THE MAGNETIC WORK OF THE CARNEGIE AND THE URGENCY 

 OF NEW OCEAN MAGNETIC SURVEYS 



The earth's surface magnetic field varies with time. 

 The description of this field at any instant hence is dif- 

 ficult because it is not practical to undertake its simul- 

 taneous measurement at all points of the earth's surface. 

 There are, in fact, some seventy magnetic observatories 

 [see figs. 1(A) and 1(B)] where continuous simultaneous 

 measurements of the earth's field are made, but the 

 complexity of the field and its changes with time do not 

 permit satisfactory interpolation of values in interven- 

 ing regions. The density of observations per given area 

 of surface is uneven, and the great oceanic areas are 

 scarcely represented. There have been established, 

 therefore, by the Department of Terrestrial Magnetism 

 of the Carnegie Institution of Washington, over 10,000 

 stations on land and sea to supplement the results at 

 observatories. Measurements at groups of these sta- 

 tions on land or sea usually are made on only one day 

 for an occupation and constitute a magnetic survey. A 

 few thousand of these are called "repeat stations" at 

 which magnetic observations have been made more than 

 once (fig. 2). 



In the reduction of magnetic observations to epoch, 

 it is aimed to determine at all stations (at epochs not too 

 far removed from that at any previous time of observa- 

 tion) the intensity and direction of the earth's field freed 

 from the effects of small superposed fluctuations in in- 

 tensity. More generally, we seek to estimate from the 

 magnetic measurements made at different times and in 

 different locations the intensity of field at all interven- 

 ing (or suitably extrapolated) times and locations. 



For practical reasons the problem of describing the 

 earth's surface field is greatly simplified by mapping 

 only the main or permanent field. This includes only the 

 largest slowly varying part, mapped to only a moderate 

 degree of accuracy (one or two orders of magnitude less 

 than that of the actual observations). As has been stated, 

 this main field undergoes secular variation, that is, a 

 gradual variation with the passing of the years. The 

 strength of the main field at any time is called its nor- 

 mal value at that time. 



Although the rough and general description of the 

 geomagnetic field in terms of its normal value (as shown 

 on magnetic charts) is comparatively easy, the deriva- 

 tion of the normal values themselves is complex and dif- 

 ficult. To obtain these normal values it is necessary 

 first to remove from the individual magnetic observa- 

 tions the contributions of extraneous fluctuations not a 

 part of the secular variation. 



Often the extraneous fluctuations are large enough 

 seriously to affect the estimates of secular change ob- 

 tained from two or more observations made in different 

 years. These fluctuations thus may render difficult or 

 impossible a reliable estimate of the normal values 

 (since the secular change may not be determined accu- 

 rately either with respect to magnitude or sign) at epochs 

 other than those of observation. When the secular change 

 has been corrected for fluctuations in field, so that it is 

 known v/ith accuracy, the magnetic observations then are 

 readily reduced to the epoch desired for a magnetic 

 chart. 



Of fundamental importance in charting the earth's 

 magnetic field for a given epoch are the charts of secu- 

 lar change per year, called isoporic charts. The strength 

 of the earth's field in some regions may change by as 

 much as one -third in the course of only one hundred 

 years, and there has been a surprising lack of attention 

 to the importance of constructing such charts on the part 

 of various organizations responsible for the preparation 

 of isomagnetic charts. This importance arises because 

 it provides the only feasible means of enhancing and ex- 

 tending the value measured at a station, say in 1920, for 

 use in obtaining a chart value for the station, say in 

 epoch 1945. 



The first comprehensive world isoporic charts 

 were prepared by Fisk at the Department for epoch 

 1922.5 (see figs. 3 to 9). These give the average annual 

 secular change in components of the geomagnetic field 

 during 1920 to 1925. 



Figure 3 shows the isopors for geomagnetic decli- 

 nation or "variation," D, east declination being reckoned 

 as positive. The values in minutes of arc per year may 

 be interpreted in terms of force changes perpendicular 

 to the horizontal intensity, H. Centers of increase in 

 east declination are shown over Europe, South Africa, 

 and the Pacific Ocean; centers of decrease appear over 

 eastern Asia, the Indian Ocean, and North and South 

 America. 



Figure 4 shows the isopors in minutes of arc per 

 year for geomagnetic inclination or dip, I, dipping of 

 north end of needle being reckoned as positive. The cen- 

 ter of most rapid decrease is in the Atlantic Ocean and 

 that of most rapid increase in northern South America. 



Figure 5 shows the isopors in gammas per year for 

 the horizontal intensity, H. Regions of increase in H are 

 shown in the Indian and North Atlantic oceans and regions 

 of decrease near the south coasts of Africa and South 

 America, North America, and Asia. 



Figure 6 gives the isopors in gammas per year for 

 the vertical intensity, Z, positive when directed toward 

 the earth's center. Centers of increase are shown over 

 Asia, the Indian Ocean, the South Pacific Ocean, and 

 western South America; marked centers of decrease ap- 

 pear in the Atlantic and North Pacific oceans. 



Figure 7 shows the isopors for the total intensity, F, 

 derived from values for H and I. Marked areas of rapid 

 annual decrease in F appear in the North Atlantic and 

 South Indian oceans. Figures 8 and 9 give corresponding 

 derived values for the geographic north and east com- 

 ponents, X and Y, respectively. 



From the cartographer's point of view the signifi- 

 cant feature is that if the form of secular change be pre- 

 served, for say twenty years for the sake of illustration, 

 the value of declination or variation D may change by as 

 much as 280' (nearly 5°), I by 5°, H by 2400 gammas 

 (24 milligauss), Z by 3600 gammas, and F by about 3000 

 gammas. 



The foregoing isoporic charts, widely used today, 

 are highly tentative in a number of regions. They were 

 not derived taking into account correction of the survey 

 data for geomagnetic fluctuations, and the error averaged 



45 



