Why a Tree-Soil-Crop Interaction Model?

Sustainable land use systems can be provided through agroforestry practices.  Agroforestry is an agricultural approach using the benefits from combining trees and crops and/or livestock.  Therefore, knowledge on selection of species combination and good management of trees and crops are needed to maximise the production and positive effects of trees and to minimise negative competitive effects on crops.

However, empirical assessment of tree crop combinations is labourious, cost expensive and time consuming.  One method for overcoming this lack of information is through development of a model which integrates soil-tree-crop interaction between the components of agroforestry.

WaNuLCAS Model

The WaNuLCAS model was developed to represent tree-soil-crop interactions in a wide range of agroforestry systems where trees and crops overlap in space and/or in time in simultaneous and sequential agroforestry.  The model is based on above and below ground architecture of tree and crop, elementary tree and crop physiology and soil science (daily water, N, P and SOM balance for four soil layers and four horizontal zones).

Figure 1. Components represented

The model was developed in the ‘Stella’ modelling platform and can be used to assess the performance in terms of profitability as well as sustainability of various agroforestry systems.

Figure 2. Two-dimensional representation of AF system

Example of Model Application

Transformations from degraded soils and landscapes to agroforestry mosaics can benefit from the potential complementary between the early stages of tree-based production systems and crop growth.  Decisions by farmers managing such transitions involve strategic (multi-year) decisions on the choice of tree species, the number of trees per ha and the spacing, while tactical (shorter term) decisions relate to the choice of intercrops, tree canopy pruning and/or tree root pruning.  Based on the current experience in Lampung (Indonesia), we use WaNuLCAS model to explore these choices.

Cassava (Manihot esculenta) was simulated as an intercrop with rubber (Hevea brasiliensis) at eight levels of tree spacing and three timber trees (Paraserianthes falcataria, Acacia mangium, and Swietenia macrophylla) at nine levels of tree spacing.

The nine levels of timber tree spacing and eight levels of rubber spacing were grouped into two comparisons: the first is the effect of widening tree row spacing on crop growth, and the second is the effect of widening spacing between and within tree row on crop growth and tree growth.

Table 1. Tree spacing scenarios tested in the model

The crop was simulated for 12 cropping seasons, once per year as long as the previous crop yield exceeded a threshold value and the trees were simulated for 10 years, being planted after harvesting the crop for two years.  Besides the intercrop systems, trees and crops were also simulated as monoculture systems.

Crop monoculture was simulated for 12 cropping season, once per year.  Tree monocultures are planted after harvesting crop for two cropping seasons and simulated for 10 years. They were supposedly kept free from weeds (other simulations involved imperata as weed).

Fertilizer of N and P were applied to each crop, at 100 kg N ha-1 and 60 kg P2O5 ha-1 respectively, as is common practice for cassava.  N was applied twice, half at planting time and half at a month after planting.  P was applied once at planting time.  Mean annual rainfall was 2641 mm.

Result 1:  Crop yield as an effect of widening tree row spacing

The four tree species tested have different growth rates and canopy development rates, resulting in different opportunities for intercropping at the default spacing, but also to differential response to widening the alleys in between tree rows. For example, cassava tuber yield intercropped in A. mangium drops to a very low value in year 3, 4 or 5 for a tree spacing of 4 x 2, 8 x 2, 10 x 2 or 12 x 2, respectively, and continuous intercropping is only feasible for 16 x 2.  With mahogany or rubber as tree species, however, prolonged intercropping is possible.

Figure 3. Cassava yields overtime

Result 2:  Crop yield as alternative of spacing design

The yield of cassava was significantly influenced by tree species grown in the systems on the wider tree spacing and the longer time available for planting crops, especially intercrops with fast-growing trees.  The highest yield was found in systems intercropped with H. brasiliensis and the lowest yield found in the systems intercropped with A. mangium. Even though the crop yield of the intercropped system with S. macrophylla was not as high as intercropped system with H. brasiliensis, they still offered intercropping opportunities until the trees were five or six years old.

Note: A. narrow spacing (timber trees: 4x2, 3x3, 4x4; non timber trees: 6x3, 5x3, 4x4); B. wide spacing
(timber trees: 8x4 and 8x8; non timber trees: 6x6 and 12x6).
Figure 4. Predicted cassava yields

Result 3: Tree performance in alternative spacing

In all tree systems studied, the crop species selected did not have a significant effect on tree growth.  After 10 years, there was a small difference in wood volume between intercropped systems and (weed-free) monoculture trees, with monoculture trees slightly larger than trees in intercropped systems at the same density, except in A. mangium.

Figure 5. Predicted wood production

Result 4: Trade-off between tree and crop yield

Increasing space between tree rows makes longer intercropping possible, but also reduces the expected yield from the trees.  An efficient way of considering the trade-off is to plot crop versus tree yield.

Most of the tree crop combinations are substantially above the straight trade-off curve, suggesting that there is indeed a benefit to be obtained by the combination when compared to separate monocultures.  However, the points for A. mangium suggest virtually no intercropping advantage.  For the slower growing trees (mahogany and rubber), maximum tree yield can be obtained at about 20 per cent of the potential long year crop yield.  After accounting for this intercept, a slight positive curvature remains when tree spacing is widened. P. falcataria has a low intercept (low crop yield opportunity when maximum wood volume is the target), but a clear intercropping advantage at lower tree population density.  This may therefore well be the most promising 'agroforestry' tree at intermediate densities.

Figure 6. Predicted trade between cassava and wood production

References

van Noordwijk M, Lusiana B and Khasanah N. 2004. WaNuLCAS 3.01: background on a model of Water, Nutrient and Light Capture in Agroforestry Systems. Bogor, Indonesia. World Agroforestry Centre - ICRAF, SEA Regional Office. 246 p
http://www.worldagroforestry.org/sea/publication?do=view_pub_detail&pub_no=BK0060-04

van Noordwijk M and Lusiana B. 1998/1999. WaNuLCAS, a model of water, nutrient and light capture in agroforestry systems. Agroforestry Systems. 43(1-3):P. 217-242
http://www.worldagroforestry.org/sea/publication?do=view_pub_detail&pub_no=JA0139-04

Walker A, Mutuo P, van Noordwijk M, Albrecht A and Cadisch G. 2007. Modelling of planted legume fallows in Western Kenya using WaNuLCAS. (I) Model calibration and validation. Agroforestry Systems. 70(2007):P. 197-209
http://www.worldagroforestry.org/sea/publication?do=view_pub_detail&pub_no=JA0267-07

Printed Flyer - English