The rangelands are most of the pastoral map and almost none of the television. They run from the Mitchell grass plains of the Barkly and the Queensland Gulf, through the Channel Country and the Kimberley, down into mulga, saltbush and the long leases of the Western Division, the South Australian pastoral country, and the Gascoyne–Murchison. Rain is an event, not a season you can set your watch by. Regeneration, in that country, is a question about rest, animals and water — and about what you will admit when the rain does not come.

I have stood in a trap-yard west of Longreach after a failed wet, and in a saltbush paddock north of Broken Hill after a good one, and heard the same word used for both. The word is doing too much work. The Land needs a smaller, more honest version.

Two rangelands, not one method

The northern rangelands, broadly, are summer-rainfall country. Storms arrive from the north-west and the monsoon, if they arrive. Mitchell grass (Astrebla) on cracking clays, spinifex on the poorer sands and gravels, and the floodouts that make the Channel Country a rumour of plenty — these are not the improved pastures of Bega or the Hunter. Cattle can put on weight in a good wet as if the country were designed for it. In a failed wet they walk the same pads to the same water and eat the last of the perennial tussocks that were supposed to be the bank.

The southern rangelands are a different climate wearing a similar hat. Winter and spring rain, when it comes, sits on mulga, belah, myall and the old-man saltbush of the Riverina and the South Australian margins. Growth is slower, recovery is slower, and the temptation to stay a month too long is the same. A method copied from a humid New Zealand dairy video will not translate. Neither will a method copied from the Barkly without looking at the shrubs.

CSIRO, Meat & Livestock Australia, and the rangeland NRM groups have been saying a consistent, unglamorous thing for years: match stocking to forage, rest country that has been hammered, and do not confuse a green year with a new carrying capacity. Landcare groups on the edges of the wheat–sheep zone already know this. The interior is a drier version of the same arithmetic.

Rest, density, and the water that writes the map

Rest is the part of regeneration that is not a brand. A paddock that is not being eaten can grow leaf, set seed, and put something back into a crown. How long that takes depends on the plant and the year. Mitchell grass may need a full growing season off. Saltbush can be browsed harder than people think, then die in a drought if the last green pick is taken. Spinifex will wait you out.

Stock density is the part that photographs well. Put a mob in a small paddock and the feed disappears evenly; the cattle do not camp on the sweet flats and ignore the ridges. Short-duration grazing can work where fencing, labour and water allow it. It can also become a theatre in which the same number of animals still eat the same annual forage, only in a more Instagrammable order. Density without destocking is rearrangement.

Water points write the real map. Before tanks, bores and poly pipe, large parts of the inland were grazed only as far as a beast would walk from a natural hole. Artificial water opened country that had been a refuge. The piosphere — the ring of bare ground, weeds and compacted pads around a trough — is as much a rangeland landform as a dune. Every new tank is a decision about where the next ring will be.

A water point is not a kindness by itself. It is a vote for which country will be used, and which country will finally get to rest.

If you are going to move cattle the way the books suggest, you need waters that let you do it, and the will to shut a valve. A regenerative plan that adds waters without adding rest just extends the worn country. The goats will thank you.

The grazers you did not buy

Feral animals do not read a grazing chart. Goats have remade stretches of the New South Wales Western Division, the Flinders Ranges, and the rangelands of Western Australia. They take browse the cattle cannot, they climb, and they keep eating when a cow has already stopped. Camels use the arid zone as a network of soaks and dunes. Donkeys still press the Kimberley and the Victoria River District. Horses — brumbies in the polite telling — concentrate on the better country and the water. Rabbits, where they have not been knocked back, still scalp seedlings that were meant to be the recovery.

A manager can rest a paddock from cattle and still have it eaten. Any honest regeneration budget in the rangelands has a feral line: mustering, trapping, aerial work where the state permits it, and the unfashionable admission that a "rewilded" goat is still a goat. State departments and the Australian Government's feral-animal programs publish the unromantic numbers. Landholders already have the romantic ones.

Kangaroos are not ferals, and they are not a rounding error. In a dry year they sit on the same remaining green as the sheep. Harvest and exclusion are political in every hall from Cobar to Carnarvon. They still belong in the forage budget. Pretending otherwise is how a rest paddock becomes a kangaroo lawn.

What regeneration can do, and what a drought will not allow

In a run of decent years, better grazing can do things you can see from the ute. Ground cover holds after a storm. Preferred grasses return in the interspaces. Litter stays attached instead of blowing to the fence. The cattle walk more of the paddock, so the creek frontage is not the only place that looks used. Soil carbon on working farms may tick up a little in the top few centimetres if the litter and the roots persist — a little, in a country that was never a carbon sponge.

In a drought, regeneration cannot grow feed. It can decide how fast you spend the last of it. The useful practices are old: early destocking, sacrifice paddocks, turning waters off, and not believing a forecast that has already been wrong twice. A cell-grazing diagram does not wet a dry profile. A compost extract does not wet it either.

There is a cruelty in the sales version of this story. It tells a family on a failed Barkly wet, or a mulga block in the third dry winter, that their country would have held if they had loved it in a more certified way. Some of those families already loved it. The rain did not come. CSIRO's rangeland work, and the pastoral boards that still remember the 1960s and the Federation Drought in the older files, are blunt about the ceiling. You cannot rest your way out of zero growth.

What you can do is refuse the practices that make the next drought worse: set-stocking the sweet country until the tussocks are stools, opening waters into the last ungrazed corners, and carrying a herd into summer on a hope. That is management. It does not need a new adjective every decade.

How a civic habit fits a lease

How Landcare actually works is relevant even where the group is thin on the ground. A fencing bee, a total-grazing-pressure count, a shared trap-yard, a conversation with the NRM body about a waterpoint that should never have been put on the dune — these are civic tools, not branding. The rangelands are too big for one hero paddock.

Households on the edge of town, reading this from a garden tank, will find a smaller rhyme in drought-ready Australian gardens. The scale changes; the refusal to pretend that enthusiasm is rainfall does not.

I like the managers who talk about rest as if it were a fence they can show you, not a feeling. I like the ones who destock without a press release. The rangelands will green when they green. The work is to still have country — and a few perennial plants worth the name — when they do.