Abstract

Soil tillage effects are not accounted for in the surface water pesticide leaching scenarios used within the EU for regulatory modelling even though tillage has been shown to influence pesticide leaching. The objectives of this study were, (i) to evaluate differences between conventional (C) and no-till (NT) systems in water flow and solute transport in the topsoil of a heavy clay soil prone to preferential flow, (ii) to calibrate key model parameters governing preferential flow in the dual-permeability model MACRO 5.1 from laboratory microlysimeter measurements, and (iii) to study differences in pesticide leaching in simulations based on the FOCUS surface water D1 scenario and calibrated parameter values for C and NT. Both C and NT microlysimeters exhibited strong preferential flow. The breakthrough was generally faster for the NT microlysimeters, but the variability within treatments was large. Five parameters, the parameter governing mass exchange between pore domains in two soil horizons, the kinematic exponent in the macropores, the macroporosity and the initial water content were included in the calibration. The observed differences in water flow and chloride leaching between treatments were reflected in the values for the parameter governing mass exchange between pore domains in the surface 6 cm of soil and in the macroporosity. Mass exchange was weaker and macroporosity was smaller in the NT soil, both of which promote faster non-equilibrium macropore flow and transport. The differences in the scenario simulations between C and NT were small for drainage rates. For a weakly sorbed easily degraded compound and a strongly sorbed slowly degraded compound, the scenario simulations predicted 3.4 and 4.6 times larger average concentrations in drainage water respectively for NT compared with C. The maximum hourly concentrations in drainage water using best parameter estimates were 3.4 and 12 times larger for NT compared with C for the same compounds. The uncertainties in the concentrations in drainage water were large. This uncertainty reflects spatial variability in the field, errors in parameter values not included in the calibration, parameter correlations and model error. More data and refined methods to better account for the spatial variability are needed in future research to reduce the uncertainty in predictions of tillage effects in regulatory modelling.

Keywords

pesticides; modelling; tillage

Published in

Emergo
2007,
ISBN: 978-91-85911-25-7
Publisher: Institutionen för markvetenskap, Sveriges lantbruksuniversitet

Permanent link to this page (URI)

https://res.slu.se/id/publ/125972