COLLINS and TENNEY: FISHERY WASTE EFFLUENTS 



salmon cannery waste effluent were taken from 

 the underflow of a Bauer screen (0.03-inch) (Table 

 4). The average values by analysis and calculation 

 are as follows: 



The NFR is 50% of the TR, but the COD of the 

 NFR is 62% of the total COD. 



DISCUSSION 



The following discussion is concerned with 

 monitoring parameters previously suggested or 

 currently in effect under EPA effluent limitations 

 for seafood processing and with the suggestion of 

 a more precise and simpler monitoring system. 

 The present EPA requirements, however, for use 

 of alternative analytical methods must be con- 

 sidered. Under EPA rules (Title 40 "Code of 

 Federal Regulations" Parts 136.4 and 136.5), any 

 person wishing to use alternative analytical 

 methods for the parameters listed must follow 

 variance procedures specified under the NPDES 

 permit system. 



Current permits require monitoring for SS, 

 COD (i.e., CODtr), TSS (i.e., NFR), O&G, flow, and 

 pH. SS is imprecise and contributes so little to the 

 pollution load in seafood processing that it has 

 relatively little value as a measure of pollution, 

 although it has merit as a check on the efficiency of 

 screen operation. As discussed later, total COD can 

 be determined more accurately in an indirect 

 manner. The O&G analysis is difficult to do, and 

 this value, too, can be obtained more accurately 

 through calculation. Data in this paper suggest 

 that the indirect analysis for NFR (i.e., TSS) was 

 more accurate than the direct method. The FR is 

 an important parameter because this fraction 

 contributed about 50% to the total COD or TR and 

 will need to be considered in the design of future 

 treatment systems. 



To develop an improved monitoring system, we 

 plotted the COD and residue data of Table 1 to 

 illustrate the correlation between the COD of the 



residue and the concentration of the residue 

 (Figure 1). The regression lines and equations 

 were determined by the method of least squares. 

 The TR and FR regression lines were obtained 

 through direct analyses, and the NFR line was 

 obtained by difference. The maximum deviation of 

 any COD value from the regression line was 260 

 mg/liter. This is slightly less than the possible 

 error of the analytical method ( ± 8%) (Moore et al. 

 1949). On the average, the individual values were 

 within 107 mg COD of the regression line. 



The correlations shown in Figure 1 can be used 

 to calculate COD and residue values. In the fol- 

 lowing, the first three equations are the regression 

 lines of Figure 1 and the next three are derived 

 equations to solve for residue rather than for COD. 

 Of course, these equations are valid only for this 

 group of data and for this particular plant. If the 

 TR and CODpK are determined by analysis, the 

 other values can be derived from the equations or 

 the regression line and from the expression 

 TR = FR -h NFR. 



CODtr 

 CODpR 



CODkpk 



1.32 TR + 113 

 1.08 FR - 96 

 1.78 NFR + 210 



(1) 

 (2) 

 (3) 



5.000 



4,000 - 



1000 



2,000 



1,000 



1000 



2000 3,000 



RESIDUE (mg/l) 



4,000 



Figure l.-Relationship between the COD of the residue and the 

 concentration of the residue from shrimp processed on Model A 

 peelers using fresh water. 



729 



