![Two distinctive magnetic components were identified by decomposition of the IRM acquisition curves using a cumulative log-Gaussian (CLG) function [Kruiver et al., 2001; Heslop et al., 2002; Roberston and France, 1994]: 1) A first component M is characterized by its low coercivity. It dominates the magnetic response, although the contribution percentage and mean coercivity field (B 1/2 ) values are different in each unit. The coercivity of this component is lower and very homogeneous in PV1 (B 1/2 : ~43.7-43.9 mT) except for the lower sample (B 1/2 : ~47.9 mT), intermediate in PV2 (B 1/2 : ~47.9-51.3mT)](https://www.researchgate.net/profile/Maria-Bogalo/publication/317552393/figure/fig2/AS:667792458137603@1536225537375/Two-distinctive-magnetic-components-were-identified-by-decomposition-of-the-IRM_Q320.jpg)
Content uploaded by María Felicidad Bógalo
Author content
All content in this area was uploaded by María Felicidad Bógalo on Oct 16, 2017
Content may be subject to copyright.
Magnetic signature of the June 22nd, 1932 tsunami deposits (Jalisco,
Mexican Pacific coast)
Bógalo, M.F.1, Ramírez-Herrera, M-T.2, Goguitchaichvili, A.3, Rey, D.4, Mohamed, K.J.4 ,
Calvo-Rathert, M.1 and Corona, N.5
1 Departamento de Física, Escuela Politécnica Superior, Universidad de Burgos, Avda.
Cantabria s/n, 09006 – Burgos, Spain.
2 Laboratorio Universitario de Geofísica Ambiental & Instituto de Geografía, Universidad
Nacional Autónoma de México, Circuito de la Investigación, Ciudad Universitaria,
Coyoacán, D.F. 04510, México.
3 Laboratorio Universitario de Geofísica Ambiental, Instituto de Geofísica, Universidad
Nacional Autónoma de México, Unidad Michoacán, UNAM-Campus Morelia, México.
4 Departamento de Geociencias Marinas y Ordenación del Territorio, Universidad de Vigo,
36310 – Vigo, Spain.
5 Centro de Estudios en Geografía Humana, El Colegio de Michoacán A.C., Michoacán,
México.
Corresponding author: María Felicidad Bógalo (email: mfbogalo@ubu.es)
Key Points:
• Successful characterization and identification of tsunami induced deposits by rock-
magnetism and mineralogy.
• Different provenance of the sediment deposits related to two distinct tsunamigenic
events PV1 and PV2.
• Magnetic properties could reflect the effect of selective transport and deposition
during each individual tsunami event.
Research Article
Geochemistry, Geophysics, Geosystems
DOI 10.1002/2016GC006752
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/2016GC006752
© 2017 American Geophysical Union
Received: Dec 05, 2016; Revised: May 26, 2017; Accepted: May 26, 2017
This article is protected by copyright. All rights reserved.
Abstract
Recent studies have demonstrated that rock-magnetic analysis may provide additional
information to distinguish and characterize extreme marine inundation events such as
tsunamis. Rock-magnetic proxies reinforce and improve the environmental evidences
supplied by other methods, adding some decisive clues for the interpretation of the origin and
genesis of the sedimentary deposits. Here we report rock-magnetic, XRD and SEM
microscopy results obtained in the Palo Verde estuary (Colima Pacific coast, Mexico) in
order to enhance the tools for identification and reconstruction of two tsunami-induced
deposits. The sedimentary sequence includes two sand units, a tsunami deposit (PV1)
associated with the 22 June 1932 tsunami and a deeper sandy layer (PV2) related to a
possible palaeotsunami that occurred around 1300 CE. Both sandy units are topped by finer
grained units.
Magnetic properties exhibit a significant correlation with the stratigraphy. High susceptibility
(χ) and high saturation isothermal remanence (SIRM) values typical of high concentrations of
(titano)magnetite are a distinctive feature of the most recent sandy tsunamigenic unit PV1
and the overlaying soil. The lower sandy tsunamigenic unit PV2 shows significantly lower χ
and SIRM values, indicating lower concentration of (titano)magnetite in this unit and the
overlaying clayey-silt unit. The latter also shows a higher coercivity component associated to
(titano)hematite. Magnetic grain-size differences are also observed between PV1 and PV2
suggesting differences in hydraulic conditions at the time of deposition. The bulk
mineralogical composition and sediment texture of these units also supports the hypothesis of
different provenances for each tsunamigenic unit as inferred from magnetic properties.
1. Introduction
Tsunami deposits are generated as a consequence of onshore inundation and
backwash flow of large water bodies and exhibit many distinct characteristics, ranging from
variations in the mineral composition and distribution to particle size range [Atwater and
Moore, 1992; Dawson and Shi, 2000; Moore et al., 2006; Morton et al., 2007; Richmond et
al., 2011; Ramírez-Herrera et al., 2007, 2012; Goto et al., 2015]. During the last decade,
several multidisciplinary studies have been carried out to identify and characterize extreme
inundation deposits. [e.g. Kortekaas and Dawson, 2007; Ramírez-Herrera et al, 2007, 2012;
Font et al., 2010, 2013; Engel and Brückner, 2011; Wassmer et al., 2011; Shanmugam, 2012;
This article is protected by copyright. All rights reserved.
Cuven et al., 2013; Goto et al., 2014, 2015]. Since 2010, magnetic parameters such as
magnetic susceptibility and anisotropy of magnetic susceptibility (AMS) have proved highly
resolving in tsunami deposit research [Font et al., 2010; Wassmer et al., 2010]. So far, several
studies have applied these techniques in environments in which tsunami or paleotsunami
deposits are present [Ramírez-Herrera, 2012, 2014, 2016; Cuven et al., 2013; Font et al.,
2013; Goguichaichvili, 2013; Schneider et al., 2014; Černy et al., 2015; Kain et al., 2014,
2016; Wassmer et al., 2011, 2015].
On the other hand, some of these studies have also included more detailed magnetic
measurements, strengthening the identification and characterization of tsunamigenic deposits
in stratigraphic records [Font et al., 2010; Goguitchaichvili et al., 2013; Veerasingam et al.,
2014; Černy et al., 2015; Ramírez-Herrera, 2016], and providing useful proxies that reinforce
the information supplied by the tsunamigenic signature detected by more traditional methods
(stratigraphical, geomorphological, sedimentological, geochemical, microfossil,
palynological, mineralogical, archaeological) [e.g. Ramírez-Herrera et al., 2007, 2012;
Shanmugam, 2012; Spiske et al., 2013]. The basis for the use of magnetic properties as an
environmental assessment tool is rooted on the principle that magnetic properties are
sensitive to the mineralogy, concentration and grain size distribution of magnetic particles,
which can be altered by geology, transport, sorting and post-deposition processes.
The Pacific coast of Mexico (Figure 1a) is seismically active, and the subduction
beneath the North American plate of both the Rivera and Cocos plates has produced large
earthquakes and tsunamis in historical time. Typically, moment magnitude (Mw) 7.3 – 8.2
earthquakes occur about every 80 years [Singh et al., 1985], and larger Mw > 8.5 have
occurred less frequently (e.g. 28 March 1787 Mw 8.6 earthquake, off the Guerrero and
Oaxaca coast [Suarez and Albini, 2009]). On 3 June 1932, as a result of the rupture of the
Rivera-North America plate interface, a surface-wave magnitude (Ms) 8.2 earthquake was
recorded, the biggest one in the last century. Two aftershocks on 18 June (Ms 7.8) and 22
June (Ms 6.9) followed the main shock [Singh et al., 1981]. More recently, on October 1995,
a Mw 8.0 earthquake hit the southern coast of Colima (Figure 1a). All of the above
mentioned earthquakes produced tsunamis [Cumming, 1933; Sánchez and Farreras, 1993;
Corona and Ramírez-Herrera, 2012]. However, the 22 June 1932 event (the second most
destructive recorded in the Pacific coast of Mexico) caused an abnormally large one,
considering its Ms 6.9 magnitude [Okal and Borrero, 2011; Corona and Ramírez-Herrera,
2012].
This article is protected by copyright. All rights reserved.
Several studies were already carried out in tsunamigenic deposits at Mexican Pacific
coast, which proved the effectiveness of environmental magnetic techniques for improving
the identification of the tsunami-induced deposits [Goguichaisvili et al., 2013; Černý et al.,
2015; Ramírez-Herrera et al., 2016]. The identification and characterization of the magnetic
minerals in the tsunamigenic units helped to solve questions about the sediment provenance
and the energy of the events. In general terms, different mineralogy as well as magnetic
grain-size characterize these energetic events and distinguish from the background sediment.
The aim of the present study is the magnetic characterization and identification of historical
tsunami deposits, specifically those of the 22 June 1932 event and other prehistoric events in
the Palo Verde (PV) estuary (Cuyutlán, Colima).The analysis of multiple proxies, including
magnetic parameters, is especially useful in regions as the Pacific coast of Mexico, because
the conservation of geologic information is affected by the inherent environmental
characteristics of the tropics and frequent storms (hurricanes). Ramírez-Herrera et al. [2014]
identified two probable tsunamis in the Palo Verde estuary analyzing different multi-proxy
data, including both magnetic susceptibility (χ) and AMS. We performed new detailed
magnetic and mineralogical measurements and re-evaluated these preliminary results, clearly
improving the characterization of the sedimentary deposit bearing tsunamigenic units.
2. Geological setting and sample materials
The Palo Verde estuary (PV), East of Cuyutlán, Colima is located behind a sand-
barrier overlain by a series of sand dunes ca. 7.5 m above sea level (masl) near the PV site
(Figures 1b and 1c). Inland, the Laguna Cuyutlán, a back-barrier lagoon, is fringed by
mangrove swamps. The NW portion of Laguna Cuyutlán is characterized by saltpans (locally
known as “Salinas”). The study site is located behind ca. 7 m sand dunes and in a coastal
plain occupied by mangrove marshes (Figure 1c). Intermediate to basic extrusive igneous
rocks characterize the north of the studied area and granites are locally exposed. [Lancin and
Carranza, 1976]. Locally, the sands from the shore are rich in iron oxides [Carranza-Edwards
et al., 2009] and the beach berm and intertidal beach zone show enrichment of heavy
minerals [Ramirez-Herrera et al, 2014]. This is because the Cuyutlán lagoon barrier beach is
constructed by sediments originating in the Armería River, whose mouth lies 7 km southeast
of Palo Verde. Volcanic rock fragments are transported by the longshore currents, which
have SE-NW directions [Carranza-Edwards et al., 2009]. The Colima volcano, the
historically most active volcano from Mexico [Sulpizio et al., 2014] is located 80 km NE of
This article is protected by copyright. All rights reserved.
Palo Verde site. This volcano belongs to the Quaternary Colima volcanic complex [Verma
and Luhr, 1993]. Lower Cretaceous andesite outcrops corresponding to this volcanic activity
are found just 3,5 km NE of the study area [SGM, 2000].
The PV estuary may be considered as well-suited for tsunami deposit studies because
it provides a vast amount of information which may be retrieved from a survey performed
after the 1932 earthquake and tsunami [Cumming 1933; Corona and Ramírez-Herrera 2012;
Corona and Ramírez-Herrera, 2015] and from an investigation of the well-preserved
geological signatures of past tsunamis that has identified the 1932 and a predecessor tsunami
[Ramírez-Herrera et al., 2014]. In addition, Corona and Ramírez-Herrera [2012] and
Ramirez-Herrera et al. [2014] reported tsunami-scour fans connected with the over-wash flow
in Cuyutlán, broken dune features formed by tsunami erosion and remnant pedestals located
in the estuary a few hundred meters inland.
In the present study, samples for magnetic experiments were collected from a trench
wall (PV) located approximately 270 m inland at an elevation of 2.0 masl (18.8987 °N,
104.0338 °W) (Figures 1b and 1c). Sampling was performed in situ in the trench using
standard acrylic cubes for magnetic property analyses. Cubes were inserted in the wall from
the surface to the bottom of the trench sequentially with no intervals in between, to recover a
complete section of the stratigraphic units (Figure 1d) (for a detailed explanation of the
technique see Wassmer and Gomez, 2011; Ramírez-Herrera et al., 2012; Cerny et al., 2015).
The samples measured in this study are the same as those previously used for the AMS
analysis by Ramírez-Herrera et al., [2014].
The Palo Verde profile can be divided into five distinct units (Figure 1d) [Ramírez-
Herrera et al., 2014]. Two of them are sand beds associated to tsunamigenic events: PV1, at
10 – 32 cm depth and PV2, between 45.5 and 77 cm. Both units exhibit geological signatures
typical of common tsunami deposits (normally graded sand beds, scattered pebbles near the
base, erosional lower contact, increase in abundance marine-brackish diatoms, low organic
content, increase in elemental concentration of paleosalinity geochemical indicators)
[Ramírez-Herrera et al., 2014]. The other units consist of an uppermost sandy soil (0 – 10
cm) and a bioturbated orange-brown clayey-silt unit (32 – 45.5 cm) interbedded between PV1
and PV2 that shows large differences in grain-size, organic content and geochemical
composition [Ramírez-Herrera et al., 2014]. The lowest silty-clayish sand unit (77 – 91.5 cm)
was not completely sampled because it is located below the water table. Ramírez-Herrera et
This article is protected by copyright. All rights reserved.
al. [2014] suggested that the sand unit PV1 was linked to the 22 June 1932 tsunami and that
PV2 is related to a possible older event that happened around 1300 CE (1284 – 1389 CE).
3. Methods
Samples were collected as explained above from a 80 cm deep trench by using 8 cm3
plastic boxes and 35 samples were obtained. Low-field magnetic susceptibility (χ) expressed
in mass-unit (m3/kg) was measured with a Kappabridge KLY4 (AGICO Ltd., applying a 300
A/m field). A dual frequency (470 and 4700 Hz) Bartington MS2 susceptometer was utilized
to evaluate the contribution of fine-grained ferrimagnetic particles near the SP/SD threshold
based on the frequency-dependent susceptibility parameter (χfd% = 100x(χfd/χlf), where χfd =
χlf – χhf, χlf and χhf being magnetic susceptibilities at low and high frequency, respectively)
[e.g. Maher, 1986, Dearing, et al. 1996]. Anhysteretic remanent magnetization (ARM) was
given to the samples by demagnetizing them in a peak alternating field (AF) of 100 mT in the
presence of a 0.05 mT direct current (DC) bias field using the 2G cryogenic magnetometer
in-line system (2G-enterprise). The ARM susceptibility (χARM) was obtained by dividing the
mass-normalized ARM by the DC bias field. A variable field translation balance (MMVFTB)
was used to perform the following series of experiments: isothermal remanent magnetization
(Mr or IRM) acquisition curves (1 T peak field), hysteresis loops (±1 T), backfield curves in
order to measure coercivity of remanence, and magnetization vs. temperature (Ms – T)
thermomagnetic curves in a 38 mT DC field. The latter experiment was performed in air,
heating up to 700ºC and cooling down to room temperature with a 30 ºC/min rate. Hysteresis
loops after subtraction of paramagnetic and diamagnetic contributions allowed determination
of hysteresis parameters like saturation magnetization (Ms), saturation remanent
magnetization (Mrs or SIRM) and coercive field (Bc). S-200 ratio was determined as (1−IRM-
200mT/SIRM)/2 [Bloemendal et al, 1992]. The RockMagAnalyzer 1.1sofware [Leonhardt,
2006] was used to visualize and interpret these parameters.
Thirteen samples were selected to perform detailed stepwise IRM acquisition curves
up to 2 T using a M2T-1 pulse magnetizer and were consolidated with non-magnetic plaster
for these experiments. Twenty five to twenty seven field steps from 17 to 2000 mT were used
to analyze the IRM curves and evaluate the magnetic coercivity components [Kruiver et al.,
2001; Heslop et al., 2002]. IRM acquisition curves were approximated by a number of log-
Gaussian (CLG) curves, using the IRM-CLG Excel spreadsheet developed by Kruiver et al.
This article is protected by copyright. All rights reserved.
[2001]. Each curve is characterized by its SIRM intensity (IRM at 2 T), mean coercivity
(B1/2) and dispersion parameter (DP) [Roberston and France, 1994]. Thereafter, a composite
IRM with 2, 0.4 and 0.12 T fields applied in orthogonal directions was imparted to the
samples and subsequently thermally demagnetized in 17 temperature steps from room
temperature up to 680ºC, following Lowrie [1990].
Several representative samples of the trench (at least 2 per unit) were selected at
depths where relevant changes were observed in geochemical and magnetic properties as well
as sedimentological features. The presence of minerals associated with these features was
analyzed through Field Emission Scanning Electron Microscopy (FE-SEM) at the University
of Vigo using a JEOL JSM-6700F microscope. The observations were performed on carbon-
coated epoxy resin embedded polished samples using the FE-SEM operating in back
scattering mode (BS). The energy dispersive X-ray spectroscopy probe from the electron
microscope was used to analyze relevant minerals and determine their chemical composition
semi quantitatively by means of an EDS Oxford Inca Energy 300 SEM.
XRD determinations and semi-quantitative analyses were performed with a
SIEMENS D-5000 diffractometer at the University of Vigo on representative samples from
the modern beach (inter and supratidal areas), and from the trench (topsoil, PV1, clayey silt
unit and PV2).
4. Results
4.1. Magnetic mineralogical results
Magnetic susceptibility (χ), saturation remanence (Mrs or SIRM at 1 T), and
saturation magnetisation (Ms) are indicators of magnetic concentration. Observed curve
shapes are similar for these studied parameters, with the exception of Ms in the soil and the
uppermost 20 cm of PV1 (Figure 2), indicating that the magnetic signal is mainly influenced
by the concentration of ferromagnetic minerals (sensu lato). The values of these parameters
are relatively high and homogeneous for most PV1. χ decreases sharply at the base of PV1
and downwards into the clayey-silt unit, where a local minimum is reached at the maximum
silt horizon. Then χ steadily increases downwards along all the PV2 unit. The sharp decrease
of the values of magnetic properties at the bottom of PV1 coincides with the sharp basal
contact between the tsunami-induced deposit and the underlying clayey-silt layer.
This article is protected by copyright. All rights reserved.
Figure 3a shows a good correlation between χ and SIRM values (r
2 = 0.9813)
confirming that ferromagnetic minerals control the magnetic susceptibility. Two clusters
associated with the upper (uppermost soil and PV1) and the lower (clayey-silt unit and PV2)
parts of the profile can be distinguished in the figure. The lowermost sample from PV1,
located at 31 cm falls into the PV2 cluster and follows the tendency showed by the samples
belonging to the clayey-silt unit. The anhysteretic remanent magnetization (ARM) is sensitive
to the concentration of single domain (SD) ferrimagnetic particles (~ 0.03 – 0.06 μm) [e.g.
King et al., 1982; Liu et al., 2012]. The ARM profile (Figure 2) displays quite a different
behavior than the parameters described above, both in the uppermost and in the lowest part of
the profile. Therefore, the observed differences from the other intensity-related magnetic
parameters may be due to variations in the concentration of fine SD ferrimagnetic grains. On
the other hand, the concentration of ferrimagnetic grains depends on the texture since SIRM
and χ shows linear correlation with the sand percentage (Figs. 2 and 3b). However, two
trends with different slopes corresponding to the upper (soil and PV1) and the lower (clayey-
silt and PV2) part of the profile can be observed (correlation parameter r2 of 0.56 and 0.623,
respectively) (Fig. 3b).
The relative concentration of ferri- and antiferromagnetic minerals can be inferred
from the S-200 ratio. Their values are close to 1 indicating that low coercive magnetic minerals
dominate the magnetic remanence (Fig. 2). However, slight differences are observed
depending on the facies. Clayey‐silt show the lowest values, indicating that the contribution
of high coercive magnetic mineral is slightly higher in this unit than in PV1 and PV2. In the
bivariate Mrs (SIRM) versus Bcr plot (Figure 3c) it is possible to clearly observe two clusters
corresponding to the different units: soil–PV1 and PV2. Samples from the “PV2 cluster”
contain less concentration of ferrimagnetic minerals but display higher coercivities compared
to those from the “PV1 cluster”. Samples from the clayey-silt unit show the most extreme
behavior, with the lowest concentration of ferrimagnetic minerals and the highest
coercivities. This slightly higher coercivity in the clayey-silt unit than in the tsunamigenic
units could be both an effect of the magnetic grain size of the ferrimagnetic minerals (SD
grains displaying a higher coercivity than MD grains) and the higher contribution of high
coercivity minerals.
4.2. Magnetic domain-state
This article is protected by copyright. All rights reserved.
Magnetic domain-state distribution along the profile can be inferred using frequency
dependent susceptibility (absolute χfd and percentage χfd%) and other combined ratios such as
Mrs/χ, χARM/χ, ARM/Mrs or the hysteresis ratios Mrs/Ms and Bcr/Bc (Figure 4). Frequency
dependent susceptibility is commonly used to determine the concentration of
superparamagnetic (SP) ferrimagnetic grains near the superparamagnetic/stable single
domain (SP/SD) boundary [e.g. Foster et al., 1994; Dearing, 1996]. χfd% values are very low
in the entire profile, varying between 3.25% in the soil layer and 0.6% in the clayey-silt unit.
Although the relative contribution of these fine grains is not significant, the absolute values
of χfd in the uppermost part of the profile (20 to 31·10-8 m3/kg) are high and comparable to
those obtained in previous environmental studies [e.g., Maher, 1998; Torrent et al., 2007;
Bógalo et al., 2014; Lyons et al., 2014] in which the pedogenic magnetic grain contribution
was important.
Small particles of magnetite yield higher Mrs/χ, χARM/χ and ARM/M
rs ratio values,
whereas low values are related to coarser grain sizes [e.g. Evans and Heller, 2003; Peters and
Dekkers, 2003; Liu et al., 2012]. Susceptibility (χ) is a parameter not only affected by
ferromagnetic minerals (sensu lato) but also by diamagnetic, paramagnetic and SP grains.
The most powerful parameter detecting the relative abundance of fine SD magnetic grains is
ARM/Mrs, since remanences exclude these paramagnetic/diamagnetic and SP contributions.
χARM/χ and ARM/M
rs profiles show similar behavior with depth. The values of these
parameters increase with depth in the tsunamigenic units PV1 and PV2 until 69 cm. The
lowest values observed in PV1 suggest that the relative amount of coarser ferrimagnetic
minerals is higher in this unit than in PV2. A significant shift is observed in the intermediate
clayey-silt unit, which could be associated not only with ferrimagnetic grain size variations
but also with a minor contribution of high coercivity minerals. The ratio M
rs/χ is quite
uniform in PV1, showing values slightly lower than in PV2 and corroborating the magnetic
grain size differences between PV1 and PV2 units observed with the previous parameters.
The values of the Mrs/Ms and Bcr/Bc ratios are rather constant in the whole profile, although in
PV1 they are slightly lower than in PV2. The relationship between these two ratios is shown
in a Day plot, together with theoretical mixing curves for magnetite [Day, 1977; Dunlop,
2002] (Figure 5). The experimental values fall into the pseudo-single-domain (PSD) area and
samples from units PV1 and PV2 can be clearly distinguished. The values of the PV1 unit are
well grouped, displaying a more homogeneous behavior, whereas those of PV2 are more
scattered following the SD+MD trend and above PV1 data. Deviations from the theoretical
This article is protected by copyright. All rights reserved.
SD+MD trend could be due to several reasons including minor contributions of high
coercivity minerals or variations in the Ti-content of titanomagnetite grains [Dunlop, 2002;
Özdemir and Dunlop, 2014].
The bivariate χARM versus χ plot or King plot [King, et al. 1982] (Figure 6a) shows
clear differences in grain size. In this diagram, changes in slope are related to grain size
variations, smaller grains yielding steeper slopes. Changes along a line of constant slope are
indicative of magnetite concentration variations. Samples from the clayey-silt unit and those
from the upper part of PV2 (up to 65 cm) follow a very good linear correlation (r2 = 0.92),
indicating a rather homogeneous grain size distribution in that part of the profile. This
behavior differs greatly from the one showed in the upper layers (soil and PV1) and in the
bottom of the profile. In this case, nonlinear correlation data are probably due both to changes
in concentration of magnetite and in grain size. In addition, soil and PV1 samples are close to
the coarser magnetite PSD-MD grain-size area whereas those from the lower units (clayey-
silt and PV2) are within the SD-PSD region. These grain size differences can also be
observed in Figure 6b. Two clusters (soil-PV1 units and the lower part of clayey-silt and PV2
unit) could also indicate that the mineralogy of the upper part of the profile mainly consists of
coarser (titano)magnetite grains with lower coercivity whereas in the lower part finer grains
of (titano)magnetite are associated with higher coercivites. Samples from the upper part of
the clayey-silt unit with the highest coercivity values are not grouped in any cluster.
4.3. Magnetic mineral identification
IRM acquisition data modeling was performed in thirteen selected samples (Table S1,
Figure 7). Two distinctive magnetic components were identified by decomposition of the
IRM acquisition curves using a cumulative log-Gaussian (CLG) function [Kruiver et al.,
2001; Heslop et al., 2002; Roberston and France, 1994]: 1) A first component M is
characterized by its low coercivity. It dominates the magnetic response, although the
contribution percentage and mean coercivity field (B1/2) values are different in each unit. The
coercivity of this component is lower and very homogeneous in PV1 (B1/2: ~43.7 – 43.9 mT)
except for the lower sample (B1/2: ~47.9 mT), intermediate in PV2 (B1/2: ~47.9 – 51.3mT)
and higher in the clayey-silt unit (B1/2: ~50.1 – 55 mT) (see Table S1).This component could
be identified as detrital (titano)magnetite of different grain size and/or Ti-content; 2) A
second component H of higher coercivity has been identified in the lowest sample of PV1,
This article is protected by copyright. All rights reserved.
just at the boundary between PV1 and the clayey-silt unit, and in the clayey-silt and PV2
units. We want to stress the fact that in PV1 this component can only be found in the
lowermost sample. The concentration of this second component varies between 2.6 and 5.9
%, being highest in the clayey-silt unit. Its mean coercivity (B1/2) lies between 371.5 and
446.7 mT in the clayey-silt unit and between 346.7 and 398.1 mT in PV2. This component
could be related to fine-grained Ti-hematite.
The results of the thermal demagnetization of 3-axial IRMs are showed in Figure 8.
Slight differences can be observed among the soft component (0.12 T) curves, showing in all
samples a wide unblocking temperatures spectrum with maxima between 550 and 580 ºC,
typical of magnetite or titanomagnetite with different Ti-content. In addition, a variable
fraction of high coercivity minerals with maximum unblocking temperature higher than 650
ºC (2 T component), probably due to a small fraction of hematite, can be observed in the inset
plots. This component appears in all measured samples, even in samples from the soil and
PV1 in which it has not been possible to identify it by statistical decomposition of IRM,
probably due to its very low concentration. The contribution of this high coercivity
component is more evident in the clayey-silt unit.
A detailed Curie temperature (Tc) analysis has been carried out along the whole
profile (Figure 9). Figure 9a displays Curie point values for every stratigraphic position
calculated with the two-tangent method proposed by Grommé et al. [1969]. Two types of
thermomagnetic behavior can be observed in the heating curves: 1) curves of samples from
the clayey-silt intermediate unit with two different phases corresponding to (Ti)magnetite
(569 – 577 ºC) and a high temperature phase, probably titanohematite (667 – 672 ºC); 2) the
remaining curves exhibit a single phase and Curie temperatures typical of magnetite (or
titanomagnetite). In this last case, the Curie temperature for samples from PV1 (between 570
and 579 ºC) is around 30º C higher than for those from PV2 (between 548 and 556 ºC),
probably due to different Ti-content. Low-Ti-titanomagnetite/magnetite is identified as the
low Curie temperature phase. The difference in Ti-content identified in the tsunamigenic
units PV1 and PV2 could be related to a different origin of the magnetic grains. Samples
from the clayey-silt units and those situated just above and just below have more variable
values (546 – 577 ºC). In the cooling curves, Tc values of the low temperature phase are very
similar throughout the profile, mainly in the PV1 and PV2 units. In addition, the differences
between the Curie temperature values from the heating and cooling curves (ΔTc) for the
(Ti)magnetite phase are dissimilar in PV1 and PV2 (supporting information Figure S1). ΔTc
This article is protected by copyright. All rights reserved.
is ~ 30 – 38 ºC for PV1 and~ 7 – 15 ºC for PV2 (with exceptions at 47 and 69 cm). A
possible explanation of this behavior could be that changes in Curie temperature could arise
from cation reordering during heating in thermomagnetic experiments [Bowles et al., 2013].
4.4. Scanning electron microscopy (SEM) and XRD analyses
In general terms, the results show that top soil, PV1 and PV2 consist of lithoarenites
formed by intermediate to basic volcanic rocks including K feldspar and plagioclase, and
abundant ferromagnesian minerals, likely olivine and pyroxenes (Figure 10). The grains are
quite angular, poorly rounded and lack of typical subaqueous and high-energy surficial
marks, particularly quartz grains (Figure 11). The composition and texture of the framework
grains are very similar to the present-day beach. Matrix is thought to be mostly primary. It is
thought that primary matrix in PV1 and PV2 has inhibited the intense alteration of lithic
framework grains observed in the clayey-silt unit, which is incipient in the topsoil. The
occurrence of montmorillonite in the clayey-silt unit is interpreted as the result of the
alteration of the friable volcanic glass and lithoclasts, and some feldspar. Fe-rich minerals are
recognized by their brighter contrast. They occur both as discrete grains and forming part of
lithic volcanic clasts (Figure 11). Iron, and iron-titanium oxides are frequent in a wide range
on grain sizes, very often as small inclusions in lithic grains that preserve their original
igneous texture. In general terms, surficial textures on silicates lack the sorting and textural
characteristics observed in the beach sediments. It is thought that either they have not been
exposed sufficiently to wave action, that their original beach characteristics have been lost by
reworking, or a combination of both. Some indication of chemical leaching is observed,
although the sparsity of quartz grains makes difficult to assess the statistical significance of
these observations (Figure 11).
Mineralogical composition between PV1 and PV2 is nearly indistinguishable and very
similar to present day intertidal beach sand and topsoil. The proportion of quartz and
composition of the topsoil and PV1 framework grains is virtually identical (Fig. 10). Major
differences reside in the occurrence and preservation of a fine-grained clayey matrix in the
topsoil. The composition of PV2 is slightly different, particularly in the iron oxides. In
particular the leucoxene-type textures typical of titanomagnetites and/or ilmenohaematite are
more abundant, suggesting a different source. Both magnetic iron oxide grains and magnetic
This article is protected by copyright. All rights reserved.
mineral inclusions in rock fragments are less abundant in this unit. Elemental mapping
showed a fairly uneven distribution of Fe in the three studied units (Figure 10).
No evidence of significant diagenesis that may have explained the observed
differences in the magnetic properties between PV1 and PV2 was found, which is consistent
with the absence of early diagenesis promotion such as high organic matter content (i.e. low
Fe/χlf ratio of Reitz et al., 2004; Hoffman et al., 2005).
5. Discussion
5.1. Characterization of tsunamigenic units
Magnetic parameters correlate with the sedimentary stratigraphy (Fig. 2 and 4),
although this behavior is most evident in those parameters connected with type and
concentration of ferrimagnetic minerals. There are two most relevant characteristics observed
from the magnetic profiles: 1) the difference between the upper part (0 – 30 cm; soil and PV1
unit) and the units beneath (30 – 78 cm; clayey-silt and PV2 units); 2) the sharp change of the
magnetic properties just in the contact between PV1 and the lower clayey-silt unit.
Regarding the former first distinctive characteristic, χ, SIRM, Ms and ARM values in
the soil and PV1 unit are more than 20 times higher than those obtained in other sedimentary
deposits at the Mexican Pacific coast that include tsunamigenic units [Černý, et al., 2016;
Ramírez-Herrera, et al., 2016]. This higher concentration of magnetic minerals in the Palo
Verde site is attributed to contribution of local volcanic material. In fact, Carranza-Edwards
[2009] showed that beach sands in this area display enhanced iron-oxide content. The good
correlation between χ and SIRM (Fig. 2 and 3a) and the magnetic mineral identification
experiments (Fig. 7, 8, and 9) indicate that these rock magnetic parameters are mainly
controlled by the contribution of ferrimagnetic minerals (magnetite and Ti-magnetite) in the
whole profile. Small differences between χ and SIRM could be associated with variations in
the magnetic grain size distribution (Fig. 4 and 6) and/or minor high coercivity mineral
contributions (Ti-hematite) (Fig. 7, 8 and 9).
The results obtained modeling IRM acquisition data have showed that the
characteristics (SIRM, B1/2 and DP) of the dominant low coercivity component identified as
detrital (Ti)magnetite (component M) in PV1 are different from those obtained in other layers
beneath. The most distinctive characteristic is its high concentration, as is showed by SIRM.
This article is protected by copyright. All rights reserved.
The different values in both remanent coercivity (Bcr) and mean destructive field (B1/2) of this
component could be related not only to its diversity in grain size (Fig. 4, 6) but also to the Ti-
content (and maybe the oxidation state). Differences in grain size distribution between both
PV1 and PV2 units have been detected. Coarser (Ti)magnetite grain contribution is more
important in PV1 than in PV2. The values of ARM/SIRM in PV1 are indicators of MD+PSD
grains, whereas coarser SD dominates PV2 [Dearing et al., 1997; Peters and Dekkers, 2003].
Pedogenic fine-grained magnetite has mainly been detected in the upper part of the profile
(soil and the uppermost PV1). From SEM experiments a secondary montmorillonite-rich
matrix as the result of alteration of volcanic grains and fine-grained Fe-rich mineral could be
detected in the clayey-silt unit. However, frequency dependent susceptibility parameters (χfd
and χfd%) which are indicative of pedogenic magnetite exhibit in the clayey-silt unit the
lowest values of the whole profile. The high coercivity component (component H) (Fig. 7c-f)
has been identified as fine-grained (Ti)hematite with low Ti content. Although this
component has not been observed in thermomagnetic curves of the tsunamigenic units
because of its low concentration (Fig. 9), its presence in the clayey‐unit and PV2 is evident
from IRM experiments (Fig. 7 and 8). Since the (Ti)hematite contribution in PV2 is constant
and slightly lower than in the clayey-silt unit and there are no changes in its coercivity values,
we consider that this magnetic phase is not the result of a diagenetic process such as
magnetite oxidation. Rather this fine-grained (Ti)hematite of igneous origin could be a
product of high-temperature deuteric oxidation of (Ti)magnetite, constituting part of the
detrital particles in this unit [Dunlop and Ozdemir, 1997].
The distinctive behavior between PV1 and PV2 might be associated with a different
origin of these tsunamigenic units and/or different transportation processes of these magnetic
minerals. Because not only the concentration but also the coercivity of the magnetic
components are nearly constant in the PV1 unit, the sedimentation of PV1 and PV2 must
have taken place under very different morphodynamic conditions. Differences in the wave
energy and propagation direction, and in their transformation as they interact with the pre-
existing topography are sufficient to explain their distinct magnetic properties. Thus the
differences in the magnetic grain-size between PV2 (finer) and PV1 (coarser) could be
related, not only to provenance but also to the specifics of the sediment dynamics during each
event. This means that the event associated to the PV2 unit must have been different enough
to facilitate the deposition of fine-grained magnetic particles. The coarser more-porous
sediment texture of PV2 may have favored the percolation of the very fine-grained magnetic
This article is protected by copyright. All rights reserved.
grains through the unit. Whilst the generally finer-grained and fining upwards sequence
observed in PV1 may have prevented such distribution of the very fine magnetic grains in the
unit. If they were ever available for deposition during that event.
Most magnetic parameters exhibit a sharp change between PV1 and the underlying
clayey-silt unit. Ramírez-Herrera et al. [2014] using a battery of multi-proxy data
(stratigraphic, particle size, geochemical, foraminifera, diatom, ostracods,
historical/ethnographic) identified a sharp erosional basal contact between PV1 and the
clayey-silt unit. It was related to probably significant erosion produced by the large energy of
the 22 June 1932 tsunami event. In addition, radiometric dating evidenced a marked hiatus of
about 500 years between PV1 and the underlying unit. The magnetic results could also
support this notable variation in detrital input in PV1 with respect the unit beneath suggested
by Ramírez-Herrera et al. [2014], thus further validating the magnetic measurements as very
useful proxies to improve the characterization of tsunamigenic units.
5.2. Provenance of the tsunamigenic deposits
Previous studies carried out in this sediment profile also identified some differences
between the tsunamigenic units [Ramírez-Herrera et al., 2014]. The AMS study performed in
these samples revealed differences between AMS fabrics along the profile, related to
different stratigraphic units. Magnetic fabrics from the uppermost soil and the clayey-silt unit
between PV1 and PV2 were construed as the result of a low-energy sedimentary
environment. However, fabrics from PV1 show a preferential ESE – WNW orientation (116º
azimuth) and those from PV2 a stronger E – W orientation (96º azimuth). Therefore, these
units were interpreted as the result of currents flowing in those directions.
Petrology and particle size analysis performed by Ramírez-Herrera et al. [2014] on
beach sand samples from the beach berm as well as from the intertidal beach area indicate
that these sands are rich in volcanic rock fragments (mainly andesites). This fact has been
also confirmed by the XRD and SEM analyses performed in this work (Fig. 10). In addition,
evidences of volcanic features were identified in the sediments during field sampling.
Taking into consideration previous results from Ramírez-Herrera et al. [2014] and the
magnetic and mineralogical results obtained in this work, and trying to narrow down the
source of the tsunamigenic sedimentary units we have also measured hysteresis and
thermomagnetic curves of: 1) pumice and ash samples from the last eruption of the Colima
This article is protected by copyright. All rights reserved.
volcano (located 80 km NE from the Palo Verde trench) occurred before the 22 June 1932
tsunami event (1913 eruption), and 2) three lava flow samples from the Colima volcano (Ar-
Ar ages ~34, 49 and 97 ka) (supporting information Figure S2). Thermomagnetic results,
observed on ash samples clearly differ from those of PV1, indicating that the presence of
windblown volcanic ashes from the last eruption of the Colima volcano before 1932 can be
ruled out from being responsible for the enhancement of the magnetic contribution in the PV1
unit. On the other hand, although Ti-magnetite is the main magnetic phase in the three lava
flow samples, additional information would be necessary to evaluate its potential influence in
the tsunamigenic sedimentary deposits.
Carranza-Edwards et al. [2009] have shown that much of the differences in beach
sand composition along the Cuyutlán barrier beach complex are due to the variability of
source area rocks and local transport conditioned by wave climate, longshore currents and
local physiography. The mineralogical composition and texture of the local beach at the study
site carried out in this work is consistent with Carranza-Edwards et al. [2009] results.
However, the compositional and textural differences observed between the PV1 and
PV2 units and the interbedded clayey-silt unit needs the contribution of other mechanisms.
The clayey-silt unit composition and texture can be explained by the action of intense post-
sedimentary edaphic processes. Montmorillonite results from alteration of volcanic lithic
clasts and feldspar, increasing the quartz ratio relative to all the other constituents.
Differences in mineralogy between PV1 and PV2 are shown by the change in slope in
the sand percentage vs SIRM diagram (Fig. 3b). Furthermore, differences in PV1 and PV2
grain sizes (whole sediment texture) clearly indicate differences in energy between the two
tsunamigenic events. Changes in the proportion of magnetic minerals in the sandy units (Fig.
3b) indicate a change in provenance and/or in selective transport. Within PV1, and especially
in PV2 both sand percentage and magnetic mineral concentration increase with depth.
Upwards grain-size fining sequences are a clear evidence of hydrodynamic sorting. This
evidence provides a physical context in which magnetic grains sorting very likely occurs. The
correlation between sand percentage and magnetic concentration (SIRM) is modulated by
differences in hydraulic behavior controlled by the higher density of magnetic minerals and
other minerals enriched in denser magnetic inclusions. These differences are particularly
accentuated in the finest grain fraction where sedimentation is likely governed by Stoke’s
law. The hydraulic behavior of sand grains will be controlled by their size/density ratio under
non-Stoke conditions (high turbulence), which results in background magnetic variability.
This article is protected by copyright. All rights reserved.
Furthermore, the denser and smaller magnetic grains deposited at the last stage of the
tsunamigenic flow may be sieved trough the coarse sand underneath, producing the magnetic
grain fining downwards trend showed by the ARM.
The complex patterns of vertical sorting observed in the magnetic mineral
assemblages is also further explained by the complex structure of tsunami-related high-speed
flows, as shown by Jaffe et al. [2012], where suspension graded beds are intermixed within
massive or inversely graded intervals. Additionally, Johnson et al. [2016] have
experimentally corroborated that differences in grain size at source or ponded water depths
influence local entrainment, transport and deposition conditions. Nevertheless, our previous
interpretation does not necessary exclude the possibility of source-area related differences in
the hydraulically sorted material. Whilst overall differences in the magnetic composition
between events may have been controlled at the source or during transport, smaller-scale
within-unit variations are more likely hydraulically controlled during deposition.
There are several processes that can be equally invoked as potential mechanisms to
explain different sources of the tsunamigenic deposits taking into account that there is no
prior information about the paleostunami occurred around 1300 CE [Ramírez-Herrera et al.,
2014]: (1) the manner in which tsunamis arrived at the shore and the backwash pattern; (2) a
change in the coast morphology between the tsunami events associated to PV2 and PV1 that
could influence the sediment origin and/or the hydrodynamic energy during sedimentation;
(3) the energy of the paleotsunami event associated with the PV2 deposit was lower than that
of the later event and the carrying capacity of the sediment-laden back-flow was not big
enough as to rework the onshore deposits accumulated during run-up flow. Indeed, Ramírez-
Herrera et al. [2014] identified much more marine and marine-brackish taxa in PV2 than in
PV1 samples.
6. Conclusions
Rock-magnetic analyses, XRD and SEM experiments were conducted on samples
from a trench located in the Palo Verde estuary (PV), near Cuyutlán, in the Colima coastal
area, complementing the previous study carried out in this site by Ramírez-Herrera et al.
[2014]. Although these authors identified two probable tsunamis related to tsunamigenic
earthquakes of local origin (sand unit PV1 with the 22 June 1932 tsunami and sand unit PV2
This article is protected by copyright. All rights reserved.
with a probable paleotsunami that happened around 1300 CE), they also opened several
questions regarding the provenance of deposits and the energy of both events.
The Palo Verde sedimentary sequence consists of five sedimentary units. Four of
them have been magnetically characterized. Strong correlation between magnetic properties
and the sedimentary stratigraphy has been observed. Ferrimagnetic minerals like Ti-
magnetite control the magnetic signal and their concentration positively correlates to the sand
content. Moreover, (Ti)hematite is also present in minor concentration mainly in the layers
beneath the PV1 unit. Differences in both concentration and magnetic grain size of the
magnetic phases have been detected between the two tsunamigenic units PV1 and PV2.
Magnetic results such as the highest concentration of (Ti)magnetite in PV1 with respect to the
lower units as well as the sharp contrast of the magnetic properties between this unit and the
layer beneath could be related to the fact that the PV1 tsunamigenic deposit is the result of a
very energetic event. Source-area differences (and/or in selective transport) between PV1 and
PV2 deposits are supported by the change in the correlation pattern between sand percentage
and magnetic concentration. On the other hand, hydrodynamic sorting could control the
magnetic grain-size distribution in each tsunamigenic unit PV1 and PV2.
Detailed rock-magnetic and SEM techniques have proved to be an interesting
complementary approach to characterize tsunami deposits. Therefore, they should be
included together with other tools in a comprehensive study to help identify source areas and
depositional process energy of tsunamigenic units.
Acknowledgments
M.F. Bógalo and M. Calvo-Rathert acknowledge financial support by projects BU0066U16
of the Junta de Castilla y León (Spain) and CGL2016-77560 from Ministerio de Economía y
Competitividad (MINECO) both with European Regional Development Fund (ERDF)
funding. M.T. Ramírez-Herrera acknowledges financial support by grants no. PAPPIT-
UNAM- IN123609 and SEP-CONACYT-129456, and DGAPA-PASPA-2015. A.
Goguitchaichvili acknowledges the partial financial support provided by DGAPA-PAPIIT
IN101717 and CONACYT n° 252149. This work was also partially supported by Spanish
projects CGL2015-66681-R and CGL2014-54117-REDT from MINECO and GRC2014/023
from Xunta de Galicia. K.J. Mohamed was funded by ED481C 2014/2 grant of the Xunta de
Galicia. We are grateful to E. Font and an anonymous reviewer as well as to associate editor
This article is protected by copyright. All rights reserved.
J. Feinberg for their useful and constructive comments and suggestions. The data used in this
work are listed in the figures, tables and references; any additional data can be made available
upon request to the corresponding author (mfbogalo@ubu.es).
Figure captions
Figure 1. Seismotectonic and geologic setting. (a) Mexican Subduction zone and seismic
rupture areas: RFZ: Rivera fracture zone, EPR: East Pacific rise, circles in red color: rupture
areas of most significant events in the last century. (b) Palo Verde estuary, white line shows
trace of topographic profile shown in "c". (c) Topographic profile and location of trench PV.
(d) Photo showing in-situ collection of samples for magnetic properties studies.
Figure 2. Palo Verde trench stratigraphy, cumulative grain size (Ramírez-Herrera et al.,
2014) and selected rock magnetic parameters related to magnetic mineralogy.
Figure 3. Bi-plots of (a) saturation remanent magnetization (SIRM) vs. susceptibility (χ). (b)
Sand percentage vs. SIRM. (c) SIRM vs. the coercivity of remanence (Bcr). The samples are
well grouped according to their stratigraphy.
Figure 4. Rock magnetic parameters which can be related to magnetic grain size versus
depth.
Figure 5. Hysteresis ratios (Mrs/Ms vs. Bcr/Bc) plotted on a Day plot [Day et. al., 1977] in
linear scale for all the studied samples according to the legend. Theoretical mixing curves
SD+MD and SD+SP for magnetite [Dunlop, 2002] are showed in the plot. PV1 and PV2 refer
to the tsunamigenic sand units.
Figure 6. (a) Relationship between ARM susceptibility (χARM) and low-field susceptibility
(c) in the so-called King plot. The dashed lines correspond to the relationship between χARM
and χ for magnetite of different grain sizes (given in microns) [King et al., 1982]. T(b) Bi-
plot of ARM/SIRM vs. the coercivity of remanence (Bcr). Samples from the different
tsunamigenic units PV1 and PV2 display different trends.
Figure 7. IRM unmixing analyses [Kruiver et al., 2001 and Heslop et al., 2002] for four
representative samples from the labelled stratigraphic units.
This article is protected by copyright. All rights reserved.
Figure 8. Representative examples of thermal demagnetization of 3-axial isothermal
remanent magnetization (IRM) experiments.
Figure 9. (a) Curie temperatures (Tc) calculated from the heating and cooling
thermomagnetic curves. (b)-(d) Thermomagnetic curves for three representative samples.
Low and high temperature components (comp. 1 and comp. 2) are associated to (Ti)magnetite
and (Ti)hematite, respectively.
Figure 10. (a) EDX colour mapping image of a polish-section from PV1. Red is Mg and
green is Fe. (b) Summary of X-Ray diffraction semi-quantitative analyses.
Figure 11. Scanning electron microscope images representative of the studied units. (a) Sand
grains from the intertidal area of the present-day beach mounted on SEM stubs. (b) Fe-Ti iron
oxide mineral showing the typical Ti-rich exolution lamellae (grey) in magnetite. The grain is
mostly replaced by hematite. (c) Polished section of Topsoil. These samples were very friable
and the fine-grained matrix disaggregated and broke during preparation. However the original
texture and composition of the matrix can be observed against the black background of the
resin polymer used for consolidation. (d) PV1 unit showing the typical polymictic
composition of the source area. (e) Fine grained Fe-rich grains included in a volcanoclast. (f)
Intense alteration of framework volcanic grains and associated occurrence of secondary
montmorillonite-rich matrix in the clayey-silt unit. Fine-grained Fe-rich mineral are very
apparent in the matrix. (g) PV2 unit showing similar textural and compositional
characteristics as PV1. The primary matrix has been better preserved during preparation
(circle). Some of the Fe-rich grains included in the large framework grain on the left have
been altered (arrows). (h) Leucoxene showing the remaining Ti-rich exolution lamellae after
alteration of titanomagnetites and/or ilmenite. See the unaltered beach sand grain in (b).
References
Atwater, B.F., and A.L. Moore (1992), A tsunami about 1000 years ago in Puget Sound,
Washington, Science, 258, 1614–1617.
Bowles, J. A., M. J. Jackson, T. S. Berquó, P. A. Sølheid, and J. S. Gee (2013), Inferred time-
and temperature-dependent cation ordering in natural titanomagnetites, Nature
Communications, 4, 1916. doi:10.1038/ncomms2938.
This article is protected by copyright. All rights reserved.
Bloemendal, J., J. W. King, F. R. Hall, and S. J.Doh (1992), Rock magnetism of Late
Neogene and Pleistocene deep-sea sediments: relationship to sediment source,
diagenetic processes and sediment lithology, J. Geophys.Res., 97 (B4), 4361–4375.
Bógalo, M.F., J.J. Villalaín, M. Calvo, M.I. González, A.I. Ortega, F. Heller, and A. Pérez-
González (2014), New rock magnetic results from Galería cave in the Atapuerca
palaeoanthropological site (northern Spain). Influence of solar irradiation in the
edaphic generation of superparamagnetic grains?, paper presented at 14th Castle
Meeting, New trends on Paleo, Rock and Environmental Magnetism, Evora, Portugal.
Carranza-Edwards, A., J.J. Kasper-Zubillaga, L. Rosales-Hoz, E.A. Morales-De La Garza,
and Lozano-Santa Cruz, R. (2009), Beach sand composition and provenance in a
sector of the southwestern Mexican Pacific, Revista Mexicana de Ciencias
Geológicas, 26(2), 433–447.
Černý, J., M.-T. Ramírez-Herrera, M.-F. Bógalo, A. Goguitchaichvili, R. Castillo-Aja, J.
Morales, J.A. Sanchez-Cabeza, A.C. Ruiz-Fernández, and A.C. Ruiz (2015),
Magnetic record of extreme marine inundation events at Las Salinas site, Jalisco,
Mexican Pacific coast, Int. Geol. Rev. doi:10.1080/00206814.2015.1075230
Corona, N., and M.T. Ramírez-Herrera (2012), Mapping and historical reconstruction of the
great Mexican 22 June 1932 tsunami, Natural Hazards and Earth System Sciences,
12(5), 1337–1352. doi:10.5194/nhess-12-1337-2012.
Corona, N., and M.T. Ramírez-Herrera (2015), Did an underwater landslide trigger the June
22, 1932 tsunami off the Pacific coast of Mexico?, Pure Appl. Geophys, 172, 3573–
3587. doi:10.1007/s00024-015-1171-1.
Cumming, J.L. (1933), Los terremotos de junio de 1932 en los estados de Colima y Jalisco:
Universidad de México, 31– 32, 68–104.
Cuven, S., R. Paris, S. Falvard, E. Miot-Noirault, M. Benbakkar, J.-L. Schneider, and I. Billy,
(2013), High-resolution analysis of a tsunami deposit: Case-study from the 1755
Lisbon tsunami in southwestern Spain, Marine Geology, 337, 98–111,
doi:10.1016/j.margeo.2013.02.002
Dawson, A.G. and S. Shi (2000), Tsunami deposits. In: Pure and Applied Geophysics, 157,
875–897.
This article is protected by copyright. All rights reserved.
Day, R., M. Fuller, and V.A. Schmidt (1977), Hysteresis properties of titanomagnetites:
grain-size and compositional dependence, Phys. Earth. Planet. Inter. 13, 260–267.
Dearing, J.A., K.L. Hay, S.M.J. Baban, A.S. Huddleston, E.M.H. Wellington, and P.J.
Loveland (1996), Magnetic susceptibility of soil: An evaluation of conflicting theories
using a national data set, Geophys. J. Int. 127, 728–734.
Dearing, J.A., P.M. Bird, R.J.L. Dann, and S.F. Benjamin (1997), Secondary ferromagnetic
minerals in Welsh soils: a comparison of mineral magnetic detection methods and
implications for mineral formation, Geophys. J. Int., 130, 727–736.
Dunlop, D.J. and Ö. Ozdemir (1997), Rock magnetism, Fundamentals and Frontiers,
Cambridge Univ. Press, Cambridge, U.K.
Engel, M., and H. Brückner (2011), The identification of palaeo-tsunami deposits – A major
challenge in coastal sedimentary research, Coastline Reports, 17, 65–80.
Evans, M.E., and F. Heller (2003), Environmental Magnetism:Principles and Applications of
Enviromagnetics, Academic Press, San Diego, California.
Font, E., C. Nascimento, R. Omira, M.A. Baptista, and P.F. Silva (2010), Identification of
tsunami-induced deposits using numerical modeling and rock magnetism techniques:
A study case of the 1755 Lisbon tsunami in Algarve, Portugal, Phys. Earth Planet.
Inter, 182, 187–198, doi:10.1016/j.pepi.2010.08.007.
Font, E., C. Veiga-Pires, M. Pozo, S. Nave, S. Costas, F. Ruiz Muñoz, M. Abad, N. Simões,
S. Duarte, and J. Rodríguez-Vidal (2013), Benchmarks and sediment source(s) of the
1755 Lisbon tsunami deposit at Boca do Rio Estuary, Marine Geology, 343, 1–14,
doi:10.1016/j.margeo.2013.06.008.
Forster, Th., M.E. Evans, and F. Heller (1994), The frequency dependence of low field
susceptibility in loess sediments, Geophys. J. Int., 118, 636–642.
Goguitchaichvili, A., M.T. Ramírez-Herrera, M. Calvo-Rathert, B. Aguilar Reyes, Á.
Carrancho, C. Caballero, F. Bautista, and J. Morales Contreras (2013), Magnetic
fingerprint of tsunami-induced deposits in the Ixtapa-Zihuatanejo area, Western
Mexico, Int. Geol. Rev., 55, 1462–1470, doi:10.1080/00206814.2013.779781.
Goto, K., K. Hashimoto, D. Sugawara, H. Yanagisawa, and T. Abe (2014), Spatial thickness
variability of the 2011 Tohoku-oki tsunami deposits along the coastline of Sendai
Bay, Mar.Geol., doi:10.1016/j.margeo.2013.12.015.
This article is protected by copyright. All rights reserved.
Goto T., K. Satake, T. Sugai, T. Ishibe, T. Harada, and S. Murotani (2015), Historical
tsunami and storm deposits during the last five centuries on the Sanriku coast, Japan,
Mar. Geol.,367, 105-117, doi: 10.1016/j.margeo.2015.05.009.
Grommé, C. S., T. L. Wright, and D. L. Peck (1969), Magnetic properties and oxidation of
iron-titanium oxide minerals in Alae and Makaopuhi lava lakes, Hawai, J. Geophys.
Res., 74, 5277–5293.
Heslop, D., M.J. Dekkers, P.P. Kruiver, and I.H.M. van Oorschot (2002), Analysis of
isothermal remanent magnetization acquisition curves using the expectation-
maximization algorithm, Geophys. J.Int.,148, 58–64.
Hofmann, D.I., K. Fabian, F. Schmieder, B. Donner and U. Bleil, (2005) A stratigraphic
network across the Subtropical Front in the central South Atlantic: Multi-parameter
correlation of magnetic susceptibility, density, X-ray fluorescence and δ18O records,
Earth and Planetary Science Letters, 240 (3-4), pp. 694-709.
Jaffe, B. E., K. Goto, D. Sugawara, B. M. Richmond, S. Fujino, and Y. Nishimura, (2012),
Flow speed estimated by inverse modeling of sandy tsunami deposits: results from the
11 March 2011 tsunami on the coastal plain near the Sendai Airport, Honshu, Japan,
Sediment. Geol., 282, 90–109, doi: 10.1016/j.sedgeo.2012.09.002.
Johnson, J. P. L., K. Delbecq, W. Kim, and D. Mohrig (2016), Experimental tsunami
deposits: Linking hydrodynamics to sediment entrainment, advection lengths and
downstream fining, Geomorphology, 253, 478–490,
doi:10.1016/j.geomorph.2015.11.004.
Kain, C.L., C. Gomez, D.E. Hart, P. Wassmer, J. Goff, and C. Starheim (2014), Assessing
topographic controls on flow direction in washover deposits using measurements of
Magnetic Fabric, Mar. Geol., 350, 16–26. doi:10.1016/j.margeo.2014.01.010.
Kain, C., P. Wassmer, J. Goff, C. Chagué-Goff, C. Gomez, D. Hart, and D. Fierro (2016),
Determining flow patterns and emplacement dynamics from tsunami deposits with no
visible sedimentary structure, Earth Surf. Process. Landforms, doi:10.1002/esp.4020
King, J., S.K. Banerjee, J. Marvin, and Ö. Özdemir (1982), A comparison of different
magnetic methods for determining the relative grain size of magnetite in natural
materials: some results from lake sediments, Earth Planet. Sci. Lett. 59, 404–419.
This article is protected by copyright. All rights reserved.
Kortekaas, S. and A.G. Dawson (2007), Distinguishing tsunami and storm deposits: An
example from Martinhal, SW Portugal, Sediment. Geol., 200, 208–221, doi:
10.1016/j.sedgeo.2007.01.004.
Kruiver, P.P., M.J. Dekkers, and D. Heslop (2001), Quantification of magnetic coercivity
components by the analysis of acquisition curves of isothermal remanent
magnetization, Earth Planet. Sci. Lett., 189, 269–276. doi:10.1016/S0012-
821X(01)00367-3.
Lancin, M., and A. Carranza (1976), Estudio geomorfológico de la bahía y la playa de
Santiago en Manzanillo Colima, Revista del Instituto de Geología UNAM, 2, p. 43–
65.
Leonhardt, R. (2006), Analyzing rock magnetic measurements: The RockMagAnalyzer 1.0
software, Comput. Geosci. 32, 1420–1431. doi:10.1016/j.cageo.2006.01.006.
Liu, Q, A.P. Roberts, J.C. Larrasoaña, S.K. Banerjee, Y. Guyodo, L. Tauxe, and F. Oldfield
(2012), Environmental Magnetism: Principles and Applications, Rev. Geophys, 50,
RG4002. doi:8755-1209/12/2012RG000393
Lowrie, W. (1990), Identification of ferromagnetic minerals in a rock by coercivity and
unblocking temperature properties, Geophys. Res. Lett., 17, 159–162.
Lyons, R., S. Tooth, and G.A.T. Duller (2014), Late Quaternary climatic changes revealed by
luminescence dating, mineral magnetism and diffuse reflectance spectroscopy of river
terrace paleosols: a new form of geoproxy data for the southern African interior,
Quat. Sci. Rev., 95, 43 – 59.
Maher, B. (1986), Characterisation of soils by mineralmagnetic measurements, Phys. Earth
Planet. Inter., 42, 76–92, doi:10.1016/S0031-9201(86)80010-3.
Maher, B.A. (1998), Magnetic properties of modern soils and quaternary loessic paleosols:
Paleoclimatic implications. Palaeogeogr. Palaeoclimatol. Palaeoecol., 137, 25–54,
doi:10.1016/S0031-0182(97)00103-X.
Moore, A., G. Gelfenbaum, and R. Triyono (2006), Sedimentary deposits of the 26 December
2004 tsunami on the northwest coast of Aceh, Indonesia, Earth Planets Space, 58,
253–258, doi: 10.1186/BF03353385.
This article is protected by copyright. All rights reserved.
Morton, R. A., G. Gelfenbaum, and B. E. Jaffe (2007), Physical criteria for distinguishing
sandy tsunami and storm deposits using modern examples, Sediment. Geol., 200, 184–
207.
Okal, E.A., and J.C. Borrero (2011), The tsunami earthquake of1932 June 22 in Manzanillo,
México: Seismological study and tsunami simulations, Geophys. J. Int., 187(3), 1–17,
doi:10.1111/j.1365-246X.2011.05199.x
Özdemir, Ö. and Dunlop, D.J. (2014), Hysteresis and coercivity of hematite, J. Geophys.
Res., 119, 2582–2594.
Peters, C. and M.J. Dekkers (2003), Selected room temperature magnetic parameters as a
function of mineralogy, concentration and grain size, Phys. Chem. Earth, 28, 659–
667.
Ramírez-Herrera, M.T., A. Cundy, V. Kostoglodov, A. Carranza-Edwards, E. Morales, and S.
Metcalfe (2007), Sedimentary record of late Holocene relative sea-level change and
tectonic deformation from the Guerrero Seismic Gap, Mexican Pacific Coast, The
Holocene, 17 (8), 1211–1220, doi: 10.1177/0959683607085127.
Ramírez-Herrera, M.T., M. Lagos, I. Hutchinson, V. Kostoglodov, M.L. Machain, M.
Caballero, A. Goguitchaichvili, B. Aguilar, C. Chagué-Goff, J. Goff, A.-C. Ruiz-
Fernández, M. Ortiz, H. Nava, F. Bautista, G.I. Lopez, and P. Quintana (2012),
Extreme wave deposits on the Pacific coast of Mexico: Tsunamis or storms? – A
multi-proxy approach, Geomorphology, 139–140, 360–371,
doi:10.1016/j.geomorph.2011.11.002.
Ramírez-Herrera, M.-T., N. Corona, M. Lagos, J. Černý, A. Goguitchaichvili, J. Goff, C.
Chagué-Goff, M.L. Machain, A. Zawadzki, G. Jacobsen, A. Carranza-Edwards, S.
Lozano, and L. Blecher (2014), Unearthing earthquakes and their tsunamis using
multiple proxies: the 22 June 1932 event and a probable fourteenth-century
predecessor on the Pacific coast of Mexico, Int. Geol. Rev. 56, 1584–1601,
doi:10.1080/00206814.2014.951977
Ramírez-Herrera, M.T., M.F. Bógalo, J. Černý, A. Goguitchaichvili, N. Corona, M.L.
Machain, A.C. Edwards, and S. Sosa (2016), Historic and ancient tsunamis uncovered
on the Jalisco-Colima Pacific coast, the Mexican subduction zone, Geomorphology,
259, 90–104. doi:10.1016/j.geomorph.2016.02.011.
This article is protected by copyright. All rights reserved.
Reitz, A., C. Hensen, S. Kasten, J.A. Funk and G.J. de Lange (2004), A combined
geochemical and rock-magnetic investigation of a redox horizon at the last
glacial/interglacial transition , Physics and Chemistry of the Earth, 29 (13-14 SPEC.
ISS.), pp. 921-931.
Richmond, B.M., S. Watt, M. Buckley, B.E. Jaffe, G. Gelfenbaum, and R.A. Morton (2011),
Recent storm and tsunami coarse-clast deposit characteristics, southeast Hawai, Mar.
Geol., 283, 79–89, doi: doi:10.1016/j.margeo.2010.08.001.
Robertson, D.J., and D.E. France (1994), Discrimination of remanence-carrying minerals in
mixtures, using isothermal remanent magnetisation acquisition curves, Phys. Earth
Planet. Inter., 82, 223–234, doi:10.1016/0031-9201(94)90074-4.
Sánchez A.J., and S.F. Farreras (1993), Catálogo de tsunamis (maremotos) en la costa
occidental de México, National Geophysical Data Center,Boulder, Colorado.
Shanmugam, G. (2012), Process-sedimentological challenges in distinguishing paleo-tsunami
deposits, Nat. Hazards, 63, 5–30, doi: 10.1007/s11069-011-9766-z.
Schneider, J.L., C. Chagué-Goff, J.L. Bouchez, J. Goff, D. Sugawara, K. Goto, B. Jaffe, B.
Richmond (2014), Using magnetic fabric to reconstruct the dynamics of tsunami
deposition on the Sendai Plain, Japan - The 2011 Tohoku-oki tsunami, Mar. Geol.
358, 89–106, doi:10.1016/j.margeo.2014.06.010.
SGM (Servicio Geológico Mexicano) (2000), Carta Geológico-Minera, Manzanillo E13-2-5,
Colima y Jalisco, Servicio Geológico Mexicano, México.
Singh, S.K., J. Havskov, and L. Astiz (1981), Seismic gaps and recurrence periods of large
earthquakes along the Mexican subduction zone, Bulletin of the Seismology Society of
America, 71, 827–843.
Singh, S.K., L. Ponce, and S.P. Nishenko (1985), The great Jalisco, Mexico, earthquakes of
1932: subduction of the Rivera plate, Bulletin of the Seismological Society of
America, 75(5), 1301–1313.
Suarez, G., and P. Albini (2009), Evidence for Great Tsunamigenic Earthquakes (M 8.6)
along the Mexican Subduction Zone, Bulletin of the Seismological Society of
America, 99, 892–896, doi:10.1785/0120080201.
Spiske, M., J. Piepenbreier, C. Benavente, A. Kunz, H. Bahlburg, and J. Steffahn (2013),
Historical tsunami deposits in Peru: sedimentology, inverse modeling and optically
This article is protected by copyright. All rights reserved.
stimulated luminescence dating, Quat. Int., 305, 31–44, doi:
10.1016/j.quaint.2013.02.010.
Sulpizio, R., E. Zanella, J.L. Macías, and R. Saucedo (2014), Deposit temperature of
pyroclastic density currents emplaced during el Chichón 1982 and Colima 1913
eruptions. In: Ort, M. H., Porreca, M. & Geissman, J. W. (eds). The Use of
Palaeomagnetism and RockMagnetism to Understand Volcanic Processes. Geological
Society, London, Special Publications, doi:10.1144/SP396.5
Torrent, J., Q.-S. Liu, J. Bloemendal, and V. Barron (2007), Magnetic Enhancement and Iron
Oxides in the Upper Luochuan Loess–Paleosol Sequence, Chinese Loess Plateau, Soil
Sci. Soc. Am. J., 71, 1570–1578, doi:10.2136/sssaj2006.0328.
Veerasingam, S., R. Venkatachalapathy, N. Basavaiah, T. Ramkumar, S. Venkatramanan,
and K. Deenadayalan (2014), Identification and characterization of tsunami deposits
off southeast coast of India from 2004 Indian Ocean tsunami: Rock magnetic and
geochemical approach, J. Earth Syst. Sci. 123 (4), 905-921,doi: 10.1007/s12040-014-
0427-y.
Verma, S.P. and J.F. Luhr (1993), Sr-Nd-Pb isotope and trace element geochemistry of calc-
alkaline andesites from Volcán Colima, Mexico, Geofis. Int., 32 (4), 617–631.
Wassmer, P., J.-L. Schneider, A. Fonfrege, F. Lavigne, R. Paris, and C. Gomez (2010), Use
of Anisotropy of Magnetic Susceptibility (AMS) in the study of tsunami deposits:
application to the 2004 deposits on the eastern coast of Banda Aceh, North Sumatra,
Indonesia, Mar. Geol., 275, 255–272, doi: 10.1016/j.margeo.2010.06.007.
Wassmer, P., and C. Gomez (2011), Development of the AMS method for unconsolidated
sediments. Application to tsunami deposits, Geomorpholie Relief Process. 3, 279–
290. doi: 10.4000/geomorphologie.9491
Wassmer, P.C., C. A. Gomez, T.Y.W.M. Iskandasyah, F. Lavigne, and J. Sartohadi (2015),
Contribution of anisotropy of magnetic susceptibility (AMS) to reconstruct flooding
characteristics of a 4220 BP tsunami from a thick unconsolidated structureless deposit
(Banda Aceh, Sumatra), Front. Earth Sci., 3:40. doi: 10.3389/feart.2015.00040
This article is protected by copyright. All rights reserved.
Figure1.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure2.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure3.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure4.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure5.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure6.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure7.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure8.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure9.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure10.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Figure11.
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
