Where You Are Starting: the Baseline Audit

#The photograph nobody took

Across 2 380 soil profiles compiled by the Agricultural Research Council, 58% of South African soils hold less than 0.5% organic carbon and only 4% exceed 2% (du Preez, van Huyssteen & Mnkeni 2011). Detecting a real change in that, against South African background variability, generally takes five years and a great many cores — and a carbon percentage measured without bulk density at a fixed depth cannot be converted into a stock at all.

Mark the points and fix the depth increments before you change anything and you get a slow answer in five years. Skip it and you get no answer at all. That applies to far more than carbon. In three years somebody — a bank, a lessor, a buyer, or you at two in the morning — will ask whether it worked, and the measurement that answers them can only be taken now.

#What the baseline actually is

Your farm file holds site facts — the polygon value, the veld type, the soil forms, the water, the measured hectares, the frost dates. Those describe the place. A baseline is a dated record of the farm's condition and performance on the day before you started. Almost every piece already exists as an artefact of a module you have worked through; the job is assembly, not discovery.

Point 5 is where the audit surprises people. Large Stock Units are not head, and the conversion is a legal one: Regulation 11(2) and Table 2 of the Conservation of Agricultural Resources Act 43 of 1983 put a medium-frame beef cow of about 525 kg at 1.21 LSU, a weaner from seven months at 0.53, a woolled ewe at 0.14 and a light-horse mare at 0.94 (CARA regulations). Sheep and goats are talked about in Small Stock Units, but Table 2 converts them straight to LSU — use those factors, keep one ledger, and label every line with the unit it is in. The often-quoted "1 LSU = 6 SSU" shorthand was not traced to a South African primary source for this course, so do not build a stocking decision on it. Game go in too, in their own currency: a Grazer Unit is a mature 180 kg blue wildebeest, a Browser Unit a mature 180 kg greater kudu — same body mass, different feeding class (Shepstone et al. 2025).

Grazers and browsers go on separate axes, because the departmental capacity map does not incorporate tree crown cover and does not include browsing capacity at all (DALRRD). No South African biome-level norm for hectares per Grazer or Browser Unit was found in the research behind this course, so count your browsers and do not divide them by a number you cannot source.

#Who is allowed to say yes

One question belongs before the grazing plan rather than after it: on this land, who can lawfully authorise a change in grazing? Four answers are common in South Africa, and they are not variations of one answer.

The title-deed owner decides alone — subject to the law. CARA Regulation 9(1) obliges every land user to protect veld effectively, using as many of the listed measures as are necessary: alternating grazing and rest with due regard to the physiological requirements of the vegetation, keeping animals of different kinds on the veld, restricting numbers to the Regulation 11 figure, and soil conservation works. Where veld is deteriorating, Regulation 9(1)(e) requires numbers to be suitably reduced, or the deteriorating portion withdrawn until recovered, or a grazing crop established in addition to the existing vegetation — and Regulation 9(2) lets the executive officer direct further measures in writing.

The lessee decides what happens this season. Year four belongs to the lessor.

A communal grazing committee, a Communal Property Association or a trust is a body, not a person. Between 40% and 50% of the national cattle herd sits in communal and emerging hands, and on open-access rangeland one household's rest period is another's grazing opportunity. The first deliverable there is a written, agreed and enforced rule about rest, not a chart. This course could not confirm the statutory citation for Communal Property Associations, so take the governing instrument from your own association's constitution and the department.

An ESTA occupier is the one people forget, and the one that turns a grazing plan into litigation.

#What a lease does to a three-year plan

Recovery works on the plant's clock, not yours. The survey that produced South Africa's best fence-line evidence could only run because it found farms under consistent management for 15 ± 0.8 years (Venter, Cramer & Hawkins 2019). Veld improvement is a slow asset, and anything bolted to the ground stays with the ground.

#The gaps, and a date against each one

An honest baseline records what you do not know as carefully as what you do, and dates each gap. Four facts about the course's own worked example remain open after a full research pass, and they are the model for your file.

Two of those deadlines are a phone call. Two are a year — which is exactly why they start the week you start, not the week you need the answer.

#What to do this weekend

Walk three transects, drive three pegs, take three photographs, dig three holes. Count every animal on the place — horses, donkeys and resident game included — and convert with the Table 2 factors, not the ones in your head. Open the notebook with the seven fields and date the first page. Then ask the question this lesson exists to force: who has to agree to this change, and have they?

The rest of this module builds a year of change on that page. If the page is blank, everything on top of it is an opinion.