E-3 

 stations. 



A general practice in offshore radio-positioning is 

 for the survey vessel to return to an exactly determined 

 location (dock, buoy, tower) once the lane count is lost. 

 We found that we could circumvent this time-consvuning prac- 

 tice by combining Loran-C measurements with satellite mea- 

 surements. We were able to obtain up to 100 simultaneous 

 Loran-C and satellite fixes during each canyon survey. 

 Propagation corrections for Loran-C were calculated so as to 

 place the Loran-C positions exactly in the center of the 

 field of scatter of a plot of the highest quality satellite 

 fixes, edited for elevation, number of doppler counts, and 

 reasonableness of computer heading and speed. We found that 

 this satellite Loran-C calibration could be used to find 

 the whole lane count of the ARGO system. The lane count 

 procedure involved a statistical matching of approximately 

 1000 simultaneous Loran-C and ARGO fixes (at 5 minute spa- 

 cings), which showed that the Loran-C standard deviation 

 from ARGO was equal to 0.64 of an ARGO lane. 



A final confirmation of the correctness of the whole 

 lane count was made by comparing the initial first-day cali- 

 bration of ARGO at the Chesapeake Light Tower with a last 

 day fix at the U.S. Coast Guard dock at Cape May, New Jersey. 

 The through-cruise tie turned out to be consistent within 

 the integrity of a single lane, and the fractional lane 

 difference amounted to less than 4 meters (most likely the 

 accuracy of the initial calibration while circling the 

 Chesapeake Light Tower at a safe distance of several boat 



