UPWELLING 197 



The Standard adopted by both these writers consists of a series of surface tempera- 

 tures more or less representative of the thermal normal at the surface. With this 

 " normal " is compared the temperature reduction caused by upwelling of lower layers. 

 This method may be applicable to measurement of upwelling in one place at different 

 times but not to measurement of upwelling in different latitudes; for at the surface, 

 temperature has a wider range from high latitudes to low than in the depths from which 

 upwelling waters originate. There may be a difference of two or three degrees between 

 the surface and 150 m. in lat. 40° S, but a difference of ten at the equator. Judged by 

 this method the same amount of upwelling would produce a more conspicuous fall of 

 temperature in the lower latitude. Vice versa, adoption of the thermal normal at the 

 upwelling depth is equally unsuitable. 



To meet this difficulty a standard has been sought which is in some way related to 

 the waters inshore. These are essentially a mixture of the deeper waters with those at 

 the surface. If we knew the ratios in which these two mixed and we knew their re- 

 spective initial temperatures it would be a simple matter to calculate a mean resultant 

 temperature and this could be used as a standard. As we do not know these facts, we 

 have endeavoured to construct an approximately similar curve by averaging the surface 

 temperatures outside the area of upwelling, e.g. say in 100° W, with the temperatures 

 at the mean depth of upwelling, e.g. say 150 m. 



In Fig. 64 the inshore surface temperatures as observed from Cape Carranza to Capo 



Blanco (curve D)^ are compared with temperatures observed at a depth of 150 m. on 



the one hand (curve A and the interpolated values A')'^ and on the other with the 



mean surface temperatures in the ocean in the meridian of 100° W (curve C)^. The 



slope of the curve D is seen to differ from curves C and A' but lies somewhere 



between them. If, as the result of upwelling, water represented by curve A' mixed in 



equal volumes with water represented by curve C, the resulting temperature could be 



A' + C 

 represented by the curve . But if the mean upwelling depth lay above or below 



150 m., and if upwelling water mixed with water cooler or warmer than that at 100° W, 



or again if more of one mixed with less of the other, then the resulting temperatures 



y4' + C 

 would of course depart from those shown by the curve . A curve of this type 



should enable a better comparison to be made than hitherto of the amount of upwelling 

 in different latitudes, though it will not afford a measurement of the absolute degree 

 of upwelling at any one place. 



DEPTH AFFECTED BY UPWELLING AND WATER LAYERS INVOLVED 



Earlier writers have been handicapped by lack of observations beneath the surface, 

 and those who first advanced arguments in favour of upwelling on this coast, made no 

 reference to the depth of water likely to be affected. Upwelling of polar-fed bottom 

 water was implied by Coker (191 8) and Murphy (1923), but Sverdrup (i 931) has shown 



1 See Table XVII. ^ gee columns A and A', Table XVI. ^ See column C, Table XVI. 



