Inversion of multi-temporal geoelectrical data sets: insights from several case studies

Time-lapse inversion of geoelectrical data is increasingly growing as remote monitoring systems
are being used in more applications such as seawater intrusions, landslides, bioremediation of contaminated sites, landfill operations, shallow geothermal systems, or water resources.
To date, several inversion strategies exist for taking into account the temporal dimension of the data. Among the most used ones are the independent inversion of multi-temporal data sets, the difference inversion, the temporally-constrained inversion, and the more recent process-based
inversion. The success of a particular time-lapse inversion scheme depends on the validity of several assumptions made by these inversion schemes. Difference inversion schemes generally
assume that part of the noise contained in the data cancels out when working with temporal data
differences. Process-based inversion requires a more advanced knowledge of the system prior the
inversion. Temporally-constrained inversion on the other hand assumes that the changes are
localized and minor. We show in this paper using data sets with different time and spatial scales,
and with different degrees of geological complexity and resistivity contrasts, that the particular
success of a time-lapse inversion scheme is highly dependent on the temporal behaviour of the
noise estimation in the time-lapse data set and of the model-dependent resolution pattern of the
survey. We attempt to provide guidelines for successful quantitative interpretation of time-lapse
data sets whenever possible.