FISHERY BULLETIN: VOL. 74, NO. 4 



TR = 0.76 CODtr - 86 



FR = 0.93 CODpR - 89 



NFR = 0.56 CODj,FR - 118 



(4) 

 (5) 

 (6) 



When salt water was used in processing, such as 

 in the second plant study (Table 2), the residue 

 values included salt. Since salt values were not 

 determined, COD and residue data were not cor- 

 related for this plant. 



In the third plant study of snow crab effluent, 

 the data (Table 3) were plotted similarly to the 

 shrimp data (Figure 2). The basic equations for 

 snow crab can also be used to calculate from two 

 analyses the other COD or residue values. The 

 equations are listed in Figure 2. 



Data for the fourth plant study of salmon-waste 

 effluents (Table 4) were also plotted, and the 

 regression lines and equations were similarly 

 determined (Figure 3). The regression lines for 

 salmon are less precise because of the variable salt 

 content of the effluent and the high levels of COD 

 and residue. Salt varied because of the erratic 

 operation of the salmon egg-processing room. 

 These regression lines (salmon) should not be used 

 to calculate or interpolate COD or residue values 

 unless a check is first made on salt content. If salt 

 content of the effluent is about normal (500 

 mg/liter), the calculation is valid since these 

 equations are derived from data with a high 

 standard deviation for salt. A check is made to 

 ensure that the level is not 1 or 2% as it could be if a 

 brine tank were dumped. A routine composite 



3,000 



\ 2,000 

 01 

 E 



o 

 O 



2.65NFR - 1 26 



1,000 2,000 



RESIDUE (mg/l) 



Figure 2.-Relationship between the COD of the residue and the 

 concentration of residue from the processing of snow crab meats 

 and sections. 



730 



9,000 ' 



epoo- 



zooo 



<lOOO 



\ 



-i 5,000 



O 



o 



<J 4.000 



JflOO- 



2.000 



1,000 • 



coo,. = 1.30TR + 4 56 



COD^„= 1.59NFR+ 330 



1000 2,000 3,000 4.000 5,000 40OO 7000 

 RESIDUE (mg/ l) 



Figure 3.-Relationship between the COD of the residue and the 

 concentration of the residue for canned salmon processing. 



sampling program for the plant, of course, would 

 reduce salt variation. 



A SIMPLIFIED MONITORING SYSTEM 



The data of the first plant study (Table 1) and 

 the six equations listed earlier may be used to 

 illustrate how a simplified monitoring system can 

 be set up for a particular plant. 



Since COD is difficult to determine on the 

 original effluent (particulate matter causes dilu- 

 tion problems) and impractical to determine on a 

 solid sample, COD should be determined on the 

 filterable residue sample before drying. Equation 

 (5) is then used to calculate FR in milligrams per 

 liter. It is not necessary to actually finish the FR 

 test. The next analysis most logically should be the 

 total residue test. It is an easy test to do and is 

 accurate. Equation (1) is used to calculate the COD 

 of the TR, and the previously calculated FR is 

 subtracted from TR to give the NFR in milligrams 

 per liter. Equation (3) is then used to calculate the 

 COD of the NFR. Thus, two analyses plus several 

 calculations give three COD and three residue 

 values. 



The two analyses recommended (CODpp and 

 TR) are logically the most accurate of the six 



