A local magnetic anomaly may change substantially even when the intrinsic magnetization of the source rocks remains constant.
The mechanism considered here is purely magnetostatic:
several independently magnetized rock bodies contribute magnetic fields at a fixed observation point;
one contribution is oppositely directed relative to the others;
the vector contributions therefore partially cancel;
physical displacement and/or rotation of one magnetic body changes its field at the observation point;
cancellation decreases;
the measured magnetic anomaly increases.
No spontaneous increase in rock magnetization is required.
No electromagnetic amplification mechanism is required.
No change in the Earth's main magnetic field is required.
The phenomenon follows directly from linear superposition of magnetic fields and the (r^{-3}) spatial dependence of a compact magnetic dipole.
The total magnetization of a magnetic rock can be represented as
\mathbf{M}{R}
+
\mathbf{M}{I},
]
with
\kappa\mathbf{H}_{0},
]
where:
(\mathbf{M}_{R}) = natural remanent magnetization;
(\mathbf{M}_{I}) = induced magnetization;
(\kappa) = magnetic susceptibility;
(\mathbf{H}_{0}) = ambient geomagnetic field.
Thus,
\mathbf{M}{R}
+
\kappa\mathbf{H}{0}
}
]
The direction of (\mathbf{M}_{R}) need not coincide with the present geomagnetic field.
This is critical.
USGS studies demonstrate that natural remanent magnetization can equal or exceed induced magnetization in volcanic rocks and may dominate observed magnetic anomalies.
Consequently, two lithologically similar magnetic rocks may produce magnetic fields having substantially different orientations if their remanent magnetization vectors differ or if the blocks have subsequently been mechanically rotated.
For a compact, approximately uniformly magnetized body whose dimensions are sufficiently small relative to the source-observer distance, the external field may be approximated by a magnetic dipole.
The magnetic dipole moment is
\mathbf{M}V,
]
where:
(M) is magnetization in A m(^{-1});
(V) is volume in m(^3);
(m) is magnetic moment in A m(^2).
The magnetic field of a dipole is
\mathbf{m}
\right]
}
]
where
4\pi\times10^{-7}\ \mathrm{H,m^{-1}}.
]
This is the standard magnetostatic dipole solution.
The principal scaling relation is therefore
[
B\propto\frac{MV}{r^{3}}.
]
Thus, source-observer geometry is highly influential.
If distance doubles,
\frac{B_{1}}{8}.
]
If distance triples,
\frac{B_{1}}{27}.
]
The field does not need to be destroyed. Moving the source away is sufficient to make its local contribution much smaller.
Magnetic fields obey linear superposition.
For (N) magnetized bodies,
\sum_{i=1}^{N}\mathbf{B}_{i}.
]
For three sources,
\mathbf{B}{A}
+
\mathbf{B}{B}
+
\mathbf{B}_{C}.
]
A scalar total-field magnetometer does not simply measure the arithmetic magnitude of every individual source. For anomalies that are small relative to the Earth's main field, the first-order total-field anomaly is approximately
[
\boxed{
\Delta T
\approx
\hat{\mathbf{F}}{0}
\cdot
\mathbf{B}{a}
}
]
where (\hat{\mathbf{F}}_{0}) is a unit vector parallel to the local main geomagnetic field.
It is therefore useful to define
\hat{\mathbf{F}}{0}\cdot\mathbf{B}{i}.
]
Then
[
\Delta T
\approx
b_A+b_B+b_C.
]
A contribution may be positive or negative.
Hence:
[
b_A+b_B>0,
]
while
[
b_C<0
]
is entirely physically permissible.
The term cancellation must not be interpreted as magnetic shielding.
Rock C does not form a barrier between the detector and rocks A and B.
Instead,
[
\mathbf{B}_{A+B}
]
and
[
\mathbf{B}_{C}
]
simultaneously exist at the observation point.
If their projections along the measurement direction are opposite,
\Delta T_{A+B}
+
\Delta T_C.
]
For example,
+50.0\ \mathrm{nT}
]
and
-40.4\ \mathrm{nT}
]
give
9.6\ \mathrm{nT}
}
]
All magnetic sources remain present.
The observed value is small because the source fields partially cancel at that particular observation coordinate.
Suppose body C is subsequently displaced or rotated while A and B remain stationary.
Then
[
\mathbf{M}_{C}
]
may remain constant in the body coordinate system, but
[
\mathbf{r}_{C}
]
and/or the orientation of (\mathbf{M}_{C}) relative to the observation point changes.
Therefore,
[
\mathbf{B}{C}^{(0)}
\neq
\mathbf{B}{C}^{(1)}.
]
If its projected contribution changes from
[
-40.4\ \mathrm{nT}
]
to
[
-1.8\ \mathrm{nT},
]
while A and B remain at
[
+50.0\ \mathrm{nT},
]
then
48.2\ \mathrm{nT}
}
]
The observed ratio becomes
5.0208.
]
The measured anomaly has therefore increased by approximately a factor of five.
However,
[
M_A,\ M_B,\ M_C
]
need not have changed at all.
The amplification is geometrical.
The following numerical model is not a claim concerning measured subsurface blocks at any specific site. It is an existence demonstration showing that the required anomaly can be obtained using magnetizations within empirically observed ranges for natural basaltic rocks.
IODP measurements of basaltic samples report natural remanent magnetization values spanning approximately
[
1.3\ \mathrm{A,m^{-1}}
\quad\text{to}\quad
73\ \mathrm{A,m^{-1}},
]
with a median near
[
6.9\ \mathrm{A,m^{-1}}.
]
Accordingly, representative values of
[
M=4\text{–}8\ \mathrm{A,m^{-1}}
]
are not extraordinary for magnetized volcanic material.
Assume each body is approximated as a sphere of diameter
[
D_{A}=D_{B}=0.80\ \mathrm{m}.
]
Radius:
[
a=0.40\ \mathrm{m}.
]
Volume:
\frac{4}{3}\pi a^{3}
]
\frac{4}{3}\pi(0.40)^3
]
0.2681\ \mathrm{m^{3}}.
]
Assume
[
M_A=M_B=4.0\ \mathrm{A,m^{-1}}.
]
Their equivalent magnetic moments are
[
m_A=m_B=MV
]
4.0(0.2681)
]
1.072\ \mathrm{A,m^{2}}.
]
For a dipole observed approximately on its magnetic axis,
\frac{\mu_0}{4\pi}
\frac{2m}{r^{3}}.
]
Converting tesla to nanotesla gives
[
\boxed{
B_{\mathrm{axial}}[\mathrm{nT}]
\approx
200\frac{MV}{r^{3}}
}
]
for (M) in A m(^{-1}), (V) in m(^3), and (r) in metres.
At
[
r=2.047\ \mathrm{m},
]
each source produces approximately
[
B_A
\approx
B_B
\approx
25.0\ \mathrm{nT}.
]
Therefore,
[
\boxed{
B_{A+B}
\approx
50.0\ \mathrm{nT}
}
]
at the selected observation point.
Assume body C is smaller:
[
D_C=0.60\ \mathrm{m},
]
therefore
[
a_C=0.30\ \mathrm{m}.
]
Its volume is
\frac{4}{3}\pi(0.30)^3
]
0.1131\ \mathrm{m^{3}}.
]
Assume a stronger natural remanent magnetization,
8.0\ \mathrm{A,m^{-1}}.
]
Then
M_CV_C
]
8.0(0.1131)
]
0.9048\ \mathrm{A,m^{2}}.
]
If C occupies a geometry in which its axial contribution is opposite to the A+B contribution, then at a centre-to-sensor distance of approximately
1.648\ \mathrm{m},
]
its contribution is
[
B_C
\approx
-200
\frac{(8.0)(0.1131)}
{(1.648)^3}
]
[
\approx
-40.4\ \mathrm{nT}.
]
Thus:
\boxed{9.6\ \mathrm{nT}}.
]
No unusual field strength is required.
No fivefold increase in magnetization is required.
For fixed magnetization and orientation,
[
B\propto r^{-3}.
]
The distance required to reduce the magnitude of C from 40.4 nT to 1.8 nT by translation alone is determined from
\left(
\frac{r_0}{r_1}
\right)^3.
]
Therefore,
\left(
\frac{1.648}{r_1}
\right)^3.
]
Hence,
1.648
\left(
\frac{40.4}{1.8}
\right)^{1/3}
]
[
\approx
4.65\ \mathrm{m}.
]
A distance-only model would therefore require an effective displacement of approximately
[
4.65-1.65
\approx
3.0\ \mathrm{m}.
]
This is possible, but it is not necessary.
Rotation provides a much more efficient mechanism.
For a general dipole geometry, define a dimensionless projection factor (G):
100
\frac{M_CV_C}{r_C^{3}}G.
]
The exact value of (G) is determined by the mutual geometry of:
the remanent magnetization vector;
the source-observer vector;
the regional geomagnetic-field direction.
In the ideal axial case,
[
|G|=2.
]
In other configurations,
[
G
]
may pass continuously through zero and may change sign.
This means that a magnetic block can retain exactly the same intrinsic magnetization while its measured total-field contribution approaches zero simply because its orientation or position changes.
For example, with
[
M_C=8.0\ \mathrm{A,m^{-1}},
]
[
V_C=0.1131\ \mathrm{m^{3}},
]
and
[
r_C=2.50\ \mathrm{m},
]
the condition
[
\Delta T_C=-1.8\ \mathrm{nT}
]
requires
\frac{
(-1.8)(2.50)^3
}{
100(8.0)(0.1131)
}
]
-0.311.
]
Since
[
-2\leq G\leq 2
]
for ordinary dipole geometries and projected components, a value of
[
G=-0.311
]
is entirely admissible.
The transition
[
G=-2
\quad\rightarrow\quad
G=-0.311
]
can be produced by a combination of translation and physical rotation of the rock.
Natural remanent magnetization is fixed to the mineral fabric of the rock.
If a boulder rotates mechanically,
[
\mathbf{M}_{R}
]
rotates with the boulder.
It does not automatically realign with the present geomagnetic field.
Consequently, a block that initially produces a strong negative projection at a surface observation point may produce:
a weaker negative projection;
nearly zero projection;
or a positive projection
after rotation.
This follows directly from the dipole term
\mathbf{m}.
]
A comparatively modest translation combined with substantial rolling or rotation can therefore alter the measured anomaly much more efficiently than translation alone.
For volcanic-rock bulk densities of approximately
2.8\text{–}3.2\times10^{3}
\ \mathrm{kg,m^{-3}},
]
body C, with
[
V_C=0.1131\ \mathrm{m^{3}},
]
would have an approximate mass
\rho V
]
of
[
317\text{–}362\ \mathrm{kg}.
]
Bodies A and B, with
[
V=0.2681\ \mathrm{m^{3}},
]
would each have masses of approximately
[
751\text{–}858\ \mathrm{kg}.
]
The cancelling source therefore need not be a multi-tonne monolith.
A sub-metre magnetic boulder can be sufficient if:
its remanent magnetization is several A m(^{-1});
it lies within approximately 1–2 m of the observation point;
its magnetic moment is favourably oriented.
The mechanism is intrinsically reversible.
Let configuration 0 be
[
\mathbf{r}{C}=\mathbf{r}{C0},
]
[
\mathbf{M}{C}=\mathbf{M}{C0}.
]
Then
[
\Delta T_0
\approx
9.6\ \mathrm{nT}.
]
After translation/rotation,
[
(\mathbf{r}{C0},\mathbf{M}{C0})
\rightarrow
(\mathbf{r}{C1},\mathbf{M}{C1}),
]
giving
[
\Delta T_1
\approx
48.2\ \mathrm{nT}.
]
If the block is subsequently returned sufficiently close to its original position and orientation,
[
(\mathbf{r}{C1},\mathbf{M}{C1})
\rightarrow
(\mathbf{r}{C0},\mathbf{M}{C0}),
]
then
[
\boxed{
\Delta T
\rightarrow
9.6\ \mathrm{nT}
}
]
to within measurement error, environmental variation, and imperfect mechanical repositioning.
The process therefore behaves observationally like a reversible magnetic switch, although no electrical switching mechanism exists.
The correct physical interpretation is:
[
\boxed{
\text{the observed anomaly increased}
}
]
but
[
\boxed{
\text{the intrinsic magnetization of the fixed sources did not need to increase}
}
]
and
[
\boxed{
\text{the fivefold ratio does not imply a fivefold increase in magnetic moment}.
}
]
The initial value
[
9.6\ \mathrm{nT}
]
is a residual:
50.0-40.4.
]
The later value
[
48.2\ \mathrm{nT}
]
is another residual:
50.0-1.8.
]
The difference is
38.6\ \mathrm{nT}.
]
Thus the observational transition can be generated by a change of approximately
[
38.6\ \mathrm{nT}
]
in the projected contribution of one pre-existing magnetic source.
Suppose a primary instrument records
[
48.2\ \mathrm{nT}
]
and a second instrument records
[
47.6\ \mathrm{nT}.
]
The discrepancy is
0.6\ \mathrm{nT}.
]
Relative to 48.2 nT,
1.24%.
]
Thus both instruments are effectively detecting the same approximately 48 nT local anomaly.
Independent replication does not determine its cause, but it substantially weakens the hypothesis that the increase is merely an idiosyncratic failure of one magnetometer.
A shallow magnetic-reconfiguration model makes a testable prediction.
Because
[
B\propto r^{-3},
]
the anomalous change should be strongly spatially localized.
Measurements displaced laterally from the original point should not necessarily reproduce the same
[
9.6\rightarrow48.2\ \mathrm{nT}
]
transition.
Instead, mapping should produce a high-gradient anomaly field containing positive, negative, or near-zero regions depending on geometry.
Therefore, a scientifically appropriate follow-up would be a dense magnetic grid rather than repeated measurement at only one station.
Such a survey could determine whether the anomaly behaves like:
a compact shallow magnetic source;
a broad geological body;
a temporal geomagnetic disturbance;
or an instrumental artifact.
The numerical configuration above is demonstrative, not unique.
An infinite family of combinations of
[
M,
V,
r,
]
and source orientation can satisfy
[
\Delta T_0=9.6\ \mathrm{nT}
]
and
[
\Delta T_1=48.2\ \mathrm{nT}.
]
Accordingly, magnetic measurements alone cannot determine whether the responsible source consists of three bodies, two bodies, a distributed magnetized fracture zone, an irregular breccia body, or another geometry.
A complete inversion would require:
multiple spatial magnetic measurements;
sensor height and orientation control;
temporal base-station correction;
rock susceptibility measurements;
natural remanent magnetization measurements;
geological constraints on source depth;
three-dimensional forward modelling.
The three-body configuration is therefore best understood as the minimum intuitive model demonstrating the principle of reversible vector cancellation.
The transition
[
9.6\ \mathrm{nT}
\rightarrow
48.2\ \mathrm{nT}
]
does not require a rock body to become approximately five times more magnetic.
It is sufficient for the initial measurement to represent a partially cancelled field:
+9.6\ \mathrm{nT}.
]
If the cancelling magnetic body is translated and/or rotated so that its projected contribution becomes approximately
[
-1.8\ \mathrm{nT},
]
the same fixed positive sources produce
+48.2\ \mathrm{nT}.
]
Returning the cancelling source to its previous geometry restores approximately the previous anomaly.
The mechanism requires no violation of conventional magnetostatics.
The unresolved question in any real field case would not be whether such a magnetic reconfiguration is physically possible.
It would be:
[
\boxed{
\text{what physical process changed the position or orientation of the magnetic source?}
}
]
U.S. Geological Survey — Books, K. G. (1962), “Remanent magnetism as a contributor to some aeromagnetic anomalies.”
USGS publication record
U.S. Geological Survey — Bultman, M. W. (2015), “Detailed Interpretation of Aeromagnetic Data from the Patagonia Mountains Area, Southeastern Arizona.”
USGS Scientific Investigations Report 2015–5029
U.S. Geological Survey — Roberts, C. W. & Jachens, R. C. (1999), “Preliminary aeromagnetic anomaly map of California.”
USGS publication record
MIT OpenCourseWare — Essentials of Geophysics: geomagnetism, magnetic dipoles, rock magnetism, and magnetic anomalies.
MIT OpenCourseWare — Essentials of Geophysics, Chapter 3
University of Virginia — Magnetostatics: magnetic dipole field equation.
University of Virginia — Magnetostatics II
Integrated Ocean Drilling Program — natural remanent magnetization measurements in basaltic samples; reported range approximately 1.3–73.0 A m⁻¹, median 6.9 A m⁻¹.
IODP — paleomagnetic and rock-magnetic measurements of basaltic samples