Key Words: Surface and Groundwater, San Juan-Argentina, Organochlorine pesticides.

The level of contamination with organochlorine pesticides and the occurrence of their degradation products in the basins of the two main rivers, San Juan and Jáchal, of the Province of San Juan, Argentina, were determined. Surface and ground water samples from both river basins were evaluated by capillary GC and results confirmed with Mass Spectrometry. Chemicals investigated were: alfa-hexachlorocyclohexane (a-HCH); beta-hexachlorocyclohexane (b-HCH); gamma-hexachlorocyclohexane (g-HCH); delta-hexachlorocyclohexane (d-HCH); Heptachlor; Heptachlor epoxide; Aldrin; Dieldrin; Endosulfan I; Endosulfan II; 1,1,1-trichloro-2,2 bis (p-chlorophenyl) ethane (p,p´ DDT; 1,1-dichloro-2,2 bis (p-chlorophenyl) ethylene (p-p´DDE); 1,1-dichloro-2,2-bis (p,p´-chlorophenyl) ethane (p-p´DDD); 1,1-dichloro-2,2-bis-(o,p´-chlorophenyl) ethane (o-p´DDD); Methoxychlor; Endrin.

Concentrations found for b-HCH and Dieldrin were above the maximum permissible values recommended by international organizations (E.C. Council Directive 1980). Mean values found in positive samples were 14.20 µg/L (0.01 ­ 208.10) and 9.35 µg/L (0.07 ­ 145.00) for b-HCH and Dieldrin, respectively.

For a total number of 314 samples, the percentage of positive samples ranged from 68.6 % for Heptachlor to 16 % for Aldrin. Concentration values and the percentage of positive samples in ground water were significantly lower than those found in surface water. In some surface samples, the values were higher than the allowed maximum (1 to 3 mg/L). Samples taken in different seasons did not show significant differences.

Global chemical pollution has become a matter of great concern with the increase in public awareness of environmental problems. Among a large number of man-made chemicals, greater attention has been focused on semi-volatile and persistent organochlorines, such as DDT, S(summation) HCH, Heptachlor, Aldrin, and their metabolites or degradation products, on account of their high bioaccumulation potential and harmful biological effects (Waliszewski, Pardio, Chantiri, Infanzón and Rivera, 1996; Beretta and Dick, 1994; Baudino, Pestchanker and Aguilar, 1985; Muir, Ford, Grift, Metner and Lockhart, 1990; Swain, Colborn, Bason Howarth, Lamey, Palmer and Swackhamer; 1992).

Even though the fabrication, commercialization and use of a large number of hazardous pesticides are prohibited, their export from industrialized countries to developing nations has been reported (Jain, 1992). Moreover, only a very limited number of countries reports to Food and Agricultural Organizations (FAO) the fabrication, sale and usage of those chemicals (Voldner and Yi-Fan Li, 1995). This problem, along with the scarce control on trading and final destination of these products, makes it possible to find high concentrations of these compounds at remote locations (Barrie, Gregor, Hargrave, Lake, Muir et al, 1992; Eisenreich, Looney and Thornton, 1981; Brun, Howell and O´Neill, 1991; Oehme, Vetter and Schlabach, 1994). Aqueous media can be a suitable compartment to study any type of contamination within a specified geographic area because of its ubiquity and peculiar physical-chemical properties (Picó, Moltó, Redondo, Viana, Mañes and Font, 1994; Fingler, Drevenkar, Tkalcevic and Smit, 1992; Miliadis, 1993; Veningerová, Prachar, Uhnák and Kova_i_ová, 1996; Barceló, Chirón, Fernández, Valverde and Alpendurada, 1996).

In the province of San Juan Argentina, the level of pollution due to those pesticides in any natural compartment has not yet been determined. On account of the intense agricultural activity near the rivers that drain the region, the presence of pesticides in high concentrations is likely to occur. The purpose of this work is to determine the concentration of several organochlorine pesticides and their transformed products, in surface and ground water from the most important basins of the Province of San Juan, Argentina. The study was carried out in different seasons in order to generate a representative database, which can improve our knowledge about the contamination status in these water systems. In addition, relationships among aqueous compartments, seasonal variations, and age of contamination in the two principal basins are discussed.

The San Juan and Jáchal are mountain rivers 320 and 80 km long, respectively, that originate in the Andes mountains (Figure 1). The area that both rivers flow through extends from latitude 30°20' to 32°40' South and longitude 6640' to 7035' West. Both river systems share similar snow regimens in their tributaries' draining area, though the snow accumulation rates are greater in the San Juan River system. Located in the southwestern part of the Province, San Juan River starts at the confluence of Los Patos Superior River and Castaño River that drain a large triangle-like area at about 4000 meters above sea level. This area is bounded on the west side by the Continental Divide, the Cordillera de los Andes that is the international frontier between Argentina and Chile. As it flows, the San Juan River passes through a long canyon before entering the Tulum Valley, where its grade is lower than 0.6 % (meters per 100 meters). After flowing out the Tulum Valley, where most of the Province's population sits, it meanders and ends up its course in an endorreic lagoon system namely the Lagunas de Guanacache at about 480-550 m above sea level. Detrital accumulation and deposits produced by glacier attritions from the Andes give rise to a stream carrying a high content of soil sediments and suspended material in the fluid mass. Likewise, the Jáchal River in Northern San Juan, originates at the confluence of the Blanco and Colola rivers. These tributaries carry the snowmelt from the North Cordillera mountain range. The Jáchal River, with similar grade to San Juan's, crosses several mountain ranges and irrigates the small oases of Iglesia, Rodeo and Jáchal.

Figure 1: Sampling locations in the San Juan and Jáchal rivers basins

The river's outflow, not used for irrigation, drains into the Bermejo River system.

Because of the low average annual rainfall (i.e., less than 100 mm), the main water sources for irrigation purposes in San Juan are the San Juan and Jáchal rivers. These rivers are used for irrigating approximately 72,000 Ha. of farming land, mostly vineyards and vegetable fields. The Ullum Dam (3,200 Ha. surface area) on the San Juan River, and the Cuesta del Viento Dam (1,200 ha. surface area) on the Jáchal River, regulate the water discharges. The mean annual flow rates in 1996 and 1997 were 26 m³/s and 42.5 m³/s (San Juan), and 7.7 m³/s and 7.9 m³/s (Jáchal), respectively.

2.2 MATERIALS AND METHODS

2.2.1 Sample collection

A total of 314 water samples, 70 from Jáchal River and 244 from San Juan River, were collected within the areas shown in Figure 1. The most frequently monitored subarea was the Ullum-Tulum oasis. In this subarea, over 90 % of the agriculture activities in the province are carried out. Samples were collected from the river itself, from the man-made irrigation channels and from the drainage devices associated to the irrigation network. Circles shown in Figure 1 represent sampling areas. Deep underground water samples from the Ullum-Tulum oasis were collected at depths ranging from 6 m to 350 m. Water samples from the Ullum Dam were taken at different depths (i.e., 0.50 m, 1 m and 10 m above the dam bottom). Samples were collected during spring-summer of 1996 (September 1996 to January 1997) and fall-winter 1997 (March to July 1997). Sampling was carried out mainly near farms or industrial facilities. Two or three samples were taken from each sampling site in each season. Surface samples were collected at a depth of 20 cm in 1-liter dark-colored flasks fitted with a TFE-lined screw cap, and stored at 4C until analysis, which was normally carried out within 24 to 72 h from sampling.

2.2.2 Reagents

The solvents used were pesticide quality (Merck, Omnisolv, tested for use in gas chromatography and residue analysis). The water was prepared by reverse osmosis and then filtered through activated charcoal. Sodium sulfate and sodium chloride, both ACS grade, were heated for 4 h at 450C to remove interfering organic substances. Florisil 60-80 mesh was activated at 676C and stored in a dark glass container fitted with a glass stopper. Before use, it was activated overnight at 130C. All glassware used in this procedure was heated at 450-500C in a glassblower furnace and was washed with hexane prior to use. Chrompack, 98 to 100 % certified purity, was used as pesticide reference standard.

2.2.3 Apparatus

Separator funnels, 2000 mL, with TFE fluorocarbon stopcock and TFE fluorocarbon stopper, were used for liquid-liquid extraction. Solutions were concentrated in a Kuderna-Danish borosilicate glass device in which the temperature was kept within the ± 2C range by using a water bath. Glass vials, 10 mL capacity, with TFE fluorocarbon-lined screw caps were used to store the concentrated samples.

A Chrompack Mod. CP 9001 Gas chromatograph equipped with Ni⁶³ electron capture detector was used for chromatographic separations. Columns: Chrompack CP Sil 8 CB 50 m, 0.25 mm ID, 0.12 mm DF (column 1); and Chrompack CP Sil 19 CB 50 m, 0.25 mm ID, 0.12 mm DF (column 2). Injection volume 2 mL on column. The operating conditions for column 1 were: Chrompack CP Sil 8 CB on column 140C (2.5 min.), 4C/min­244C (0 min.), 20C/min.­290C (4 min.). Ultra pure nitrogen was used as the carrier gas at a flow rate of 0.85 mL/min. The make up gas was nitrogen (25 mL/min.). Head column pressure was 200 kPa. For column 2 (confirmation) were: Chrompack CP Sil 19 CB on column 80C (1 min.), 35C/min­185C (0 min.), 3C/min.­245C (20 min.). Ultra pure nitrogen was used as the carrier gas at a flow rate of 0.99 mL/min., and the make up gas was nitrogen (25mL/min.). Head column pressure was 200 kPa.

A Mass Spectrometer, Quadrupole Hewlett-Packard HP 5972 with GC and column CP Sil 8 CB 30 m, 0.25 mm ID, 0.12 mm DF, make up gas helium 1mL/min, 37.6 cm/s linear velocity, was used to confirm the chromatographic peaks. Operating conditions for the column in the GC-MS system were: Injector temperature: 250C; Detector temperature: 250C; Program temperature: Initial Oven temperature: 140C (2.50 min.), 4 C/min ­244C (0 min), 20C/min ­290C (4 min).

The analytical procedure applied was the liquid-liquid extraction-gas chromatographic method of the Standard Method for the Examination of Water and Waste Water (American Public Health Association, 1992), using 15 % methylene chloride in n-hexane and capillary columns. After filtration with Wathman 6 F/A glass microfiber filter to remove sand and debris, one-liter samples were extracted with a solvent mixture and then concentrated in a Kuderna-Danish apparatus. Clean up using Florisil column was performed when necessary. Finally, the extracts were dried under nitrogen and were re-dissolved in 1 mL of isooctane.

Table I: Detection limits of organochlorine pesticides

All sample extracts were analyzed in two different gas-chromatographic columns. Only the peaks observed in both columns were evaluated. External Standard Method was used for quantification. Because in column CP Sil 8 CB, Dieldrin and pp'-DDE have the same retention times, their quantification was performed in column CP Sil 19 CB. Component confirmation was achieved through Mass Spectrometry. The recovery of the method for the samples spiked in the concentration range 0.01 to 1.00mg/L was 70-105 %, and the relative standard deviation was below 15 %. Reported concentrations were not corrected for the recovery percentage. Detection limits for each pesticide are given in Table I.

Results obtained from 314 samples collected from September 1996 to July 1997 for both basins are summarized in Tables II and III, and Figures 2 and 3.

Figure 2: Pesticide concentration ranges distribution for surface and ground water (Overall Mean Values of both basins: San Juan and Jáchal rivers).	Figure 3: Comparison between pesticide and their metabolites or degradation products in the San Juan River basin (ground and surface water).

Table II shows mean values of concentration of pesticides in the Province of San Juan-Argentina. It shows the overall (ground water and surface water) values, mean value in positives samples (surface and ground water) and the comparison between surface water of the two main basins in the area.

Table II: Mean Values of concentration of pesticides in the Province of San Juan-Argentina. It shows the overall (ground water and surface water) values -see figure 2- , mean value in positives samples (surface and ground water) and the comparison between surface water of the two main basins in the area.

The values shown in the Table II represent the average for two or three samples taken at each point and season. A significant difference in contaminant concentration can be observed. In the San Juan River, a higher level of contamination was found for all chemical species studied except for b-HCH. However, observing the overall mean values, only b-HCH and Dieldrin were above the internationally acceptable concentrations of 1-3 µg/L for raw superficial waters and for all the pesticides (E. C. Council Directive 1980).

In some of the 66 sampling sites chosen, very high contamination values were found for several pesticides. Those sites were relatively close to waste-water drainage systems of vegetable canning industries.

Table III shows the overall percentage of positive samples, i.e. those samples with concentration levels above the respective detection limits which includes all samples and basins, and positive samples in surface and ground water occurring in the Ullum-Tulum oasis within the San Juan River basin.

Table II (p=0.09) clearly shows that contamination levels observed in the San Juan River are higher than those found in the Jáchal River. This is due to a higher population density, and more intensified agricultural and industrial activities (90% of the total). Figure 2 represents overall values found for surface and ground water in the San Juan and Jáchal river basins.

From Figure 2 and Table III (p<0.001 for both basins), it was apparent that there were differences in contamination levels between surface and ground water for both basins. Ground water was less contaminated than surface water. Exceptionally, high contaminant (heptachlor) concentrations in ground water were detected in shallow drillings and in areas (e.g. Médano de Oro) where the watertable is high. This finding is important because in many places of the Ullum-Tulum oasis the water supply for human consumption originates in ground water, either from free or confined aquifers. In most cases, water is consumed without further treatment, excepting for some cases in which chlorine is used for disinfection purposes. It is worth pointing out that the concentration of several pesticides in ground water was above the maximum limit suggested by the European Community (0.1 µg/L for tap water).

Table III: Overall percentage of positive samples for both basins and percentages for surface and groundwater in the San Juan River basin.

A comparison of concentration for pesticide and their metabolites or degradation products in the San Juan River basin (ground and surface water), is shown in Figure 3. A high concentration of heptachlor is observed as compared with that of its main metabolite, heptachlor epoxide. This suggests recent contamination, mainly from agrochemicals. A different situation is found in the cases of isomers of HCH, where a high proportion of the persistent b-isomer would indicate chronic contamination . A high Dieldrin to Aldrin ratio was observed. This can be partially due to Aldrin degradation. Despite the larger levels of Dieldrin, the levels for Aldrin and g HCH are relatively high as well, thus suggesting a recent contamination from these pesticides.

Although the consumption of pesticides for agricultural, domestic, and sanitary purposes, was higher in spring and summer, a larger river discharge due to an increase of snow melt flow and consequent dilution of the pesticides might explain the reason why the contamination levels found in different seasons were not significantly different (p=0.44).

After analyzing the overall mean value for the concentration of organochlorine pesticide in water sources of San Juan Province, only HCH and Dieldrin were above the maximum permissible value recommended by international organizations (E. C. Council Directive 1980). As expected, due to a higher population density, a larger cultivated area and industrial sites, the San Juan River basin was the most contaminated one. In this basin, the superficial water of the main oasis (Ullum-Tulum) was more polluted that its corresponding groundwater, excepting for some individual cases.

Point contaminations by effluents from canning industries were detected in the Ullum ­ Tulum oasis area.

By comparing the results reported here with other regional studies on water contamination (Luco, Aguilar, Silva, Baudino and González, 1992), similar contamination levels with HCH isomers and lower ones with DDT and their degradation products were found. Both recent and chronic contaminations with Aldrin, a recent contamination from pesticides heptachlor and HCH, and a chronic contamination with HCH, due to a high concentration of the persistent beta isomer, can be suspected by analyzing the pesticides/degradation product relationship.

It can be concluded that, in spite of the prohibition to manufacture, trade, and use these chemicals, their frequent occurrence, as well as that of their degradation products or metabolites in the different compartments of the biosphere, is due to their persistence and to the scarce control on the import, sale and usage in the various agricultural production zones.

The authors would like to thank the CICITCA (National University of San Juan) for the financial support, and Ing. Alberto Salmuni and Ing. Lucía Petkovic for their suggestions and valuable help with the English version.

1- Waliszewski, S.M., Pardio, V.T.S., Chantiri, J.N.P., Infanzón, R.M.R. and Rivera, H., 1996. Organochlorine pesticide residues in adipose tissue of Mexicans. The Science of the Total Environment. 181, 125-131.         [ Links ]

2- Beretta, M. and Dick, T., 1994. Organochlorine Compounds in Human Milk, Porto Alegre, Brazil. Bull. Environm. Contam. Toxicol. 53, 375-360.         [ Links ]

3- Baudino, O, Pestchanker, M.J. and Aguilar E., 1985. Niveles de pesticidas organoclorados en leche materna en la Provincia de San Luis. Rev. Anal. Asoc. Quim. Arg. 73(4), 421-432.         [ Links ]

4- Muir, D.C.G., Ford, C.A., Grift, N. P., Metner, D. A. and Lockhart, W.L., 1990. Geographic variation of chlorinated hydrocarbons in Burbot (Lota Lota) from remote lakes and rivers in Canada, Arch. Environ. Contam. Toxicol. 19, 530-542.         [ Links ]

5- Swain, W., Colborn, T., Bason, C., Howarth, R., Lamey, L., Palmer, B. and Swackhamer, D., 1992. Exposure and effects of airborne contamination. Great Waters Program Report, U.S. Environmental Protection Agency, Washington, D.C..         [ Links ]

6- Jain, V. , 1992. Disposing of pesticides in the third world. Environ. Sci. Technol., 26: 226-228         [ Links ]

7- Voldner, E.C. and Yi-Fan Li, 1995. Global usage of selected persistent organochlorines. The Science of the Total Environmental. 160/161, 201-210.         [ Links ]

8- Barrie, L.A., Gregor, D., Hargrave, B., Lake, R., Muir, D., Shearer, R., Tracy, B. and Bidleman, T., 1992. Arctic contaminants: source occurrence and pathways. Sci. Total Environ., 122, 1-74.         [ Links ]

9- Eisenreich, S.J., Looney, B.B. and Thornton, J.D.. 1981. Airborne organic contaminants in the Great Lakes ecosystem. Environ. Sci. Technol., 15,30.         [ Links ]

10- Brun, G.L., Howell, G.D. and O´Neill, H.J., 1991. Spatial and temporal patterns of organic contaminants in wet precipitation in Atlantic Canada. Environ. Sci. Technol., 25, 1249.         [ Links ]

11- Oehme, M., Vetter, W. and Schlabach, M., 1994. Quantitative analysis of persistent organochlorines in samples from polar regions. Intern. J. Environm. Anal. Chem. Vol. 55, 261-266.         [ Links ]

12- Picó, Y., Moltó, J. C., Redondo, M. J., Viana, E., Mañes, J. and Font, G., 1994. Monitoring of the pesticide levels in natural waters of the Valencia community (Spain). Bull. Envir. Contam. Toxicol. 53, 230-237.         [ Links ]

13- Fingler, S., Drevenkar, D., Tkalcevic, B. and Smit, Z., 1992. Levels of polychlorinated biphenils, organochlorine pesticides and chlorophenols in the Kupa river water and in drinking waters from different areas in Croatia. Bull. Environ. Contam. Toxicol. 49, 805- 812.         [ Links ]

14- Miliadis, G. M., 1993. Gas chromatography determination of pesticides in natural waters of Greece. Bull. Environ. Contam. Toxicol. 50, 247-252.         [ Links ]

15- Veningerová, M., Prachar, V., Uhnák, J. and Kova_i_ová, J., 1996. Levels of chlorinated pesticides in water of the Danuve River. Fresenius Envir. Bull. 5, 361-368.         [ Links ]

16- Barceló, D., Chirón, S., Fernández Alba, A., Valverde, A. and Alpendurada, M.F., 1996. Monitoring pesticides and metabolites in surface water and groundwater in Spain. ACS Symposium Serie N630. Herbicide metabolites in surface water and groundwater. Chap. 18. M. T. Meyer and E. M. Thurman, editors. American Chemical Society.         [ Links ]

17- Luco, J.M., Aguilar, E.G., Silva, P., Baudino, O.M., Gonzalez, D. M., 1992. Niveles de pesticidas organoclorados en aguas de la provincia de San Luis República Argentina. Acta Farm. Bonaerense 11(3), 121-8.         [ Links ]

18- American Public Health Association, 1992. Standard methods for the examination of water and wastewater. 17 ed. APHA, AWWA, WEF. Washington DC.         [ Links ]