Aug 26 – 30, 2024
The Couvent des Jacobins
Europe/Paris timezone

Optimization of agricultural land use to support sustainable and healthy diet: a Bayesian approach coupled with soil-crop modelling

Aug 30, 2024, 11:50 AM
15m
Les Horizons (2nd floor) (The Couvent des Jacobins)

Les Horizons (2nd floor)

The Couvent des Jacobins

Rennes, France
Oral Synergies of technologies Future & optimization

Speaker

Tom Desmarez (Gembloux Agro-Bio Tech)

Description

In recent decades, a significant global shift in our food system was observed, marked by increased consumption of high-calorie diets, processed foods, and animal products. This trend led to a surge in obesity, cardiovascular diseases, and non-communicable diseases [1,2]. To sustain this food system and meet the needs of a growing population, agriculture underwent transformation with the adoption of high-yield crop varieties, chemical fertilizers, pesticides, and mechanization [3]. This transformation brought forth environmental challenges, notably by contributing to 30% of global greenhouse gas emissions [4] and 70% of freshwater use [5]. Conversion of natural habitats to farmland led to biodiversity loss [6], while overuse of fertilizers and pesticides resulted in eutrophication zones [7], and health issues [8]. Various proposals emerged to address these challenges [9], with a key focus on shifting towards sustainable and healthy diets [10]. Reference diets, such as the one proposed by the EAT-Lancet [11] commission or the TYFA project [12], propose universal guidelines for a healthy food supply which respect planetary boundaries.

Taking Wallonia as a case study, the aim of this work is to design cropping systems aimed at locally supporting sustainable and healthy diets, while activating agroecological levers.

Considering i) the estimated needs of the Walloon population based on reference diets [11,12] and ii) the various pedoclimatic contexts and agronomical potentials of Wallonia, the Bayesian optimization algorithm DREAM [13] was used to optimize the allocation of the different agricultural productions within the available land use. From the optimization, cropping systems were designed for each region, in co-construction with stakeholders (scientists, farmers, food chain actors, etc.). The mechanistic model STICS [14] was then used to simulate the agronomic and environmental performances (e.g., yield, carbon storage, nitrogen cycle) of some of these cropping systems, under reference pedological contexts, for current and future climatic conditions, to evaluate their sustainability and resilience.

On the one hand, preliminary results indicated that optimizing land use to meet the population's needs should be feasible under current climatic scenarios, with still some spared arable lands. Since the optimization of land use was based upon actual yield, it seemed that the available spared land could mitigate the yield decrease that usually accompanies the implementation of more environmentally sustainable but less productive techniques (e.g., reduction of chemical pesticides or mineral fertilizers).

On the other hand, the initial predictions from the STICS model indicated significant yield increases of 10 to 20% by 2050 (for RCP 4.5 and 8.5 respectively) for potato cultivation, and by 30% for wheat, while no changes were observed for legume crops, and conversely, rapeseed cultivation sees its yields decreased by 20 to 30% compared to the current scenario.

Regarding cereal crops and potatoes, the increases in yields due to CO2 fertilization is likely to compensate for yield losses due to activation of agroecological levers. However, the predicted yield losses for some other major crops indicate that a redesign of rotations and the implementation of adapted management practices will be crucial to adapt to future scenarios.

References
1. Anand, S. S. et al. J. Am. Coll. Cardiol. 66, 1590–1614 (2015).
2. Popkin, B. M. et al. Nutr. Rev. 70, 3–21 (2012).
3. Matson, P. A. et al. Science 277, 504–509 (1997).
4. Whitmee, S. et al. The Lancet 386, 1973–2028 (2015).
5. Molden, D. Earthscan, London, 48 p. (2007).
6. Newbold, T. et al. Nature 520, 45–50 (2015).
7. Conley, D. J. et al. Science 323, 1014–1015 (2009).
8. Van Maele-Fabry, G. et al. Cancer Causes Control 21, 787–809 (2010).
9. Springmann, M. et al. Nature 562, 519–525 (2018).
10. Tilman, D. et al. Nature 515, 518–522 (2014).
11. Willett, W. et al. The Lancet 393, 447–492 (2019).
12. Poux, X. et al. ddri-AScA, Study N°09/18, Paris, 78 p. (2018).
13. Vrugt, J. A. et al. Hydrol. Earth Syst. Sci. 15, 3701–3713 (2011).
14. Brisson, N. et al. Eur. J. Agron. 18, 309–332 (2003).

Keywords Agroecology;Food systems;Co-construction;Optimization;Modelling

Primary author

Tom Desmarez (Gembloux Agro-Bio Tech)

Co-authors

Jérôme Bindelle (University of Liège) Benjamin Dumont (ULiege - Gembloux AgroBio-Tech, Plant Sciences Axis, Crop Science lab., B- 5030 Gembloux, Belgium.)

Presentation materials

There are no materials yet.