Research ArticleGEOLOGY

Geomorphic expression of rapid Holocene silicic magma reservoir growth beneath Laguna del Maule, Chile

See allHide authors and affiliations

Science Advances  27 Jun 2018:
Vol. 4, no. 6, eaat1513
DOI: 10.1126/sciadv.aat1513
  • Fig. 1 Maps of postglacial lava flows in the LdM volcanic field [modified from (25, 34)].

    Shown are postglacial rhyolites (pink), rhyodacites (orange), and andesites (pale green). The presumed vent for the plinian eruption of the Rhyolite of LdM (unit rdm) is beneath the lake. The star in the inset shows the location of LdM in South America. Important map units include (i) rle, Rhyolite of Los Espejos that dammed the lake at 19 thousand years; and (ii) rcb, the many rhyolitic lavas of the Barrancas complex that were emplaced during the last 14 thousand years. These and other map unit abbreviations follow those in (25, 34).

  • Fig. 2 Map of preserved highstand paleoshoreline (color-contoured for elevation) and postglacial silicic lava flows.

    Map unit abbreviations follow those in (25, 33, 34). ka, thousand years ago (ka ago).

  • Fig. 3 Outcrops illustrating geomorphic expression of highstand paleoshoreline in select localities around the LdM basin.

    (A) Rounded boulders of 19.0 thousand years old Espejos rhyolite deposited on an angular outcrop of the same unit to define the highstand of the paleoshoreline (dashed yellow line). (B) Wave-cut terrace on 990 thousand years old Bobadilla rhyodacite ignimbrite north of LdM. The block below the hammer yields a 36Cl exposure age of 9.4 ± 0.4 thousand years. (C) Beach deposit of rounded boulders of Pleistocene andesite partially buried by rhyolitic ash and lapilli at the southeastern end of the lake basin (Fig. 2). The rhyolite lava from the Barrancas complex flowed over this shoreline terrace at 5.6 thousand years ago. (D) View north of the southern margin of the rhyodacite of Colada Dendriforme (unit rdcd in Figs. 1 and 2) where it flowed over a highstand paleoshoreline terrace (arrows). A glassy tower on the levee of the flow lobe on the right yields a 36Cl exposure age of 8.4 ± 1.2 thousand years that provides a lower limit for the age of the paleoshoreline. Photo credits: B.S.S. [taken on 1 January 2016 (A), 16 January 2015 (B), 8 January 2015 (C), and 29 March 2015 (D)].

  • Fig. 4 Model of single magma intrusion to explain Holocene surface deformation at LdM.

    (A) Map of interpolated elevations of the paleoshoreline (color scale in meters) relative to the lowest measured level to the northwest (NW) (2372 m), as measured at 64 GPS sites (black circles) on the highstand paleoshoreline around LdM (blue outline). The white circle indicates the location of the best-fit prolate spheroid deformation source estimated from inversion of the GPS data set. (B) The analytical solution for a pressurized prolate spheroid source (54) has seven free model parameters: location (x0, y0), depth (z0), dimensionless pressure change from which we calculate the total volume change (ΔV), spheroid aspect ratio (A = b/a), dip angle (θ, positive downward), and strike angle (ϕ, clockwise from north). Gray histograms show the initial random distribution and bounds for each parameter. Blue histograms show the best-fit values for each parameter and each of the 250 runs. Final best-fit value (minimum χ2) for each parameter and uncertainty estimates from Jackknife resampling are indicated on top of each panel. (C) Vertical displacement data (red error bars) and modeled values (blue squares) as a function of the radial distance from the source.

  • Fig. 5 Sketch illustrating a conceptual model of episodic deformation at LdM.

    Triangles on the upper x axis indicate the eruption ages of the rhyolite flows mapped in (25, 33) that issued from the Nieblas, Barrancas, Divisoria, and Cari Launa eruptive centers. Red filled triangles indicate flows dated by 40Ar/39Ar or 36Cl in (33) and this study, and gray filled triangles indicate ages inferred from field relationships; these flows have been placed at regular intervals between the bounding ages. The black square shows the cumulative vertical displacement estimate of 60 m over 9.4 thousand years. If the uplift rate had been constant over this time interval, then the long-term average rate would be 6.4 mm/year (dashed black line). The inset shows the temporal evolution of the vertical displacement as observed during the uplift episode that is ongoing since 2007 (red line) (32) and the projected rate assuming no further perturbations (dashed line). If all the deformation episodes share the same temporal and spatial characteristics, then 16 of these 50-year-long episodes would be needed to reproduce the 60-m total uplift of the paleoshoreline over the Holocene, with a magma intrusion average recurrence interval of 624 years.

  • Fig. 6 Schematic evolution of the upper crustal magma reservoir beneath LdM during the last 22 thousand years.

    The current Bouguer gravity low (36) is interpreted to include a smaller melt-rich body embedded within a larger, mostly crystalline, body centered under the lake. The gravity data suggest that the magma reservoir has begun to solidify into relatively dense crystal-rich mush beneath the southern end of the lake basin. Petrologic data (33, 34) imply that low-density rhyolitic melt segregates from crystal mush at depths of 4 to 6 km. The source of current inflation is best modeled as a sill opening at a depth of 4.5 km (31, 32). Thus, in the lowermost panel, we show that the partially molten zones and locus of magma injection coincide with the deeper portion of the inferred source of the Bouguer gravity low.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/6/eaat1513/DC1

    table S1A. Data from cosmogenic 36Cl dating samples.

    table S1B. Major element compositions of cosmogenic 36Cl dating samples.

    table S1C. Trace element compositions of cosmogenic 36Cl dating samples.

    table S2. Static GPS measurements of highstand paleoshoreline.

    fig. S1. Histogram of elevation differences between GPS and DEM.

    fig. S2. One-meter-resolution DEM and identification of highstand paleoshoreline outcrops.

    fig. S3. Estimated vertical displacement in response to crustal unloading from lake draining.

    fig. S4. Map of highstand paleoshoreline surface where it intersects the trace of the Troncoso fault.

  • Supplementary Materials

    This PDF file includes:

    • table S1A. Data from cosmogenic 36Cl dating samples.
    • table S1B. Major element compositions of cosmogenic 36Cl dating samples.
    • table S1C. Trace element compositions of cosmogenic 36Cl dating samples.
    • table S2. Static GPS measurements of highstand paleoshoreline.
    • fig. S1. One-meter-resolution DEM and identification of highstand paleoshoreline outcrops.
    • fig. S2. Histogram of elevation differences between GPS and DEM.
    • fig. S3. Estimated vertical displacement in response to crustal unloading from lake draining.
    • fig. S4. Map of highstand paleoshoreline surface where it intersects the trace of the Troncoso fault.

    Download PDF

    Files in this Data Supplement: