Gazing into the future of agroforestry: which models work?

Climate change effects are harder to model in multi-functional landscapes. Photo by Neil Palmer (CIAT), of ploughed fields in Kibirichia, Mount Kenya region. Via Flikr -

Climate change effects are harder to model in multi-functional landscapes. Photo by Neil Palmer (CIAT), of ploughed fields in Kibirichia, Mount Kenya region. Via Flikr –

What if we could peer into the future and see clearly how climate change will affect agriculture in the coming decades and centuries? We would then be able to confidently design, now, the farming systems that would work in that future.  With climate change modeling it is possible to forecast future scenarios with variable accuracy.

Modeling has been used extensively to plan ahead for single-crop systems, but predicting how agroforestry systems will behave in the environments of the future presents special challenges.

World Agroforestry Centre (ICRAF) scientist Eike Luedeling says in complex, multifunctional agroforestry systems, which typically include various components—trees, crops, livestock, etc—in myriad configurations, climate modeling is complicated because “these different components interact with one other and with the climate.”

“Furthermore, beyond altering individual components in an agroforestry system, climatic change will probably also alter these interactions between the components, and affect additional factors such as pest and disease pressures.”

In a new review in the journal Current Opinion in Environmental Sustainability, Luedeling and co-authors summarize three classic modeling approaches and discuss each one’s prospects for predicting the impacts of climate change in agroforestry systems.

According to the article,

  • Process-based modeling is feasible where all major processes (such as water and nutrient uptake  and competition for light) of a particular system are reasonably well understood. It will take “major additions to existing frameworks”  to model agroforestry systems with process-based models, they add. And “it seems unlikely that the model for agroforestry can become as robust as those for single-crops.”
  • Species distribution modeling is currently the most sophisticated approach to project climate change effects. It is based on a statistical method to determine the environmental niche of a species, system, biome or genotype. For agroforestry, however, the method has several pitfalls related to samples and data, as well as species variations and adaptation over time. Given these limitations, “we expect that species distribution modeling can provide a conservative projection of the potential distribution of the ‘climatic niche’ of a particular agroforestry system. This projection could either be considered an ‘optimistic’ or ‘pessimistic’ view of the future distribution of systems,” says the article.
  • Climate analogue analysis is used when there is not enough data about a system to allow the use of the first two methods. By looking at successful farming systems at an ‘analogue location’ of a site (a place currently experiencing the predicted future climate of the site), it might be possible to predict what farming at the site will look like in the future, and plan for it. In reality, however, it is difficult to find two locations with the same non-climatic characteristics (such as soils, farm size, markets, and culture) to allow this comparison; this is one of the main setbacks for climate analogue analyses. Furthermore, the cost of monitoring at analogue sites can be high. And analogue analysis only provides information about particular sites, limiting extrapolation on a large scale.

Luedeling and colleagues recommend the development of process-based models for agroforestry, supported by experimentation. This type of modeling, they write, is “the only approach that can comprehensively capture the effects of both CO2 and changing climates.”

The authors call for more efforts in agroforestry modeling—both in application and model development—for planning future farming and natural resources management.

Modeling can give us the power to ‘see’ into the future, and allow the design of farming systems that will stand the test of time. Such systems will be able to meet the growing demand for food and products while protecting the environment against further damage.

Read full article:

Eike Luedeling, Roeland Kindt, Neil I. Huth, Konstantin Koenig, Agroforestry systems in a changing climate—challenges in projecting future performance, Current Opinion in Environmental Sustainability, Volume 6, February 2014, Pages 1-7, ISSN 1877-3435, (

This article appears in a special issue of the journal Current Opinion in Environmental Sustainability on the theme ‘Sustainability challenges.’  The full special issue is available Open Access at

The issue is edited by Cheikh Mbow, Henry Neufeldt, Peter Akong Minang, Eike Luedeling and Godwin Kowero

It will be launched during the World Congress on Agroforestry, Delhi, February 2014.

At the Congress 2014, Luedeling and co-researchers will discuss “ensemble suitability mapping” methods, which have been integrated into BiodiversityR



Related stories

Where will trees grow in a new climate?

امير فراهاني'

Daisy Ouya

Daisy Ouya is a science writer and communications specialist with the World Agroforestry Centre (ICRAF). Over the past 15 years she has been packaging and disseminating scientific knowledge in the fields of entomology, agriculture, health, HIV/AIDS research, and marine science. Daisy is a Board-certified Editor in the Life Sciences ( and has a Masters’ degree in chemistry from the University of Connecticut, USA. Her BSc is from the University of Nairobi in her native Kenya. She has worked as a journal editor, science writer, publisher, and communications strategist with various organizations. She joined ICRAF in July 2012. Twitter: @daisyouya

You may also like...