Nothing here
is dead

The runoff problem

Stand on degraded rangeland in southern Kenya at the start of the rains and you will see something that looks like a mistake. Water falls on ground that has been grazed bare and baked hard, and the ground refuses it. Instead of sinking, the rain gathers into sheets, finds the slope, and leaves — carrying the finest, most fertile particles of soil with it.

The land is not short of rain so much as short of time with rain. A downpour that would transform a field somewhere else runs off a sealed surface in minutes. What is left behind is drier than before the storm, because the storm took soil with it on the way out.

This is the problem a Dutch foundation called Justdiggit has been attacking since 2009, with something close to the simplest tool available. Their entire theory is that if you can make the water stop for long enough, everything else the landscape needs is already there, waiting.

Scaling simplicity

One bund holds a bathtub and a half of water. That is not a climate intervention. What makes it one is repetition: Justdiggit reports more than 800,000 bunds dug across its project sites in Kenya and Tanzania, alongside 260 kilometres of Fanya Juu and Fanya Chini trenches. The communities dig them themselves — which the organisation frames as the point rather than a cost saving, because a bund nobody feels ownership of is a bund nobody maintains.

Below is a hillside of roughly one hectare. Right now it sheds most of what falls on it. Tap the slope to dig, choose a rainfall year, then run six months and watch what the ground does. Switch to thermal view to see the thing that is hardest to believe until you measure it: green ground is cooler ground.

Two of the controls are there to keep you honest. Drought is one. The other is livestock: regrowth that isn't protected from grazing gets eaten, and no amount of digging fixes that.

Dormant life

Here is the part that changes how the landscape looks once you know it. Scattered across farmland all over the region are grey, cut stumps — the remains of trees taken for charcoal, timber and clearance, sometimes decades ago. They read as evidence of loss. Many of them are alive.

A felled tree with an intact root system doesn't necessarily die. It sends up shoots, which get cut, grazed or burned, and it sends up more. In Swahili the technique for working with this is called Kisiki Hai — living stump. You don't plant anything. You choose a stump, protect it, and prune the shoots so the root system's energy goes into a few strong stems instead of thirty weak ones.

A single stump can throw between 10 and 30 stems in a year. Managed this way, survival exceeds 80% — far better than planted seedlings, which need a nursery, transport, watering and luck. Since 2017, Justdiggit and the LEAD Foundation have run this at scale in Tanzania under the name Treecovery; farmers there have brought back more than 17.5 million trees.

The technique is not theirs and not new. In Niger, farmer-managed natural regeneration restored roughly five million hectares and some 200 million trees, adding an estimated $900 million a year to the country's agricultural output — at a cost measured in tens of dollars per hectare, most of which is the farmer's own labour.

Grass banks and grazing rules

A bund is a hole. Anyone can make one in an afternoon. The reason regreening projects fail is almost never that the hole was dug wrong — it is that nothing protected what grew in it, or nobody had a reason to come back next year.

So a surprising share of the actual machinery here is agreement. In Kenya, regrowth is protected inside olopololi — grassland enclosures kept closed to livestock so that grass can set seed before it is grazed. That is a decision a community makes about its own animals, and it costs something in the short term.

The clearest example is the grass seed banks. Small parcels of communal land are given over to growing grass for seed, and they are managed by Maasai women's groups who are trained in sowing, weeding and harvesting. The seed is sold on local markets or on to other restoration sites. Justdiggit reports 43 seed banks involving 879 women — an income stream that exists because the grass is worth more standing than grazed.

Read the intervention that way and the shovel stops being the invention. The invention is a reason to leave the grass alone.

The honest ledger

Justdiggit is a Dutch foundation with public-benefit (ANBI) status since 2010 and recognition from the CBF, the Netherlands' charity regulator — which requires independent assessment of governance, fundraising and reporting. The regulator's published figures for 2024 show €6.52 million raised against €7.00 million spent, of which €5.44 million — a little over 77% — went directly to the objective. It spent more than it raised that year, drawing on reserves of €4.32 million. The mechanism it uses has decades of evidence behind it. None of that makes every claim on its website equally solid, and four are worth holding up to the light.

What survives all four objections is the core observation, and it is a good one. On crusted ground, the difference between water that leaves in minutes and water that stays long enough to be used is a shape you can cut with a hand tool. The seeds are already in the soil. The roots are already under the stumps. Most of the landscape is not gone — it is waiting, and it is cheaper to wake than to rebuild.