Temporal Trends of Selected Agricultural Chemicals in Iowa's Groundwater, 1982-95: Are Things Getting Better?

Dana W. Kolpin, Debra Sneck-Fahrer, George R. Hallberg, and Robert D. Libra

---

DISCUSSION

In this study, we have demonstrated that a general decrease in median atrazine concentrations and a general increase in median metolachlor concentrations has occurred (particularly between the 1987-91 and 1992-95 time periods) in a subset of IGWM wells that had the longest record of water-chemistry data. These temporal concentration changes appear to coincide with their corresponding trends in long-term chemical-use. A fundamental question that remains to be answered, however, is whether the severity of groundwater contamination by combined agricultural chemicals is increasing or decreasing through time. In other words, are the overall groundwater quality conditions improving (is the problem getting better or worse), or is there a simple replacement of chemicals in groundwater (decreasing concentrations of one chemical is balanced by an increase for another).

One measure of the overall severity of contamination by agricultural chemicals is the total combined concentration in groundwater. For this study, a summation of alachlor, atrazine, cyanazine, and metolachlor concentrations (SUM) was used. Concentrations below the 0.10 µg/L analytical reporting limit were set to zero for the calculation of SUM.

No significant difference (P = 0.065) in median SUM through time was identified in the subset of 89 wells using a Kruskal-Wallis test, but a significant decrease (P = 0.002) in median SUM through time was found using a one-tailed, Wilcoxon signed-rank test. A substantial decrease in the median SUM concentration occurred during the 1992-95 time period (Fig. 6). Thus, a decreasing trend in the overall severity of groundwater contamination from pesticides might be occurring. This decreasing trend appears coincident with progressive decreases in the overall combined use of alachlor, atrazine, cyanazine, and metolachlor in Iowa (Table 1) -- with over a 20% decrease occurring during this study. There is, however, an offset between the greatest period of change in the combined chemical use (1982-86 and 1987-91, Table 1) and the greatest period of change in the median SUM concentration (1987-91 and 1992-95, Fig. 6). This suggests that, as noted in the literature (e.g. Hall, 1992), a lag time likely exists between changes in chemical use at the land surface and measured concentrations in groundwater.

This simplistic measure of severity, however, has various shortcomings. First, it does not take into account differences in toxicities between these four compounds (all chemicals have equal weighting). Currently, maximum contaminant levels are only determined for alachlor (2 µg/L) and atrazine (3 µg/L); with non-regulatory health advisory levels set for cyanazine (1 µg/L) and metolachlor (100 µg/L) (U.S. Environmental Protection Agency, 1996). Second, this measure of severity does not take into account the presence in groundwater of other pesticides that are known to be used in Iowa. Other pesticides beyond the four examined for this study have been found in Iowa's groundwater (e.g., Kross et al., 1990; Kolpin et al., 1995; Kolpin, Zichelle, and Thurman, 1996). Third, this measure of severity does not take into account the presence in groundwater of pesticide degradation products, some of which have been found to be prevalent in groundwater (Barrett, 1996; Kolpin et al., 1996; Kolpin et al., 1997; Potter and Carpenter, 1995; Roy and Krapac, 1994). If degradation products are not quantified, the total concentration in groundwater is substantially underestimated. Finally, this measure of severity assumes the toxicological interactions between all combinations of agricultural chemicals are additive and not synergistic. Research has shown that some combinations of pesticides have synergistic toxicity (Marinovich et al., 1996; Thompson, 1996). Thus, to adequately determine if there is an actual decreasing trend in the overall severity of contamination, the collection of additional water-chemistry data and the investigation of other measures of severity are required.

Next: Conclusion