Earthquakes and Water

“Earthquakes and Water” Course

Lecturer: Professor Shih-Jung Wang, Ph.D (Assistant Professor in Graduate Institute of Applied Geology, National Central University)


Earthquakes and Water [Link]

Earthquakes and Water in Encyclopedia of Complexity and Systems Science [Link]

Earthquake Hidrology, book chapter in Treatise on Geophysics [Link]

Hydrologic Responses to Earthquakes and a General Metric, book chapter in Frontiers in Geofluids [Link]

Continental-scale water-level response to a large earthquake, book chapter in Crustal Permeability [Link]

Fluid Induced Seismicity [Link]

Involvement of Fluids in Earthquake Ruptures [Link]


Scientists in this topic (earthquakes and water)

Prof. Michael Manga [Link 1] [Link 2]

Prof. Chi-yuen Wang [Link 1] [Link 2]

Prof. Emily Brodsky [Link 1] [Link 2]

Prof. Evelyn A. Roeloffs [Link 1] [Link 2]


Scientists in particular topic of fluid induced seismicity:

Prof. Paul Segall [Link 1] [Link 2]

Prof. Serge A. Shapiro [Link 1] [Link 2]

Prof. Teruo Yamashita [Link 1] [Link 2]

Prof. Shemin Ge [Link 1] [Link 2]

Prof. Stephen Andrew Miller [Link 1] [Link 2]


My paper reviews (Presentation will not be shown)

Paper Review 1

Paper Review 2

Paper Review 3

Paper Review 4

Final Report


Some notes and resume:

Chapter I: Introduction

– studied about interaction between earthquakes and water (fluids)

– Stress from earthquake can be large; mainly two types i.e. static stress and dynamic (transient) stress [see Freed et al., 2005, Link]

– Static stress ΔCFS, decrease rapidly with distance (time independent) caused by permanent displacement of fault (near field) [decay with distance as 1/R^3]

– Dynamic stress ΔCFS(t) (time dependent) [decay with distance as 1/R^1.66]

– Stress perturbation from earthquake can enhance crustal permeability

– Concept of permeability enhancement can be seen in Rojstaczer et al. (Nature, 1995) [Link]

– Empirical formula from Elkhoury et al. (Nature, 2006) [Link]

ΔK = R. v/c where Δk is the change in permeability at the time of the earthquake, v is PGV in the vertical and c is the phase velocity of the seismic waves. The ratio v/c is approximately the imposed strain on the system, so R measures the permeability response to strain. The parameter R is a property of the well–aquifer system and is different for each well.

– Stress changes also can induced fluid pressure changes.

– All concept related to: dynamic strain and seismic energy density (dissipated energy density).

– Basic theories: groundwater flow, groundwater transport and hydro-mechanical coupling.


Chapter II: Liquefaction

Keypoints: rigidity is reduced to zero, then sediments become fluid-like.

– Mainly based on the principle of effective stress: [Download here about The Terzaghi Theory of Effective Stress]

σ’zz = σzz – αP

effective stress = vertical stress – [alpha x pore pressure]

alpha = empirical parameter

or can be rewritten σ’ = σ – μ   –>  if σ = μ, then σ’ = 0

Means: pore pressure generated by surface load acts to reduce σ’zz by an amount of αP.

It means, when sediments consolidate in an undrained condition during earthquake, pore-pressure increase and then liquefaction occurred.

When loading from earthquake come, pore pressure become high (P increase) and then σ’zz decrease to ~ 0. Sediments become fluid like (liquefy).

Important notes: In most cases, the liquefied sediments are sand (pasir) or silty sand (lumpur). Sometimes + well-graded gravel (kerikil).

(a) Liquefaction in near field: Terzaghi’s principle of effective stress; undrained consolidation.

[Seismic shaking can disturb grain-to-grain contacts to cause sediments to consolidate; vertical stress decrease, pore pressure increase, effective stress vanishes; sediments become fluid-like]

– Laboratory studies: Cyclic loading experiments, dissipated energy for liquefaction by undrained consolidation

– Threshold of diss. energy density to induce liquefaction: 30 J/m^3

(b) Liquefaction beyond near field: factors: hypocentral distance, style of Eq faulting, directivity effect, sediment type and grain size.

– Seismic energy density (e) as a metric for liquefaction distribution; note: e ~ PGV^2 [consistent with field data]: make empirical relation among e, r and M.

– Beyond near field: e below the threshold (30 J/m^3) to induce undrained consolidation

Possible mechanism beyond near field: mechanism of pore-pressure spreading from nearby sources to sediment sites!

– For mitigation (as a mid-term question): (a) assess potential/vulnerability of soil liquefaction (liquefaction zoning); (b) microbial denitrification; (c) analysis and well-designed stucture/facilities; (d) understanding soil ageing; (e) densification, drainage and solidification.

– An unresolved issue is the complex relationship between liquefaction and the frequency of seismic waves. Current results from the field and laboratories are in conflict. Future work is needed to resolve these conflicts.


Chapter III: Mud volcanoes

Paper to read: Bonini et al., (2016) [Link]

(a) Static stress changes can trigger eruptions through:

(1) Expanding magma reservoirs

(2) Compressing magma reservoirs

(3) Unclamping magma ascent paths (Indonesian; unclamping dalam bahasa Indonesia artinya seperti melepaskan jepitan)

(b) Dynamic stress changes can trigger eruptions through:

(1) Changing permeability

(2) Nucleating or mobilizing bubles

(3) Mobilizing magma

* In near field, triggering maybe combination of static and dynamic stress changes.

Other related papers:

Florian Fuchs, Dissertation, Dynamic triggering: The effects of remote earthquakes on volcanoes, hydrothermal systems and tectonics. University of Bonn, 2015 [Link]

Miguel Neves, Dissertation, Dynamic triggering of seismic activity in rifting and volcanic settings, University of Lisboa, 2016 [Link]

Maxwell Rudolph, Dissertation, Mechanical Controls on Eruptions, University of California Berkeley, 2012 [Link]

Matthew Weingarten, Dissertation, On the interaction between fluids and earthquakes in both natural and induced seismicity, University of Colorado Boulder, 2015 [Link]

Hill et al. (2002), Earthquake-Volcano Interaction [Link]

Manga and Brodsky (2006), Seismic Triggering of Eruptions in the Far Field: Volcanoes and Geysers [Link]

Manga et al. (2009), Earthquake triggering of mud volcanoes [Link]

Linde and Sacks (Nature, 1998), Triggering of volcanic eruptions [Link]

New papers: Maestrelli et al. (2017] [Link]; Nishimura et al. (2017) [Link]

Difficulty to assess static stress changes to volcanoes: how to define feeder Dyke as receiver of stress.

Solution: Nakamura, 1997 [Link]; Nostro et al. (1998) [Link]; Link to Coulomb 3.3 software: click here

Calculation of static stress changes: Click Here to see my tutorial


Chapter IV: Increased Stream Discharge

Paper to read: Mohr et al. (2017) [Link], Wang and Manga (2015) [Link]

What we need: extensive networks of stream gauges, exact location of the upstream of particular gauge(s), weather observation (weather station).

Some phenomenons: instant waterfall created by earthquake faulting, decreases of downstream discharge caused by the damming of mountain valleys by landslides and rockfalls, increases of discharge in regions of high relief caused by the avalanche of large quantities of snow to lower elevations, increasing the supply of melt water.

Significant example: Stream flow increase after 1989 Mw 6.9 Loma Prieta earthquake.

Problems that need to be solved: How to calculate the amount of “excess” stream discharge after earthquakes.

Need to be eliminated: influences of local precipitation.

Characteristics of increased discharge:

(a) Recession Analysis (postseismic baseflow recession analysis and its characteristic time)

Analyze the baseflow: the component of stream discharge from groundwater seeping (Indonesian: merembes/mengalir keluar) into the stream.

ln Q = a – c t

where Q = discharge of the stream, a and c = empirical constants for the linear fit.

τ ≡ 1/c (time scale that characterizes the rate at which the groundwater discharge decreases), where c as the recession constant.

One dimensional model of aquifer: extend from one end of the aquifer at x=0 to the other end at x=L (I have resumed this model in a Pdf file, click here to download it).

c ≡ – ∂ log Q / ∂t ≈ π^2 D / 4L^2   OR   τ ≡ 4L^2 / π^2 D  [ D is hydraulic diffusivity D = K/Ss where K = horizontal hydraulic conductivity and Ss = specific storage).

(b) Estimate excess discharge

The solution to calculate the excess of discharge based on one dimensional model proposed by Wang and Manga (2015) can be seen in a Pdf file that I have prepared [Link]. The amount of earthquake-induced excess discharge can be estimated by fitting the stream flow data with the equation.

*Proposed mechanisms for river discharge after earthquake:*

(a) Expulsion of deep crustal fluids resulting from coseismic elastic strain

(b) Enhanced near-surface permeability

(c) Coseismic consolidation and liquefaction of near-surface deposits (water liberated from consolidation of soils or water mobilized from the unsaturated zone).

Debate about mechanisms: geochemical and temperature constraint, multiple earthquakes analysis, constraints from recession analysis, multiple stream gauges analysis, role of anisotropic permeability.

Other consideration: detail geology, climate, rupture directivity of earthquake, other surface water source.


Chapter V: Groundwater Level Change

# The most widely documented changes among all earthquake-induced hydrologic phenomena.

Paper to read: Continental-scale water-level response to a large earthquake, book chapter in Crustal Permeability [Link], Professor Shih-Jung Wang paper’s [Link 1; Link 2].

Observation: (1) A single well that responds to many earthquakes  OR  (2) many wells that respond to single earthquake.

What we need: many groundwater well measurements (dense network), weather observation (weather station), geology of the well site.

# Mechanisms:

(a) Near field: step-like changes

Hypothesis: (1) Static strain, (2) Undrained consolidation, (3) Energy to initiate undrained consolidation, (4) Seismic energy density.

See my tutorial to calculate static strain changes due to earthquake [Link]

Common empirical formula: log r = 0.48 M – 0.33 log e (r) – 1.4 where e (r) is seismic energy density (see Chapter 2).

(b) Intermediate field: Sustained changes (more gradual and can persist for hours to weeks)

Hypothesis: also use (1) static poroelastic volumetric strain induced by earthquakes (similar to near field), (2) Undrained consolidation to explain pore-pressure changes OR (3) Enhanced rock permeability by seismic wave, (4) Local pressure or production of localized liquefaction.

(c) Far field: Groundwater (transient) oscillations

Usually associated with long-period Rayleigh waves –> hydroseismogram!

Large water level oscillations occur when the geometric (water depth in well, well radius, aquifer thickness) and the hydrogeologic (transmissivity) properties have the right combination.

Commonly accepted mechanism: (1) Seismic waves may cause aquifers to expand and contract which in turn may cause pore-pressure to oscillate; (2) Enhancement of the permeability of a fractured aquifer that associated with an increase in the magnification factor, defined as the ratio of the spectrum of groundwater oscillation to the spectrum of the particle velocity of Rayleigh waves.

#Role of S-wave and Love Waves on Groundwater Oscillations: a special case example in Taiwan after M7.9 Wenchuan earthquake.

The association of water-level changes with Love and S-waves could be due to an anisotropic poroelastic effect.

#Pore-pressure changes on the sea floor: a special case example in Juan de Fuca ridge

Postseismic Groundwater Recession: 

Following the step-like changes in the groundwater level in the near field, post-seismic flow in the aquifer leads to a new equilibrium state (time-dependent recovery).

(a) Recession Analysis (similar to Chapter 4)

log h = a – bt

where h = postseismic residuals of the groudwater level; t = time; a and b are empirical constants to be evaluated from least-square fit to well data.

τ ≡ 1/b

(b) Interpretation of the Postseismic Recession

Build of simplified one-dimensional aquifer (similar with Chapter 4).

I have prepared a Pdf file to explain the postseismic recession [Link] that taken from Wang et al. (2017).

b ≡ – ∂ log h / ∂t ≈ π^2 D / 4L^2


Other Consideration: pattern of earthquake faulting (rupture), sampling rate of water measurement


Chapter VI: Temperature and Composition Changes

#skipped deliberately


Chapter VII: Geysers

A new review about geyser system has been done by Hurwitz and Manga (2017) [Link].

Paper to read: Hurwitz et al. (2014) [Link]

Geysers are springs that intermittenly [Indonesian: sebentar-sebentar, sekali-sekali] erupt mixtures of steam and liquid water.

Changes in the behavior of geysers is usually characterized by the interval between eruptions (IBEs).

There is no systematic pattern to responses after earthquakes.

Difficulty: A best approach to measure the discharge of a geyser (need additional measurements).

Hypotheses of geyser’s response to earthquake (Hurwitz et al., 2014):

(a) Affected by altering the permeability structure of the geyser’s conduit or reservoir, which could result in either an increase or a decrease of IBE and could be either irreversible (e.g., a new IBE state is attained) or reversible, where after some period with anomalous IBE, the intervals will return to the pre-earthquake values.

(b) Bubble nucleation from steam-saturated water can decrease the period required to attain conditions for an eruption. This process is expected to occur only when the water in the geyser reservoir is at, or near, boiling conditions and when seismic waves are reverberating locally.


Chapter VIII: Earthquakes Influenced by Water

This is a special chapter for me. There are some different text books for this chapter:

(1) Fluid Induced Seismicity [Link]

(2) Involvement of Fluids in Earthquake Ruptures [Link]

Two way relationship between earthquake and water.

Changes in pore-pressure can also induce earthquakes.

Note: Term of “induced” and “triggered”.

Trigger: adds a small contribution to the stress

Induce: plays a dominant role

(1) Fluids and Rock Failure

Related textbooks about poroelasticity:

(a) Poroelasticity, by Professor Alexander H.-D. Cheng [Link]

(b) A Method of Fundamental Solutions in Poroelasticity to Model the Stress Field in Geothermal Reservoirs, by Professor Matthias Albert Augustin [Link]

(c) Theory of Linear Poroelasticity with Applications to Geomechanics and Hydrogeology, by Professor Herbert F. Wang [Link]

(d) Elastic Storage of Aquifers by Professor Arnold Verruijt [Link]

Youtube link to a course named “Reservoir Geomechanics


(2) Earthquakes induced by fluid injection and extraction

(3) Reservoir-induced seismicity

(4) Natural hydrological triggering of earthquakes

(5) Earthquake triggering via hydrological processes


… to be continued