hand. All contoured sections were reviewed and 

 smoothed by hand before final drafting. 



The Rockaway data involved an STD lowering and a 

 Nansen bottle cast on each station. Separate numbers 

 were assigned the lowerings and casts on the same 

 stations, so the oxygen and phosphate sections bear 

 different station numbers, but the positions are the 

 same. 



LITERATURE CITED 



SCARBOROUGH, J B 



1958 Numerical mathematical analysis 4th ed Johns Hopkins 

 Press, Baltimore, 576 p 



STRICKLAND, J D H . and T R PARSONS 



1965. A manual of sea water analysis (with special reference to 

 the more common micronutnents and to particulate organic 

 material). 2nd rev. Fish. Res. Board Can., Bull 125, 203 p. 



1968 A practical handbook of seawater analysis. Fish, Res 

 Board Can , Bull 167, 311 p 



TASHIRO, J E., and J F HEBARD 



1969 A modification of the plankton volume gauge for use 

 aboard ship Limnol Oceanogr 14 794-796. 



VAN LANDINGHAM. J W 



1958. Stabilizing heteropoly color in the estimation of 

 phosphate-phosphorus in sea water Abstract reprinted from 

 Proceedings of the Hawaiian Academy of Science thirty- 

 third annual meeting, 1957-1958, p 22 



APPENDIX 1 



Station Data Compute Program Equations 

 Standard Depth Interpolation Routine 



The following 3-point Lagrange interpolation equa- 

 tion (Scarborough, 1958:74-75) is used to interpolate 

 temperature, salinity, and oxygen at standard depths: 



(d-dz) (d-d 3 ) 

 (dy-dz) (d,-d 3 ) 



(d-d,) (d-d 3 ) 

 (dz-d,) (dz-ds) 



+ 



where 



(d-d,) (d-d 2) 

 (d 3 -d,) (ds-dz) 



X 3 



d =standard depth in question 

 Cm =observed depth #1 

 dz = observed depth #2 

 d3 =observed depth #3 

 X =parameter at standard depth 

 Xi =observed parameter at depth di 

 Xz =observed parameter at depth dz 

 X3 =observed parameter at depth d3. 



However, there is one case where linear interpolation 



X =Xi + 



d-d, 

 dz-d. 



(xz-x\) 



is used. That is where the 3-point interpolated value 

 does not lie within the interval between the valid ob- 

 served values immediately above and below: if there 



are additional interpolated values with in this interval, a 

 new value is also computed (by linear interpolation) for 

 these standard depths. 



The observed values used to interpolate at a given 

 standard depth are normally chosen as one above the 

 required standard depth and two below. However, if 

 there are standard depths between the last two ob- 

 served depths for a station then the values are chosen 

 as two above and one below for the required standard 

 depths. No doubtful data are used for interpolation. 

 Doubtful data are bypassed until valid data are en- 

 countered. 



If there is no valid observation at the surface, the 

 surface interpolated value is made equal to the ob- 

 served value at di, if di=£9 m. If di>9 m. then no 

 standard depth interpolations are computed until 

 dsdi. 



In addition, no interpolation for temperature, salin- 

 ity, and/or oxygen is performed where the following 

 conditions are encountered: 



1) The first depth (di) with a valid observed value 

 (card type 3, 4) is 400 m or less and the next depth 

 (dz) with a valid observed value is greater than d, + 

 200 m. 



2) di exceeds 400 m but not greater than 1,200 m 

 and dz is greater than di + 400 m. 



When di exceeds 1,200 m, interpolation proceeds 

 according to normal procedure. 



