as one meter from the surface to the bottom of 

 the hydrographic cast. This quality controlled 

 data set forms the input to the Data Processing 

 Program which prepares the data for use by the 

 Labrador Current Model. 



Input data must be in the form of temperature 

 and salinity vertical averages for each of the six 

 model layers. In addition, these averaged values 

 must also be located at each grid point of the 

 model matrix (2025 grid points). The Data 

 Processing Program takes the sampled data, 

 positioned by latitude and longitude, calculates 

 averages for the six layers at the sampled sta- 

 tions, locates these averages in the matrix grid 

 system and proceeds to scale the data at all grid 

 points based solely on those locations sampled 

 and the bottom bathmetry. The program is con- 

 structed so that if only two sampled data points 

 existed in any layer, a complete data field would 

 be generated based on those two points. Ob- 

 viously, the more sampled data, the more ac- 

 curate will be the scaled data field. The scaling 

 program uses a system developed by several 

 Coast Guard Academy cadets and myself while 

 working on 1975 model tests. This routine 

 iterates for a maximum of 150 cycles or until the 

 temperature and/or salinity data ceases to change 

 by more than .01. The number of iterations used 

 is printed out for both temperature and salinity 

 and for each layer for quality control. 



When the sampled data are initially entered 

 into the Data Processing Program, the location 

 of the area to be modeled is also entered. The 

 sampled data locations are checked and are dis- 

 carded if they lie more than 5 km outside the 

 desired model area. The program gives as a 

 printout; model location, a list of stations used 

 showing their number, grid location, and latitude 

 and longitude. Following this output is the 

 bathymetric data for the area modeled. The next 

 series of outputs is supplied sequentially for each 

 layer: the base matrix indicating the intersection 

 of the continental shelf or the location of open 

 boundaries; location and value of the sampled 

 temperature averaged over the layer followed by 

 a similar one for salinity; the number of itera- 

 tions used to complete the scaling process; the 

 properly formated data for model input (first 



temperature then salinity) ; and lastly a complete 

 temperature and salinity data matrix for that 

 layer which can be quickly scanned or contoured 

 for quality control. The program is presently 

 being run on a UNIVAC 1108 at the Underwater 

 Systems Center, Xew London, CT. It requires 

 less than 50,000 words of core memory and can 

 be run in less than 15 minutes of CPU time. 



A complete system for determination and pre- 

 diction of the Labrador Current is envisioned as 

 a future goal. This would include the air de- 

 ployable expendable conductivity-temperature- 

 depth probes under development by the Navj\ 

 These instruments could supply sufficient data 

 from the area to be modeled in a timely manner. 

 A peripheral quality control program could 

 ready this raw data for input into the primary 

 Data Processing Program, which could have 

 available to it as a data bank, the complete 

 bathymetry of the Grand Banks region and upon 

 command select the proper bathymetric input for 

 the desired model location. This data program 

 would then produce a complete data set for input 

 into the main model program. At this same 

 time, present and predicted winds would be 

 entered along with satellite information concern- 

 ing the slope of the sea surface. The model 

 could then produce predictions of the absolute 

 current system which would be valid for up to 

 10 days, i.e., flights updating data of the bound- 

 aries where the Labrador and the Gulf Stream 

 enter the model w T ould be required for extended 

 predictions. These would consist of short flights 

 using the air deployable data probes to check on 

 the location, salinity anil temperatures of the 

 major currents entering the modeled area. A 

 summary of this IIP Labrador Current Determi- 

 nation System is shown in Figure B-3. The 

 shaded area reflects the work which has already 

 been accomplished. Completion of the other 

 parts of the system must wait for the technology 

 to develop. However, the bathymetric data bank 

 for the entire region could be prepared at this 

 time to facilitate present use of the model as a 

 substitute for the Dynamic Height method used 

 by IIP. 



Further model development possibilities have 

 emerged in the form of vorticity modeling work 



B-4 





