Understanding MEMS-based digital seismic sensors

Over the last few decades almost all the electronic devices we use in our daily lives have switched from analog to digital owing to the numerous benefits they offer, such as miniaturization, enhanced functionalities or reduced power consumption, etc. This revolution also reached the seismic industry, primarily when fully digital recorders and telemetries became available in the late 1970s. However, measurement of the Earth’s displacement by sensors remained analog. The last step towards full digital recording was made in the early 2000s with the launch of digital seismic sensors based on MEMS (Micro Electro Mechanical Systems) accelerometers that had the potential to replace analog geophones that had been used since the early days of seismic dating back to the 1930s. However, although digital sensors have established a foothold and indeed even become a reference for a number of different applications, the anticipated revolution has however not lived up to the expectations of the seismic industry, the current situation being perhaps analogous to using audio cassette tapes alongside MP3 music files. More than a decade after the release of the first digital seismic sensors, several explanations for such a situation have become clear: • the operating principles of digital sensors are somehow perceived to be more complicated than those of geophones, and trickier to understand; • the quality of data acquired with digital sensors is sometimes judged to be no better than equivalent to data acquired with geophones (more often than not, this is owing to a comparison of single digital sensors with strings of geophones at the same trace interval); • the cost of digital sensors is often seen to be higher (at least for one-component – 1C – acquisition) than the cost of a conventional Field Digitizing Unit (FDU) connected to a string of geophones. This article aims firstly to review the main operating characteristics of MEMS-based Digital Sensor Units (DSU), especially when compared to geophones, and then discuss the quality and cost issues in the light of experience gained after more than a decade of field operations. We conclude that their use with an adapted high-density geometry makes it possible to achieve much better imaging than with geophone arrays, for an equivalent cost.