Nonlinear and minor ocean tides in the Bay of Biscay from the strain tides observed by two geodetic laser strainmeters at Canfranc (Spain)

Abstract

This work presents and discusses the tidal analysis of about 2.5 years of strain data recorded by two 70 m‐long high‐resolution laser strainmeters, which are operating in the Canfranc underground laboratory (Central Pyrenees, Spain). Spectra show clear tidal peaks whose frequencies range from the diurnal band to at least 8 cycles per day; strain amplitudes (relative change in length for each strainmeter) range from few 10−12 to 10−8. The reliability of observations and corrections for local distortions are testified by the excellent agreement between measurements and computations for all the diurnal and semidiurnal tides included in the TPXO8 and FES2012 tidal atlases, with the exception of L2. Observed higher‐frequency strain tides are mostly attributed to loading by nonlinear shallow‐water constituents in the Bay of Biscay, more than 120 km from the measurement site. The signals are quantitatively compared with computations using TPXO8 (MN4, M4, and MS4) and FES2012 (M3, N4, MN4, M4, MS4, and M6). Computations fully agree with M4 observations for one strainmeter, overestimate M4 by about 30% for the other strainmeter and M6 by about a factor of two, and underestimate the other tides. This work shows that data from high‐sensitivity strainmeters installed in high‐quality sites may provide valuable additional information for studying the nonlinear tidal dynamics and energetics of coastal waters and minor ocean tides, at spatial resolutions of tens to thousands square kilometers, depending on the strainmeter location with respect to the coastline.

Parameters of the Earth’s Free Core Nutation from Diurnal Strain Tides

“…3) is probably of anthropic origin. As for the LSC interferometers (LAB780 and GAL16), we already demonstrated that air temperature and barometric pressure effects on strain are actually negligible on time scales of one day or shorter 28,32 .…”

“…Φ , N 2 and M 2 at each interferometer, obtained (i) computing the theoretical tidal strain components by using the reference 3 frequency-dependent Love and Shida numbers, T FCN , and Q FCN , (ii) combining them by using reasonable local effect coefficients 24,32 , and (iii) assigning them the uncertainties we got from the tidal analysis of the real extension records. Parameters…”

Different Couplings Between Precipitation and Deformation at the Same Site: A Case Study at Central Pyrenees (Spain)

“…The karst aquifer is probably very heterogeneous, but the schist unit may be considered approximately homogeneous; we expect mechanical anisotropy to be of little influence on strainmeter data because compressional forces act horizontally and schist layering is almost horizontal (Figure 1f). Moreover, as for larger-scale phenomena, tidal strain (diurnal band to at least 8 cycles per day) recorded by the two strainmeters does not evidence any anomaly attributable to mechanical heterogeneity or anisotropy (Amoruso & Crescentini, 2016). Thus, for simplicity, we assume homogeneous isotropic conditions in the whole computational domain.…”

“…Although strain noise amplitude is inversely proportional to frequency, we do not filter low frequencies out, but for easier visualization, we detrend strain data whenever we plot them. On time‐scales longer than tens of seconds and shorter than several days, the largest signal on a strain record is the Earth (body and ocean loading) tides and occasionally a large earthquake or a hydrology‐associated deformation event (e.g., Agnew, 2007); loading by nonlinear shallow‐water tides in the Bay of Biscay also gives non‐negligible deformation in the study area (Amoruso & Crescentini, 2016). Therefore, we subtract linear and nonlinear tides from the strain time series before further analysis (Amoruso & Crescentini, 2020).…”

Using Distributed Temperature Sensing for Long-Term Monitoring of Pockmark Activity in the Gulf of Patras (Greece): Data Processing Hints and Preliminary Findings

“…The above signifies that a missing-values prediction method based on spectral analysis would most likely make accurate predictions for most of the data variance, but it may not account for any random irregularities, such as extreme meteorological events, other anthropogenic factors (e.g., marine traffic), or events related to earth dynamics (i.e., earthquakes and sediment dynamics). Missing value imputation methods based on the frequency domain are either using the Fourier [55][56][57] or the wavelet (e.g., in [58]) spectrum of the time series.…”