Linking local ecological knowledge to plant functional traits in coffee and cocoa uri icon


  • Local knowledge and ranking of trees per function During August and October 2013, semi-structured interviews and a ranking exercise were carried out with 50 coffee AFS farmers in Turrialba and 50 cocoa AFS farmers in Talamanca, Costa Rica. We selected (i) three ecological functions of the shade component based on prior work on coffee AFS in Turrialba (Cerdan 2012), that were: (1) Maintenance of soil fertility, (2) Maintenance of fresh microclimate in the plantation and (3) Control of soil erosion, and (ii) one 'utilitarian' function, defined by the farmers, that was the 'suitability' of the shade component for the plantation (best shade). A semi structured interview was carried out to collect information on criteria that farmers used to assess shade species in terms of their performance for the 4 selected functions. After the interview, each farmer was presented photos of 18 (Coffee AFS) to 20 (Cocoa AFS) commonest shade species found in each study site with clear print of the local name and they were asked to divide the cards between those species they knew well and those species they did not know. From the species they knew well, 10 species were selected at random for the ranking exercise. In the few cases that less than 10 species were available, the ranking exercise included only with the species farmers knew. Then farmers were asked to rank the selected species for each selected function. Comments of the farmers explaining the ranking were noted. Plant functional traits and measurement: Three functional traits related to production, accumulation of standing biomass and litter decomposition were selected: plant height (PH, m), Specific leaf area (SLA, m2kg-1) and physical strength of leaves, measured as the force to tear (PSL-FT, N mm-1). Measurement for all traits was carried out according to (Pérez-Harguindeguy et al. 2013). Ten individuals of each species in each agroforestry system were sampled. We visited farms where the selected species were more abundant to allow selecting suitable individuals for sampling. For each individual, PH of plants less than 5 m was estimated with a stick of known longitude. For plants taller than 5 m, we used a clinometer. After measuring height, two leaf samples per individual were taken for measurement of PSL-FT. To avoid dehydration, PSL-FT was measured directly in the field immediately after collection of the leaf samples with a tearing apparatus (Pérez-Harguindeguy et al. 2013). For SLA measurements, 5 mature leaves with minimal damage from pest, diseases or parasites were selected from each individual, including those selected for the tearing test. The projected area of the leaves included petioles and was estimated from photographs, taken in the field, using Image J software and protocols ( After taking the photographs, leaves were stored in paper bags and left in a ventilated place, upon more than two weeks in Talamanca, until being transported to the lab where they were air dried at 70°C for 72 hours and weighed. The total area of the five leaves was divided by their total weight to obtain SLA

publication date

  • 2015