TABLE 11 



Comparison of pesticides in microlayer and surface water. 

 Station 3: 57°56'N, 175°04"W. 



ng/1 



Sample volume 3.1-3.5 I layer thickness = I20|im. 

 Sample volume 9.5 1, depth = 2 m. 



in the Beaufort Sea using experimentally determined Henry's 

 law constants (Equations 2,3) resulted in slight changes in the 

 estimated SI to 7 1 % and 28% for a-HCH and y-HCH. Thus in 

 the Bering. Chukchi, and Beaufort Seas the 

 SKa-HCH )>S1(Y-HCH). 



The Sr s of a a-HCH and y-HCH at Nome Station (64°29'N, 

 165°24'W)were 155%and68%. These Si's are greater than at 

 other stations, and for a-HCH indicate a sea to air flux. This 

 station was very shallow (15 m) and warm (I8°C), and the 

 salinity was low (25 parts-per-thousand) due to Yukon River 

 input into Norton Sound. We could find no report of HCH's in 

 the Yukon River, but average concentrations of a-HCH and 

 y-HCH in five rivers draining the Hudson Bay lowlands were 

 6.4 ng 1 ' and 0.9 ng 1 ' (McCrea & Fischer, 1986). These are 

 above average HCH levels in the open Bering Sea. Influx of 

 HCH's via cold Yukon River water and advection of cold 

 Bering Sea water into Norton Sound followed by solar heating 

 could lead to higher Si's, and in the case of a-HCH, to a 

 reversal of the flux direction. 



An explanation for the lower SI of y-HCH at all stations is 

 unknown, but our results suggest a more rapid disappearance 

 of y-HCH than a-HCH in the upper water column. The second- 

 order base hydrolysis of lindane follows the equations (Ellington 

 etai, 1987): 



and 



dC/dt = -kJOH] (7) 



In k, (L mol ' min ' ) = -8895/T -i- 30.46 . (8) 



A half-life of over 1,600 days was calculated for lindane 

 at the average temperature found in the Bering Sea (5°C) and 

 pH 8. The application of this freshwater rate constant to 

 seawater is uncertain; however, hydrolysis alone probably 

 cannot explain the deficiency of y-HCH in surface water. 



More rapid breakdown of the HCH's appears to occur by 

 photolysis. Saleht'/ a/. (1982) determined first-order photolysis 

 rate constants for y-HCH in purified water (Milli-Q) and three 

 fresh waters in the pH range 7.3-9.2. Adjusted midwinter half- 

 lives were given as 65 d in Milli-Q water and 14-150 d in the 

 natural waters. MalyandieM/. ( 1982) reported a 48-d half-life 

 for y-HCH in distilled water with 5-25 mg/1 added fulvic acids. 

 Isomerization of a small percentage of y-HCH to a-HCH 

 occurred after 15-35 d irradiation. In distilled water alone, 

 y-HCH degraded slightly more rapidly than a-HCH. The 

 relative photolytic stability of the two HCH isomers under 

 arctic conditions is unknown. 



Fluxes of HCH's were calculated by Equation 5 for all 

 stations with concurrent air and surface water concentrations. 

 It is important to note that these fluxes were estimated from kj 

 values calculated from wind speed (Equation 6) and were not 

 directly measured. How closely these estimates represent the 

 actual situation is unknown. Despite model predictions, Peng 

 et al. ( 1979) found no correlation between the on-station wind 

 speed and the flux of radon between the atmosphere and 

 surface water in the open ocean. 



Exchange rates in the Bering and Chukchi Seas ranged 

 from -99 pg cm- yr ' (air to sea) at Station 7 (60°28'N, 

 177°50'W) to -1681 pg cm- yr ' at Station 45 (67°44'N, 

 172°50'W) (Figs. 2-5). Station 45 was different from others in 

 this region in having sea ice and a low salinity of 24 parts-per- 

 thousand from melting ice. Average fluxes were 

 -290pgcm-yr ' in the Bering Sea and -SlOpgcm'yr ' in the 

 Chukchi Sea. More dynamic fluxes were estimated from 

 St. Lawrence Island to the Bering Strait, ranging from 

 1 ,620 pg cm- yr -' sea to air flux at Nome (64°29'N, 165°24'W) 

 to -4.800 pg cm - yr ' air to sea flux at Station 100 (64°23'N, 

 169°09'W) (average = -1,290 pg cm - yr '). This is a shallow 

 area (40-50 m) with a complicated geometry and a relatively 

 high flow of 1.2 Sv (Coachman & Aagaard, 1988) producing 

 a high degree of vertical mixing that may explain the variability 

 of flux. 



Average fluxes over the entire cruise were 

 -880 pg cm - yr ' for a-HCH and -965 pg cm - yr ' for y-HCH. 

 Atlas's (1988) "best" estimate of HCH fluxes for the North 

 Atlantic and North Pacific Oceans were -466 to 

 -715 pg cm - yr ' for a-HCH and - 100 to -240 pg cm - yr ' for 

 y-HCH. Atlas arrived at these fluxes using literature values of 

 air and surface water HCH concentrations, Henry's law 

 constants estimated as functions of temperature, and 

 k,=7.5x 10 'ms'. Atlas assumed an SI of 90% for both HCH 

 isomers. Our higher estimate of the y-HCH flux results from 

 the considerably greater undersaturation of this isomer. 



This work was done with support of the US Department of the 

 Interior. Fish and Wildlife Service; the US Environmental Protection 

 Agency. Great Lakes Program Office. Grant No. R0005027-()l; and 

 the USSR Academy of Sciences in accordance with Activity 02.07- 

 2101 "Comprehensive Analysis of Marine Ecosystems and Ecological 

 Problems of the World Ocean" under Project 02.07-21 of the US- 

 USSR Environmental Agreement. We wish to thank Joseph Kovelich 

 of the Canadian Atmospheric Environmental Service for doing the 

 five-day back air trajectories shown in Fig. 1 . Contribution No. 862 

 of the Belle W. Banich Institute. 



278 



