No easy fix

Limits to growth

JOSE TOLEDO PISA looks out over the Cremaq farm in remote north-eastern Brazil. Thirty-tonne trucks have finished spreading lime fertiliser to reduce the acidity of the soil. He is about to start planting soya beans first developed by the Brazilian agricultural-research institution, Embrapa, that are suited to the sweltering climate (soyabeans were originally a temperate plant and did not grow well in the tropics). The computer in the farmhouse is checking the temperature, the water and the level of organic material in the soil. Five years ago much of this farm was scrubland. This spring Mr Pisa will reap around 3 tonnes of soyabeans per hectare.

Land, water, fertiliser: three basic components of farming. At Cremaq, Mr Pisa has harnessed new supplies of them to grow abundant crops. But is that the rule or the exception?

Try making deserts bloom

If crop yields are to match the rise in population, then some of them will have to go up dramatically. The world's population is growing at just over 1% a year, so—allowing something extra to feed animals because of rising demand for meat—staple yields will have to rise by around 1.5% a year. This may not sound much, but it is a great deal more than current growth rates. CIMMYT reckons that, to keep prices stable, the growth in rice yields will have to increase by about half, from just under 1% a year to 1.5%; maize yields will have to rise by the same amount; and wheat yields will have to more than double, to 2.3% a year.

Since the 1960s the traditional way of growing more food—by ploughing more land—has been out of favour. That is partly for environmental reasons—much irreplaceable Amazon jungle has already been lost—and partly because many countries have used up all their available farmland. So though the population has soared, the supply of land has not.

However, the potential is not exhausted yet. The biggest agricultural success story of the past two decades has been Brazil, largely because it was able to increase its usable acreage by making its vast cerrado (savannah-like grassland) bloom. By reducing the acidity of the soil (as at Cremaq), Brazil has turned the cerrado into one of the world's great soyabean baskets.

A new study by the World Bank says the world has half a billion hectares of land with fewer than 25 people per square kilometre living on them (this excludes land on which farming would be impossible, such as deserts, forests and rainforests or the Antarctic). The area currently under cultivation is 1.5 billion hectares, so if all that extra land could be used it would represent an increase of one-third. In fact a lot of it either should be left alone for environmental reasons or would be too expensive to farm. But that would still leave plenty that could be useful for farming.

Most of it is concentrated in a few countries in Latin America, including Brazil and Argentina, and in Africa in the so-called “Guinea belt”, a vast loop of land that stretches round the continent from west Africa to Mozambique. In 11 countries less than half the usable land is farmed. These countries could presumably boost food output by taking in some new land.

But estimates of land availability are contentious. Some put available virgin land at only 10-12% of the current total, not over 30%. The difference depends on cost and politics, not just the physical characteristics of the soil. The cerrado itself was once deemed useless for farming.

And some of this extra land is offset by soil erosion. Africa has some of the most exhausted soils in the world, with less than 1% of organic matter in them, half the level required for good fertility. For centuries African farmers allowed for this by letting the land lie fallow for eight or nine years after a harvest. But with more people to feed they have to squeeze in more harvests, and the soil is no longer recovering.

The chemistry of the soil—the presence in it of phosphorus, nitrogen and so on—is being degraded. That at least can be corrected by fertilisers. But the biology of the soil is also being damaged by the loss of organic matter, which can take five to ten years to recover. Worst of all, the physical structure changes if the top soil erodes, making it harder for the land to retain water or fertiliser. Top soil can take hundreds of years to replace.

And the more land is turned over to agriculture, the greater the loss of biodiversity. Three-quarters of all the world's plant genetic material may have gone already, mostly by habitat destruction, says Pasquale Steduto of the FAO, and more is going every day. This is a worry because some of the most desirable characteristics of plants are in the wild gene pool and might be needed again one day.

According to the World Bank, “land grabs” (deals in which capital-rich food importers buy up supposedly spare land in poor countries, farm it and ship the produce back home) have had much more impact than expected. Only three years after the first deals, says the bank, they already run to 65m hectares—an eighth of the bank's own estimate of total available land (and a third of the more modest estimates). So a lot of virgin land is already coming under the plough.

On balance, concludes the FAO's Parviz Koohafkan, land is not a decisive problem for world agriculture. But nor, except in a few countries, will it allow big increases in production.

Drink sparingly

Water, on the other hand, is crucial. At the moment it is probably agriculture's critical limiting factor.

According to Nestlé's Peter Brabeck, roughly 4,200 cubic kilometres of water could be used each year without depleting overall supplies. Consumption is higher, at about 4,500 cubic kilometres a year, of which agriculture takes about 70%. As a result, water tables are plummeting. The one in Punjab has fallen from a couple of metres below the surface to, in parts, hundreds of metres down. The rivers that water some of the world's breadbaskets, such as the Colorado, Murray-Darling and Indus, no longer reach the sea.

Scarce and precious

By 2030, on most estimates, farmers will need 45% more water. They won't get it. Cities are the second-largest users of water, and those in the emerging world are growing exponentially. They already account for half the world's population, a share that will rise to 70% by 2050. In any dispute between cities and farmers, governments are likely to side with cities. Agriculture's share of the world's water used to be 90%, so it has already fallen a long way. It will surely decline further.

The reason water matters so much is that irrigated farming is so productive. It occupies only one-fifth of the world's farmland but contributes two-fifths of the world's food output. Rice, the world's most important crop in terms of calories, is mostly irrigated, and is especially sensitive to shortage of water, stopping growth at the first sign of getting dry.

Water problems will worsen both because irrigated areas will suffer disproportionately from the effects of climate change and because diets are shifting towards meat, which is “thirsty”. Arjen Hoekstra, of the University of Twente, says it takes 1,150-2,000 litres of water to produce 1kg of wheat, but about 16,000 litres of water for 1kg of beef. As more people eat more meat, rising demand by farmers will collide with contracting water supplies.

There are things farmers can do. Roughly a third of the water used in fields with ordinary gravity-fed irrigation is reckoned to be wasted (more accurately, it recharges the aquifers without being taken up by plants, which is not quite the same thing). Switching to drip-feed irrigation means that watering becomes more precise, cutting consumption per unit of output. Jain Irrigation, the largest drip-feed company in India, has shown the technology can work for smallholders, cutting their water usage by about 40%. Drip-feed irrigation also boosts overall yields because the plants are watered at the right time and get the right amounts.

Overall efficiency gains in the use of water could be large. Israel wastes only about a tenth of its water, and if everyone were equally efficient, the world's water problem would be much less pressing. Israel makes widespread use of low-volume irrigation such as drip-feed and micro-sprinklers, which is expensive. The FAO reckons that over the next 40 years irrigation will require cumulative investment of almost $1 trillion. That may be forthcoming eventually, but it won't be soon.

No-till agriculture, an agronomic practice in which farmers do not plough up the land but leave part of the previous year's crop on it, also preserves water. The residue acts like a blanket, lowering the soil temperature by a degree or so in the tropics (and thus helping to combat the effects of global warming). It also prevents water run-off and reduces evaporation by 30-40%, reckons Patrick Wall of CIMMYT. As a bonus, adds Shivaji Pandey of the FAO, no-till and low-till farming sequester about 200kg of carbon per hectare per year. In parts of India, the time saved by not ploughing after harvest also makes it possible to grow an extra crop.

So why hasn't this miracle cure been adopted universally? Because of weeds. They like to grow in the mat as much as crops do. It helps to have plants that are genetically engineered to resist weedkillers, but Europe has banned those. This has meant that no-till was used on only 6% of farmland in developing countries and hardly at all in Europe in 2008.

Agriculture's third basic input is nitrogen. Historically, lack of nitrogen, not lack of land or water, has been its biggest constraint. The invention of a process to synthesise nitrogen cheaply into ammonium, a fertiliser, paved the way for the huge increase in food production in the 20th century. Vaclav Smil of the University of Manitoba in Winnipeg argues that this process, rather than the transistor or computer, was the century's most important invention, and that 2.5 billion people would not be alive without it.

African farmers use an average of 10kg of fertiliser per hectare. Indians use 180kg. India is richer than Africa, but not hugely so. IFAD's Mr Nwanze thinks Africans could double yields by doubling their fertiliser use.

Don't overdose

But there are limits, as China's example shows. Since 1990 Chinese grain production has been roughly stable but the use of fertiliser—which is heavily subsidised— has risen by about 40%. China could cut fertiliser use by at least a third without ill effects. In fact, it would be a blessing. At the moment excess gunk runs off into rivers, gathers in lakes and produces toxic blooms of algae. Likewise, the “dead zone” of the northern Gulf of Mexico is caused largely by overuse of fertiliser in the American Midwest that is making its way down the Mississippi.

So increased fertiliser use would boost yields in some countries and be counterproductive in others. But globally there is little prospect of a big rise because of the expense. Fertiliser prices spiked even more dramatically than food prices in 2007-08. Phosphorus prices soared and have stayed high, reflecting fears that the stuff may be running out. Making fertiliser is energy-intensive, so unless oil prices fall, increasing food production by slathering ever more fertiliser on the land would be inefficient.

Similar considerations apply to dealing with pests and diseases. At the best of times, farmers face the curse of the Red Queen in “Alice Through the Looking-Glass”(“A slow sort of country! Now, here, you see, it takes all the running you can do, to keep in the same place.”). Predators wage a constant war on plants, and if farmers do nothing the output of a new seed will decline by a percentage point or so every year. This is why new seeds are needed all the time.

There are signs that the burden of disease may be increasing. Rothamsted Research, home to the Broadbalk experiment, has been tracking aphid infestations for 50 years. In 2000 no aphids had a particular resistance mechanism called mace. Now 70-80% do. The aphid that causes potato blight now appears a month earlier than it used to, so it feeds on the plant at a more vulnerable point in its life.

The greater incidence of disease may be caused by many things: more insects surviving winter; the banning of dangerous pesticides; cuts in the budgets of institutions that conducted research into diseases; even globalisation. The corn-borer moth, native to Central and North America, first appeared in Europe in 1999 in Kosovo, presumably on the boots of American peacekeepers. It has since spread in concentric circles each year and is now eating into maize crops in Germany and Italy.

These problems are onerous, but most are probably tractable. Climate change is not. Global warming upsets the world's water cycle, increases the burden of pests, desiccates soil and reduces yields. In 2010 the world got an unpleasant taste of what climate change might bring. During the summer the jet stream (air currents at 7,000-12,000m above sea level which affect the winds and weather) changed its course. That seems to have been linked to the catastrophic floods in Pakistan and huge forest fires in Russia which help explain the big food-price rises in the second half of last year.

Agriculture is itself a big contributor to climate change. According to the Intergovernmental Panel on Climate Change, farming directly accounts for 13.5% of greenhouse-gas emissions, and land-use changes (often cutting down jungle for fields) are responsible for a further 17.4%. That adds up to almost one-third.

The day of the locust

Agriculture is responsible for between half and two-thirds of emissions of two especially toxic greenhouse gases, methane and nitrous oxide. These stay in the atmosphere for years, absorb a lot of radiation and, weight for weight, have many times the impact of carbon dioxide. So even if nothing else were happening, farmers would be under pressure to cut emissions.

But a lot else is happening. An increase of 2°C in global temperatures, says Hans-Joachim Braun, the head of CIMMYT's wheat programme, could cause a 20% fall in wheat yields. This would exceed any possible gains from warming in areas currently too cold to grow crops and would also offset the benefits of rising carbon-dioxide concentrations. Plants eat CO₂, so if there is more of it in the atmosphere, photosynthesis should increase and yields rise. But no one knows by how much.

Climate change also affects the rhythm of the seasons. Winters arrive later or spring earlier. Rainy seasons become shorter, milder or more intense. All living things depend on the heartbeat of seasonal change. In spring, caterpillars time their emergence to coincide with the bud burst of trees; birds start nesting when they can feed those newly emerged caterpillars to their fledglings. Any disruption to the seasonal rhythm tugs at the web of life. For example, in parts of Mozambique where villagers cultivate maize on the flood plain of the Zambezi river, the rainy season now begins later, so the crop is sown later, shortening its growing period.

Out of synch

In 2009 Oxfam, a British charity, asked thousands of farmers in a dozen countries what worried them most about climate change. Their biggest concern was not higher temperatures but disruptions to the natural cycle. “I know I am supposed to sow by a certain time or date,” said Mohammed Iliasuddin, a farmer in Bangladesh. “That is what my forefathers have been doing. But then for several years the temperature and weather just does not seem right for what we have been doing traditionally. I do not know how to cope with the problems.”

When the International Food Policy Research Institute (IFPRI) tried to work out the impacts of climate change on the main cereal crops, almost all its results suggested that yields in 2050 are likely to be lower than they were in 2000, sometimes much lower. Almost half the forecasts showed yield reductions of 9-18% by 2050. One came up with a drop in rainfed-maize yields of 30%. The most vulnerable crop turned out to be wheat, with the largest losses forecast in developing countries. The Indo-Gangetic plain, home to a seventh of mankind and purveyor of a fifth of the world's wheat, is likely to be especially hard hit.