Abstract

The Greenhouse Emission Model has recently been adopted as a model package for assessing emissions to and concentrations in groundwater and surface water after use of plant protection products in greenhouse crops. Stakeholders advised that the model be tested against experimental data.

In October 2014, facilities of WUR Plant Research were used to perform a pilot experiment in which cucumber plants on stone wool substrate were treated with three plant protection products, using a drip irrigation method. Concentrations of the active substances were measured in both the water flowing to and draining from the substrate. GEM was tailored to the experimental conditions and used to predict concentrations in parts of the experimental system. Measured and simulated concentrations of imidacloprid and fluopyram were comparable from approximately 36 hours after the start of the experiment onwards. Prior to this, concentrations in the inflowing water were underestimated and concentrations in the drain water were overestimated, probably because of incomplete mixing. For dimethomorph, agreement between the measured and calculated concentrations was reached after approximately 80 hours. This more lengthy period may be due to exceeding the solubility of the substance, causing precipitation or settling on the tube walls, and redissolving later on; the model does not account for these processes.

Degradation of all three substances was found to be negligible over the duration of the experiment. Plant uptake was the major dissipation process. Experimental results show that uptake of substances was lower than uptake of water, thereby supporting the transpiration stream concentration approach proposed by Briggs et al. (1982); this approach is often applied however experimental evidence is scarce. Transpiration stream concentration factors far below one were found to fit experimental results best.