an easy resolution to these concerns. The Agency should continue to consult with USGS and USDA for the most detailed and up-to-date sources.
5.) EFED recommends collecting raw water and finished water samples with reactive analysis of the finished water if pesticides are detected in the raw water. Finished water more directly reflects drinking water exposure, while raw water better reflects watershed and usage characteristics allowing better risk management decisions.
Given the analytical costs for paired raw and finished water samples and the lack of
comprehensive information on water treatment effects on the removal and transformation of most pesticides, does the panel have alternative recommendations on assessing raw and finished water?
Strong arguments have been made for the value of having both raw and finished water concentration information. Since human exposure is the ultimate goal of this exercise, an appropriate characterization of the human risk requires an understanding of how watershed processes combine to control transport from fields to streams. Thus, both raw and finished water should be tested. This addresses the issue of multiple objectives.
The SAP agrees in principle with the general approach of reactive analysis - to lead with the analysis of raw water and follow up with analysis of finished water. However, some members felt that it was not practical given the large numbers of samples of diverse analytes - some with relatively short sample holding times compared to expected analytical turn-around time. The SAP did express concern about logistical problems related to the functioning of each water-supply system, such as trying to time sampling in the raw and finished water to try to insure that the same parcel of water was being sampled (i.e., by trying to estimate process/holding times through the treatment system). This is necessary to evaluate specific treatment effects. Another idea suggested for consideration was to conduct a unit design study; after the water systems are selected, assess what treatment is in place and then select key, representative systems for paired sampling.
Given the cost constraints, the Agency might consider using some threshold for reactive analysis other than the non-detect level. For example, finished water might only be analyzed if a raw water concentration exceeded 1/10 of the MCL for one of the target compounds. Where to set the threshold would be a difficult balancing act.
Preliminary analyses by EFED
indicates pesticide fate properties, sampling
frequency, and hydrologic residence time impact the accuracy of calculated annual mean
Does the SAP have recommendations on balancing the characteristics of the pesticides
and the water body being sampled with the number of samples needed to adequately measure the annual mean at different CWSs?