highveld veld is far more frequent, and growth stops well before the thermometer reaches zero. The 284-day frost-free window looks generous for the Highveld, where frost commonly runs May to August, and it has not been checked against a station record. Nobody has published the ground-frost frequency for that property. Nor for yours. This one you can only measure.

#The file itself

Keep it physical — a folder you could hand to a bank, a valuer or the next generation.

It grows by roughly a page per foundation module. This one contributes the six facts and a provenance note saying which of your numbers are gauge-measured, which are map-read and which are borrowed. Soil and water adds the spade records and the water analysis; the grazing ledger adds the veld-condition survey and your honest LSU count; later modules add the fodder flow, the written drought triggers, the compliance calendar and the enterprise shortlist. In the last module you assemble those pages into one worked plan — and the plan will only ever be as good as the file underneath it.

Two habits carry the whole thing, both free. Read a rain gauge on your own farm daily and write it down: in five years you own something no portal can sell you, and in ten you own the variability figure that decides your stocking policy. Write down the first and last frost every season. Those two records will outlast every number in this course.

Get the map value this week: a copy is held for inspection at the extension office. Until you have it, every stocking figure you have ever been told about your farm, including the ones in these lessons, is a guess wearing a decimal point.