TABLE 2 



Statistical analysis for linear regressions of plots in Figs. 5 and 6. 



Variable abbreviations: Thy (thymidine incorporation), Bac 



(bacteria), T (temperature), and Spa (specific activity of thymidine 



incorporation) are the parameters for the regression analysis. 



Statistics Abbreviations: F-value (value for data variance), r- 



(correlation coefficient), C.V. (coefficient of variation). Intercept 



(y-intercept). Slope (slope of the best linear regression), T-test (test 



of null hypothesis (Ho), Slope >0). 



Anadyr Waters 



Regres. F value r- C.V. Intercept Slope T-test 



Alaskan Coastal Waters 



Reeres. 



F value 



C.V. Intercept Slope T-test 



Bering Shelf Waters 



Regres 



F value 



C.V. Intercept Slope T-test 



lower than Anadyr and ACW" s (T-test. P < 0.05 ). The specific 

 activity again was unrelated to frequency of dividing cells 

 (T-test, P> 0.05). 



Bering Sea water : The highest rates of thymidine 

 incorporation and numbers of bacterioplankton were measured 

 in the surface mixed layer of the south Bering Sea (Fig. 6). 

 Incorporation rates ranged from 5 pmole 1 ' h ' above the 

 thermocline to less than 1 pmole 1' h ' below the thermodine 

 and less than 0.2 pmoles 1 ' h ' below 100 m. Rates were 

 strongly correlated with temperature and specific activity 

 (Fig. 2, Table 2). However, this strong relation could be 

 attributed to any number of factors (e.g., phytoplankton 

 productivity or biomass in the upper mi.\ed layer). 



Surface mixed layer processed the highest number of 

 bacterioplankton (1 x 10" cells 1 '), and numbers decreased to 

 less than l-2x 10'*cellsl' below l()()m(Figs. 5a,b). Likewise, 

 bacteria correlated strongly with temperature and specific 

 activity (Fig. 3, Table 2). 



Surface waters also had the highest frequency of dividing 

 cells (4 to 7% dividing), but frequency of dividing cells 

 decreased with depth to less than 2% dividing below 500 m 

 (Fig. 6). At Station 1 10, bacterioplankton showed a secondary 

 peak in dividing cells at 1 ,500 m, and below 2,000 m dividing 

 cells increased with depth to nearly 12% dividing cells. Specific 

 activity showed a similar distribution to frequency of dividing 

 cells in the water column. In the mixed layer, specific rates 

 averaged 2.5 x 10-' mole cell ' h ' (Table 1) and decreased to 

 0.1 1 X 10 -' mole cell ' h ' below 500 m. At Station 1 10, both 

 specific activity and the frequency of dividing cells increased 

 with depth below 1,500-2,000 m. At Station 113, specific 

 activity, but not frequency of dividing cells, increased with 

 depth below 1,500 m (the old GEOSECS Station). 



Spatial Distribution of Bacterioplankton 



In the Chirikov basin, bacterioplankton dynamics showed 

 considerable variability through the water column and across 

 the water types. The regional depth distribution of 

 bacterioplankton parameters are summarized in Table 1 . In 

 nutrient-rich AW, the highest rates of thymidine incorporation 

 occurred in the surface and near bottom waters. 

 Bacterioplankton, frequency of dividing cells, and specific 

 activity, generally covaried with the rates of thymidine 

 incorporation, even though the water column was isothermal. 



In nutrient-poor ACW, the thermocline was generally a 

 dynamic region in the water column for bacterioplankton 

 activity. Bacterioplankton, thymidine incorporation, frequency 

 of dividing cells, and specific activity peaked within the region 

 of the thermocline at water depths of 10-20 m, whereas in 

 BSHW, the upper mixed layer contained highest 

 bacterioplankton activity. 



Like BSHW, the highest bacterioplankton activity occurred 

 in the upper mixed layer in the deep waters of the south Bering 

 Sea. Below the thermocline, bacterioplankton uptake of 

 thymidine diminished greatly even though measures of 

 population growth rate increased with depth in bottom waters. 

 In these deep waters, bacterioplankton populations were an 

 order of magnitude or two lower than upper mixed layer. 



Comparison to other Marine Ecosystems 



Thymidine incorporation data reported here for Chirikov 

 basin and south Bering Sea (0.0 to 4.7 pmoles 1' h') fell within 

 the range of values reported for other coastal-shelf waters and 

 adjacent and marginal seas in both high and low latitudes of the 

 Northern and Southern Hemispheres. In polar waters of 

 McMurdo Sound and the ice edge zone of the Ross Sea, 

 Antarctica, where temperatures range from -1.8 to 5°C year 

 round, Fuhrman and Azam (1980) found similar rates of 

 thymidine incorporation of 0.2 to 1 1 .3 pmoles 1 ' h"' (calculated 

 from values in Table 1, Fuhrman & Azam, 1980). 



Within the Antarctic Polar Front (2.5°C) of the Drake 

 Passage, rates were also on the order of 0.1 to 

 10 pmoles 1 ' h ' over the upper mixed layer, but within the 

 productive marginal ice edge zone (-1 to 2°C) off the Palmer 

 Peninsula, rates as high as 200 pmoles 1' h ' were measured 

 (Hanson & Lowery, 1983). In northern latitudes off Nova 

 Scotia, Canada, Douglas et al. (1987) found thymidine 



72 



