Francisco Jiménez

Species and site characteristics that permit the association of fast-growing trees with crops: the case of Eucalyptus deglupta as coffee shade in Costa Rica

The benefits that tropical farmers derive from associating trees with crops may be reduced or outweighed by competition for light and soil resources, and the use of incompetitive trees in agroforestry associations is therefore a standard recommendation. However, under certain conditions tropical farmers use very fast-growing and presumably competitive trees in tree–crop associations. We studied the factors which allow the use of Eucalyptus deglupta as coffee shade in parts of Costa Rica, considering three possible explanations: (a) the availability of soil resources matches the requirements of the two species so that competition cannot arise; (b) complementary resource use reduces competition; and (c) coffee plants possess sufficiently competitive root systems to cope with the competition of fast-growing trees. We measured coffee and tree growth, soil nutrient availability, root distribution and dynamics of 4- to 5-year-old coffee and E. deglupta shade trees on a private coffee farm with recommended fertilization in a high-rainfall area in Costa Rica. There was no evidence of a negative effect of the trees on coffee growth, yields and mineral nutrition despite fast tree growth. However, a significant small-scale partitioning of the soil space between the root systems, with most of the coffee roots close to the coffee rows and most of the tree roots in the interrow spaces, was evidence for root interactions between the two species, resulting in complementary use of soil resources. Apparently, the root system of coffee was sufficiently competitive to restrict the rooting space of the trees. Trees and coffee also differed in the depth distribution of their roots, with the trees having the shallower root systems. The compatibility of coffee with fast-growing shade trees at this and similar sites can be explained with a combination of all three hypothesized mechanisms.

Fine-root dynamics of coffee in association with two shade trees in Costa Rica

Fine-root dynamics (diameter < 2.0 mm) were studied on-farm in associations of Coffea arabica with Eucalyptus deglupta or Terminalia ivorensis and in a pseudo-chronosequence of C. arabica-E. deglupta associations (two, three, four and five years old). Coffee plants were submitted to two fertilisation types. Cores were taken in the 0–40 cm soil profile two years after out-planting and subsequently in the following year in depth layers 0–10 and 10–20 cm, during and at the end of the rainy season, and during the dry season. Fine root density of coffee and timber shade trees was greater in the coffee fertilisation strip as compared to unfertilised areas close to the plants or in the inter-rows. Coffee fine roots were more evenly distributed in the topsoil (0–20 cm) whereas tree fine roots were mostly found in the first 10 cm. Although the two tree species had approximately the same fine root length density, lower coffee / tree fine root length density ratios in T. ivorensis suggest that this shade tree is potentially a stronger competitor with coffee than E. deglupta. Coffee and tree fine root length density for 0–10 cm measured during the rainy season increased progressively from two to five-year-aged associations and coffee fine root length density increased relatively more than E. deglupta fine root length density in the four and five-year-aged plantations suggesting that contrary to expectations, coffee fine roots were displacing tree fine roots.

Root interactions between young Eucalyptus deglupta trees and competitive grass species in contour strips

To reduce soil erosion on sloping fields grass strips are commonly planted on the contours. The integration of timber trees into such strips can be economically beneficial and at the same time increase the anti-erosion effect of the strips. Root competition between grasses and trees within such strips may affect tree development, but might also reduce root competition between the trees and adjacent field crops if the grass strips restrict the lateral development of the tree root systems. The present experiment, carried out on a fertile site (Aquandic Dystrudept) in a high-rainfall region in Costa Rica (2700 mm per year), was designed to evaluate whether the lateral extension of the root system of Eucalyptus deglupta could be restricted by strips of the grasses Panicum maximum, Brachiaria brizantha or Saccharum officinarum, planted either in single, double- or triple-rows on both sides of rows of the trees. E. deglupta growth was significantly reduced in the association with P. maximum, the most competitive grass species, and to a lesser extent, by the double- and triple-row B. brizantha strips. In these treatments, the total number of lateral tree roots, as evaluated in profile walls, was reduced by up to 40% compared to the no-grass control. Furthermore, the lateral restriction of the tree root systems led to an increased density of tree roots between the grass strips. S. officinarum did not reduce the total number of tree roots and only slightly restricted lateral tree root extension. Grass strips composed of multiple grass rows appeared to restrict lateral tree root development more than strips with single grass rows but this effect was not statistically significant. None of the grasses forced the tree root systems to develop at a greater depth; on the contrary, tree roots were more superficial in the profiles beyond as well as between the grass strips in comparison to the no-grass control. There was some evidence of an increased root barrier effect of the grass strips in areas with a compacted topsoil. The barrier effect tended to decrease with increasing age of the trees. These results suggest that strips of competitive grasses can reduce the lateral root development, especially of young trees, when planted in contour strips, and this may reduce and/or delay root competition with neighboring field crops. The observed trends are likely to change as trees grow older and longer-term studies are needed.