Separating climate‐induced mass transfers and instrumental effects from tectonic signal in repeated absolute gravity measurements

Abstract

We estimate the signature of the climate‐induced mass transfers in repeated absolute gravity measurements based on satellite gravimetric measurements from the Gravity Recovery and Climate Experiment (GRACE) mission. We show results at the globe scale and compare them with repeated absolute gravity (AG) time behavior in three zones where AG surveys have been published: Northwestern Europe, Canada, and Tibet. For 10 yearly campaigns, the uncertainties affecting the determination of a linear gravity rate of change range 3–4 nm/s2/a in most cases, in the absence of instrumental artifacts. The results are consistent with what is observed for long‐term repeated campaigns. We also discuss the possible artifact that can result from using short AG survey to determine the tectonic effects in a zone of high hydrological variability. We call into question the tectonic interpretation of several gravity changes reported from stations in Tibet, in particular the variation observed prior to the 2015 Gorkha earthquake.

Geophysics From Terrestrial Time‐Variable Gravity Measurements

“…This is especially true for volcanoes, most of which are located along or surrounded by oceans, hence experiencing strong rainfalls, where water, either meteoric or marine, mixes with hot magmatic fluids and convects within porous rock and sediments.The separation can be done by three methods, as discussed hereafter: (1) the hydrological signal is known with a precision sufficient to allow subtraction of the hydrology signature from the gravity data, (2) one disposes of sensors with a response to hydrological load and tectonic effect different from that of the gravimeter, or (3) the space‐time behavior of the two signals differs to such an extent that it is possible to use data processing technique to separate them. The third method is not practical because of the sparsity in time and space of microgravimetric surveys (Van Camp, de Viron, & Avouac, ).…”

“…Gravity measurements are cumulative observations and have a particular space‐time transfer function. Hence, these measurements are extremely sensitive to local water storage changes, ocean loading, and atmospheric effects, which induce time‐correlated signatures in the gravity time series (Van Camp, de Viron, & Avouac, ; Van Camp et al, ). Neglecting these signatures leads to underestimating the uncertainties on the computed gravity rates of change (Agnew, ; Van Camp et al, ; Williams, ).…”

Resolving Geophysical Signals by Terrestrial Gravimetry: A Time Domain Assessment of the Correction‐Induced Uncertainty

“…Furthermore, the uncertainty of the long‐term trend was estimated by looking at the slope of a degree 1 polynomial fitted to the deviation between models. The uncertainty of trend estimation in absence of the hydrological correction is discussed in Van Camp et al (, ). In the differential mode, each period of interest was separated from the previous one (with shorter period) by an antialiasing zero‐phase low‐pass filter (see supporting information Figure S1).…”

“…These studies, however, did not quantify the uncertainty of the corrections that were applied to reduce the environmental noise. A different approach to assess the uncertainty of long‐term trends in terrestrial gravity measurements was applied in Van Camp et al () by considering the impact of water storage estimates of one model or observations (i.e., Gravity Recovery and Climate Experiment) on the trend estimate. Here we present the statistical uncertainty of a set of gravity corrections commonly applied to terrestrial gravity observations by comparing correction models that meet criteria as discussed in the following section.…”

Research Article. A new gravity laboratory in Ny-Ålesund, Svalbard

“…It has an accuracy of better than 10 µGal (1 µGal = 10 −8 m/s 2 ) at field sites, assuming that instrument standards are checked [5]. In contrast, the FG5 provides measurements with an accuracy of typically 1-2 µGal and allows gravity changes to be estimated with a precision of 0.5 µGal/yr after 10 years of annual measurements [35]. The raw AG-observations were processed with the gsoftware (version 9) provided by Micro-g LaCoste, the manufacturer of the A10 and the FG5 gravimeters.…”