Gubbins D., Herrero-Bervera E. Encyclopedia of Geomagnetism and Paleomagnetism

Подождите немного. Документ загружается.

Thermal remagnetization and demagnetization

Rocks lose their magnetization when temperatures exceed the Curie Point

(580



C for magnetite) and become remagnetized again as the tempera-

ture drops below this value. At shallow depths in volcanic regions,

particularly in recently emplaced extrusions and intrusions, cracking from

thermal expansion allows gas and fluid movement to rapidly transport

heat. Large local anomalies of hundreds of nanoteslas from remagnetiza-

tion of the upper few tens of meters can be generated quickly (Dzurisin

et al., 1990). At greater depth these effects occur on very long timescales

because of the small thermal conductivity of crustal rocks.

Observations

Volcanomagnetic effects

Unambiguous observations of magnetic field changes have been

obtained on many active volcanoes. The best examples clearly related

to volcanic activity were obtained during (1) the 18 May 1980, erup-

tion from Mount St. Helens, (2) the 1987 eruption sequence from Piton

de la Fournaise, Reunion Island, (3) the 16–22 November 1986, erup-

tion of Izu-Oshima volcano, Japan, (4) the 1980–1990 intrusion epi-

sode in Long Valley caldera, California, (5) the 2000 eruption of

Miyake-Jima, and (6) the 2003 eruption of Mt. Etna, Italy. Figure V9

shows an example of data from Mount St. Helens during the first stage

of the eruption on 18 May 1980. A final magnetic field offset of

9  2 nT was permanent and consistent with pressure decrease within

the mountain at the time of the eruption. The net positive signal con-

trasts with the negative change expected from the removal of magnetic

material from the mountain (Johnston, 1997). The oscillatory signals are

magneto-gas dynamic effects that couple into the ionosphere and propa-

gate many thousands of kilometers round the world. Figure V10 shows

this signal propagating across a 20-station magnetometer array between

1000 and 1500 km from the volcano. These TIDs are commonly gener-

ated by explosive volcanic eruptions and earthquakes.

Volcanoelectric effects

A number of important new electric field experiments have focused on

obtaining measurements on, or near, active volcanoes. In particular,

impressive self-potential observations have been made that delineate

the changing electric potentials around erupting volcanoes. SP anoma-

lies show apparent correlation with episodes of intrusive activity,

active venting, and eruptions. The pertinent physical mechanisms in

this case are electrokinetic effects, fluid vaporization and thermoelec-

tric effects. These are all generated by the injection of hot magma,

gas, and hydrothermal fluids into the volcano.

The best documented examples are recorded on:

1. Unzen, Japan, during 1990–1992 where large positive self-potential

anomalies were identified in the vicinity of the new lava dome.

Potentials generated by fluid flow in the hydrothermal system

beneath the array is suggested as the most likely cause for this

anomaly.

Figure V10 Magnetic field as a function of time and great circle

distance following the 18 May 1980 eruption of Mount St. Helens

(from Mueller and Johnston, 1989).

Figure V11 Self-potential profile across Miyake-Jima volcano Japan (from Sasai et al., 1997). The solid line in the upper plot shows the

data corrected for elevation effects.

986 VOLCANO-ELECTROMAGNETIC EFFECTS

2. Piton de la Fournaise volcano on Reunion Island, where similar

positive self-potential anomalies have been observed above the

active magma chamber.

3. Oshima Eruption, Japan, in 1986, where continuous measurements

of vertical electric fields in the frequency range 0–3 KHz before,

during and after a minor eruption from Mt. Mihara show a series

of rapidly rising slow-decaying pulses. These effects are attributed

to electrokinetic phenomena in the hydrothermal system.

4. Miyake-Jima, Japan, in 2000, where resistivity on scales of

0.5–1.5 km show up to a 300% change before eruptions. Immedi-

ately before the onset of caldera formation on 8 July 2000, self-

potential near the summit reversed sign from positive to as much

as 225mV. Figure V11 shows the SP record from Miyake-Jima

about 3 years before the eruption from Sasai et al. (1997). Zlotnicki

et al. (2003) suggest the preeruption observations resulted from

electrokinetic effects in a changing hydrothermal system. Remea-

surements in 2003 indicate the SP anomalies are much reduced in

amplitude and are localized to the edges of the new crater.

Malcolm J.S. Johnston

Bibliography

Dzurisin, D., Denlinger, R.P., and Rosenbaum, J.G., 1990. Cooling

rate and thermal structure determined from progressive magnetiza-

tion of the dacite dome at Mount St. Helens, Washington. Bulletin

of the Seismological Society of America, 95: 2763–2780.

Ishido, T., 2004. Electrokinetic mechanism for the “W”-shaped self-

potential profile on volcanoes. Geophysical Research Letters, 31:

L22601 (doi: 10.1029/2004GL020964).

Johnston, M.J.S., 1997. Review of electrical and magnetic fields

accompanying seismic and volcanic activity. Surveys in Geophy-

sics, 18: 441–475.

Johnston, M.J.S., Mueller, R.J., and Dvorak, J., 1981. Volcano-magnetic

observations during eruptions May-August 1980. US Geological

Survey Professional Papers, 1250:183–189.

Mueller, R.J., and Johnston, M.J.S., 1989. Large-scale magnetic field

perturbations arising from the 18 May 1980, eruption from Mount

St. Helens, Washington. Physics of the Earth and Planetary Inter-

iors, 57:23–31.

Sasai, Y., Zlotnicki, J., Nishida, Y., Yvetot, P., Morat, P., Murakami, H.,

Tanaka, Y., Ishikawa, Y., Koyama, S., and Sekiguchi, W., 1997.

Electromagnetic monitoring of Miyake-Jima volcano, Izu-Bonin

Arc, Japan: a preliminary report. Journal of Geomagnetism and

Geoelectricity, 49: 1293–1316.

Yukutake, T., Yoshino, T., Utada, H., Watanabe, H., Hamano, Y., and

Shimomura, T., 1990. Changes in the electrical resistivity of the

central cone, Mihara-yama, of Oshima volcano observed by a

direct current method. Journal of Geomagnetism and Geoelectri-

city, 42: 151–168.

Zlotnicki, J., 1995. Geomagnetic surveying methods. In McGuire, W.,

Kilburn, C., and Murray, J. (eds.), Monitoring Active Volcanoes.

London: University College London Press, pp. 275–296.

Zlotnicki, J., Sasai, Y., Nishida, Y., Uyeshima, M., Takahashi, Y., and

Donnadieu, G., 2003. Resistivity and self-potential changes asso-

ciated with volcanic activity: the 8 July 2000 Miyake-Jima erup-

tion (Japan). Earth and Planetary Science Letters, 205: 139–154.

Cross-references

Depth to Curie Temperature

Geomagnetic Spectrum, temporal

Gravity-Inertio Waves and Inertial Oscillations

Ionosphere

Magnetotellurics

Seismo-Electromagnetic Effects

VOYAGES MAKING GEOMAGNETIC

MEASUREMENTS

Introduction

The magnetic compass (q.v.) has been used in Western Europe since

the early or middle 12th century (see History of Geomagnetism). By

1187 a form with pivoted needle was described by Alexander Neckam,

a monk at St. Albans, in his De naturis rerum. At this time, however,

and for the next two and a half centuries, it was believed that the com-

pass pointed toward true north. As long as this was the case, there was

no reason to make or record any form of magnetic measurement.

Even when (probably sometime about the middle of the 15th century)

it was realized that the magnetic needle did not, in general, point true

north the situation with regard to making measurements did not change.

The discrepancy in pointing was regarded more as a fault of the needle

itself or of the method of magnetizing it rather than as being caused by

some external agency.

And yet, some glimmering of the truth seems to have begun to

dawn. Instrument makers in different parts of Europe began to com-

pensate for the difference between true north and magnetic north (the

magnetic variation or declination) by offsetting the needle from the

true north point of the compass card by the appropriate angle for

the place of manufacture. This worked pretty well near where the

instrument had been made but caused problems when used further

away. On his second voyage to the New World Christopher Columbus

took compasses manufactured in Genoa and in Flanders. Flemish com-

passes had their needles aligned approximately 11.5



east of true north,

this being the approximate value of declination at the time in Flanders;

Genoese instruments had theirs aligned with true north since the

declination was close to zero in the Mediterranean.

Columbus has been credited with the discovery of magnetic declina-

tion. Mitchell (1937) made a very detailed study of the surviving ver-

sions of the journals of Columbus’s four transatlantic voyages and

concluded that (a) Columbus was aware of the existence of declination

before setting out on his first voyage, and (b) he may have discovered

that it varies with position on the Earth’s surface. The journals are so

vague and have been edited so extensively that these conclusions are

by no means certain. It would perhaps be stretching a point to say that

Columbus’s journals contain geomagnetic measurements; they do

however contain comments and some rather obscure discussion of

the behavior of his compasses.

As it became obvious that, for accurate navigation, declination had

to be allowed for, so the practice developed of measuring and record-

ing values of declination during a voyage. These measurements were

useful at the time to navigators (usually of the same nationality or trad-

ing company) following the same or similar routes. Later, such knowl-

edge was systematized by drawing declination contour charts using the

measurements as a basis. Later still measurements of declination (and,

eventually, other magnetic elements such as inclination and field

strength) were collected by those wishing to produce charts and/or

mathematical models of the historical geomagnetic field. Among such

collections are those of Hansteen (1819), van Bemmelen (q.v. ) (1899),

Sabine (q.v.) (1868, 1872, 1875, 1877), Veinberg (1929–1933), and

Veinberg and Shibaev (1969).

The most recent, and by far the most comprehensive, data collection is

described by Jonkers et al. (2003). It incorporates the collections men-

tioned above as well as others. In this database, the measurements have

been derived where possible from the original sources and errors in long-

itude, which could be very large before the late 18th century, have been

corrected. Many of the sources are logbooks and journals (published

and unpublished) kept on board ship during the voyages. Others are col-

lections of voyages of a broader nature than the data collections discussed

above: collections that summarize or give edited accounts of voyages.

Two important examples are Hakluyt (1598–1600) and Purchas (1625).

VOYAGES MAKING GEOMAGNETIC MEASUREMENTS 987

Since the database of Jonkers et al. (2003) contains over 177000

measurements, of declination, inclination and field strength, from over

2000 voyages, space does not allow us to discuss every voyage that

made magnetic measurements. We shall instead select a few particu-

larly interesting voyages from each century from the 16th to the

20th, both inclusive.

Sixteenth century

The earliest voyage from which we have geomagnetic measurements

is that of the brothers Jean and Raoul Parmentier. They sailed from

Dieppe in March 1529, rounded the Cape of Good Hope and arrived

in Sumatra in November, having made a brief stop in Madagascar.

Soon after their arrival in Sumatra the brothers and many of their crew

perished, possibly murdered, and the ships returned, with great diffi-

culty, to France. During the outward voyage declination measurements

were made in the South Atlantic and six of these were collected by van

Bemmelen (1899), the earliest surviving measurements made during a

sea voyage.

The earliest voyage to have scientific work as one of its planned

objectives left Lisbon in the spring of 1538 bound for Goa in India.

In command of one of the eleven ships that made up the fleet was

Dom João de Castro who had been given the task of testing a new

form of magnetic compass and its associated method of observation.

Harradon (1943) describes the instrument and Harradon (1944) dis-

cusses de Castro’s declination measurements, 38 made on the outward

voyage and five on the return voyage. On the outward voyage de

Castro noted that iron objects such as cannon in the vicinity of his

measuring position adversely affected some of his measurements.

Later, he was the first person to investigate the local attraction of mag-

netic rocks (what would now be called an intense local magnetic

anomaly). He was also able to cast severe doubt on the idea that differ-

ent lodestones, used in the magnetization of compass needles, could

result in different values of declination being measured.

The Portuguese had pioneered and, by the early 16th century,

controlled the route to the Far East via the Cape of Good Hope;

the Spaniards from their settlements in Central and South America

controlled access across the Pacific. The seafarers of northern Europe

tried to find routes around the northern fringes of North America and

Eurasia, expending vast amounts of energy, and many lives, in search-

ing for the North-West and North-East Passages. An early voyage to

the north-east yielded six declination measurements. This voyage, in

the years 1556 and 1557 was under the command of Stephen Borough,

later Chief Pilot under Queen Elizabeth I. He had sailed with Richard

Chancellor in 1553 as part of the earliest English attempt to find the

North-East Passage, an attempt that failed to find the Passage but did

pioneer a new route to Moscow via the White Sea.

One of the earliest voyages to search for a North-West Passage was

that of Martin Frobisher in 1576. He entered Frobisher Bay and fol-

lowed it inland sufficiently far to convince himself (wrongly) that this

was the Passage. Because of the lateness of the season he returned

home, taking with him some samples of rock that he hoped might con-

tain gold. This voyage yielded three declination measurements. On his

next two voyages, in 1577 and 1578, the collection of rocks took pre-

cedence over exploration and magnetic measurements. The ore, alas,

proved to be pyrites and did not contain any gold.

The early circumnavigations led by Magellan and Drake have not,

apparently, provided any magnetic measurements, but that of Thomas

Cavendish (or Candish) between 1586 and 1588 did (five declination

measurements). This voyage was planned as a repeat of Drake’s and,

like Drake’s, had an element of piracy about it.

Another English voyage that furnished declination measurements

was essentially a privateering expedition in 1589 commanded by the

third Earl of Cumberland with the intention of capturing any Spanish,

Portuguese, or French ships it might meet. The fleet of seven ships

ended up in the Azores and returned with more prizes than they

could really manage. Edward Wright, a mathematician and expert on

navigation, wrote an account of the voyage in which he included the

30 measurements that he managed to make, presumably in the inter-

vals between the various fights that the ships were engaged in. Wright

was the first person to develop the mathematical basis of Mercator’s

projection. Mercator had devised this map projection, of great use in

navigation because a course at a constant bearing becomes a straight

line on the chart, in an empirical way in about 1569.

The Dutch also tried to find the North-East Passage. Willem Barentsz

made three voyages, in 1594, 1595, and 1596. On the last, in company

with Jacob van Heemskerck, he discovered Bjrnya (Bear Island) and

Svalbard. He and his crew wintered on Novaya Zemlya but Barentsz

died during the voyage home. The Dutch also began to challenge the

Portuguese control of the East India trade via the direct route. A Dutch

fleet under the command of Cornelis de Houtman sailed in 1595 and

returned, after trading successfully, in 1597. This was followed in

1598 by a second Dutch fleet under Jacob van Neck. Magnetic declina-

tion measurements were made on these two Dutch East India voyages

and on the third voyage of Barentsz. They, along with the measurements

made by Borough, Frobisher, Cavendish and Wright, are included in the

collections of van Bemmelen (1899) and Jonkers et al. (2003).

Seventeenth century

The English had broken the Portuguese monopoly in the East Indies a

little earlier than the Dutch. Sir James Lancaster sailed in 1591 and

returned laden with booty in 1594. As a result of this voyage the

English East India Company was formed in 1600 and Lancaster was

given command of the first Company fleet that sailed in 1601 and

returned in 1603. Declination measurements were presumably made

on these voyages but only one value has survived, from the 1601 to

1603 voyage. A potentially beneficial discovery was made on the latter

voyage. Lancaster took with him bottles of lemon juice and adminis-

tered three teaspoonfuls every morning to all his crew. His ship

remained relatively free of scurvy compared with the other vessels in

the fleet. The antiscorbutic effect of lemon juice, due to its vitamin C

content, was, however, largely forgotten until the time of James Cook

almost 200 years later.

The search continued for the North-West Passage. Two English

expeditions visited Hudson Bay at about the same time, believing

(wrongly) that there was an exit to the west. Luke Foxe (or Fox) sailed

from London at the end of April 1631 and made a thorough explora-

tion of the shores of Hudson Bay, the islands to the north and of Foxe

Basin to the south of Baffin Island. He returned safely at the beginning

of November without the loss of a single member of his crew. Thomas

James sailed from Bristol a month later than Foxe, met and dined with

him off the southern shore of Hudson Bay at the end of August and

entered and explored James Bay at its southeast extremity. There, on

the mainland near Charlton Island, James and his crew were forced

to spend a very difficult winter, but they managed to repair their

damaged ship and returne d to Bristol in October 1632. The voyages

of Foxe and of James both provided declination measurements in a

part of the world where few, if any, had previously been obtained.

The East India Companies of England and Holland sent out fleets

annually. These have provided large numbers of declination measure-

ments, helped by the fact that the English East India Company ordered

each captain, master, mate, and purser on their vessels to submit a

journal at the end of each voyage. Two Dutch voyages that provided

declination measurements were those of Willem Corneliszoon

Schouten and Jakob Le Maire between 1615 and 1616, the first Eur-

opean voyage to round Cape Horn, and of Abel Janszoon Tasman

between 1642 and 1643, the first European to visit Tasmania and

New Zealand.

The discovery by Henry Gellibrand (q.v.) in 1634 that declination at

a particular place varied with time (a phenomenon known as secular

variation) made the collection of geomagnetic measurements even

more important. It was now crucial to record the date of observation,

something that had not always been done before.

988 VOYAGES MAKING GEOMAGNETIC MEASUREMENTS

It may not have escaped the attentive reader that all the magnetic

observations mentioned so far have been of declination, despite the

fact that Robert Norman (q.v.) first measured the angle of inclination

or dip (the angle a freely suspended magnetic needle makes with the

horizontal plane) in 1576. One reason is that inclination is more diffi-

cult to measure accurately than declination, and this is particularly so

on board a moving ship. The earliest surviving inclination measure-

ments made at sea are seven values measured by Benjamin Harry in

the Atlantic and Indian Oceans during a voyage to the East Indies in

1680. This was followed by a more extensive set of 34 measurements

made by James Cunningham, a Scotsman more famous as a botanist,

on a voyage to China for the English East India Company in 1700.

We cannot leave the 17th century without mentioning the voyages

of Edmond Halley (q.v.). These were the first voyages planned with

scientific endeavors as the main part of their program. With the back-

ing of the Royal Society Halley persuaded the British Admiralty to

build a ship in which he could survey the Atlantic Ocean making mea-

surements of declination and, as a further aid to navigation, of the

longitude of his ports of call. The ship was named the Paramore,

Halley was commissioned as a Captain in the Royal Navy and was

given command of the vessel. The first voyage, from October 1698

to July 1699, was cut short because of trouble with Halley’s officers.

A second voyage was therefore ordered. This lasted from September

1699 to September 1700 and surveyed the Atlantic as far south as lati-

tude 52



S. During the two Atlantic voyages Halley made 187 mea-

surements of declination (though, strangely, none of inclination) and

used them as the basis of the first published contour chart, for the

Atlantic Ocean, published in 1701. In the summer of this year Halley

made a third voyage in the Paramore, sailing around and across the

English Channel studying the tides. He also, when opportunity arose,

measured declination, 13 values in all. The logbooks of Halley’s three

voyages have been edited by Thrower (1980).

Eighteenth century

Although by no means commonplace, voyages of circumnavigation

became more numerous during the 18th century. Among those that pro-

vided geomagnetic measurements were those of Woodes Rogers

between 1708 and 1711; Jacob Roggeveen (1721–1722); George Anson

(1740–1744); John Byron, grandfather of the poet, (1764–1766);

Samuel Wallis (1766–1768) and Philip Carteret (1766–1769); Louis

de Bougainville (1766–1769); the three voyages of James Cook

undertaken between 1768 and 1780; Jean-François de la Pérouse

(1785–1788); Antoine d’Entrecasteaux (1791–1794); and George

Vancouver (1791–1795).

The voyages of Rogers and Anson were repetitions of Drake’s cir-

cumnavigation in that one of their primary purposes was the harrying

of Spanish possessions in South and Central America and the disrup-

tion of their sea-borne trade. Rogers’ voyage was financed by a group

of Bristol merchants; Anson’s was an official naval project. Both were

successful financially and politically. Rogers managed, in large mea-

sure, to avoid the scourge of scurvy but the disease almost wrecked

Anson’s ill-equipped expedition.

The voyages of Roggeveen, Byron, Wallis and Carteret, and

Bougainville all had as one of their main objectives a search for Terra

Australis Incognita, the great southern continent postulated by arm-

chair geographers. All failed to find the continent but instead added

to the list of Pacific islands known to Europeans.

Cook’s first and second voyages also searched for the southern con-

tinent and their failure to find it after meticulous searching proved that,

if it existed, it must lie entirely south of the Antarctic Circle. During

the first voyage observations of a transit of the planet Venus across

the Sun’s disc were made from Tahiti. Cook’s third voyage included

a search for the western end of the North-West Passage north of the

Bering Strait. All three voyages were noted for the accuracy of their

positional data (the second and third carried chronometers made by

Kendall and Arnold). As well as declination measurements inclination

data were also collected.

The voyage of la Pérouse ended in tragedy. He left New South

Wales in February 1788 and was never seen alive again. The relief

expedition of d’Entrecasteaux helped to solve the mystery of the disap-

pearance and also made some of the earliest magnetic intensity mea-

surements, though these were in relative rather than absolute units.

Beaglehole (1934) summarizes the voyages of discovery in the Pacific

from Magellan’s circumnavigation to the end of the 18th century.

Nineteenth century

Polar exploration flourished during the 19th century, spurred on by

commercial interests such as the whale fishery and seal hunting. The

search for the North-West Passage also continued.

Naval voyages led by John Ross in 1818 and by William Edward

Parry in 1819–1820, 1821–1823, 1824–1825, and 1827 (the latter

an attempt to reach the North Pole across the ice from Svalbard) all

provided geomagnetic observations, at sea and on land, inclination as

well as declination. None of them achieved their major objective but

greatly increased geographical knowledge of the Arctic. From mag-

netic observations made at Parry’s winter quarters during his first three

expeditions it became obvious that the North Magnetic Dip-pole

(where the dip attains its maximum value of 90



) was to the south

of the archipelago of islands that lie to the north of the Canadian

mainland.

John Ross, having failed to persuade the Admiralty to give him

command of another attempt to find the North-West Passage, found

a private sponsor: the gin distiller Felix Booth. Accompanied, as sec-

ond-in-command, by his nephew James Clark Ross, who had been

on the elder Ross’s previous expedition and on all four of Parry’s,

Ross established winter quarters on a peninsula that he named, in

honor of his sponsor, Boothia Felix. The expedition ship, a paddle

steamer named Victory, was beset by ice and the explorers had

to spend four winters on or near Boothia Felix. During June 1831

the younger Ross led a sledging party to the North Dip-pole, which

he claimed for King William IV. Other, more useful, magnetic

measurements were also made.

Antarctic waters also began to attract attention, largely because they

were rich in sea mammals such as seals and whales. Voyages were also

sent out with more scientific aims. The Russian expedition under

Baron von Bellingshausen, during his circumnavigation of Antarctica

between 1819 and 1821, was the first to sight the continent. He also

made geomagnetic measurements.

During the years 1837–1843 three expeditions were exploring the

Antarctic: a French Expedition under Captain Dumont d’Urville

between 1837 and 1840, the US Exploring Expedition under

Lieutenant Charles Wilkes (1838–1842) and the British National

Expedition to Antarctic Seas under James Clark Ross (1839–1843).

D’Urville discovered Adélie Land and Wilkes also charted the coast

of the continent in this same region. Both made geomagnetic observa-

tions but neither expedition landed on Antarctica.

Ross’s expedition was largely devoted to geomagnetic studies.

Three permanent and three temporary magnetic observatories were

established during the voyage. Ross’s personal objective was to be

the first person to set foot on both North and South Magnetic Dip-

poles. He was thwarted in this ambition because the magnetic mea-

surements made at sea clearly indicated that the South Dip-pole lay

well inland, most probably beyond the mountains of Victoria Land.

He did, however, discover the Ice Shelf that now bears his name.

Another unsuccessful attempt to find the North-West Passage was

led by Sir John Franklin. It sailed in 1845 but was beset in the ice

off King William Island and every member of the expedition perished.

Although no results of geomagnetic measurements survived, the expe-

dition is important to our story because of the large number of expedi-

tions that were sent out over the next three decades to try to find news

VOYAGES MAKING GEOMAGNETIC MEASUREMENTS 989

or relics of Franklin; many of these search expeditions did make geo-

magnetic measurements in regions where such data were scarce or

nonexistent.

Two important oceanographic expeditions included geomagnetic

measurements in their observational programs. Both the British Chal-

lenger Expedition between 1872 and 1876 and the German Expedition

on board SMS Gazelle (1874–1876) made hundreds of measurements

of declination, inclination and total intensity during their circumnavi-

gations. And now, thanks to Gauss’s pioneering work (Gauss, 1833),

the intensity measurements were in absolute units (see Gauss’ determi-

nation of absolute intensity).

Twentieth century

The North-West Passage was finally discovered and traversed by

Roald Amundsen between 1903 and 1907. During the winters of

1903/1904 and 1904/1905 a magnetic observatory was operated at

Gjöahavn on King William Island. The measurements made there

served to fix the position of the North Magnetic Dip-pole for the first

time since Ross’s visit in 1831. A second observatory was established

at King Point near the Canada-Alaska border during the winter of

1905/1906.

Interest in exploring the Antarctic continent began to increase dur-

ing the final years of the 19th century, but the early years of the

20th century saw a veritable assault on the South Pole in particular

and Antarctica in general. Over a dozen major expeditions visited

the southern polar regions between 1900 and 1918 and the majority

made geomagnetic measurements during the voyage south or on Ant-

arctica or both. Somewhat surprisingly Amundsen’s 1910–1912 expe-

dition, which was the first to attain the South Pole, was not one of

these. On Shackelton’s British Antarctic Expedition of 1907–1909,

David, Mawson, and Mackay became the first to visit the South

Magnetic Dip-pole. Mawson retained his interest in the dip-pole and

during his Australasian Antarctic Expedition of 1911–1914 detailed

geomagnetic measurements by Webb enabled a revised position to

be calculated, although adverse conditions prevented the party from

reaching it.

Of much greater practical importance was the global magnetic sur-

vey undertaken by the Carnegie Institution of Washington under the

leadership of Louis A. Bauer (q.v.) from 1905 onward. The oceanic

part of this was begun using a converted brigantine, the Galilee , and

was continued by the specially built nonmagnetic ship Carnegie (q.v.)

from 1909 until her destruction by fire in 1929.

Plans by the UK Government to build a nonmagnetic ship, the RRS

Research, to carry on the work of the Carnegie were delayed by World

War II and were eventually abandoned after the War. Geomagnetic

measurements of the complete field vector were continued, however,

by the Russian nonmagnetic ship Zarya in a series of cruises begin-

ning in 1956.

The development of the proton precession magnetometer revolutio-

nized marine magnetic surveying. This type of magnetometer does not

need to be accurately oriented with respect to the geomagnetic field

and can be towed in a container behind a ship so that it is not affected

by the intrinsic magnetic field of the ship. A disadvantage is that it is

only able to measure the total intensity of the geomagnetic field rather

than the complete vector. Proton magnetometer surveys have therefore

been mainly used for mineral exploration and for studying magnetic

anomalies associated with plate tectonics. Measurements of total inten-

sity can be used to supplement vector measurements in producing

charts and models of the main geomagnetic field and they have proved

to be a valuable source of data, particularly over large areas of deep

ocean were few other measurements exist.

So sparse have such measurements been in the past that routine

observations made to check the reading of the magnetic compass on

board merchant vessels have also been used in the production of global

magnetic charts and models. These measurements, although of lower

accuracy than those made on board specially designed magnetic

survey vessels, have been invaluable in ensuring that the charts and

models were of acceptable accuracy in remote ocean regions such as

the Southern Ocean and the central Pacific Ocean.

Discussion

The Portuguese and Spanish were the first to venture across the major

oceans in search of new trading opportunities and the conquest of

newly discovered territories. It is therefore strange that very few Iber-

ian voyages feature in the datasets of magnetic measurements that

have been collected. van Bemmelen (1899) pointed out the potential

riches that he thought should be available in the Spanish and Portu-

guese archives but was not able to confirm that this was the case.

Jonkers and his colleagues (2003) have made comprehensive

searches in seven of the main Spanish archives. Apart from material

relating to Malaspina’s expedition in the Atlantic and Pacific between

1789 and 1794, which supplemented observations already published,

they found disappointingly few logbooks. In particular, they had hoped

to find a manuscript legacy of the Manila galleons that, over many

decades, sailed annually across the Pacific between the Philippines

and South America. No logbooks from these voyages were found in

the Spanish archives. Jonkers et al. (2003) speculate that manuscript

sources containing magnetic measurements may exist in former

Spanish colonies in South America or the Philippines.

In their discussion of possible future work on collecting magnetic

measurements made on board ship Jonkers et al. (2003) mention the

Portuguese archives as one of the most promising of hitherto untapped

sources “even though much Portuguese material is known to be no

longer extant.” The remarks made above concerning possible archives

in former overseas possessions also apply here. If found such sources

may provide valuable early measurements in the Indian Ocean.

Mention has already been made of the improved accuracy of the

measurements made by Cook on his three voyages in the third quarter

of the 18th century compared with those made on other contempora-

neous voyages. The biggest improvement was in the determination

of position at sea, particularly that of longitude and this was primarily

due to the employment of some of the first marine chronometers.

Cook’s use of these was experimental, part of the tests that proved

the worth of this method for finding the longitude at sea, and it

was some time before their use became general. The Hydrographic

Office of the British Admiralty was founded in 1795 and, together

with the Royal Observatory at Greenwich, began to test and issue mar-

ine chronometers. By the end of John Pond’s time as Astronomer

Royal in the 1830s complaints were being made that too much of

his time and that of his Observatory’s staff was being taken up with

chronometer testing, to the detriment of the astronomical work. By

the middle of the 19th century most major navies and merchant mar-

ines were using chronometers and the accuracy of positional data

approached modern values.

Conclusion

Since the earliest European voyages of exploration the importance of

the magnetic compass as an aid to navigation has been realized along

with the necessity to collect information about the behavior of the

geomagnetic field that controls it. Beginning with measurements

of the most obvious geomagnetic element, declination, observations of

the other geomagnetic elements have gradually been added. Because

about 70% of the Earth’s surface is covered by oceans, measurements

made at sea are of the greatest importance in making mathematical

models and charts of the global geomagnetic field. This article has

summarized the history of such measurements by highlighting a selec-

tion (no doubt very subjective) of the more important and interesting

voyages that have made geomagnetic measurements since the begin-

ning of the 16th century.

David R. Barraclough

990 VOYAGES MAKING GEOMAGNETIC MEASUREMENTS

Bibliography

Beaglehole, J.C., 1934. The Exploration of the Pacific. London: A & C

Black.

van Bemmelen, W., 1899. Die Abweichung der Magnetnadel; Beo-

bachtungen, Säcularvariation, Wert-und Isogonensysteme bis zur

Mitte des XVIIIten Jahrhunderts. Observations of the Royal Mag-

netic and Meteorological Observatory, Batavia, 21(Suppl): 109 pp.

Gauss, C.F., 1833. Intensitas vis magneticae terrestris ad mensuram

absolutam revocata. Göttingen: Dieterich.

Hakluyt, R., 1598– 1600. The Principal Navigations Voyages Traffi-

ques & Discoveries of the English Nation Made by Sea or Over-

Land in the Remote and Farthest Distant Quarters of the Earth

at any Time Within the Compasse of these 1600Years. London:

George Bishop, Ralph Newberie, and Robert Barker (Reprinted:

1903–1905 in 12 Volumes. Glasgow: J. Maclehose & Sons).

Hansteen, C., 1819. Untersuchungen über den Magnetismus der Erde.

Christiania (Oslo): Lehmann & Gröndahl.

Harradon, H.D., 1943. Some early contributions to the history of geo-

magnetism—V. Terrestrial Magnetism and Atmospheric Electri-

city, 48: 197–199.

Harradon, H.D., 1944. Some early contributions to the history of geo-

magnetism—VII. Terrestrial Magnetism and Atmospheric Electri-

city, 49: 185–198.

Jonkers, A.R.T., Jackson, A., and Murray, A., 2003. Four centuries of

geomagnetic data from historical records. Reviews of Geophysics,

41(2): 1006 (doi: 10.1029/2002RG000115).

Mitchell, A.C., 1937. Chapters in the history of terrestrial magnetism.

II. The discovery of the magnetic declination. Terrestrial Magnet-

ism and Atmospheric Electricity, 42: 241–280.

Purchas, S., 1625. Purchas his Pilgrimes. In five books. The first, con-

taining the voyages (...) made by ancient kings (...) and others, to

and throw the remoter parts of the known world, etc. 4 Parts.

London: W. Stansby, for H. Fetherstone (Reprinted: 1905–1907

as Hackluytus Posthumus (or) Purchas His Pilgrimes: Containing

a History of the World in Sea Voyages and Land Travels by Eng-

lishmen and Others. 20 Volumes. Glasgow: J. Maclehose & Sons).

Sabine, E., 1868. Contributions to Terrestrial Magnetism, No. XI.

Philosophical Transactions of the Royal Society of London, 158:

371–416.

Sabine, E., 1872. Contributions to terrestrial magnetism, No. XIII.

Philosophical Transactions of the Royal Society of London, 162:

353–433.

Sabine, E., 1875. Contributions to terrestrial magnetism, No. XIV.

Philosophical Transactions of the Royal Society of London,

165:

161–203.

Sabine, E., 1877. Contributions to terrestrial magnetism, No. XV.

Philosophical Transactions of the Royal Society of London, 167:

461–508.

Thrower, N.J.W. (ed.), 1980. The Three Voyages of Edmond Halley

in the “Paramore” 1698–1701, Second Series, Nos. 156, 157.

London: The Hakluyt Society.

Veinberg, B.P., 1929–1933. Catalogue of Magnetic Determinations in

USSR and in Adjacent Countries from 1556 to 1926, Parts 1, 2 &

3. Leningrad (St. Petersburg): Central Geophysical Observatory.

Veinberg, B.P., and Shibaev, V.P. (Editor-in-Chief: Pushkov, A.N.),

1969. Catalogue. The Results of Magnetic Determinations at Equi-

distant Points and Epochs, 1500–1940. Moscow: IZMIRAN (Eng-

lish translation: No. 0031 by the Canadian Department of the

Secretary of State, Translation Bureau, 1970).

Cross-references

Bauer, Louis Agricola (1865–1932)

Bemmelen, Willem van (1868– 1941)

Carnegie, Research Vessel

Compass

Gauss’ Determination of Absolute Intensity

Gellibrand, Henry (1597–1636)

Geomagnetism, History of

Halley, Edmond (1656–1742)

Norman, Robert (1560–1585)

Sabine, Edward (1788–1883)

VOYAGES MAKING GEOMAGNETIC MEASUREMENTS 991

WATKINS, NORMAN DAVID (1934–1977)

Norman Watkins (Figure W1) was an innovative and dynamic paleo-

magnetist who in just a few years attained international recognition

and high praise for his work. As a student in geophysics and paleo-

magnetism during the period when the polarity timescale was first

being developed he experienced first hand the plate tectonics revolu-

tion in the earth sciences. He was among the first to develop a modern

paleomagnetism laboratory, initially with spinner magnetometers

developed during the middle-late 1960s, and then adding one of the

first cryogenic magnetometers to his laboratory in 1972. In his short

lifetime he published over 100 refereed publications (and hundreds

of other works), received six academic degrees, rose to the rank of

professor of oceanography at the Graduate School of Oceanography,

University of Rhode Island, was continuously funded by NSF from

1963 until his death in 1977 (grant funding starting before he received

his Ph.D.), and served as the division director of the Earth Science

Division of the National Science Foundation.

Born in Sunbury-on-Thames in Surrey, England, in 1934, of Welsh

parents, Watkins later became a naturalized citizen of the United

States. His first college degrees were from the University of London

where he received a B.Sc. in mathematics, geology, and geography

in 1956, and a second B.Sc. in mathematics, physics, and “geology

special (Part 1 only)” in 1957. He then attended the University of

Birmingham, England, where he received an M.Sc. in applied geophy-

sics in 1958. Watkins moved to Alberta, Canada, in 1958, where he

enrolled at the University of Alberta, eventually receiving another

M.Sc., this time in geology with a thesis titled Studies in Paleomagnet-

ism. During his time in Alberta, he took a job with Shell Oil Company

of Canada where he worked in various areas of geophysics (gravity,

magnetic, and seismic interpretation) as well as geological mapping.

In 1961, he moved to California where he accepted a position at

Menlo College in Menlo Park, California, as a lecturer in physics.

He served in that position until 1964 when he was named the acting

dean of science at the college. He enrolled as a graduate student at

Stanford University in 1962, serving there as a research assistant. At

Stanford he worked with Allan Cox and Richard Doell in paleomag-

netism, Richard Doell nominally being his advisor. In 1963, he

received an appointment upgrade at Stanford to acting assistant profes-

sor in the Department of Geophysics, and in 1964, he became a

research associate there. Because of very strong personal conflicts with

Doell, Watkins moved his dissertation research problem back to the

University of London where he defended and was granted his Ph.D.

in geophysics in 1964. The title of his dissertation was Paleomagnet-

ism of the Miocene Lavas of Southeastern Oregon. After receiving

his Ph.D. he took a position as visiting senior fellow in the Chadwick

Physics Laboratory at the University of Liverpool in England. In 1965,

he returned to the United States as an associate professor in the

Department of Geology at Florida State University in Tallahassee,

Florida, where he received tenure in 1968. In 1970, he accepted the

position of professor of oceanography at the Graduate School of Ocea-

nography, University of Rhode Island (URI). In 1972, he was granted

tenure there. Later still, he received a doctor of science degree from the

University of Birmingham in recognition for his contributions to geo-

physics. He took a leave of absence from URI in 1976 to serve as the

director of the Earth Science Division of the National Science Founda-

tion, but after 8 months, illness forced his resignation from NSF and a

return to URI.

During his career, Watkins mainly taught courses in geophysics and

paleomagnetism. While he was at Florida State University he super-

vised six students who received their MS degrees and one Ph.D. stu-

dent who finished after Watkins moved to Rhode Island. While at

URI he supervised three Ph.D. students, one who graduated after

Figure W1 Norman Watkins, on board the USNS Eltanin in 1970.

his death, and five MS students. He also served as chief scientist on a

cruise of the USNS Eltanin to the Indian Ocean and sent students out

on cruises of the RV Trident (URI), RV Chain (Woods Hole Oceano-

graphic Institution of MIT), and other ships as his representatives

during various cruises. The primary aim of this work was to collect pis-

ton cores of deep-sea sediment, to dredge for marine basalts, and for

other related oceanographic studies. Sediment samples from piston cores

were returned to his laboratory for paleomagnetic measurement. These

samples were also shared with James Kennett (a foraminifera specialist)

and his students for biostratigraphic analysis. His relationship with

Kennett started when they were professors at Florida State University.

Both men moved to the Graduate School of Oceanography at URI in

1970, as a “package deal.”

Watkins’ contributions to research in paleomagnetism began with a

paper published in Nature in 1961, where he was a co-author with

Ernest Deutsch. This work examined the direction of the geomagnetic

field recorded in Triassic rocks in Siberia, and was published at that

exciting time in paleomagnetism during the early development of the

Plio-Pleistocene polarity timescale. Another paper in Nature in 1963

examined the behavior of the Miocene geomagnetic field recorded in

Columbia Plateau basalts and led to his important paper on the polarity

transition identified at Steens Mountain. He was very concerned with

the effects of oxidation on the magnetic properties of rocks. This con-

cern was reflected in a number of papers with Stephen Haggerty and

others published in Nature, Science, and other journals between 1965

and 1968.

In 1968, he published his first paper on the magnetic properties of

deep-sea sediments recovered from piston cores. While he and his stu-

dents continued to work actively on the magnetic properties of basaltic

materials, this shift to marine sediments reflected a new direction in

the overall field of paleomagnetism. It was facilitated by the availabil-

ity at about this time of easy-to-use and reliable spinner magnet-

ometers and adequate instruments that allowed rapid demagnetization

of samples. While measurements on deep-sea sediments were impor-

tant, the magnetometers were still only marginally sensitive enough

for many samples. It was not until the availability of cryogenic mag-

netometers that many of these materials could be adequately studied.

Therefore, he continued to concentrate primarily on basaltic rocks

and polarity studies.

Much of his work in the 1970s involved measurements of basaltic

rocks from oceanic islands and characterization of secular variation

and polarity of Earth’s magnetic field during the last 10–15 Ma.

Watkins and his students collected samples from islands all over the

world including Reunion Island, Kerguelen, and Amsterdam Island

in the Indian Ocean, and the Azores Islands in the Atlantic. Much

of this work was concentrated in Iceland with Leo Kristjansson (an

Icelandic physicist and paleomagnetist) as well as with the late well-

recognized physical volcanologist, professor George P. L. Walker,

where they had revolutionized the understanding of the growth of

volcanoes in Iceland by means of volcanic and magnetic stratigraphy,

and with Ian McDougal (a good friend and potassium-argon dater

from Australia), where his group collected thousands of cores from

stratigraphic sequences of lavas all over the island. Many of these

samples and his notes are archived at the Graduate School of Oceano-

graphy, URI.

Watkins was a very aggressive scientist who enjoyed stimulating

debate. For example, he would always sit in the front row during pre-

sentations at national and international meetings and ask one question

of every speaker that was designed to elicit discussion. He would do

this while casually resting his arm on the chair next to him and turning

sideways toward the audience, as if to get their opinion. His students

observed that at these meetings he would usually be dressed in a

three-piece suit. That is, except when it was time for him to give a pre-

sentation. Then he would appear in his field “traveling” cloths, a light

blue dungaree shirt, tan pants, and a white ship-captain’s hat with a

black bill. Minutes later he would be seen again wearing his three-

piece suit.

He was a very big man. One of his students once commented that

when sitting at the back of a classroom on the first day of geophysics

class he noticed there was something strange about the appearance of

Prof. Watkins. The student suddenly realized that Watkins was stand-

ing with his left forearm on the top of the blackboard, bending down

so that he could write on the board. “This guy is really big!” commen-

ted the student to a classmate. Watkins enjoyed impressing his students

at Florida State University with the medal he earned throwing the

Javelin for Wales at the 15th Commonwealth Games.

Besides being big, Watkins was also powerful. Basaltic paleomag-

netic samples were collected by his group using chain saws that he

had converted to rock drills. This was before such conversions were

commercially available, and these drills were heavy. To impress his

students and to demonstrate how the drilling method worked, he

would drill several, perfectly straight, six-inch long cores, upward at

a45

angle using only one hand! On his best, in a single day he drilled

and collected samples from all 55 lavas contained in a very steep,

1250 m thick stratigraphic sequence. His students were never able to

keep up with him in the field.

When working in the field he was a very careful scientist and was

always well prepared. He stressed back-up systems and redundant

parts so that there would be no delays that might limit the number of

samples returned to the laboratory. Early in his work he used a Sun

compass to correct problems associated with local or anomalous mag-

netic field directions. To protect the collection and avoid problems he

demanded that every evening the notes be recopied into a separate

notebook, thus producing a redundant copy. He insisted that all sam-

ples be checked for marking errors or other problems. As a result there

were very few ambiguities associated with these field collections.

In summary, Norman D. Watkins made important contributions

to paleomagnetism during the early years of that developing field

and helped to establish the techniques we use today. He actively

stimulated thought and interaction among all who really knew him,

and is missed by all of them. During his lifetime, Watkins and his wife

Patricia had two sons, the oldest, Paul Watkins, currently living

in the United States, became a successful teacher and author of

many well-received novels. The youngest, Clive Watkins, now living

in Luxembourg, received an MBA and became a successful business-

man in Europe.

Brooks B. Ellwood

WESTWARD DRIFT

The time variation (secular variation) of the magnetic field was first

recognized in the late 17th century. In attempting to provide a theory

to explain it, Halley (1693) (qv) noticed that a large part of the secular

variation could be explained in terms of a “westward drift” of the

field. If maps of declination are produced over the Earth’s surface at

different epochs, then there is a clear westward movement of field

patterns, particularly lines of zero declination (so-called agonic lines)

(e.g., Langel, 1987). Using his map of Atlantic secular variation,

Halley estimated that a full revolution of the field would take about

700 years, giving a rotation rate of just over 0.5



per year. To explain

the drift, he posited a model of the interior of the Earth consisting of

concentric shells of magnetic material rotating at varying rates with

respect to the Earth’s surface. In many ways, his theory has remarkable

similarities to our current understanding. The Earth does indeed con-

sist of different layers, and it is one of these layers (the fluid core)

which gives rise to the secular variation, of which a large part can

indeed be represented by westward motions at the surface of the core.

As understanding of the Earth’s interior improved, it became

clear that Halley’s theory was inadequate both observationally and

WESTWARD DRIFT 993

theoretically, not least because the thermal state of the Earth does not

allow permanent magnetization at depth. With the advent of dynamo

theory, and the understanding that the magnetic field originates from

fluid motions in the molten iron core, the first “modern” theory of

westward drift was proposed by Bullard et al. (1950). They argued that

cooling of the Earth would lead to a pattern of large-scale convection

in the core. As a parcel of fluid rose, it would move further from the

Earth’s rotation axis. Therefore, to conserve angular momentum, the

angular velocity of the parcel would reduce. Similarly, a sinking fluid

parcel would reduce its moment arm, and so need to increase its angu-

lar velocity. Bullard et al. suggested that this would lead to a net east-

ward flow with respect to the solid Earth at the base of the core near

the inner core boundary, and westward flow at the core surface, which

if advection by flow dominates diffusion (the frozen-flux approxima-

tion (qv)) would then carry the magnetic field in a westerly direction,

giving rise to westward drift. While the basic physical ideas of this

theory are very attractive, modeling of rotating convection has shown

that the interaction of convection and rotation is much more compli-

cated, and this pleasingly simple explanation for the drift does not

work (although, interestingly, an eastward inner core rotation (qv)

has more recently been suggested both from numerical dynamo models

(qv) and seismological observations).

An alternative to Bullard’s mechanism was provided by Hide

(1966). While Bullard et al. had argued for an origin of westward drift

in large scale flow, Hide instead suggested that it could arise as a result

of wave motion. Magnetic waves called Alfvén waves (qv) were

known to be supported in a fluid penetrated by a magnetic field. Add-

ing the effect of rotation, additional families of diffusionless magneto-

hydrodynamic waves (qv) were shown to exist. Some have periods of

order days (similar to inertial waves), but others, named magnetic

Rossby waves or planetary waves, have much longer periods, for rea-

sonable estimates of the magnetic field strength perhaps 300 years.

Initial analysis based on propagation in a thin shell (similar to analysis

for the atmosphere) unfortunately suggested that these waves would

propagate eastward, not westward. Hide provided an intuitive argu-

ment, later confirmed by more detailed analysis, as to why in the core

(a thick shell) such waves would propagate westward instead. Further

examination has suggested that westward motion will dominate when

the toroidal magnetic field is stronger than the poloidal magnetic field,

as is thought to be the case within the core (Hide and Roberts, 1979).

However, such simple models, while attractive, can in no way

explain the full complexity of secular variation. Estimates of westward

drift obtained show considerable variation depending on just how it is

defined—should it involve a fit to the whole field, just the equatorial

dipole, the secular variation, or some combination? The value of

Bullard et al. (1950) of 0.2



per year has entered the geomagnetism

consciousness as a standard, but estimates have varied from 0.08



east-

ward per year, up to 0.733



westward (for a summary, see Langel

1987). What this range of values demonstrates is that the picture of

westward drift is much too simplistic. Even in the 18th century, it

was realized some features of the field drift northwards rather than

westward, and once Gauss had developed a method for measuring

magnetic intensity, the observed decay of the dipole could clearly

not be explained by drift alone. Further, westward drift is much more

clearly visible in Europe and North America than in Asia and the

Pacific hemisphere. Yukutake and Tachinaka (1969) argued that other

features in the field did not move at all, and proposed a division of the

field into drifting and standing components. Yukutake (1969) further

suggested that the equatorial dipole field could be separated into two

components, drifting in opposite directions; such a process is reminis-

cent of wave motion, consistent with Hide’s ideas. Wave theory further

allows drift rate to vary with location, as the wave velocity is a func-

tion of the background field strength, which will vary over the core’s

surface.

However, this confused picture was largely swept away by the

advent of the Magsat satellite (qv), and the impetus it gave to a pro-

gram of high-resolution mapping of the field at the core-mantle

boundary. If we want to understand the secular variation, we must

really examine it at its source; upward continuation filters the field, com-

plicating considerably the interpretation of the field at the Earth’s sur-

face. Field models constructed for different epochs, culminating in the

time-dependent main field model (qv) “ufm” of Bloxham and Jackson

(1992), allowed detailed features of field evolution to be examined at

source. Their model from 1690 to 1990 shows apparent upwelling of

field under Africa, followed by a westward motion under the Atlantic.

However, such features are not observed under the Pacific, leading to

the weak westward drift observed in the surface field in the Pacific hemi-

sphere. Other features, particularly two pairs of high-latitude flux

patches of opposite signs in the northern and southern hemisphere,

are observed to remain stationary over historical time, contributing to

Yukatake’s standing field. Models of surface core motion (qv) have been

generated, using the magnetic field as a tracer of fluid motion. While

such models uniformly demonstrate a westward flow under the Atlantic,

elsewhere the pattern is more complex, and many suggest eastward flow

under the Pacific. An alternative view of drift has also recently been pre-

sented: instead of the drift of field or secular variation, the underlying

flow pattern is considered steady but drifting either westward or east-

ward (Holme and Whaler, 2001). Such a model can explain the secular

variation well with either a westward or eastward drift, again suggesting

that the problem may be best posed in terms of wave motion.

In conclusion, the concept of westward drift has largely been super-

ceded by more complex models of the secular variation, with maps of

the field evolution at the core-mantle boundary allowing the secular

variation to be examined at source. However, the simplicity of west-

ward drift means that it is still used for interpretation of archeo- and

paleomagnetic data. Because insufficient data have been available to

enable calculation of high-resolution field models as has been done

for historical time, we are limited to considering the change of the field

at the Earth’s surface. Therefore, interpreting changes in the field in

terms of drift, and relating such drift to our understanding of the histor-

ical field, is still extremely valuable. Furthermore, the underlying con-

troversy as to whether secular variation is dominantly generated by

large scale core convection (as proposed by Bullard et al.) or wave

motion at the surface of the core (as proposed by Hide) remains alive

to this day.

Richard Holme

Bibliography

Bloxham, J., and Jackson, A., 1992. Time-dependent mapping of

the magnetic field at the core-mantle boundary. Journal of Geo-

physical Research, 97: 19537–19563.

Bullard, E.C., Freeman, C., Gellman, H., and Nixon, J., 1950. The

westward drift of the Earth’s magnetic field. Philosophical Trans-

actions of the Royal Society of London A, 243:61–92.

Halley, E., 1693. On the cause of the change in the variation of the

magnetic needle; with an hypothesis of the structure of the internal

parts of the Earth. Philosophical Transactions of the Royal Society

of London A, 17: 470–478.

Hide, R., 1966. Free hydromagnetic oscillations of the Earth’s core

and the theory of the geomagnetic secular variation. Philosophical

Transactions of the Royal Society of London A, 259: 615–650.

Hide, R., and Roberts, P., 1979. How strong is the magnetic field in

the Earth’s liquid core? Physics of the Earth and Planetary Inter-

iors, 20: 124–126.

Holme, R., and Whaler, K., 2001. Steady core flow in an azimuthally

drifting reference frame. Geophysical Journal International, 145:

560–569.

Langel, R.A., 1987. The main field. In Jacobs, J.A. (ed.), Geomagnet-

ism, Volume. 1. New York: Academic Press (Chap. 4).

Yukutake, T., 1979. Review of the geomagnetic secular variations on

the historical time scale. Physics of the Earth and Planetary Inter-

iors, 20:83–95.

994 WESTWARD DRIFT

Yukutake, T., and Tachinaka, H., 1969. Separation of the earth ’s mag-

netic field into drifting and standing parts. Bulletin of the Earth-

quake Research Institute, 47: 65.

Cross-references

Alfvén Waves

Alfvén’s Theorem and the Frozen Flux Approximation

Core Motions

Dipole Moment Variation

Halley, Edmond (1656–1742)

Inner Core Rotation

Magnetohydrodynamic Waves

Time-Dependent Models of the Geomagnetic Field

Upward and Downward Continuation

WESTWARD DRIFT 995