u 



apparent changes in calibration. A report on this instrument has appeared in 

 Journal of Marine Research by Anderson (1951). 



The 16-channel temperature recorder consists essentially of 16 TPR's 

 feeding into a single temperature recorder. It requires 16 cables and 16 beads 

 which can be spaced at any desired depth or horizontal distances up to 200 feet. 

 The Brown recorder used has one indicator that automatically shifts from one 

 channel to the next, printing an indication of the temperature for one bead after 

 another. This instrument was required for the study of thermal-pockets and 

 temperature variability in the Arctic and the central pacific. 



The CTD was developed at NEL for the study of simultaneous tempera- 

 ture and salinity changes in regions of convergence and during advection proc- 

 esses in the Arctic. The temperature-sensing element in this instrument con- 

 sisted of a group of ten 14B type thermistors. The response and accuracy of 

 the temperature element was surprisingly good. However, the speed of the 

 servo mechanism was a little slow; the speed has since been increased by the 

 Oceanographic Department of the University of Wasington. 



The problem of measuring temperatures in ice in the study of ice forma- 

 tion was also solved by thermistor beads mounted in plastic with leads extending 

 through the ice to a junction box. A portable, heated power supply and galvano- 

 meter were taken to the site to make readings for temperature. 



The list of recorders which are needed seems to have been adequately 

 covered by Vine; however, there are a couple of points about uniformity and 

 availability of instruments which should be emphasized. If various organiza- 

 tions could point out their needs for temperature-measuring equipment and agree 

 on a few standard types, it would be possible to have these manufactured in quan- 

 tity and even stock-piled for major operations. It may be possible to standard- 

 ize on units of temperatures as well. 



DISCUSSION: R.O. Reid 



Surface layer - It might be well to expand upon the discussion of the surface 

 layer as introduced by Vine. A knowledge of the fine structure in the layer of 

 water immediately adjacent to the surface can be of great value in our under- 

 standing of boundary transfer processes. Existing instruments show essential- 

 ly isothermal conditions in the water within mixed layers; however, knowledge 

 of the detail near the surface is deficient. From physical considerations, if 

 there is to be heat conducted to the surface to replenish that which is lost by 

 back radiation and evaporation, one should expect to find a decrease in water 

 temperature with approach to the surface. The development of micro-oceano- 

 graphic gear would be of value in disclosing such gradients very near the sur- 

 face, if they exist, and in procuring quantitative data on these gradients for use 

 in surface heat budget studies. Such instruments would also be useful in ascer- 

 taining the pattern of surface or near surface temperatures which reflect per- 

 haps the pattern of evaporative convection cells in the surface layer. Tempera- 

 ture inhomogeneities would be most pronounced at the surface where unstable 

 conditions of density structure could exist in the presence of processes of cool- 

 ing. 



Thermal Response Factors - It has been stressed by Vine that short response 

 time is essential in certain types of temperature measuring instruments in the 

 sea. This is certainly true for any moving instrument, such as the BT, which 

 is essentially a temperature change measuring device. However, instruments, 

 such as the reversing thermometer, which have long response times are useful 

 and essential if one wishes to eliminate the short term fluctuations of tempera- 



