Sequential use of the STICS crop model and of the MACRO pesticide fate model to simulate pesticides leaching in cropping systems

Abstract

The current challenge in sustainable agriculture is to introduce new cropping systems to reduce pesticides use in order to reduce ground and surface water contamination. However, it is difficult to carry out in situ experiments to assess the environmental impacts of pesticide use for all possible combinations of climate, crop, and soils; therefore, in silico tools are necessary. The objective of this work was to assess pesticides leaching in cropping systems coupling the performances of a crop model (STICS) and of a pesticide fate model (MACRO). STICS-MACRO has the advantage of being able to simulate pesticides fate in complex cropping systems and to consider some agricultural practices such as fertilization, mulch, or crop residues management, which cannot be accounted for with MACRO. The performance of STICS-MACRO was tested, without calibration, from measurements done in two French experimental sites with contrasted soil and climate properties. The prediction of water percolation and pesticides concentrations with STICS-MACRO was satisfactory, but it varied with the pedoclimatic context. The performance of STICS-MACRO was shown to be similar or better than that of MACRO. The improvement of the simulation of crop growth allowed better estimate of crop transpiration therefore of water balance. It also allowed better estimate of pesticide interception by the crop which was found to be crucial for the prediction of pesticides concentrations in water. STICS-MACRO is a new promising tool to improve the assessment of the environmental risks of pesticides used in cropping systems.

Life Cycle Assessment of agri-food systems

“…• urban and peri-urban production systems to feed ever-expanding cities, operating in highly constrained conditions with a generally excessive and inappropriate use of chemical and organic inputs� In terms of LCA modelling, tropical contexts generate specific issues� Most existing direct emission models used in LCA were calibrated for field conditions of crops growing in temperate environments (practices, soil characteristics, temperature, rainfall, etc�)� Hence, their validity domain pertains to the conditions for which they were initially calibrated� It is notably true for the Swiss model suite SALCA (Swiss Agricultural LCA) used in ecoinvent, which encompasses the modelling of all primary field emissions, e�g� nitrogen, phosphorus and trace element emissions, while relying on field data collected in Switzerland only� Other commonly used empirical models for nitrogen and carbon compounds are the IPCC guidelines (IPCC 2006, Volume 4, Chapter 11)� These guidelines are regularly updated to account for state of the art� For instance, in the latest version (IPCC 2019), models from Stehfest and Bouwman (2006) or Cardinael et al (2018) were updated� But the coverage of tropical conditions in the background datasets is still limited (Bouwman et al 2002c)� Existing direct field emission models were not designed -or calibrated -to properly consider specific tropical conditions nor developing and emerging contexts, i�e� the pedoclimatic conditions or the substantial variability in practices (e�g� the high diversity of field inputs, agroforestry systems, etc�) (Table 6�1, more details in Appendix B p� 122)� This issue was also recently demonstrated for pesticide emission models by Gentil et al (2019) and for N emission models by Avadí et al (2022)� Other process-based models exist, such as APSIM (Holzworth et al 2018), STICS (Brisson et al 2003) and combinations of models (Constantin et al 2015;Lammoglia et al 2017), that make it possible to calibrate the models to very specific site conditions� However, calibrating process-based models requires specific expertise and extensive datasets� Moreover, such models are not available for all cropping systems, nor can all process-based models model the field emissions in a mechanistic way� The same limitations apply to impact assessment models which are either too generic or valid only for temperate conditions� For instance, Gentil et al (2019) highlighted in their review the lack of validity of ecotoxicity data for tropical species that show a specific sensitivity to the exposure to pollutants� Avadí et al (2022) demonstrated that direct field nitrogen emissions modelling is to date not well adapted to tropical conditions, organic fertilization, or short-cycle crops such as market vegetables�…”

Common ground: restoring land health for sustainable agriculture

“…Conservation agriculture generally refers to a farming approach built on three main principles: It is important to note that, while conservation agriculture generates a number of benefits, it most often makes use of herbicides to manage weeds, which has detrimental effects on soil biodiversity, water quality and farmers health (Lammoglia et al, 2017). Furthermore, the contribution to mitigating climate change, through accumulation of organic C in soil, has sometimes been overstated.…”

Punto de encuentro: restaurar la salud de las tierras para una agricultura sostenible

“…Es importante señalar que, si bien la agricultura de conservación genera una serie de beneficios, en la mayoría de los casos, hace uso de herbicidas para controlar las malas hierbas, lo que tiene efectos perjudiciales sobre la biodiversidad del suelo, la calidad del agua y la salud de los agricultores (Lammoglia et al, 2017). Además, su contribución a mitigar el cambio climático, a través de la acumulación de C orgánico en el suelo, a veces ha sido exagerada.…”