This project is studying data from seismometers and borehole strainmeters in southern California to understand better what signals they can and cannot detect. The detection levels are largely controlled by noise in the earth and in the instruments, and the project aims to characterize this, especially in two areas. One is at the frequencies of seismic waves, where it is important to understand the variations in noise level from natural sources and human activities. The other is at periods of days to months, over which the strainmeters can be affected by groundwater changes. The project will be examining strainmeter and pumping data to develop models of how this effect works.
As part of the NSF-funded Plate Boundary Observatory, about 75 instruments, called borehole strainmeters (BSM's) were installed to make very precise measurements of how the ground changes shape. Our research looked at data from these instruments with the primary goal of finding out how they performed, and in particular what created noise on them and how this affected their ability to measure the smallest changes in the ground. We found that for recording earthquakes (not their primary purpose) these instruments were not as good as the seismometers already used. We also looked in detail at a natural experiment, where two BSM's were installed, unintentionally, in an area where water was being pumped from several wells. We monitored the well pumping and could relate this to signals seen on the strainmeters, to fund out about how water flow and changes in pressure took place in the surrounding rocks. To better determine the calibration of the BSM's, we compared data from them with data from long-base strainmeters located nearby. These other data showed that the earthquake waves seen were often different from what we would expect, in ways that come from the uneven distribution of different types of rick deep in the Earth. In the spring of 2013 a magnitude 4.7 earthquake caused all these instruments to record an unusual signal that started at the time of the earthquake and tapered off over the following weeks, a sign that the nearby San Jacinto fault was rapidly slipping at depth without produding eathquakes, a somewhat unusual behavior. This is still being studied.
