An experimental study of the role of subsurface plumbing on geothermal discharge

Abstract

In order to better understand the diverse discharge styles and eruption intervals observed at geothermal features, we performed three series of laboratory experiments with differing plumbing geometries. A single, straight conduit that connects a hot water bath (flask) to a vent (funnel) can originate geyser‐like periodic eruptions, continuous discharge like a boiling spring, and fumarole‐like steam discharge, depending on the conduit length and radius. The balance between the heat loss from the conduit walls and the heat supplied from the bottom determines whether and where water can condense which in turn controls discharge style. Next, we connected the conduit to a cold water reservoir through a branch, simulating the inflow from an external water source. Colder water located at a higher elevation than a branching point can flow into the conduit to stop the boiling in the flask, controlling the periodicity of the eruption. When an additional branch is connected to a second cold water reservoir, the two cold reservoirs can interact. Our experiments show that branching allows new processes to occur, such as recharge of colder water and escape of steam from side channels, leading to greater variation in discharge styles and eruption intervals. This model is consistent with the fact that eruption duration is not controlled by emptying reservoirs. We show how differences in plumbing geometries can explain various discharge styles and eruption intervals observed in El Tatio, Chile, and Yellowstone, USA.

The Fascinating and Complex Dynamics of Geyser Eruptions

“…Laboratory studies have been used to show that steady heating and recharge can lead to episodic eruptions (Munby 1902, Forrester & Thune 1942, Steinberg et al 1982, to show how increasing complexity of plumbing geometries results in greater variation in discharge styles and eruption intervals (Namiki et al 2016), to understand the effects of geometry on convection and hence temperature in the conduit (Sherzer 1933), to show how increasing reservoir temperature increases the vigor of eruptions (Toramaru & Maeda 2013), to show how the decrease in reservoir pressure over the course of the eruption leads to recharge and the end of eruption (Lasic 2006), to show how bubble formation and collapse generates weak high-frequency tremors (Anderson et al 1978), and to show how intermittent modulation of the rate of boiling and the closely coupled accelerations and decelerations of the water column generate strong low-frequency tremors (Anderson et al 1978).…”

Geometry of Geyser Plumbing Inferred From Ground Deformation

“…These studies show how seismic and geodetic measurements over a range of frequencies can be used to characterize the plumbing systems of, and pressures within, geysers. Additional imaging tools include using microphones (e.g., Namiki et al, ), forward looking infrared (e.g., Karlstrom et al, ), and ground‐penetrating radar to characterize the shallowest subsurface (e.g., Lynne et al, , ).…”

Origin and Properties of Hydrothermal Tremor at Lone Star Geyser, Yellowstone National Park, USA

“…In contrast, a deeper and larger reservoir boils and evacuates in main eruptions with an interval strictly controlled by long-term recharge and heating rates. Significant pressure and temperature gradients might be present in a long and narrow conduit leading to variability in the preplay sequence (Namiki et al, 2016), whereas the larger reservoir is expected to be more uniform in thermodynamic state due to convection, resulting in a stable recurrence time for the main eruption. There are similarities among type-1, type-2 preplays, and main eruptions at LSG compared to small, medium, and large eruptions, respectively, observed in the Adelstein et al (2014) experiments.…”

“…Here we use seismic signals recorded for 4 days by multiple broadband seismometers around Lone Star Geyser (LSG), Yellowstone National Park (Figure 1), to improve the understanding of multiphase fluid storage and transport, before, during and after eruptions. We build on previous geophysical investigations at LSG (Karlstrom et al, 2013; Namiki et al, 2016; Vandemeulebrouck et al, 2014). These studies revealed complex eruption cycles in which the main eruptions that occur every ~3 hr are preceded by a variable number of preplay episodes.…”

“…To accomplish these goals, five broadband seismometers (Nanometrics Trillium P/PA sensors with a corner frequency of 120 s; data recorded by Tauras digitizers at 250 Hz sampling rate) were deployed at distances of ~9 to ~30 m from the geyser cone (Figure 1 and supporting information Text S1 and Figures S1–S4) between 14 and 17 April 2014. An infrared (IR) sensor directed toward the geyser vent was used as a proxy for liquid and steam discharge (Namiki et al, 2016).…”