Grassy ecosystems are under fire. Aisling Irwin investigates the world's booming tree campaigns.
When Madagascans flocked to the country's barren Central Highlands in January this year to help plant a million trees in a day, botanist Cédrique Solofondranohatra stayed at home.
Vehicles jammed the highway from the capital Antananarivo, the mood was jovial and an estimated 12,000 people turned up in Ankazobe, north-west of the capital. But Solofondranohatra was not in the party mood: she was convinced the tree-planting was destroying an ancient ecosystem.
Solofondranohatra says that an ancient grassland was being planted - a place whose rich biodiversity is only now being elucidated, and which serves its people best without the disruption of trees. Ancient Madagascan grasslands are, she believes, falling victim to a modern frenzy to afforest the world that has gripped political leaders from Madagascar to Ethiopia and from Turkey to the United States.
Solofondranohatra, a PhD student at the University of Antananarivo and Kew Madagascar Conservation Centre, has just published work suggesting that the common view that the grasslands are degraded forest, razed by humankind and in need of replanting, may be wrong.
Instead, at least parts of the grasslands appear to be crammed with species that have evolved over millions of years to live there and nowhere else in the world. That is, it is not somewhere to try and set trees growing, particularly if they are fast-growing alien species such as eucalyptus or pine.
"Knowing that where they were going to put the plantation might have been ancient grasslands, I was not okay with it," she says.
As presidents, corporations and billionaires scramble to turn hundreds of millions of hectares globally into forest, ecologists fear that less fashionable ecosystems, and the people who use them, are under threat. They are particularly worried about the open canopy woodlands, savannahs and grasslands that together make up what are known as ancient grassy biomes. Already depleted by agricultural expansion, they may face a second wave of encroachment as the planting of forests gathers pace, with damage to water supplies, long-term social and cultural consequences as grassland ways of life become harder -- and even a net loss of carbon.
Forests are seen as "intrinsically more valuable," says Caroline Lehmann, tropical biologist at the Royal Botanic Garden Edinburgh.
But, they are not the "ultimate ecosystem", says Guy Midgley, a biologist and grasslands specialist at Stellenbosch University in South Africa.
"There are parts of the world where ... they're not the richest ecosystems in play ... In fact, some of these open ecosystems, these grasslands, are incredibly rich with ancient lineages of diversity that have existed for millennia."
The number of trees pledged to be planted worldwide can be hard to grasp.
The Bonn Challenge aims to bring 350 million hectares of the world's deforested and degraded land into restoration by 2030. The European Commission is aiming to plant three billion trees.
The organisation 8 billion trees plans to "plant and save" that number; YouTube stunt philanthropist Jimmy MrBeast Donaldson is aiming for 20 million. National presidents are also planting.
For context, there are thought to be three trillion trees already in the world.
But where are these new trees to be planted? Last year, scientists from the Swiss research institute ETH-Zürich and the Food and Agriculture Organization (FAO) argued there were 900 million hectares of space on the planet where trees could be planted without affecting existing cities or agriculture. A few years earlier, scientists at the World Resources Institute (WRI), in the United States, published an influential digital atlas showing that two billion hectares of degraded land worldwide provided an "opportunity" for restoration.
But critics say vast, empty, plantable spaces mostly do not exist.
Grassland experts have calculated that of the 23 million square kilometres identified in that atlas as deforested or degraded land that could be restored, nine million square kilometres were actually ancient grassy biomes. These "could be destroyed by misinformed forest restoration projects," a research group, led by Joseph Veldman, an ecologist at Texas A&M University in the United States, wrote in a 2015 paper. Far from being empty or degraded, these grasslands are supplying the needs of up to one billion people.
"In order to get to the numbers [in the atlas] you have got to afforest the Serengeti," says Midgley.
Research such as the WRI atlas has heightened ambition. In Africa, for example, the New Partnership for African Development, NEPAD, has worked with the WRI, the World Bank and the German government to create the African Forest Landscape Initiative (AFR100), committed to restoring 100 million hectares by 2030. Since its creation in 2015 it has been over-pledged by 26 million hectares.
Among some spectacular commitments, Cameroon has pledged to restore one quarter of its land, some 12 million hectares. Madagascar, meanwhile, has committed to four million hectares. This year is a special year for tree-planting in Madagascar with a goal of 60 million trees -- a million for each year of independence -- of which the 1.2 million seeds and seedlings folded into its red soils on January 19 were part.
While there are many reasons to restore ecosystems, mostly to do with bringing back their ability to supply human needs such as wood, fish or water, one new imperative stands out: to absorb more carbon.
As the effects of climate change become more tangible by the day, leaders are turning to nature in the hope that it can be enticed to absorb some of the excess carbon dioxide. The ETH-Zürich and FAO scientists controversially calculated that such efforts could absorb the equivalent of two-thirds of the carbon dioxide we need to remove from the atmosphere in order to keep global temperature rises below 2 degrees Celsius. They have now admitted this figure is too high, but it remains a seductive offer for all of us, including companies with budgets targeted at offsetting their production or use of fossil fuels.
"Conservation has moved from ideas of biodiversity to how much carbon can you capture -- and then to the idea that forests should be the focus of carbon sequestration," says Veldman.
"Every new minister wants to do something big and giant especially when Ethiopia announced that they had been planting 350 million trees in a day," says Lucienne Wilmé, national coordinator for WRI in Madagascar.
In the face of this hunger for more trees, there is a massive search going on for the land on which to plant them. Grasslands can seem the obvious choice. To many, they look like they once held forest and are thus a symbol of humanity's environmental destruction, says Lehmann.
"They are perceived as trashed ecosystems," says Midgley. "People say, 'Oh, there's enough rainfall here to plant a forest let's put a forest here!'."
But grasslands are important in their own right, say their advocates.
Grassland vs tree land
Grassy biomes are shaped by the sun. Where life on the forest floor craves the dim and the moist, life within grasslands thrive on drought, fire or herbivory -- depending on latitude.
Just like forests, they vary richly across latitudes. In Africa's south they can be empty, arid savannahs, pocked by the occasional, fire-hardy tree. Further north, towards the wetter forest boundaries in places such as Zambia and Cameroon, trees are more common, increasing to as much as 60 per cent canopy cover. Crucially, there is always grass -- by definition -- and there are always gaps between the trees.
And below the height of the grass live tiny plants with miniature flowers whose differences can only be distinguished with a magnifying glass.
Grasslands are not as charismatic as forests, Lehmann concedes.
"Trees are bigger and have big flowers and they often have animals sitting up in them. In grasslands it's very hard, if you're not actually a little bit short-sighted, to differentiate one species of grass from another."
Put together the search for somewhere to plant trees with the perception of grasslands as being degraded land and you have a powerful agenda.
Madagascar President Andry Rajoelina says he wants to turn his country from the so-called "Red Island" into a "green island" -- covering it with forest being the symbol of environmental good, says Veldman.
But, grassy biomes offer important ecosystem services to those who live in and around them -- estimated to be hundreds of millions of people worldwide -- and use them for grazing, fuel, medicines and water.
"I'm not sure our modern civilisations would be here without grasses, as they underpin global food security," says Lehmann.
It may seem counterintuitive but, in water-scarce catchments, healthy grassy biomes are often better than forests at providing a good water supply to people. Rain streams off grasslands and into rivers, while trees suck water up through their roots and whisk it into the atmosphere, a process known as evapotranspiration.
A study published earlier this year looked at the consequences, over decades, of tree-planting near rivers. Trees planted on natural grassland where the soil is healthy decrease river flow significantly. And the system does not bounce back: the reduction in river flow lasts for at least half a century, found the scientists from Cambridge University.
South Africa has reaped the consequences of just such planting in its grasslands. Years ago, to supply its timber needs, fast-growing alien trees were planted on extensive areas of grasslands. Many have escaped into surrounding land.
"They grow very fast and they shade out a vegetation type that is adapted to high light conditions," says Midgley. "It has devastating effects on biodiversity, but also these trees use a lot more water than the endemic... vegetation. So, huge landscapes that were previously yielding a lot of beautifully filtered water were being encroached by invasive trees."
That is why, since the mid-1990s, the Working for Water programme has employed thousands of people to dig out the invasive trees -- though it gets criticised these days for removing carbon-absorption devices from the landscape, says Midgley.
Solofondranohatra has concerns that the same could happen in Madagascar. "Food security in Madagascar is highly precarious and agriculture in the Central Highlands is dependent on abundant stream flow for rice production," she and colleagues, including Madagascan grass taxonomist Maria Vorontsova, also of Kew Madagascar Conservation Centre, argued in a paper published in May.
"If grasslands are an extensive ancient component of these Central Highlands landscapes ... not only is planting of exotic tree species damaging, but at scale will probably reduce stream flow with unforeseen environmental consequences in a changing climate."
Carbon and fire
In addition to their water services, grassy biomes withstand fire and drought.
Fire may race through them, but it is part of their natural cycle that helps, for example, to disperse seeds. And these grassland fires are cooler than forest fires, whose intense temperatures can sterilise the soil and delay recovery for years.
They are also often a more secure carbon store than trees in climates where drought and rising temperatures make fire ever more likely. While they often store less carbon than forests, it is mostly safe in the ground in their roots and surrounding soil.
A grassland fire will release a year's worth of carbon accumulation in the leaves and stems, but it will not release what is in the roots and soil. In contrast, trees store their carbon above ground and this is easily lost back to the atmosphere when fire sweeps through - or, in the case of plantations, when the trees are harvested.
"If you go and afforest landscapes in the middle of flammable grassland are you seriously expecting that to be a significant carbon store, reliable over multiple decades?" says Midgley. "It's naïve, especially if you plant things like eucalyptus, which are themselves flammable and can themselves then introduce a very destructive fire regime."
Midgley knows this from experience. In South Africa's Western Cape Province in 2017, the Knysna fires "tore through these planted and escaped forests of pines and eucalyptus, killed a bunch of people, destroyed a lot of property". Scientists later concluded that the severity of the fire was "significantly higher in plantations of invasive alien trees and in fynbos [a form of grassland] invaded by alien trees, than in uninvaded fynbos".
When scientists modelled the fire problem for planted forests in California, they concluded that, when these risks are taken into account, it is a more reliable option to conserve grasslands than plant them with forests, from a carbon curation point of view.
The services offered by grassy biomes are just some of the reasons why Madagascar should reconsider the extent of its tree planting campaign, say advocates.
A forest by any other name
But how did high-minded ideals about the restoration of ecosystems, universally regarded as vital and soon to be promoted under the United Nations' Decade on Ecosystem Restoration, transform into planting alien invasive trees?
It is to do with language -- genuine misunderstandings as well as deliberate obfuscations -- say critics.
"By exploiting broad definitions and confused terminology, policymakers and their advisers are misleading the public," Simon Lewis, a geographer at University College London, and Charlotte Wheeler, a forest researcher at Edinburgh University, wrote in a commentary, published in the journal Nature, that argued that tree restoration had partially morphed into the creation of monoculture plantations.
Here is how the language of restoration is slippery.
For a start, while most of us think we know what is meant by 'forest,' there are multiple major and minor classifications of forest -- and this can include land with a tree density of only ten per cent. Savannah, meanwhile, can have a tree density of up to 60 per cent, which is why it can end up classified as forest, which opens the door, says Veldman, to it being regarded as degraded forest.
Then there are differences in meaning between 'restoration' and 'rehabilitation', 'reforestation' and 'afforestation', 'planting' trees and 'growing' trees.
"Sometimes the 'restoration' term is not the same for a politician as it is for a scientist," says Solofo Rakotoarisoa, ecologist at Kew Madagascar Conservation Centre. "For a politician they say restoration is like tree plantation -- planting whatever they want to plant, like eucalyptus ... . In terms of science [that's] not really restoration but afforestation and reforestation: forest restoration is to restore the original forest in place by using indigenous species as much as possible."
Because 'landscape restoration' can easily become 'replanting forest', which can become 'planting trees', it can soon become any old trees, any old where - so long as what is achieved is the arboreal equivalent of bums on seats.
Trillions and trillions
When United States President Donald Trump announced in February he was supporting what he called the "One Trillion Tree Initiative" -- the World Economic Forum platform 1t.org -- he said it was an effort "to plant new trees in America and all around the world". But 1t.org actually aims to "connect reforestation champions" for conservation and restoration.
"Most of the time the funding agencies don't even know what they are talking about," says WRI Madagascar's Wilmé.
Even a judiciously placed 'and' can make a world of difference. There is 'forest and landscape restoration' -- and then there is 'forest landscape restoration'. This particular 'and' has been a topic of "great debate" because, without it, the phrase "would imply just tree-based restoration", says Katie Reytar, a spatial analysis specialist at the WRI in the US capital Washington DC and the manager of the controversial restoration atlas.
"We've pivoted towards using 'and' to be more inclusive of [other restoration activities] -- not just tree-planting ... like terracing of hillsides," she says.
But the African Forest Landscape Restoration Initiative does not include that 'and' in its title. And, when it comes to its acronym -- AFR100 -- the letters, chosen to be evocative of the continent, lost their 'landscape' reference altogether.
Open to interpretation
The interpretation of scientific data is another slippery issue. A set of results might mean one cautious and caveat-filled conclusion for experts who understand their limitations -- but it could mean something far greater for those of a campaigning mindset or with other agendas.
This was the case with the WRI atlas, say critics. Satellite images tell us something about tree cover, but it is a crude impression of what is actually happening on the ground, they say.
"I think they could do better, we have got the technology to do the fine-scale stuff at a global level," says Midgley.
But Reytar rejects the criticism. The atlas is no more than a map of restoration "opportunities", she says.
"The idea was to give, at the highest level, a rough estimate of what the total area was that could be restored. It wasn't saying 'plant forests everywhere'," she adds.
Critics have mistakenly assumed that "just because an area is an opportunity it means that something should be done there", Reytar argues.
"But the biggest misinterpretation," she adds, "is people assuming we mean 'forest' as in 'dense forest' -- as in either converting grasslands to forest or excluding people, excluding agriculture because we need more forest -- and that's not at all what was meant. The Forest and Landscape Restoration interventions that we talk about always incorporate the existing use of the landscape."
Enhancing the productivity of cropland by adding trees would be an example, she says.
"I guess if someone had a very, very cursory understanding of what restoration is they might see it that way [just planting trees] but if you explore the 'forest and landscape' definition and what all the different interventions entail it would be clear that that's not what it meant."
She is satisfied that the atlas has not set restoration on erroneous paths.
"It has led to countries making restoration a priority, but I have not seen any evidence of it leading to inappropriate tree-planting," she says.
Reytar says the atlas informed AFR100's target to restore 100 million hectares: "It did seem to be a reasonable ambition based on the ... map."
The case for restoration
Restoration is the ideal, says Mamadou Diakhite, who leads the AFR100 secretariat based in Johannesburg, South Africa.
Africa desperately needs to fight desertification and deforestation, and there are excellent projects working to satisfy these goals, he says.
But he also agrees that the language of restoration is delicate. According to the International Union for Conservation of Nature (IUCN), Diakhite points out, 'restoration' is "action that gives back the primary function of the land or forest -- ecological, economic and social".
AFR100 has examined the issue of restoration at length and posted its own guidelines on its website, which include the advice to "avoid strategies that lead to the conversion of natural ecosystems".
However, Diakhite points out that the financial costs of true restoration can be prohibitive. For example, just assessing a patch of land to understand what was originally there could cost tens of thousands of dollars. Instead, smaller steps that aim to "regenerate" or "rehabilitate" the land may be advisable at first.
"We try not to take the restrictive side of the definition," he says. "Restoration is ideal, but the perfect is the enemy of the good.
"If you take bare land that was not occupied or did not have any activities on it, that was dry, that was totally degraded and the government says 'yes, try to revive it and plant bamboo,' the land might not be restored like it was 200 years back, but it is regenerated and rehabilitated and even may be contributing to the local economy," he says, stressing that he is not endorsing any particular project.
"So, we try to be really inclusive, we debate. We tend to accept all the activities unless it is really, really damaging and hurting people and the environment," he adds.
AFR100 endorsed Madagascar's January planting day in a newsletter, saying that it was "in line with the Government's goal to make Madagascar a green country by reforesting 40,000 hectares a year".
Pat Wright, lemur expert and founder of Founder of Centre ValBio, a research centre in Madagascar, laughs when she hears the claims being made about the ancestry of desolate Madagascan grasslands -- that ancient megafauna once roamed pristine, but forest-free, savannahs stretching across as much as half of Madagascar. She is one of the conservationists and palaeontology experts who are demanding more evidence.
"If that area, that looks like savannah now, was a true biome -- meaning that it was something that had evolved in Madagascar -- [then] I would be seeing animals [there now] that had evolved to be on savannah," says Wright. "If you go to Nairobi and Tanzania and go to the savannah then there's lions and giraffes and antelopes. If you go to Madagascar ... there's just nothing.
"You might get a chameleon that's wandered in trying to get to the next forest."
Solofondranohatra and colleagues think that Madagascar might just have been ahead of mainland Africa in the extinction of its megafauna, but Wright's work involves searching for ancient lost forests on the island and recent findings she has made propel her further towards the forest theory.
"We're getting more and more evidence now, I'd say, that Madagascar was probably all forested. There are some native grasslands, but what covered most of Madagascar was not native grassland," she says.
Yet Solofondranohatra's research presents evidence that Central Highland grasslands have tell-tale signs of ancient adaptation to fire and to grazing -- indicating that both phenomena were around before there were humans to set fire to grass or livestock to graze them. She and her colleagues argue that this and other genetic evidence indicate that millions of years ago some sort of ancient megafauna -- giant tortoises or hippopotamuses -- grazed the Central Highlands.
Wilmé is also dissatisfied with the grassy biome theory, asking what the purpose of it is.
"I know what [the grasslands experts] are doing and I know how many species of plants [they have] discovered and how many they want to discover and the richness of grasslands in Madagascar etcetera. But for the time being we have no idea what the extent of the native grasslands was.
"When you do restoration, you have to decide what kind of restoration you want to do and to which stage. Is it to when humans arrived in Madagascar? What do you want to achieve?"
Wilmé believes AFR100 goals should be met through activities that allow destroyed forest to regrow - she advocates an expansion of the country's system of protected areas. But, for practical reasons, she is in favour of planting the grasslands with trees for fuel and other needs.
"The demand is large for that so why should they not? I mean that's their own responsibility as a country to produce whatever the population needs," Wilmé says.
Local communities are capable of deciding where to plant so as to avoid damaging their water supply, she adds.
Up for debate
This range of views was aired at a lively workshop hosted by Kew Madagascar Conservation Centre last year, says centre ecologist Rakotoarisoa. There was a big debate, he said, and by the end of it he took refuge in the view that "it's not wise to say it all used to be grassland and it's not wise to say it used to be forest".
Madagascar is rich in endangered biodiversity, subject to increasing drought and has a rising population demanding, at a minimum, food and fuel. It may be a microcosm of what is happening in mainland Africa, or it may be uniquely different. Across Africa, it remains unclear whether grassland biomes are being rampantly converted to forest or whether, for now, these are just plans.
Even AFR100's Diakhite does not know to what extent the group's pledges are materialising on the ground -- or what they look like. So far, the scheme has focussed on agreeing its terms of reference and recruiting countries to the cause. How to monitor progress is something that is to be sorted out this year, he says. He also points out that the secretariat has little control over the nature of restoration activities that individual countries agree with private firms, non-profits or other donors.
Midgley has his scare stories. He has heard of one company dropping eucalyptus seeds from drones across central Africa -- "a terrible idea," he says. Such projects are "all over the place", he says: "It's very poorly conceived and it's driven by a lot of money."
Nicola Stevens, an ecologist at Witwatersrand University in Johannesburg, South Africa, has similar worries, highlighting a project in which balls of tree seedlings are being dropped by helicopter on ancient Kenyan grasslands. They are native tree seedlings, she says, but they can be just as invasive if planted in the wrong biome.
Motion to protect
Mozambique had a last-minute reprieve in 2015, says Natasha Ribeiro, a forest specialist at Eduardo Mondlane University in Maputo, Mozambique. The Miombo Woodland -- a type of grassy biome that forms a broad belt across south-central Africa, running from Angola in the west to Tanzania in the east -- was due to be planted with commercial plantations. A plunge in the carbon credit market led to the abandonment of the plans.
But there are many precedents for the conversion of grassy biome to forest, says the ecologist Veldman, including forests grown in old-growth grasslands in India, China, United States, the Cerrado region of southern Brazil and Colombia.
"Some of the concern over Africa is what we've seen happen elsewhere ... If these commitments are in fact realised, it's quite clear what they are likely to become," Veldman says.
"If it's economically advantageous to clear old-growth forests ... [and] if you can then go to a savannah and plant trees, you could achieve [officially] no net deforestation, or reduce your rate of tree cover loss."
The fight to protect grasslands continues at an international level. Next year, the postponed IUCN World Conservation Congress 2020 will hear a motion to protect and restore endangered grassland and savannah ecosystems.
Where now for the trees?
But if grasslands are not to be planted, where can the trillion tree enthusiasts plant their trees?
As forest destruction continues around the globe, the focus should be on reducing that and allowing natural forest to grow back, almost all ecologists agree. Above all, new planting projects should put local communities at the heart of decision-making and those communities must perceive them to be just, because that is the only way that trees will stay in the ground, says the World Rainforest Movement's Jutta Kill.
In Madagascar, Solofondranohatra says it would be possible to restore the country's target four million hectares by focussing on using native trees in depleted forest areas in the east of the country. There are some ingenious projects, adds Wright, which with toil and innovation are strengthening forests and catering to the needs of local people.
Many projects do not need high finance or armies of planters, says Wilmé. After a forest fire, "passive restoration" in which you just "let nature do its job" will often suffice.
"It's unbelievable how nature is doing a good job, you just have to leave it alone because the forest can recover itself. That's normally something that leaders or powerful people, or the people from the [global] North don't want to listen to because they like to manage," Wilmé says.
One thing that might increase the incentive to restore depleted forests, is if this restoration could be measured more easily, so that regrowth could count towards targets.
Satellites are good at spotting that a tree has gone, but not that a sapling is growing back over a longer time frame, explains Reytar. The WRI has a new Global Restoration Observatory initiative to try and fix that, she says.
"We're coming up with a globally consistent way of measuring tree cover gain, which is something that doesn't exist."
Diakhite says there are examples of real restoration progress in Ethiopia, Rwanda, Malawi and Niger, among others. "We see many, many African countries putting their own resources into restoring their land and water systems and forests to become not only productive from an economic point of view, but to restore ecosystems and for communities also," he says.
He is particularly enthusiastic about a partnership by Malawi, Cameroon, Kenya and Rwanda funded by Germany's International Climate Initiative (IKI). The €20 million project has spent the past two years identifying areas for restoration work. "It is not only about planting and grasslands, it is a really participatory approach to work with the local communities to do the type of restoration that is suitable for their biome, for their ecosystems and that can be also productive in terms of financial revenue for the smallholder farmers who will be involved," Diakhite says.
Reprieve and restoration
But perhaps the most profound solution, say some, would be to drop the fixation on trees as the foremost solution to the problem of excessive carbon in the atmosphere.
"Why should South Africa [for example] grow [trees] in ecosystems that are rich and diverse and which yield water, in order to take up CO2 that is being emitted in much richer countries? What kind of equation is that?" asks Midgley. Lewis and Wheeler last year calculated that a commercial tree plantation will, over the course of a century, absorb 40 times less carbon than a regrown natural forest. And since one third of the known international tree planting pledges under the Bonn Challenge are commitments to plantations, it is much less powerful as a carbon drawdown tool than the headlines suggest.
So, a reprieve for old growth grasslands -- and the people who inhabit them -- might, therefore, come with a shift to 'real' restoration and the acceptance that tree planting is not going to fulfil the most ambitious hopes for carbon sequestration.
For now, Solofondranohatra draws solace from another source -- the poor quality of the planting on that day back in January.
"It was not well-organised," she says. "People were just dropping seeds on places that were not ready for it.
"I am okay with that ... I know that many of these seedlings won't survive up there on the grassland."