The AlphaLab Earth Magnetometer costs $700 complete, with battery and probe rod. It weighs only 18 ounces (= 510 grams), and runs about 8 hours on the included standard 9-volt alkaline battery. It measures difference in field strength to accuracy of +/-0.5% of the difference, either for point-to-point or time-to-time measurements. In other words, if one location is 2.00 milligauss (200 nanotesla) stronger than another location, the Earth Magnetometer will indicate that difference to accuracy of +/- .01 milligauss or 1 nanotesla.
This magnetometer has a magneto-resistive sensor, which uses the spins of electrons flowing through a circuit, rather than their charge, to measure magnetic field. This type of device most closely resembles a “flux gate” magnetometer, but the magneto-resistive technology is newer, smaller, less expensive and intrinsically more stable over temperature.
Measuring the Earth's Magnetic Field
Geomagnetometers, such as this meter, are usually used to measure the strength of earth's field at several points on the land surface. These points are sometimes arranged in a square grid (with spacings such as 10 x 10 feet). In this case, a measurement is recorded at every point on the grid, and a map can be made from these numbers. The map might show for example, that the field is stronger in a circular area about 20 feet in diameter in the center of the grid than it is at the edges of the grid. This would suggest that a magnetized object is buried roughly 20 feet underground.
Instead of measuring on a square grid, measurements are sometimes made at several points along a straight line. The relative strength of field is measured at each point (relative to the field strength at the starting point). If there is an area with a stronger (or weaker) field than average, then the place along that line where the field is strongest (or weakest) can be found with the meter. Then measurements are made along a straight line that is perpendicular to the original line, starting at that strongest (or weakest) point on the original line. If you're looking for the location of the strongest (or weakest) field, this method is faster than measuring every point of a grid.
There is one complication involved with measuring the field strength: magnetic field is a vector, which means it has both a strength (“magnitude”) and a direction. At each point (or position) that a measurement is taken, both the strength of field and the direction may be different from that of the previous measured point. Most geomagnetic mapping is concerned with the strength only, and not with the direction. When using the more expensive meters such as proton magnetometers, the direction cannot be determined, because these types of meters measure the strength only. (In comparison, our Earth Magnetometer is capable of showing the field direction, but this additional information is difficult to quantify, because it would require that you measure the angle between the sensor rod and the vertical direction. A measurement like this could be done using a plumb-bob.)
All types of geomagnetic meters require that you point them, at least approximately, in a certain direction (with respect to the earth's field direction). Otherwise the readings will be inaccurate. With a proton magnetometer, the sensor can be pointed as much as 30° off from the preferred direction before the readings start to become noisy. It's easy to keep the sensor pointed with +/- 30° of a certain direction when measuring in the field, so this is not much of a problem. However, the AlphaLab Earth Magnetometer is a vector magnetometer, and it must have its sensor pointed (at least momentarily) within one-third of a degree of the correct field direction in order to measure the actual field to .01 milligauss precision. This kind of precision required in pointing the sensor had been a problem when using vector magnetometers to measure the exact earth field strength. A “fluxgate” magnetometer (which is also a vector magnetometer) could have been available a long time ago as a less expensive alternative to the proton magnetometer, had it not been for the accuracy of pointing needed.
The expensive geomagnetic field meters now in use measure the magnitude of the field and are accurate if pointed roughly within +/- 30° of a preferred direction (the “preferred direction” is actually a certain amount of angle off-axis from the earth's field direction, but that detail is not important here). These meters will read the same number if pointed anywhere within that +/-30°. In contrast, vector magnetometers, like fluxgate or the AlphaLab Earth Magnetometer, use sensors that read a field strength that is correct only when the sensor is pointed in the same direction as the field; if pointed in a different direction, they will read a lower number. The number they read is the actual field strength multiplied by the cosine of the angle between the sensor direction and the actual field direction. For example, if the actual field is 500.00 milligauss, and a vector sensor is pointed in exactly the direction of the field, it will read “500.00” (subject, of course, to the sensor's accuracy). If the sensor is 1/3 of a degree off from the correct direction of the earth's field, it will read 500.00 milligauss x cos(1/3°) = 499.99 milligauss (about 1/500 percent low). If the sensor is 3 degrees off, it will read 500 x cos (3°) = 499.31 milligauss, or about 1/7 percent low. The sensor will always read low unless pointed in exactly the direction of the field. In theory, this fact could be used to find the actual field strength. You could just look at a digital display of the field strength that the sensor measures, and then “tweak” the direction of the sensor until you see the highest possible strength. If the sensor is correctly pointed in the direction of highest field, then any direction you tip the sensor away from this correct direction will make the digital display read a lower number. That “highest” number is in fact the true strength of the field. The problem with this method is that it is slow—it may take literally a minute to determine the field strength this way.
The Earth Magnetometer solves this problem by using an analog meter (needle-type) to tell you if you're tipping the sensor closer to or farther from the correct position. This analog meter does a continuous auto-zero to keep changes as small as .01 milligauss clearly visible while still being able to handle a several hundred milligauss dynamic range. The digital display shows the highest number that occurred (since the last reading) and holds that number. Therefore, if at any time, the sensor was briefly pointed within 1/3° of the correct direction, the digital display will show the correct strength of the field and will hold that number until it's time for the next reading. The correct reading will continue to be held even if you point the sensor in a completely wrong direction. To take the next reading, just press the “reset” button. A reading can be done every 5 seconds this way.
This is not as fast as the more expensive geomagnetic meters; they generally require one second to make each reading. (Each measurement is a timed internal process, but some of the more expensive versions can read faster than once a second). However, one reading every 5 seconds is not extremely slow, because it usually requires 5 seconds or more to walk to the next measurement position.
The Earth Magnetometer also has a rapid scan function. This is only sensitive effectively down to 1 milligauss (but theoretically down to 0.1 milligauss, or 1/50 percent of the earth's field). It allows you to walk along, continuously reading the field. You can see “blips” as fast a 1/10 second.
The Earth Magnetometer also has an “offset” function that adjusts the display to a reading of zero +/- .01 milligauss at whatever time you want to. Then you can measure how many nanoteslas stronger or weaker the field is, compared to your starting point. (This offset function only works on the more sensitive range, not the rapid scan range).
This meter has maximum error of .5% of the reading +/- .5 milligauss. Because .5% of 500 milligauss is 2.5 milligauss (=250 nanoteslas), this may seem to be a problem when reading changes in the earth's field from one point to the next with precision of .01 milligauss (= 1 nanotesla). However, if you're measuring how much the field varies from point to point, this amount of variance is accurate to +/-0.5% of the variance. For example, if you use the offset controls to set your first point at “0.00” milligauss, and you measure the next point as “-1.53” milligauss, then that next point really is 1.53 +/- .01 milligauss weaker (note “minus” sign) than the field at the first point. A change of up to 2.00 milligauss is actually correct +/- .01 milligauss, which is just the rounding error of the number. A change of 4.00 milligauss is accurate to +/- .02 milligauss. When doing these variance or differential measurements, the .5 milligauss maximum offset error will cancel itself and so does not introduce any inaccuracy. The other source of error is due to temperature change of the sensor; that error is +/- 1 milligauss/°C (.0056 milligauss/°F) at maximum. The biggest problem with measuring the earth's field is that some magnetism is created in the ionosphere, which is a layer of charged particles near the top of the atmosphere. There is no way to distinguish between the small contribution from the ionosphere and the big contribution from the earth; all types of meters simply read the sum of both fields. The earth's field stays fairly constant, but the ionosphere's field varies randomly from minute to minute. The variation is faster during the day when the sun is contributing solar wind to the ionosphere. A typical variation between day and night is +/- .30 milligauss. Typical random fluctuation during the day is .10 milligauss in one hour, and during the night it is .03 milligauss in an hour. However, these numbers can be much higher during strong solar activity. To see plots of these variations at specific locations, look up “magnetogram” on the internet. These are usually plotted in nanoteslas.
Fortunately, the variation contributed by the ionosphere is fairly uniform over distance on the surface of the land. That is, if the ionosphere is currently adding .13 milligauss at a certain location, it will probably be adding .13 +/- .01 milligauss everywhere else within a square kilometer, because the ionosphere is roughly 100 kilometers up. This is sufficiently distant that the ionosphere cannot create "high-resolution" magnetic patterns on the land (with significant changes between points separated by only 1 km). This means that if you need an accurate grid map of how the field varies from point-to-point, go back to the reference point and re-measure it periodically, no matter which type of geomagnetic meter you're using. The actual map of variance of the field from point-to-point remains the same from day to night, even though the field strength at every point may rise or fall. This is because the field strength at all locations rises or falls by the same amount from hour to hour. If you periodically return to re-measure the same point, you can make a graph of how much variation the ionosphere is actually contributing. This will clarify how reliable the measurements are.