Occasionally the two radar beacon ranges will produce an imaginary- 

 position. When this occurs, one of the coordinates of the radar fix is 

 defined by the square root of a negative number. When such an imaginary 

 fix occurs, the line of data producing the imaginary fix is eliminated 

 from the file being processed. 



e. The north and east coordinate arrays are searched for ex- 

 treme values by Chauvenet's criterion and any data set containing extreme 

 values is eliminated. Chauvenet's criterion is based on the normal dis- 

 tribution, and a value is rejected if the probability of occurrence of 

 such a deviation from the mean of the n_ measurements is less than 1/2 n 

 (P. R. Rider, 1933) . 



f. Plot the track of the survey vessel by a call to subroutine 

 BENLH2 . 



g. Compute the position of the mobile underwater detector for 

 each fix by correcting for the distance the detector is towed astern of 

 the survey vessel. Cable length, mean water depth, and length of survey 

 vessel, are all input on the program control card. Assuming a position 

 for the detector vehicle at the beginning of the survey, the position of 

 the detector for each fix is computed by linear interpolation from the 

 present position of the vessel to the last interpolated position of the 

 detector. 



YD„ = YV r 



XD n * XV n 



CABLE (YV n - YD^) 



/ (YVn-YD^i) 2 + (XV n - XD^) 2 



CABLE (YV n - W n -i) 

 / (YV n - YD n _!)2 - (XV n - XD n . 1 )2 



where XD and YD are the coordinates of the detector and XV and YV are 

 coordinates of the vessel. CABLE is the horizontal distance from the 

 detector to the survey vessel. 



h. Radiation data are converted to counts per second, and the 

 mean and standard deviation of the background count rate is computed. 

 An estimated background count rate is entered on the data control card 

 for each channel. This estimated background count rate is used to 

 compute the extreme values of the range of the background radiation 

 level, again by means of Chauvenet's criterion. All counts between 

 these extreme ranges are averaged to obtain the mean background count 

 rate for each channel. 



i. Plot uncorrected radiation values by a call to BENLH2 

 through RDPLT2 entry. 



