The strip of desert between the Andean mountains and the Pacific Ocean has an annual average rainfall as low as 1.5mm.
By way of comparison, London enjoys around 650mm a year.
It's not an obvious place to choose if you're looking for somewhere to plant trees, but for restoration ecologist Oliver Whaley the harsh environment of the northern fringes of the Atacama desert is part of the point.
By helping to restore the shrinking native forests, the aim is to benefit local people and wildlife, prevent soil erosion, and help alleviate climate change.
"If we can get trees established here, and learn how to do it with as little water as possible, then it is a model for the rest of the world," he says.
While the plight of the world's rainforests are well known, the same cannot be said of tropical dry forests. These less biodiverse, but equally remarkable forests, face threats every bit as severe as their better known cousins.
The Atacama dry forest "is really an ecosystem on its last legs," says Mr Whaley, of London's Kew Gardens - an internationally renowned botanical research institution.
The tree under threat is the huarango, Prosopis limensis, found only in the Ica region of Peru.
In this parched landscape, the hardy huarango is no stranger to thirst. Although rain seldom falls, it is able to capture moisture from other sources - trapping fog on its leaves, directing the water downwards towards its roots. The roots themselves are among the longest of any plant - 50m to 80m - and seek out underground water sources that flow from the Andes.
Burning bush
The huarango is also a valuable source of food and fuel, and a keystone of the local ecosystem. Whaley estimates that when he arrived in Peru, just 1% of the original local forest habitat remained - much of it consumed in charcoal production.
The problems facing dry forest habitat are not unique to Peru. Restoration expert James Aronson says, in general, these are more critically endangered than wet tropical forests, in respect of the total percentage already lost.
Mr Aronson, who heads the Restoration Ecology Group in the Centre for Functional and Evolutionary Ecology in Montpellier, France, adds there are more than 1,000 species of tree that grow in desert areas.
"In many desert areas of the world there used to be enough trees to constitute to a real canopy - not in the sense of an English forest, where the forest is everywhere you look - [in deserts] they were restricted to areas where there was sufficient water."
With so little forest in existence (see map of original forest range ) conservation, on its own, would not be enough to preserve the trees, leaving restoration as the only option.
Hard times
It's a situation made more complicated by the global recession.
Some local agro-industries growing asparagus, grapes and oranges for foreign markets are laying off workers. With hard times on the way, Mr Whaley worries that what little vegetation remains may be used for firewood.
Whatever his project's success, Mr Whaley is certain that the future of the tree rests firmly in the hands of the local people. They are encouraged to help with planting, and tree nurseries and seedbanks have been sent up in communities and schools.
"We are not going to Peru saying we are going to reforest the whole of the coast. We are developing a model that we can replicate and hopefully we can get that to be so interesting, or fun, or useful that it's contagious."
Hence the idea of an annual huarango festival, started in 2006 and held in April. The festival is a chance to celebrate the tree and the ecosystem it helps support. But it is also about food. The fruit of the tree can be used to make syrup, similar to molasses.
"Fill people's tummies," says Mr Whaley. "Where do you get social science and biodiversity overlapping? In the stomach. That is the best place to do it."
The project has already had some successes, with a reserve set up in Tunga (see map, below), and more planned.
"We are at the beginning of habitat restoration - it's only a science that's been around for a couple of decades. Particularly in arid areas, we are only just learning how to do it.
"If I am able to come back in 500 years, then I would know if it has been a success. I will never know in my lifetime if its been a success."BBC News
