Water-level monitoring for crustal deformation studies was initially viewed as a
low-cost substitute for strainmeters: poroelastic theory implies that fluid pressure in an
isotropic porous rock varies in proportion to volumetric strain, with a coefficient in the
range of 30-100 cm of water/ppm. Observations of water-level variations induced by
earth tides and coseismic static strain support the idea that fluid pressure tracks crustal
strain, and can be monitored via water-level measurements in wells. (Unless otherwise
stated, fluid-pressure and water-level will refer here to the same physical variable.)
Nevertheless, as water-level and strain data have accumulated, primarily in
California and Japan, a new view of fluid pressure has evolved. Water-wells, as strain
sensors, have limitations, and acquiring reliable water-level data is not as cheap as
initially hoped. On the other hand, it is also now clear that subsurface fluid-pressure
changes induce strain changes that are recorded by borehole strainmeters, complicating
strain data interpretation. Observations of either borehole strain or water-level changes,
especially for pre-earthquake signals, are more readily accepted by the scientific
community when both types of data are available. There is increasing evidence that
fluid-pressure changes of tectonic origin are not necessarily proportional to volumetric
strain of the solid rock, implying that fluid-pressure is an independent physical variable
that cannot be inferred from, or substituted for, strain. The concept of the water-well as a
strainmeter has thus evolved into a view of fluid pressure and strain as two separate and
essential components of a complete crustal deformation measurement.