|Main authors:||R.K. Laursen, F. Bondgaard, P. Schipper, K. Verloop, L. Tendler, R. Cassidy, L. Farrow, D. Doody, F. A. Nicholson, J. R. Williams, I. Wright, J. Rowbottom, I. A. Leitão, A. Ferreira, B. Hasler, M. Glavan, A. Jamsek, N. Surdyk, J. van Vliet, P. Leendertse, M. Hoogendoorn and L. Jackson-Blake.|
|Source document:||R.K. Laursen et al. (2019) Evaluation of Decision Supports Tools. FAIRWAY Project Deliverable 5.2 216 pp|
A comprehensive review and survey of Decision Support Tools (DSTs) currently in use in the FAIRWAY case studies is described in »Survey and review of existing decision support tools. Of the 36 DSTs identified as most relevant, 12 were selected for further investigation to see if a tool developed in a particular national context could be used or provide inspiration elsewhere (»Evaluation of decision support tools). Here we describe the tool evaluated for potential use in the La Voulzie case study.
|1. Selection of DST to evaluate in La Voulzie case study|
|[Note: Because of the resolution of the images, it is difficult to see the detail in some of the figures and tables. See the »full report for more legible originals.]|
The SIRIS tool was tested on the La Voulzie catchment about 70 km west of Paris (»La Voulzie, FR case study). The dominant crops grown are wheat followed by oilseed rape and barley. Wheat and barley are well spread throughout the catchment. To collect the input data, the BNV-d (Banque Nationale des Ventes des distributeurs) database was used. The BNV-d is supplied since 2009 with the declarations of the annual reports of the sales of phytosanitary products by the »authorized distributors.
SIRIS-Pesticides is a DST that allows classification of pesticides according to their potential to reach surface water and groundwater. SIRIS classifies pesticides from ideal to worst. The core system is a penalty grid. SIRIS-Pesticides helps to organize the monitoring of pesticides in waters at the regional or local scale. It is a software tool developed around a simple interface.
There are many DSTs in France which can perform diagnostics at a farm/field scale, but SIRIS is one of the few tools available for predicting pesticide loss at the catchment scale, but has not yet been tested in the La Voulzie catchment. It is therefore possible, from the results, to coordinate actions on several farms. The lack of mitigation measurement simulations is one of the main limitations for stakeholder use.
Parameters are grouped into classes. The class represents the importance of a process compared to the other. Two equivalent parameters are in the same class. Each parameter is affected to one “level” reflecting its contribution to exposure (Table 17).
For example, a substance, with a DT50 (Half-life of the substance representing the degradation) of less than 30 days is classed "o" for this parameter. A substance, with a Koc (Sorption coefficient of the substance representing the sorption on organic carbon of soil) of less than 100 L.kg-1 is classed "d" for this parameter.
The penalty grid assigns a score for each possible combination of parameters. The ideal substance has all its parameters at an “o” level and its score is 0. Penalties are attributed to substances that have parameters affected to levels different from “o”.
This grid lists the scores for every possible combinations of levels. Using the grid, the classes and the levels attributed to the parameters, it is possible to suggest which pesticides should be monitored in priority in freshwaters.
To collect the input data, the BNV-d database was used. The data specific to the catchment have not been gathered, the data for the department of Seine-et-Marne had to be used instead. In this web portal, the user can select when and where data are needed. A specific SIRIS export format is available. An average of the quantities provided over the period 2008-2017 has been used for this test.
Table 18 shows results from SIRIS using the BNV-d input data. The table was adjusted from the original to make it easier to understand for this report. For instance, a level column for each parameter has been included to make the level explicit. Atrazine was added for comparison even though it is no longer sold. Only the top 25 substances with the highest scores are presented.
Table 19 shows all the pesticides detected at least once in spring water. For the FAIRWAY project, data from several springs were analysed. In this report, data from the Durthein spring are presented. The analyses analysis records for the La Voulzie springs are generally too short to be assessed. The first column shows the names of the products, the second column shows the number of detection instances between 2008 and 2017, and the last column indicates if the product has a rank above 30. Pesticides above rank 30, according to SIRIS, could reach the ground water. In the La Voulzie case study, 88 pesticides (out of 280) had a rank above 30.
A concordance between the concentrations actually observed and the results of the SIRIS tool are shown in Table 19. Explanations of the mismatch between the SIRIS forecasts and the measurements in groundwater include:
- Some products have only been measured for a few years so it is difficult to make comparisons. For instance, in La Voulzie spring water, fluroxypyr was measured only 15 times in 4 years (between 2008 and 2017).
- There is delay between pesticide application and measurement of the pesticide in groundwater. Many of the products detected are not sold anymore. The French researchers made a study on the site and calculated that the travel time for water is 7 years. It is known that the travel time for pesticides is always longer than for water.
Table 20 summarizes the advantages and disadvantages of SIRIS used in a French context.
Table 20. Advantages and disadvantages of SIRIS seen in a French context.
|The model is suitable for working at the watershed scale.||The working scale of the model is not suitable for farmers.|
|The tool is very easy to use and in France, the input data are easy to obtain thanks to the BNV-d database.||Input data is easily available in France from the BNV-d database, but the total amount of pesticides is difficult to obtain at smaller scale.|
|The tool is very easy to use. It is possible for a manager, non-specialist modeller, to use it quickly.||The BNV-d + SIRIS association is not able to simulate the impact on water of unauthorized products and metabolites.|
|The tool identifies some of the pesticides that must be restricted.||The tool is very easy to use but knowledge relating to the transfer of pesticides is necessary. For instance, notion on pesticide sorption (Koc) and degradation (DT50) could be needed to understand the tool.|
|Comparisons between the measured data and the predicted data show differences that are difficult to explain.|
|Some features of the model systematically prevent it from correctly reproducing the behaviour of certain pesticides (for example, products with high sorption always have low ranks).|
|Apart from the reduction of doses, no mitigation measures can be tested|
SIRIS was easily applied on the catchment area using the data from the BNV-d. It can easily be applied on other basin watersheds in France. There are two opportunities to use it: a) by using the quantities actually applied or b) using pesticides doses (approved doses). This information is usually available (at least the approved doses).
SIRIS only allows classification of the products, and its minimal scale of operation is the watershed. The tool does not propose mitigation measures at the farm scale or across the basin watershed. Because of this, it may not be used for creating scenarios where practices are changed. The possibility to change the doses is not really usable because SIRIS react by threshold (e.g. parameter surface). It is possible to multiply or divide the dose by three and see no impact if one remains in the same level (Table 17), whereas a minimal change in the dose can have effects if it allows to pass a level ("o" towards "m" for example).
Basin watershed managers and water company managers could use SIRIS but it would be for farmers to access. For managers, it can help to select plans for monitoring, for animators; it can be used to know which products to reduce. By slightly modifying the original output template to make table appear clearer (o m d, see Table 17), it is easier to know why each product gets its ranking. Adding a column showing the threshold value will determine if the product will see its rating change with a small change of input data.
For full references to papers quoted in this article see » References