PS 24-6 - Preliminary modeling of light availability in diverse agroforestry systems using a spatially-explicit forest model

Thursday, August 11, 2016
ESA Exhibit Hall, Ft Lauderdale Convention Center
Erik Stanek1, Kevin J. Wolz2 and Sarah Taylor Lovell1, (1)Crop Sciences, University of Illinois at Urbana-Champaign, Urbana, IL, (2)Program in Ecology, Evolution & Conservation Biology, University of Illinois at Urbana-Champaign, Urbana, IL
Background/Question/Methods

Concerns over the long-term sustainability of modern agriculture in the United States have brought about increased interest in agroforestry systems that incorporate trees into annual production systems. The practice of alley cropping, where short-term crops are planted between tree rows, represents a template for agroforestry adoption but has been relatively understudied with species yielding fruits and nuts included within the tree rows. These diverse agroforestry systems (DAS) contain multiple species canopies and densities which introduce considerable variation in light availability, making design optimization difficult. The objective of this study aims to address this issue by determining how various species, densities, and canopy levels affect light availability in DAS. The spatially-explicit forest simulator, SORTIE-ND, was used to model light transmission across four layouts based off research at the University of Illinois which includes treatments combining the following species: Ribes nigrum, Corylus avellana, Malus domestica, and Castanea mollissima. We used the model to produce General Light Index (GLI) per 1m2 ­ at heights corresponding to each species canopy level across the individual layouts. We then modeled the GLI received by each individual tree or shrub at mid-canopy crown. Finally, we analyzed the output for trends in light availability across DAS layouts and species.

Results/Conclusions

The results show that GLI within mature DAS treatments varies based on plant spacing, opacity, and the number of canopy layers. Treatments containing tree rows with three canopy levels of distinctly separate heights produced the lowest mean GLI for each canopy level (highest light capture), while single species tree rows had the highest (lowest light capture). The addition of a mid-canopy tree, Malus domestica, and doubling within-row spacing achieved nearly 40% more light capture than single species tree rows. On a plant scale, the GLI per individual was positively correlated with canopy height for all species and layouts. GLI was greatest for Castanea mollissima, the tallest tree, and lowest for Ribes nigrum, the shrub layer. In high density layouts where overstory canopies were separated by 15ft or less, the shrub layer had a GLI less than 20%. Overall, the addition of canopy layers at distinctly separate heights, paired with decreased spacing between plants resulted in more light being captured by the system as a whole. Further research on light requirements per species and changes in system GLI through time is necessary to continue to fill knowledge gaps related to the optimal design of DAS for long term agricultural sustainability.