THOMAS and OWEN : ESTIMATING PHYTOPLANKTON PRODUCTION 



ammonium and where nitrate was undetectable. 

 One hundred and five such production stations 

 were available from 10 EASTROPAC cruises 

 in this nutrient-poor water. 



Production calculated from equation 3 is com- 

 pared with measured ^*C production in Figure 

 1. There is a highly significant (P<.01) rela- 

 tionship between the two sets of values. The 

 slope of the regression line is 1.057, which is 

 very near to the value 1.0 which would be ex- 

 pected if agreement were perfect. Nevertheless, 

 there is a large amount of scatter in the values 

 of Figure 1; that is, the calculation overesti- 

 mates in some cases and underestimates in 

 others. 



Table 2. — Errors in the calculation of production. 



Errors in the values used in the calculation 

 are given in Table 2. To figure total error these 

 have been converted to variances and summed. 

 The 95% confidence limit shows that any cal- 

 culated production value can vary by ± 1.5 fold. 

 Thus, one would expect quite a large scatter in 

 Figure 1. 



Most of the error is in Ks. When only the 

 Ks values of Thomas (1970b) are used the cal- 

 culation generally underestimates the observed 

 •''*C production. Use of the mean of the Ks 

 values of Maclsaac and Dugdale (1969) results 

 in an overestimation. Since there is no reason 

 to doubt either set of Kg values, we have used 

 the overall mean Kg from Table 1. In applying 

 this method to any other nutrient-limited waters, 

 it would be well to obtain several values of Kg 

 so that the error due to lack of precision in 

 measuring Kg can be recognized. 



Part of the scatter in Figure 1 may also be 

 due to the fact that the parameter Kg is species 

 — and temperature — dependent (Eppley, Ro- 

 gers, and McCarthy, 1969) and that variations 

 in species composition of the crop or slight var- 



iations in temperature may have affected the 

 calculation. The parameters ^max and R are 

 also probably dependent upon the species com- 

 position of the crop and on temperature. Be- 

 cause of these factors, which are unknown, it 

 is perhaps surprising that the relationship be- 

 tween calculated and observed production is so 

 good when constant values of Kg, ^ma.x.and R are 

 used. 



This evidence supports the hypothesis that 

 phytoplankton production in the upper mixed 

 layer is controlled by the limiting nutrient, 

 ammonium, and shows that the hyperbolic model 

 describes this control very well. In this latter 

 connection it should be noted that if a linear 

 model having a term "S/Smax" in equation 3 

 (where Smax is that concentration supporting 

 a maximum growth rate and which has a value 

 near 10.0 ^M from the data of Thomas, 1970b) 

 is used rather than the term "S/ (Kg + S)," 

 the calculation very much underestimates the 

 ^^C production. The linear model was used 

 previously by Riley (1963) and Steele (1958) 

 but should now be considered obsolete in view 

 of more recent work using the hyperbolic model. 



ACKNOWLEDGMENTS 



We appreciate the assistance of many persons 

 in gathering these data. Ammonium analyses 

 were performed by Mr. Edward Renger, and 

 Mrs. Anne Dodson aided in the determination 

 of /-max and Kg. Sampling and incubation for 

 production measurements and determination of 

 chlorophyll concentrations were carried out by 

 the following: Messrs. Tapuni Mulitauaopele, 

 Michael Kruse, David Justice, James McCarthy, 

 Lawrence Klapow, David Judkins, Gerald John- 

 son, Eric Forsbergh, and Jack Metoyer. Dr. 

 Bernt Zeitzschel and Mr. Michael Kruse helped 

 to process and edit the ^'*C and chlorophyll data. 

 Most of these data were collected aboard the 

 NMFS vessel David Starr Jordan and we appre- 

 ciate the assistance of Capt. C. W. Forster and 

 his crew. 



91 



