The aim of this study was to perform a field experiment to collect a high quality data set suitable for validating and improving pesticide leaching models and nutrient leaching models for drained and cracking clay soils. The transport of water, bromide, nutrients and the pesticides bentazone and imidacloprid was studied on a 1.2 ha experimental plot. Moisture profiles and groundwater tables were measured, starting in November 1997. Winter wheat was sown on 23 October 1997 and harvested on 20 August 1998. Bentazone and bromide were applied at 7 April 1998; imidacloprid was applied at 27 May when the soil was almost completely covered by the crop. The amount present in soil was measured within 2 days after application (32 sampling cores) and was found to vary between 80% of the nominal dose (imidacloprid) to 110 % (for bentazone). Manuring and soil cultivations were as usual for the wheat crop. Soil profiles were sampled at eight times (16 cores at each date, last in April 1999). Drain flow was continuously recorded and the water flow proportionally sampled for analysis of the test compounds. Groundwater was sampled periodically from sets of permanently placed filters at four depths at 16 sites. Sorption isotherms of the pesticides were measured with soil from 0-25 cm. Transformation rates of the pesticides were measured at different temperatures in soil material from topsoil and subsoil layers. Soil hydraulic properties and shrinkage characteristics were measured in the laboratory. Meteorological data (i.e. rainfall, air temperature, global radiation, air humidity etc.) groundwater levels and soil temperatures at three depths were monitored continuously. After 56 days, about 80% of the bromide dose was taken up by the crop, which demonstrates that bromide is not a suitable tracer in cropped soil during the growing season. After that time the bromide was gradually released again into the soil. Preferential transport through cracks and macropores of all test compounds was measured both in summer and in winter. This resulted in the highest concentration of bromide and bentazone measured in drain water already 21 days after application following 56 mm rainfall. Imidacloprid was already detected in groundwater at 1.3-1.5 m depth, 11 days after application, following 65 mm rainfall. High peaks in nitrate concentrations in the groundwater at 1.00-1.50 m depth and in the drain water were detected within 14-18 days after the first fertilizer application, following 94 mm of rainfall. Extreme high peaks in concentrations of ortho-P and soluble organic-P were measured in the drain water at respectively 2 days and 37 after slurry application (the only phosphorus application during the experiment). For nitrate concentrations in the drain water there were indications for bypass by preferential flow of 'clean' rainwater to the drains.

In 2016 a greenhouse scenario for soil-bound crops has been adopted in the Dutch authorisation procedure for plant protection products to assess the exposure of aquatic organisms. However, two important shortcomings resulting from the parameterisation of the surface water scenario were discovered. These were due to the assumption of an exact repetition of the irrigation scheme for a fully-grown chrysanthemum crop and that of a top layer with a well-developed macropore system. The first assumption resulted in an extreme sensitivity of the exposure concentration to the application date and the second assumptions was considered unrealistic, because the soil in the greenhouse is frequently ploughed. The above two issues were addressed in the revised exposure scenario presented in this report. The first important improvement was the division of the greenhouse into 24 cultivation sections with a description of the sequence and growth development of all crop cycles in each cultivation section. The second shortcoming was remedied by the change of the topsoil layer with a well-developed macropore system into a topsoil layer without macropores. The target concentration for the assessment of the exposure in surface water strongly depends on the half-life of the substance in the greenhouse soil. Should data on the DegT50 obtained from measurements in greenhouse soils not be available, we recommend to apply provisionally the same factor as included in the tiered assessment scheme proposed by Wipfler et al. (2014), i.e. multiplying the DegT50 value derived from measurements in field soils by a factor 10. The new version of the Greenhouse Emission model (GEM) contains the option to specify an application day relative to the date of planting or harvest. If the GAP (Good Agricultural Practice) of a plant protection product specifies a number of applications per crop cycle, this option should be used. If the GAP specifies a number of applications per year, the application option to specify an absolute date should be selected. For the selection of the application date resulting in the required percentile of the PEC90 in surface water, we recommend an update of the SAFE (Select Application For Evaluation) tool. Suggestions are made to increase the validation status of the model and for the further development and use of the GEM model within the EU.

Door middel van modelberekeningen, literatuuronderzoek en analyse van grasrassenonderzoek is een schadedrempel voor herinzaai van grasland op droge zandgrond vastgesteld. Daarbij is de toename van herinzaai door extra droogtestress gekwantificeerd voor twee niveaus van grondwateronttrekking voor waterwinning. Het melkveeproefbedrijf De Marke dat in het waterwingebied ’t Klooster ligt, is hierbij als uitgangspunt genomen. De toename van de opbrengstdepressie en de extra kosten voor herinzaai zijn vertaald in normbedragen voor extra herinzaai.

Et simulatiemodellen is de diffuse nutriëntenbelasting van het oppervlaktewater vanuit/vanaf de bodem van vier proefgebieden in Laag Nederland berekend. De studie is onderdeel van een project gericht op de berekening van de effecten van diffuse belasting op de oppervlaktewaterkwaliteit en de toesting van een modelinstrumentarium hiervoor. Naast bemesting blijken bodem en kwel belangrijke bronnen van nutriëntenuitspoeling, vooral bij de twee veenweidegebieden die overwegend de grootste nutriëntenbelasting hebben. De bijdrage van mest aan de uitspoeling bedraagt bij deze gebieden slechts 23-32%. Bij de twee andere gebieden met meer minerale bodems is de mestbijdrage meestal groter, tot 74%. Gemiddeld spoelen bij de veenweidegebieden 2,0% en 1,4%, en bij de twee andere gebieden 3,3% en 0,7% van de stikstof- respectievelijk fosformeststoffen uit en af.