The presence of pesticides at concentrations exceeding the EU drinking water limit has been reported in many aquifers used for drinking water supplies in Europe. 
      
The PEGASE research project (Pesticides in European Groundwaters: detailed study of representative Aquifers and Simulation of possible Evolution scenarios; 
EU contract number EU contract EVK1-CT1999-00028) was funded by the European Commission       
to: i) investigate pesticide contamination in a number of representative aquifers;       
ii) elucidate transport processes from the soil surface to and in groundwater; and iii) develop advanced tools (deterministic mathematical models, but 
also socio-economic instruments) supporting the management of pesticide usage with regard to the sustainable management of groundwater (GW) quality. 
            

Six contrasted aquifers representative of European GW resources (from a sandy aquifer with a GW table <2 m below ground level to a 50 km2 karst aquifer with GW 
at >10 m depth) were extensively monitored for water and pesticide fluxes for up to three years. The six study areas were Brévilles (Normandy region, France), 
Les Trois Fontaines (Centre region, France), Zwischenscholle (North Rhine - Westfalia, Germany), Roswinkel (Drente region, The Netherlands), Martigny 
(Valay, Switzerland) and Havdrup (Sealand, Denmark). The intensity of data collection was adjusted according to the aquifers and variables monitored. Data 
collected included information on soils, land use, climate, pesticide usage, hydrogeology and geology. Knowledge about the transport of water and pesticides 
in soil and in the unsaturated and saturated zones was obtained through i) field leaching experiments in soil; ii) soil and subsoil coring; and iii) sampling 
of groundwater at various locations and depths. In addition, laboratory studies were conducted on sorption and degradation of five pesticides in batch systems 
with solids recovered from the saturated zone of five aquifers. The monitoring exercise demonstrated that i) there is a potential for significant and rapid 
transport of selected pesticides to depth under unfavourable climatic conditions (i.e. significant rainfall events shortly after application); ii) except 
for selected compounds which were found throughout the year (e.g. atrazine, deethylatrazine), detections of pesticides in groundwater were generally limited 
to a few weeks following application; and iii) concentrations of pesticides in aquifers are very variable in space and time. The laboratory studies revealed 
that a significant degradation of MCPP and isoproturon in samples of aerobic saturated zone occurred. Redox conditions were found to influence potentials 
for sorption and degradation. Mechanistic or semi-empirical tools were developed for the modelling of pesticide contamination in GW at various spatial scales. 
Approaches investigated included the refinement of a screening tool (PESTGW) and 1D root zone models (MACRO and ANSWERS), the addition of pesticide fate and 
crop subroutines in integrated models (thereby allowing the prediction of pesticide fate in the soil-unsaturated zone-saturated zone continuum; MARTHE, 
TRACE and POWER) and the coupling of different models (TRACE+3DLEWASTE, MACRO+FRAC3DVS, MACRO+MODFLOW, ANSWERS+ MODFLOW). The numerics of the models were 
upgraded within the framework of the project, which allows the future deployment of advanced modelling activities, such as automated calibration against 
field data or sensitivity and uncertainty analyses. The majority of the new modelling tools developed within the project were applied to reference case studies, 
numerical solutions or to the datasets collected as part of the project. First investigations demonstrated that new subroutines had been correctly implemented 
in the models. The large inter-annual and spatial variability in pesticide concentrations noted at most study sites was difficult to reproduce with the models 
in some instances. Inadequate simulation of field data by models was generally attributed to i) the lack of long-term spatialized information on pesticide 
application practices in the catchments; ii) the lack of spatialized information on the structure and properties of the unsaturated and saturated zones; 
and iii) the inability of most models to account for preferential flow processes. Overall, the use of integrated models enabling the simulation of pesticide 
fate in the soil and in the subsoil in 1, 2 or 3 dimensions was considered to be more appropriate than combinations of root zone and groundwater models for assessing 
the transfer of pesticides from the soil surface to and in groundwater. Still, the modelling work undertaken within the project confirmed the challenging 
nature of simulating water and pesticide fluxes in the root zone - unsaturated zone - saturated zone continuum. Alternative management scenarios to ensure 
the sustainability of groundwater in relation to quality issues were developed on the basis of the driving forces and solutions envisaged for pesticide management 
strategies. These scenarios were implemented within a software tool referred to as 'PEG@SE'. The system is designed to create a bridge between scientific and 
non-scientific knowledge, integrating technical, scientific and social information in a single tool. The system has been evaluated through interactive 
demonstrations in schools, universities and local authorities and proved a useful tool in helping users in formulating their expectations with regard to 
water quality. 
            

The PEGASE project was designed by 11 research institutes from 8 European countries to i) characterise the presence and transfer of pesticides from the soil 
surface to groundwater in a number of locations representative of European aquifers; ii) elucidate processes involved in the transport of pesticides from 
the soil surface to and in groundwater; and iii) develop advanced tools (mathematical modelling tools, but also socio-economic instruments) supporting 
the management of pesticide usage with regard to the sustainable management of groundwater quality. The extensive monitoring and characterisation of water 
flow and pesticide transport, which covered the root, unsaturated and saturated zones, supported the development, refinement and evaluation of the modelling 
tools.       
       
A significant amount of work within PEGASE was dedicated to the refinement and improvement of modelling tools in an effort to further the capabilities of 
models in the simulation of pesticides to and in the groundwater. This involved i) the refinement of root zone models; ii) the coupling of models, e.g. models 
simulating water and pesticide fluxes in the root zone and groundwater models; and iii) the addition of new subroutines to integrated models thereby allowing 
the simulation of water transport and pesticide fluxes from the soil surface to and in the groundwater.        
       
A revised version of the preferential flow model MACRO was developed. It and other models can be downloaded from http://www2.brgm.fr/pegase/modeling_appr ... 
     
       
Pesticide fate routines from the GLEAMS model were integrated into the ANSWERS model. MACRO and ANSWERS were combined to two groundwater flow models (MODFLOWT 
and FRAC3DVS), but difficulties in the coupling of the models were experienced in some instances. The couplings between the various models were restriced 
to movement of water and pesticide from the root zone model to the groundwater model only. Three 3D integrated groundwater models were adapted to simulate the 
water transport and pesticide fate from the soil surface to and in the groundwater (MARTHE, TRACE and POWER). This involved the addition of new subroutines 
to account for pesticide degradation, crop and root development and water and solute uptake by plants. Finally, a simple screening tool with low input requirements, 
PESTGW, was adapted. Numerics of the various models were improved as part of the refinement work and running times were greatly reduced in numerous instances. 
      
       
The final report of the project can be downloaded from the project web page. 
            

A number of PEGASE research findings have a direct relevance to current policy initiatives, notably the Water Framework Directive and the Daughter Groundwater 
Directive. The strong variability in pesticide concentrations measured in groundwater, which translated into differences in neighbouring piezometers 
and significant variations in concentrations from one sampling date to the other, is of importance with regard to the design of monitoring activities aimed 
at assessing the chemical status of water resources within the scope of the Water Framework Directive. Laboratory experiments undertaken in the project suggested 
that there is a significant potential for degradation of selected pesticides in aerobic aquifers. This natural remediation process has rarely been reported 
before and could form the basis of a long-term management strategy for groundwater quality. Finally, a range of advanced modelling tools enabling the simulation 
of the fate of pesticides from the soil surface to and in the groundwater have been developed. Although further evaluation of the models is required, they offer 
great expectations with regard to their use as groundwater quality management tools. A total of four priority research needs have been identified. First, 
there is a need for additional information on the distribution of pesticides in the unsaturated and saturated zones. A number of recommendations with regard 
to sampling schemes and data collection have been made to support the design of future groundwater monitoring activities. Secondly, laboratory experiments 
undertaken within the project have revealed that there is a significant potential for pesticide degradation in the unsaturated and saturated zones depending 
on redox conditions of the aquifer and the nature of the pesticide. These innovative results deserve further investigation as they could lead to a better understanding 
of the functioning of the aquifer with regard to pesticide contamination and the development of remediation strategies. Thirdly, there is a need to undertake 
further research into the application of integrated models which can simulate water transport and pesticide fluxes from the soil surface to and in the groundwater. 
 
 
Finally, an aspect which should be researched is the connection between preferential flow phenomena in soil and flow processes in the unsaturated and saturated 
zones. The significant contribution of preferential flow to the presence of pesticides at the bottom of soil profiles has been established, but little is known 
about      
the fate of these ‘hot spots’ in the deeper environment and hence the overall contribution of preferential flow phenomena to the contamination of European 
aquifers by pesticides.