Pesticide Concentrations in Surface Waters of New York State in Relation to Land Use - 1997

by Patrick J. Phillips , Gary R. Wall, David A. Eckhardt, Douglas A. Freehafer, and Larry Rosenmann

Table of Contents

Abstract
Introduction
Methods
Pesticides in Surface Waters of New York
Water-Quality Criteria
Use of Low Detection Limits
Summary and Conclusions
References Cited

Abstract

Analyses of water samples collected from 64 streams and rivers across New York State in June 1997 indicate that patterns of pesticide detection are largely related to the predominant upstream land use and pesticide-application patterns. Of the 47 pesticides for which the samples were analyzed, 25 were detected. Concentrations of most pesticides detected were low and generally did not exceed 0.1 ug/L (microgram per liter). Herbicides used on cornfields, including atrazine, metolachlor, cyanazine, alachlor, and the atrazine degradate deethylatrazine, were detected in samples from 41 to 97 percent of the 64 sites sampled. The highest concentrations (greater than 0.10 ug/L) of these compounds were in streams in western New York State, where corn production is the greatest. Two insecticides-carbaryl and diazinon-were detected in 20 and 14 percent of the samples, respectively. Carbaryl was detected most frequently in streams whose drainage basins either contain extensive vineyards or orchards, or are widely urbanized. Diazinon was detected most frequently in streams that drain urban or residential watersheds. Concentrations of four insecticides-azinphos-methyl, p,p'-DDE, diazinon, and dieldrin- and one herbicide-simazine-exceeded New York State water-quality criteria. Some Federal or State criteria were exceeded at 10 sites. These results represent an initial assessment of the status of pesticide concentrations in surface waters of New York State and, when combined with data collected in the future, will help water managers to assess the status, trends, and health impacts of pesticide contamination of ground and surface waters of New York State including Long Island. This information also will be useful to researchers and water managers who require such data to define the health and environmental effects of pesticide use in the State.

Introduction

In 1997, the State of New York and the U.S. Geological Survey (USGS) began a cooperative effort to monitor pesticides in State waters as required under the New York State Pesticide Reporting Law (Environmental Conservation Law Section 33-0714). The initial monitoring phase entailed a statewide survey of pesticide concentrations in surface waters, particularly in areas where pesticides are applied and in areas where surface water is used for water supply. Samples were analyzed for 47 pesticides, including herbicides, insecticides, and their degradation products. Herbicides are used to control weeds in agricultural fields as well as lawns, commercial land, and other open areas in urban and residential settings. Insecticides are used to control insects in agricultural and urban settings

This report presents the results of the June-July 1997 statewide pesticide survey of 64 streams and rivers across New York State, and discusses the methods used to collect and analyze the data. Detection rates for several pesticides are presented, and pesticide concentrations are discussed in relation to (1) Federal and State water-quality standards, (2) results of previous water-quality investigations in New York State, and (3) predominant land-use and pesticide-use patterns in the watersheds investigated.

Methods

Water samples were collected from a statewide network of 64 sites (table 1, fig. 1) from early June through early July 1997, by which time most agricultural pesticides had been applied. Each site was sampled once, and samples were collected under base-flow (dry-weather) conditions except at five sites, where they were collected during periods of stormflow runoff. In general, concentrations of pesticides in streams in June and July are highest during stormflow conditions (Wall and Phillips, 1997), but inclusion of these few samples in the analysis had negligible effect on the results. Samples from six sites on Long Island were collected as part of the USGS Long Island/New Jersey National Water-Quality Assessment program. Together, the 64 sites represent a wide range of land uses-forested, agriculture (cropland, orchards, and vineyards), urban, and residential. The watersheds represented by these sites range in size from less than 1 mi2 (square miles) to more than 10,000 mi2.

Water samples were collected and filtered in accordance with methods described by Shelton (1994) and were analyzed for 47 pesticides through methods described by Zaugg and others (1995). Detection limits (technically known as Method Detection Limits) for pesticides analyzed ranged from 0.001 to 0.018 ug/L. Analyses of quality-assurance samples indicate that these laboratory results accurately represented concentrations in the streams. The laboratory methods used in this study resulted in low and (or) inconsistent recovery for five pesticides (carbaryl, carbofuran, deethylatrazine, terbacil, and azinphos-methyl). Thus concentrations reported for each of these compounds are considered estimates (Chris Lindley, U.S. Geological Survey, written commun., 1994). Detection rates are reported as a percentage of the total number of samples analyzed, and include samples in which concentrations were reported as being below the method detection limit. This reporting is common when a compound can be conclusively identified (Jeffrey W. Pritt, U.S. Geological Survey, written commun., 1994). These concentrations indicate the presence of pesticides in the sample; these concentrations are considered estimates. The data discussed in this report are available in Butch and others (1998) and on the Internet at http://ny.usgs.gov/htmls/pub/nypesticides/index.html. Each site was classified in one of five categories, depending on the predominant land use in the watershed. These categories were Forested, Urban/Residential, Orchard/Vineyard, Low row-crop agricultural and High row-crop agricultural. Watershed boundaries were overlain on mapping-data imagery generated from satellite data collected in 1994 (U.S. Geological Survey, 1997). Forested watersheds were defined as those in which forests and wetlands cover more than 88 percent of the watershed area. Urban/residential watersheds are those in which more than 13 percent of the land is urban (including residential, commercial and industrial land, parks, lawns, and golf courses). Low row-crop agricultural watersheds were those in which row crops occupy less than 20 percent of the land, and high row-crop watersheds are those in which more than 20 percent of the land is planted in row crops. The remote-sensing data were inadequate for delineation of orchards and vineyards. Therefore, these watersheds were not classified according to remote-sensing data, but through field reconnaissance, as having substantial orchard or vineyards.

Pesticides in Surface Waters of New York

The most commonly detected pesticides were the herbicides that are commonly applied to cornfields. The herbicides atrazine, metolachlor, and the atrazine-degradation compound deethylatrazine were detected in 80 percent of the streams (figs. 1 and 2). Other frequently detected herbicides that are commonly used on cornfields include alachlor and cyanazine, which were detected in 50 and 41 percent of the streams sampled, respectively. The highest concentrations of these compounds were found in western New York streams that drain areas with the greatest corn production in the State. These four herbicides also are frequently found in streams and rivers of the Midwest, which drain the nation's major corn-producing regions (Goolsby and Battaglin, 1993). Atrazine was detected in all but two of the streams; yet the concentrations of atrazine in the four forested watersheds were extremely low. The presence of atrazine in streams draining forested watersheds is probably due to atmospheric transport and deposition. The herbicide EPTC, which is commonly used on corn and dry beans, was detected in slightly more than 10 percent of the streams, most of which are high row-crop watersheds in western New York . The herbicide simazine was detected in 72 percent of the streams sampled. This compound is commonly used in orchards and vineyards, and many of the streams and rivers with the highest concentrations of simazine drain watersheds in western New York that are clasified as orchard/vineyard.

Two insecticides-carbaryl and diazinon-were detected in 20 percent and 14 percent of the samples, respectively. These compounds were most often detected in streams draining areas in which these compounds are commonly applied-carbaryl in orchard/vineyard and urban/residential watersheds, and diazinon in urban/residential watersheds. The highest carbaryl concentrations were found in streams draining two types of watersheds-orchard/vineyard watersheds in western New York, and urban/residential watersheds in southeastern New York (including Long Island). The highest concentrations of diazinon were found in urban/residential watersheds in southeastern New York, including Long Island.

In general, concentrations of most pesticides detected in this statewide survey were low, and few exceeded 0.1 ug/L. The largest exceptions to this generalization were atrazine, metolachlor, cyanazine, and simazine; more than 10 percent of the streams contained these compounds in concentrations greater than 0.1 ug/L. Of the 47 pesticides studied, 22 were not detected in any sample (table 2).

The pesticide concentrations measured in this survey probably do not reflect maximum annual concentrations because most of the samples were collected during base-flow (low-flow) conditions. Previous sampling for pesticides in a small agricultural watershed in the Hudson River Basin during 1994-96 indicated that concentrations of pesticides are lower in base flow than in stormflow (Wall and Phillips, 1996a, 1997). Base flow consists mostly of ground water that discharges from the underlying aquifer to streams. Thus, the presence of pesticides in base flow samples suggests that these pesticides may be present in ground water.

The similarity of results from the 1997 survey to results from a 1994 survey of pesticides at 46 sites in the Hudson River Basin (Wall and Phillips, 1996b) indicates that most of these pesticides have been present in New York streams for at least 3 years. Both surveys used identical sample-collection and analytical methods; therefore, the results of the two surveys can be compared directly. The most commonly detected pesticides in both surveys were atrazine, metolachlor, and deethylatrazine. Other compounds that were commonly detected in both years were simazine, cyanazine, and alachlor. The most commonly detected insecticides in both years were carbaryl and diazinon.

Water-Quality Criteria

Concentrations of only a few compounds exceeded applicable State or Federal water-quality standards. Concentrations, detection limits, and water-quality criteria are summarized in figure 2. No pesticides exceeded Federal MCL (maximum contaminant levels) or health advisory levels (HA), and four insecticides (azinphos-methyl, p,p'-DDE, diazinon, and dieldrin) and only one herbicide (simazine) exceeded a New York State water-quality criterion. (New York State water quality criteria are given in New York State, 1998; Federal standards are given in United States Environmental Protection Agency, 1996). Three types of State criteria were exceeded-those for consumption of fish (for the persistent organochlorine compounds p,p'-DDE and dieldrin), those for the protection of aquatic life (azinphos-methyl and diazinon) and for surface water (simazine). One or more State criteria were exceeded in samples from 10 sites. The State criterion for consumption of fish (6.0 x 10-7 ug/L) was exceeded at four sites (sites 11, 37, 38, and 42, see table 1) for p,p'-DDE; most of these sites are in orchard/vineyard watersheds in western New York. The State criterion for consumption of fish (7 x 10-6 ug/L) was exceeded at four sites (sites 2, 38, 64, 66) by dieldrin. Three of these sites are on Long Island in urban/residential watersheds; the other (38) is in western New York in orchard/vineyard watershed. Use of DDT (the parent compound of p,p'-DDE) and dieldrin is prohibited in New York State.

The State criterion for protection of aquatic life (0.005 ug/L) for azinphos-methyl was exceeded at four sites (35, 36, 37, and 38). All of these sites are in western New York in orchard watersheds. The state criterion for protection of aquatic life (0.070 ug/L for diazinon ) was exceeded at three sites (sites 7, 35, and 64). Two of these sites are in urban/residential watersheds on Long Island or southeastern New York, and the other (site 35) is in an orchard/vineyard watershed in western New York. The State guideline for surface water and class GA ground water (0.50 ug/L for simazine) was exceeded at one site (site 40) . This site is in an orchard/vineyard watershed in western New York.

Use of Low Detection Limits

This study used detection limits that are generally (1) much lower than Federal or State water-quality criteria, and (2) below those used in most other studies and monitoring programs. The reasons, paraphrased from Ryker and Williamson (1996), are explained below:

1. Use of low detection limits (for pesticides) allows detection of temporal trends and identification of streams that need protection to prevent concentrations of pesticides from increasing to levels that could threaten the water quality or the ecological health of a stream. Although detection limits close to the established water quality-criteria are suitable for compliance monitoring, they would provide less useful data than do low detection limits for early warning of increasing pesticide concentrations. Use of low detection limits over a long period can help indicate whether pesticide concentrations are increasing, decreasing, or remaining constant.

2. Low detection limits allow researchers to discern correlations between pesticide exposure and human health or ecological health. If detection limits were higher, most pesticide concentrations would be reported as below those limits and could not be used in statistical correlations between pesticide exposure and human health.

3. Low detection limits maximize the number of samples that can be used to relate pesticide concentrations to environmental factors. Large numbers of samples decrease the uncertainty in predicting pesticide contamination.

4. Low detection limits can increase the likelihood that pesticides not detected in analyses are truly absent from waters sampled.

Summary and Conclusions

Results of an initial assessment of the status of pesticide concentrations in surface waters of New York State indicate that, of the 47 pesticides studied in a statewide survey of 64 streams and rivers in New York State in June-July 1997, 25 pesticides were detected, and most detected pesticides were at concentrations below 0.10 ug/L. The most commonly detected pesticides (detected at more than 80 percent of the sites sampled) were herbicides that are commonly applied to cornfields, including atrazine, metolachlor, and the atrazine-degradation compound deethylatrazine. The highest concentrations (between 0.1 and 1.0 ug/L) for these three compounds were found in streams in western New York that drain areas with the greatest corn production in the State. Two insecticides-carbaryl and diazinon-were detected in 20 percent and 14 percent of the samples, respectively, and were most frequently detected in streams draining watersheds dominated by orchards or vineyards or in watersheds dominated by urban or residential land use. Insecticides, that were detected, were mostly at concentrations below 0.01 ug/L. In general, patterns of pesticide detections corresponded to patterns of use.

Concentrations of only a few compounds exceeded applicable State water-quality standards, and no concentrations exceeded federal health advisory or maximum contaminant levels. New York State water-quality criteria were exceeded at 10 sites by four insecticides (azinphos-methyl, p,p'-DDE, diazinon, and dieldrin) and one herbicide (simazine).

References Cited

Butch, G. K., Lumia, Richard, and Murray, P. M., 1998, Water Resources data - New York, water year 1997, Volume 1. Eastern New York excluding Long Island: U. S. Geological Survey Report NY-97-1, 400 p.

Goolsby, D. A., and Battaglin, W. A., 1993, Occurrence, distribution and transport of agricultural chemicals in surface waters of the midwesten United States, in Goolsby, D. A., Boyer, L. L., and Mallard, G. E., (compilers), Selected papers on agricultural chemicals in water resources of the midcontinental United States: U.S. Geological Survey Open-file Report 94-418, p. 1-25.

New York State, 1998, Water Quality Regulations for Surface and Groundwaters, Title 6 Chapter X (Parts 703.5, Table 1), 10 NYCRR Subpart 5-1 New York State Health Department Public Water Systems Regulations effective March 12, 1998.

Ryker, S. J., and Williamson, A. K., 1996, Pesticides in public supply wells of the Central Columbia Plateau: U.S. Geological Survey Fact Sheet 205-96, 4 p.

Shelton, L. R., 1994, Field guide for collecting and processing stream-water samples for the National Water-Quality Assessment Program: U.S. Geological Survey Open-File Report 94-455, 42 p.

U.S. Environmental Protection Agency, 1996, Drinking water regulations and health advisories: Washington DC, Environmental Protection Agency Office of Water, EPA 822-B-96-002, Oct., 1996, 11 p.

_______, 1986, Ambient water quality criteria for chlorpyrifos - 1986: Washington DC, Environmental Protection Agency Office of Water, EPA 440/5-86-005, September 1986, 2 p.

U.S. Geological Survey, 1990, Land use and land-cover digital data from 1:250,000-scale and 1:100,000-scale maps, data users guide 4: Reston Va., U.S. Geological Survey, 33 p.

_____, 1997, Digital map file of Land Cover for the Environmental Protection Agency Region II, Version 1: EROS Data Center, Sioux Falls, S.D., 1:100,000-scale, 1 sheet.

Wall, G. R., and Phillips, P. J., 1996a, Pesticides in surface waters of the Hudson River basin - Mohawk River Subbasin: U.S. Geological Survey Fact Sheet 237-96, 4 p.

_____, 1996b, Pesticides in surface waters of the Hudson River basin, New York and adjacent States: U.S. Geological Survey Fact Sheet 238-96, 4 p.

Wall, G. R., and Phillips, P. J., 1997, Pesticide concentrations in Canajoharie Creek, New York, 1994-96, U.S. Geological Survey Fact Sheet 131-97, 4 p.

Zaugg, S. D., Sandstrom M. W., Smith, S. G., and Fehlberg, K. M., 1995, Methods of analysis by the U.S. Geological Survey National Water Quality Laboratory-Determination of pesticides in water by C-18 solid-phase extraction and capillary-column gas chromatography with selective-ion monitoring: U.S. Geological Survey Open-File Report 95-181, 49 p.

Back to the New York State Pesticide Monitoring Home Page