Research ArticleGEOLOGY

Short-lived pause in Central California subsidence after heavy winter precipitation of 2017

See allHide authors and affiliations

Science Advances  29 Aug 2018:
Vol. 4, no. 8, eaar8144
DOI: 10.1126/sciadv.aar8144
  • Fig. 1 Subsidence rates and seasonal amplitudes.

    (A) Secular rates of displacement in the satellite’s line of sight (LOS) inferred from 40 Sentinel-1 acquisitions (November 2014 to November 2016), as well as the Corcoran clay extent (gray, white outline), rivers (black lines, dotted where inferred), major canals (black dashed lines), and faults (49) (red lines). Inset indicates location of (A) and (B) (red square), Tulare Basin (blue), and Corcoran clay extent (gray). (B) Amplitude of best-fit sinusoidal term. SAF, San Andreas Fault.

  • Fig. 2 Temporal variability of ground deformation.

    Displacement history at location of GPS station CRCN (Fig. 1) for Sentinel-1 (red dots) and GPS (black dots) projected into Sentinel-1 LOS (~38° from vertical), and best-fit model of secular + seasonal signal (green line), with daily stream discharge (light blue curve) at the U.S. Geological Survey (USGS) monitoring site at Deer Creek (https://waterdata.usgs.gov). The light gray–shaded area indicates the approximate time range of drought in California, with purple shading highlighting the rate changes that began in early 2017. We use GPS position solutions in the NA12 reference frame, available for download on the data products pages of the Nevada Geodetic Laboratory (http://geodesy.unr.edu).

  • Fig. 3 Displacement profiles and time series reference area.

    (A and B) Profiles of cumulative displacements colored by acquisition time [profile location in (C)]. Every fifth date is plotted. (A) Cumulative displacements after removing a planar function and mean offset from each date. Note that the large region of subsidence has biased this estimate, resulting in an artificial signal of apparent uplift along the margins, particularly near point P. (B) Cumulative displacements after removing mean offset calculated on the basis of subset of pixels defined by SD criterion. (C) SD of time series at each pixel after applying a low-pass filter (allowing only periods greater than 180 days). The black contour line (10 cm) represents the threshold SD used to separate deforming areas from nondeforming areas in offset correction. Nondeforming areas were used as reference pixels in the displacement time series. Note that even some of the smaller subsidence features associated with hydrocarbon extraction in the southwest are automatically identified using our approach.

  • Fig. 4 Spatiotemporal progression of deviation from the multiyear trend (α).

    Filled contours colored by date when LOS displacements deviated by more than 13 cm from model fit during peak drought. Time ranges are shown in inset plot (October 2016 to July 2017). Inset plot: α versus time at points 1 to 5. Light blue curve is a time series of daily stream discharge at the U.S. Geological Survey (USGS) monitoring site at Deer Creek (https://waterdata.usgs.gov) filtered with a 2-week moving average.

  • Fig. 5 Year-to-year differences in subsidence behavior.

    (A) Time series at two representative points as a function of the day of year for each of the three years, beginning on the date of the first available SAR acquisition (November 8; day of year, −53). Locations of examples 1 and 4 are the same as in Fig. 3 and are shown in map view in (B) and (C). Cumulative difference of first and second years (B) and second and third years (C) at each pixel, averaged over shaded time period shown in (A). Map symbols are the same as in Fig. 1. Differences less than 2 cm are transparent. This analysis assumes no model (for example Eq. 1) to characterize the data but still illustrates the spatial and temporal variability in subsidence, indicating that the effects that we observe are not just an artifact of the time period used in fitting Eq. 1.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/8/eaar8144/DC1

    Fig. S1. Data quality metrics.

    Fig. S2. GPS time series.

    Fig. S3. Time series at points.

    Fig. S4. Example seasonal signal.

    Fig. S5. Deviation from multiyear trend (α) averaged over March 2017.

    Fig. S6. Continued subsidence, 2017.

    Table S1. Previous observations of subsidence in the Tulare Basin.

    Movie S1. Spatiotemporal progression of deviation of subsidence from the multiyear trend.

    Data file S1. Text file containing a list of all dates of SAR acquisitions used in this study.

    References (5053)

  • Supplementary Materials

    The PDF file includes:

    • Fig. S1. Data quality metrics.
    • Fig. S2. GPS time series.
    • Fig. S3. Time series at points.
    • Fig. S4. Example seasonal signal.
    • Fig. S5. Deviation from multiyear trend (α) averaged over March 2017.
    • Fig. S6. Continued subsidence, 2017.
    • Table S1. Previous observations of subsidence in the Tulare Basin.
    • Legend for movie S1
    • References (5053)

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.avi format). Spatiotemporal progression of deviation of subsidence from the multiyear trend.
    • Data file S1 (.txt format). Text file containing a list of all dates of SAR acquisitions used in this study.

    Files in this Data Supplement:

Navigate This Article