Speakers
Description
Introduction
Germany aims to achieve carbon neutrality by 2045, with focus needed on the agriculture sector where agroforestry systems (AFS), such as alley-cropping, can serve as effective carbon sinks by cultivating perennial woody plants alongside annual crops or grasslands. According to (Beillouin et al., 2023), AFS are the most effective agricultural measure for increasing organic soil carbon content in proportion to area.Furthermore, the establishment, maintenance, and harvesting of the tree strips incurs additional expenditures and greenhouse gases (GHGs) emissions. AFS feature reduced wind speeds and altered evapotranspiration dynamics compared to normal arable farming systems (Markwitz et al., 2020). The Hi-sAFe model is a 3D model considering competition and facilitation, which are significant mechanisms explaining positive biodiversity-productivity relationships in biodiversity ecosystem functioning research, even though it does not explicitly incorporate biodiversity as a driver for outcomes like crop yield. Machine learning approaches are increasingly being used as data-driven tools to extract patterns and insights from the ever-increasing stream of geospatial data (Reichstein et al., 2019), but they have received less attention thus far in AFS.
Materials and Methods
Following extensive review, we found that the existing DSS lacks decision-making aids for cultivation recommendations and assessing AFS’s actual climate change mitigation potential, gaining different limitations (Figure 1). The Hi-sAFe agroforestry model (Dupraz et al., 2019) outperformed other models offering a unique 3D and spatially explicit framework to analyze tree-crop competition for light, water, and nitrogen. Incorporating factors such as climate, soil characteristics, species interactions, and management practices, Hi-sAFe provides a comprehensive platform for understanding the complexities of agroforestry systems. Microclimatic impacts, such as radiation (shading), temperature, humidity, and wind, depending on the distance to the wood strip and strip orientation, as well as the width, height, and density of the wood strip, are simulated in daily time steps for different woody features.
Results and Discussion
Multi-year (2009-2016) crop yield of oilseed rape and winter wheat in the narrow and wide crop alleys at different spaces from trees (0, 1, 4 and 7 m) were used in parameterizing Hi-sAFe. Since the model couples the pre-existing STICS crop model (Brisson et al., 2003) with a new tree model, we first calibrated STICS for diverse crops under conventional arable farming conditions. To ensure, deploying the model at spatial explicit in different environment, we integrated the Hi-sAFe model with machine learning algorithms such as Artificial Neural Network (ANN) and Convolutional Neural Network (CNN) as well as partial life cycle assessment for simulating yield and GHG in AFS at a robust and spatially explicit scale, allowing for a global assessment of GHG reduction potential.
Conclusion
The current approach combines Hi-sAFe with machine learning to fill the limitations of the past data driven tools, dataset, and models, creating a robust DSS for agroforestry systems. The EUS is designed to allow farmers and consultants across Germany to virtually create a wide range of agroforestry systems on their property, as well as find the best sites and methods for maximizing productivity and climate change mitigation.
Keywords
Agroforestry, alley cropping, poplars, crop rotation, life cycle assessment, winter wheat, oilseed rape
References
Beillouin, D., Corbeels, M., Demenois, J., Berre, D., Boyer, A., Fallot, A., Feder, F., and Cardinael, R. (2023). A global meta-analysis of soil organic carbon in the Anthropocene. Nature Communications 14, 3700.
Brisson, N., Gary, C., Justes, E., Roche, R., Mary, B., Ripoche, D., Zimmer, D., Sierra, J., Bertuzzi, P., Burger, P., Bussière, F., Cabidoche, Y. M., Cellier, P., Debaeke, P., Gaudillère, J. P., Hénault, C., Maraux, F., Seguin, B., and Sinoquet, H. (2003). An overview of the crop model stics. European Journal of Agronomy 18, 309-332.
Dupraz, C., Wolz, K. J., Lecomte, I., Talbot, G., Vincent, G., Mulia, R., Bussière, F., Ozier-Lafontaine, H., Andrianarisoa, S., Jackson, N., Lawson, G., Dones, N., Sinoquet, H., Lusiana, B., Harja, D., Domenicano, S., Reyes, F., Gosme, M., and Van Noordwijk, M. (2019). Hi-sAFe: A 3D Agroforestry Model for Integrating Dynamic Tree–Crop Interactions. Sustainability 11, 2293.
Markwitz, C., Knohl, A., and Siebicke, L. (2020). Evapotranspiration over agroforestry sites in Germany. Biogeosciences 17, 5183-5208.
Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., and Prabhat (2019). Deep learning and process understanding for data-driven Earth system science. Nature 566, 195-204.
| Keywords | Agroforestry; alley cropping; poplars; crop rotation; life cycle assessment; winter wheat; oilseed rape |
|---|
