Everyday millions of people around the world enjoy chocolate. Whether as part of a snack, drink or dessert, chocolate is among the best-known and best-loved foods. However, while the global demand for cocoa beans – the raw ingredient for chocolate - is increasing, cocoa productivity is decreasing due to a range of challenges such as degraded soils, pests and diseases, and climate change. To meet growing needs in the past, the area devoted to cocoa cultivation expanded, leading to extensive deforestation, thereby reducing carbon stocks and biodiversity. Many people believe that including shade trees in cocoa production systems – agroforestry – might simultaneously solve all of these problems. But despite wide spread optimism, it is not known how many shade trees should be included in an agroforest to meet production, climate, sustainability, and conservation goals. Many smallholders do not really have the luxury to get the level of shade wrong because cocoa remains their main source of cash income.

We tested how much shade is optimal for both production and other ecosystem services in cocoa agroforests. Previous studies have shown that farms with more shade trees can supply higher levels of non-production ecosystem services, but optimal agroforests need to balance agricultural yield versus these other ecosystem services. While there have been some really nice studies on the trade-off between agricultural yield and biodiversity, or agricultural yield and carbon sequestration in agroforests, there has been a lack of definitive answers on the appropriate level of shade that can best meet multiple goals at the same time. Expectations are high, but can agroforests truly meet those expectations? This is why we decided that measuring as many of the effects of agroforests as possible in a single study, and doing so along a gradient of shade-tree cover, was necessary to determine just how effective agroforests can be. Our motivation was to work out exactly what agroforestry can and cannot do, and to provide clear management recommendations for optimum levels of shade-tree cover that can simultaneously capitalize on the potential benefits of shade trees for the provisioning of multiple ecosystem services, while continuing to produce cocoa for smallholder incomes and to meet rising demand.

The fieldwork was challenging and immensely rewarding. Our small team endured car accidents, drone crashes, covered many miles on foot and on bicycles (imagine our Strava stats) and had the opportunity to spend a lot of time with local farmers. However, after lots of late night Kelewele (spicy fried plantain), lots of sweat and laughs we ultimately found a relative simple result: Low-to-intermediate shade agroforests are ideal for optimizing the trade-off between agricultural yield and the provisioning of other ecosystem services and have major advantages over monocultures. As shade-tree cover increases above approximately 30%, agroforests become increasingly less likely to generate win-win scenarios. This demonstrates that careful management of the level of shade could yield more favorable outcomes for smallholders than the current ad hoc implementation of agroforestry or the trend towards full sun monocultures.

To make further improvements in sustainability will likely require additional effort beyond simply managing for optimal shade levels. For example, judicious use of organic and inorganic fertilizers and pest control techniques, and increased attention to weeding and pruning might provide important additional benefits for increasing cocoa productivity, while retaining the benefits of shade-trees. Combined with the diversification of crops that are more heat and drought adapted, these methods of sustainable intensification, could help farmers produce more cocoa on less land, and also adapt to climate change by being less dependent on cocoa alone.

Finally, it is important to recognize that our work does not highlight a whole different set of trade-offs between agricultural production and biodiversity, and agricultural production and climate. For example, while our work shows that agroforests have clear benefits for climate mitigation and biodiversity conservation relative to monocultures, agroforests do not come close to the carbon storage capacity and levels of biodiversity in undisturbed forests. Even the best designed agroforests may be unable to maintainall of the ecosystem services provided by the natural ecosystems that they replace. Nevertheless, the benefits of agroforests are substantial and they represent a viable solution for meeting production and sustainability goals in landscapes with little remaining primary habitat.

Our paper titled "Climate-smart sustainable agriculture in low-to-intermediate shade agroforests" can be found here: http://dx.doi.org/10.1038/s41893-018-0062-8