Climate and Coconuts

Millions of people in the tropics depend on coconuts for food, raw materials and livelihood. Coconuts are also a high value commercial crop. But like any crop, coconuts are at risk of drought and other prolonged events. By using climate science and better agricultural forecast models, the IRI has helped increase the resilience of coconut plantations to climate variability in one of the world’s major producers, Sri Lanka.

“The uncertainty that characterizes agriculture – much of it due to climate variability- is among the greatest challenges facing farmers and others in the agriculture value chain in much of the world. Any information that reduces the uncertainty about future production or prices has immense value to them,” says IRI research scientist James Hansen.

IRI’s partners in this work have been the Coconut Research Institute of Sri Lanka (CRI), the Foundation for Environment, Climate and Technology (FECT), and the Sri Lanka Department of Meteorology. Together, the institutions have developed an improved prediction scheme that generates annual coconut production forecasts fifteen months in advance. The CRI has been using these forecasts since 2005 to help calculate projected coconut yields for the upcoming year. In a country such as Sri Lanka, more accurate projections are critical.

“Coconut cultivation sustains the livelihood of large numbers in the tropics and is the most important crop for food security after rice in Sri Lanka,” says IRI scientist Lareef Zubair, who has been working with the CRI and other Sri Lankan colleagues since 2003.

“When we started out on this project, we were not sure whether we could provide an adequate prediction using climate information, as climate is one of many factors that affect production,” Zubair says. “In addition, we had to assume that the climate and coconut data were of adequate quality.”

Sri Lankans love their coconuts. The country has the highest annual per capita consumption of coconuts in the world- about 110 nuts. Sri Lankans dedicate about 400,000 hectares, more than a fifth of the agricultural land in the country, to grow the crop. These efforts produce on average 2.5 billion nuts every year, which account for 2% of Sri Lanka’s gross domestic product, 2.5% of its export earnings and 5% of its workforce.

“One concern is about how coconut plantations can cope with climatic variability and adapt to climate change,” says Neil Fernando, an agricultural economist at the CRI, with whom Zubair and Hansen collaborated. The work was funded through the Assessments of Impacts and Adaptations to Climate Change (AIACC) project from the Global Change System for Analysis, Research and Training (START).

Fernando and Zubair coauthored a paper in 2007 which calculated the economic impacts of climate variability on the Sri Lankan economy. In years of extreme crop shortage, income losses to the economy could be $32 million to $73 million. However, in years of extreme crop surplus, the economy could realize income gains between $42 million and $87 million.

Much like any agricultural commodity, coconuts are bought and sold months ahead of time via futures, or forward contracts. Accurate forecasting is critical for national agricultural planning and for negotiating forward-contract pricing with foreign buyers.

Coconut is a perennial crop with a long development period of about 18 months. The long maturation period makes it vulnerable to severe or prolonged weather events, especially during Sri Lanka’s two dry seasons (January to March, June to August).

“Extended dry spells can reduce yield,” says Sanathanie Ranasinghe, the head of plant physiology at the CRI. “However, excessive cloudiness during the wet season can reduce photosynthesis, which can diminish the ‘dry matter’ production needed for nut growth”.

“Coconut is quite different from short-season cereal and pulse crops that we’ve worked with,” says Hansen. “We initially thought our ability to forecast rainfall would be the most important factor in forecasting coconut yields months before harvest. Instead, we learned that the history of weather as far as two years before harvest provides quite a bit of information about future yields.”

Simply by monitoring rainfall during a dry season, growers are able to predict future yields up to 15 months in advance with good skill. If there’s too little rainfall, then they can make management decisions, such as irrigating or increasing fertilizer use, to try to mitigate the impacts on yield.

“Seasonal climate forecasts allow us to extend the lead time of yield predictions from 15 months to 21 months,” says Sarath Peiris, the head of the biometry division of the CRI. Growers don’t have to wait until the dry season is upon them in order to adjust their yield predictions, he says.

Peiris and Fernando came to the IRI for three weeks of training in 2004 and have been in regular consultation since. Zubair has been to Sri Lanka to help scientists in the country’s agricultural research institutions, meteorological services and other institutions on using the IRI’s Data Library, and the use of monitored and forecast information.

The predictions are distributed to national policy makers and to the agricultural extension officers in the 12 coconut growing sub-regions, and through them, to 270 major coconut plantations in the country. The predictions have helped the government and growers make informed decisions during droughts. Users have started requesting predictions by season and by region, and the CRI scientists are now working towards this.

“By investing modest resources, we helped them incorporate state-of-the art knowledge on climate variability into their prediction system,” says Zubair, “and it turned out to have outcomes that were very useful for the end users.”

Numbers and figures for this story taken from:
Use of seasonal climate information to predict coconut production in Sri Lanka (abstract)
Economic Value of Climate Variability Impacts on Coconut Production in Sri Lanka (pdf)
Predictability of Sri Lankan Rainfall based on ENSO (abstract)