See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/236013725
On the analysis of short-and long-term deformation processes at Tenerife,
Canary Islands
Article · January 2010
CITATIONS READS
0 304
7 authors, including:
Pietro Tizzani Andrea Manconi
Italian National Research Council Swiss Federal Institute for Forest, Snow and Landscape Research
149 PUBLICATIONS 2,686 CITATIONS 164 PUBLICATIONS 3,886 CITATIONS
SEE PROFILE SEE PROFILE
A. Pepe Mariarosaria Manzo
Italian National Research Council Italian National Research Council
225 PUBLICATIONS 5,974 CITATIONS 96 PUBLICATIONS 3,405 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by A. Pepe on 16 August 2017.
The user has requested enhancement of the downloaded file.
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 115, B12412, doi:10.1029/2010JB007735, 2010
Long‐term versus short‐term deformation processes at Tenerife
(Canary Islands)
Pietro Tizzani,1 Andrea Manconi,1,2 Giovanni Zeni,1 Antonio Pepe,1
Mariarosaria Manzo,1 Antonio Camacho,3 and Jose Fernández3
Received 28 May 2010; revised 4 August 2010; accepted 2 September 2010; published 16 December 2010.
[1] Several geophysical investigations have identified that the Tenerife volcanic complex
is affected by crustal deformation processes occurring at timescales of millions of years.
Recently, space‐based geodetic observations have also detected a short‐term surface
deformation, characterized by a broad subsidence pattern with maximum ground velocities of
about 4 mm yr−1. For the purpose of investigating the relationship between these long‐term
and short‐term deformation processes, we performed an advanced fluid dynamic analysis
(FDA). We first carried out a standard dimensionless FDA to discriminate the deformation
style of Tenerife and found that, at million year timescales, basement flexure mainly controls
its long‐term structural evolution. Secondly, to highlight the driving forces of the short‐term
deformation process, we simulated a numerical FDA based on finite element models that
include topography as well as vertical and lateral material heterogeneities. Our results show
that the recent surface deformation is mainly caused by a progressive sagging of the denser
(less viscous) core of the island onto the weaker (but more viscous) lithosphere. Moreover,
over periods comparable to the hypothesized age of loading of the oceanic crust beneath
Tenerife, this tendency would result in a total flexure of about 3–4 km, which is in agreement
with independent estimations based on geophysical analyses. Our study shows that a unitary
physical model may explain both the deformation recorded in deep geological structures and
the current active ground deformation processes occurring at the Tenerife volcano.
Citation: Tizzani, P., A. Manconi, G. Zeni, A. Pepe, M. Manzo, A. Camacho, and J. Fernández (2010), Long‐term versus
short‐term deformation processes at Tenerife (Canary Islands), J. Geophys. Res., 115, B12412, doi:10.1029/2010JB007735.
1. Introduction mostly submerged) was built upon a basement of submarine
extrusive rocks, which outcrop west‐southwest and northeast
[2] Tenerife, the largest volcanic complex among the
(see Figure 1). The Las Cañadas composite volcano (from
Canary Islands (Spain), is the result of the coalescence of
more than 3.5 Ma to 0.18 Ma) and the currently active Teide–
several shield volcanoes [Martí et al., 1994; Ablay and
Pico Viejo stratovolcano (from 0.18 Ma to present) form the
Kearey, 2000]. Different hypotheses about the origins of
central volcanic complex. The latter mostly consists of lavas
Tenerife have been proposed over the past decades. Among
that evolved from basaltic to phonolitic composition, and it
the several theories, the most debated ones propose a hotspot
has been characterized by abundant explosive eruptions and
or mantle plume [Pérez et al., 1994; Carracedo et al., 1998;
several vertical collapses following the explosive withdrawal
Perlock et al., 2008; Dañobeitia and Canales, 2000], a region
of shallow magma chambers (4–6 km deep), sporadically
of compressional block faulting [Araña and Ortíz, 1991], a
accompanied by lateral collapses [Martí et al., 1997]. Lava
rupture propagating from the active Atlas Range [Anguita and emissions from Teide–Pico Viejo and along the Las Cañadas
Hernán, 1975], or a unifying model [Anguita and Hernán,
caldera rim characterized the recent volcanic activity of
2000]. Since about 10 Ma to the present, the ascent of
Tenerife [Ancochea et al., 1990].
mantle‐derived basaltic magmas has been focused along two [3] Several authors have analyzed the geodynamic and
main rift zones, trending northeast and northwest, and on a
tectonic processes affecting the Tenerife volcano and, more
third subsidiary rift trending south [Carracedo et al., 2007].
in general, the Canary Islands [Watts, 1994; Watts et al., 1997;
The oldest part of the island (the shield volcanic complex,
Watts and Zhong, 2000; Collier and Watts, 2001; Minshull
and Charvis, 2001]. The main results propose a complex
1
Istituto per il Rilevamento Elettromagnetico dell’Ambiente, CNR, interaction between the subvolcanic and regional structures
Napoli, Italy. at the Myr timescale, which produced a large flexure of the
2
Now at IRPI‐CNR, Torino, Italy.
3
Instituto de Astronomia y Geodesia, Consejo Superior de Investigaciones
oceanic lithosphere [Minshull and Charvis, 2001], as con-
Científicas, Universidad Complutense de Madrid, Madrid, Spain. firmed by seismic and gravimetric studies [Watts et al., 1997;
Araña et al., 2000; Gottsmann et al., 2006, 2008]. On the
Copyright 2010 by the American Geophysical Union. other hand, recent space‐based geodetic observations based on
0148‐0227/10/2010JB007735
B12412 1 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Figure 1. Sketch of the geological map of Tenerife Island (redrawn from Fernández et al. [2009]).
Differential Synthetic Aperture Radar Interferometry (DIn- [5] We analyzed the long‐term deformation processes
SAR) and GPS measurements revealed, in the 1992–2005 occurring at Tenerife by following the guidelines proposed
period, a deformation pattern characterized by a broad subsi- by van Wyk de Vries and Matela [1998]. We calculated the
dence with maximum velocities of about 4 mm yr−1 dimensionless numbers Pa and Pb by integrating different
[Fernández et al., 2009]. The understanding of possible re- geophysical data to constrain the structural parameters and the
lationships between the long‐term deformations recorded in oceanic crust age of loading (see Table 1). Our dimensionless
geological structures and the short‐term deformations mea- FDA places Tenerife in the field with low Pa and Pb values,
sured at the surface via geodetic techniques can be very defined as “flexure” (Figure 2). In particular, this result also
important for understanding active volcano behavior. For this remains constant when considering different values for the
reason, this work aims at clarifying whether the long‐term and model parameters. Thus, our result suggests that “basement
short‐term deformations observed at Tenerife might be ex- flexure” mainly controls the long‐term structural evolution of
plained by considering a unitary physical model. In the fol- the Tenerife volcanic complex at the Myr timescale.
lowing, we first describe the approach used to discriminate the
2.2. Short‐term Volcano Deformation Analysis
long‐term deformation style of the Tenerife volcanic complex.
Secondly, we extend the work done by Fernández et al. [6] Fernández et al. [2009], by considering GPS and
[2009], investigating the short‐term surface deformation pro- Small Baseline Subset (SBAS) DInSAR analyses [Berardino
cesses by means of advanced Finite Element (FE) models. et al., 2002], detected a broad surface deformation pattern at
Tenerife characterized by subsidence with maximum velocities
2. Methods of about 4 mm yr−1. To model this data, they proposed a
simplified analytical approach (mass loading of a point source
2.1. Long‐term Volcano Deformation Analysis embedded in a layered medium, with a top elastic layer over-
[4] Gravitational processes in volcanic areas are mainly lapping a viscoelastic half‐space) and interpreted the ground
controlled by the rheologies of the substratum underlying the displacements as gravitational sinking caused by the denser
main edifices [Merle and Borgia, 1996; van Wyk de Vries and core of the island descending onto the weaker lithosphere.
Borgia, 1996]. According to van Wyk de Vries and Matela However, while such a model seems to reasonably fit the
[1998], the deformation style due to volcanic loading de- observed displacements and is substantially in agreement with
pends on the presence of a ductile layer beneath the volcano the geodynamic scenario, this analytical approach fails to
edifices, which might in turn cause substratum flexure, base- accurately interpret the spatial variability of the deformation
ment extrusion, or volcano spreading. More specifically, these signal. This lack of fit is probably related to the simplified
studies propose the characterization of the deformation styles modeling settings (e.g., point source assumption, flat surface,
of volcanoes in different geodynamic contexts by considering lateral homogeneity, etc.) and/or to its physical assumptions
a dimensionless Fluid Dynamic Analysis (FDA). The latter (elasticity, viscoelasticity). We extended the analysis presented
relates the geometrical and structural parameters of the vol- by Fernández et al. [2009] and considered FE models that
cano under consideration (such as the radius of the volcanic include the topography of Tenerife as well as the vertical and
complex, the age of loading of the crust, the thickness and lateral variations of the physical characteristics of the crust and
viscosity of the ductile layer, etc.; see details in van Wyk the upper mantle. Because the ground displacements detected
de Vries and Matela [1998]) to a set of dimensionless num- at Tenerife are mainly due to tectonic strain, the deformation
bers (Pa and Pb) derived by the Buckingham P theorem processes can be well‐represented in a fluid dynamics context
[Buckingham, 1914, 1915]. [Kennett and Bunge, 2008]. Moreover, the displacements
show a linear dependence with time (see Figures 2b–2f of
2 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Table 1. Dimensionless FDA Parametersa
Radius (R)b (km) D Layer (D)c (km) EP Layer (E) (km) Cv (Pa m3) m (Pa s) t (Myr) Pa R/D Pb m/Cv D t
d
MODEL A
110 15 0.41 7.0E‐03
105 20 3 0.43 5.0E‐03
100 25 0.45 4.0E‐03
110 15 0.41 3.0E‐03
45 105 20 1.0 E+05 1.0 E+21 8 0.43 2.0E‐03
100 25 0.45 2.0E‐03
110 15 0.41 2.0E‐03
105 20 13 0.43 1.0E‐03
100 25 0.45 1.0E‐03
MODEL Be
80 45 0.56 1.0E‐04
77 48 7 0.58 9.0E‐04
74 51 0.61 9.0E‐04
80 45 0.56 7.0E‐04
45 77 48 1.0E+05 1.0E+21 10 0.58 7.0E‐04
74 51 0.61 6.0E‐04
80 45 0.56 5.0E‐04
77 48 13 0.58 5.0E‐04
74 51 0.61 5.0E‐04
MODEL Cf
95 30 0.47 6.0E‐04
90 35 17 0.5 5.0E‐04
85 40 0.53 5.0E‐04
95 30 0.47 5.0E‐04
45 90 35 1.0E+05 1.0E+21 20 0.5 5.0E‐04
85 40 0.53 4.0E‐04
95 30 0.47 5.0E‐04
90 35 23 0.5 4.0E‐04
85 40 0.53 3.0E‐04
a
Model Groups A, B, and C were assembled using three different values of crust elastic thickness (EP Layer) and loading time (t) according to Watts
[1994], Filmer and McNutt [1989], and Dañobeitia et al. [1994], respectively. The thickness of the ductile layer (EP Layer), the cohesion of the
elastoplastic layer and the volcano edifice (Cv), and the model viscosity (m) were defined by considering the study of Watts and Zhong [2000], whereas
geometric properties were constrained using information derived by bathymetric data and seismic profiles [Watts, 1994; Watts and Masson, 2001;
Krastel et al., 2001].
b
Data from Watts and Masson [2001] and Krastel et al. [2001].
c
The bottom of the ductile layer is fixed at a depth of 125 km, according to Watts and Zhong [2000].
d
EP‐Layer, 20 ± 5; t, 8 ± 5; Watts [1994].
e
EP‐Layer, 48 ± 3; t, 10 ± 3; Filmer and McNutt [1989].
f
EP‐Layer, 35 ± 5; t, 20 ± 3; Dañobeitia et al. [1994].
Fernández et al. [2009]); thus, we can assume a steady state approximated topographic profile of the island above the
viscous flow (Newtonian fluid) in our FDA and solve for the seafloor (see Figure 3). The domain discretization (about 6000
incompressible Navier‐Stokes set of differential equations triangular elements) was validated through several resolution
tests, which evidenced that the use of a finer mesh would affect
r ru þ ðruÞT þ ðu rÞu þ rp ¼ F the results by less than 1%.
ð1Þ [7] Body forces on the domain are mostly due to the vertical
r u ¼ 0; component of lithostatic loading; thus,
where u is the velocity vector, F is the body force term, r is FðrÞ ¼ 0; FðzÞ ¼ ðr;zÞ g; ð2Þ
the density, p is the pressure, and h is the dynamic viscosity
(hereafter referred to as viscosity). We first simplified our where g is the acceleration due to gravity (9.81 m s−2). As
simulations by considering an axisymmetric domain. The boundary conditions, we set the pressure reference level (p =
model dimensions were 160 km and 85 km in the radial (r) and 0) at the top of the simulated volcano topographic surface
vertical (z) directions, respectively. The distribution of densi- (see inset in Figure 3), and, because of the incompressibility
ties was derived from both seismic and gravity data [Watts et and stationarity assumptions of our analysis, we assumed
al., 1997; Camacho et al., 2000; Camacho et al., 2002; velocities of u(r) = u(z) = 0 on all model boundaries, excluding
Gottsmann et al., 2008]. The density distribution was included the symmetry axis. These constraints allowed for momentum
within the FE models as a two‐dimensional (2D) function with conservation in the analyzed domain and, as the area of
a 500 × 500 m resolution (r(r,z)), which also considers an
3 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Figure 2. Plot of Pa vs. Pb, illustrating the fields for flexure‐dominated, spreading‐dominated and
extrusion‐dominated systems (after van Wyk de Vries and Matela [1998]). The values calculated in
this study for the Tenerife volcano are represented by red, blue, and green triangles which refer to models
A, B, and C, respectively (see parameters in Table 1).
interest is far enough from the boundaries, did not affect the FE models, the deformation pattern is more accurately
FE modeling results. represented. This is also confirmed by the lower value of
[8] The viscosity distribution was evaluated through an the RMS associated with the heterogeneous FE models. In
advanced procedure that allowed the implementation of particular, the best fit viscosity distribution derived by our
the FE models within nonlinear optimization algorithms optimization procedure is reported in Figure 5. Interestingly,
[Manconi et al., 2010]. With this approach, we searched for while the vertical viscosity contrasts appear not to be very
the best fit viscosity model explaining the observed surface high (∼5 × 1022 versus ∼1023 Pa s for the upper mantle and
deformation velocities. We first allowed the algorithm to the oceanic crust, respectively), the viscosity strongly varies
search for simplified homogeneous and vertically layered laterally by about two orders of magnitude (from ∼1.2 ×
(crust, upper mantle, lower mantle) viscosity models. Sec- 1021 Pa s for the (denser) cumulitic complex to ∼1023 Pa s
ondly, we also allowed for vertical and lateral hetero- for the oceanic crust).
geneities in the viscosity distribution associated with the 2D [10] The axisymmetric approximation of the above‐
density function defined a priori (Figure 3). The model described FE models might be considered as an oversim-
velocities were evaluated at the topographic surface, pro- plification because Tenerife has a complex triangular shape,
jected along the satellite line‐of‐sight (LOS), and compared and the analysis of a single profile might be misleading. On
with the measured SBAS‐DInSAR data. The best fit vis- the basis of this consideration, we implemented a three‐
cosity model was finally selected by considering the Root‐ dimensional (3D) forward model considering the 3D density
Mean‐Square (RMS) error as the cost function. structure derived again by the seismic and gravity data,
[9] In Figure 4, we show the LOS ground velocities whereas the viscosity distribution was constrained through
resulting from the FE models versus the velocities measured the best fit heterogeneous model obtained within the opti-
via the SBAS‐DInSAR analysis on a profile along the radial mization procedure. Moreover, in the 3D case, we also
distance. We note that the solutions for the homogeneous simulated a more realistic topographic relief of the volcanic
case, as well as those calculated within models that consider complex by implementing the digital elevation model
simplified vertical layering in the viscosity distribution, fit (DEM) (90 × 90 m) derived from the Shuttle Radar
the deformation pattern within the expected accuracies, Topography Mission (SRTM) (Figure 6a). The difference
although they are not able to suitably explain the lateral between the observed and simulated LOS‐projected veloc-
variability of the deformation signal. This finding is in ities shows that the 3D model is able to suitably explain the
agreement with the analytical solutions proposed by features of the surface deformation related to tectonic strain
Fernández et al. [2009]. In contrast, when horizontal and in the volcanic environment (Figures 6b–6d). This good fit
vertical variations of viscosity are also included within the occurs despite the 3D viscosity model being constrained by
4 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Figure 3. Setup of the axisymmetric FE model used in the numerical optimization. The density model
(color bar) is derived from gravimetric and seismic data [Watts et al., 1997; Gottsmann et al., 2008]. The
dashed white curve indicates the Moho discontinuity [Longpré et al., 2008, and references therein]. The
inset describes the boundary and the loading conditions of the FE model (see text for more details).
considering the results of the axisymmetric models. The viscous) lithosphere. In addition, the implementation of our
larger values of residuals are mostly located in places where results in a 3D forward model shows that the spatial vari-
an anthropogenic component of the surface deformation has ability of the deformation signal is well‐compensated by the
previously been recognized (see also Figure 3f of Fernández lateral variation of the crustal and upper mantle rheological
et al. [2009]). properties. The latter seems to control both the amplitude
and the spatial orientation of the ground velocity field.
3. Discussion and Conclusions Indeed, the model in general fits very well with the detected
broad subsidence, while the localities with larger residuals
[11] In this work, we analyzed the active deformation are mainly associated with man‐made deformation features,
processes at the Tenerife volcano. In particular, to highlight such as water exploitation [see Fernández et al., 2009].
the origin of the driving forces that controlled the structural Moreover, the surface deformation is not constrained by the
evolution of the volcano and the currently observed defor- Las Cañadas caldera rim, as the deformation continues well
mation field, we carried out a two‐scale FDA. We first per- beyond its rim (see Figures 4 and 6b). After the removal of
formed a dimensionless FDA by applying the Buckingham the model, some localized residuals still remain in the cal-
P theorem and found that basement flexure controls the dera (P3 in Figure 6d), but this feature is probably associ-
long‐term structural evolution of Tenerife, thus confirm- ated with the local dynamics of the caldera itself. Indeed,
ing the general hypothesis proposed by van Wyk de Vries rocks within the Las Cañadas caldera can present very low
and Matela [1998] for the deformation style of the Canary viscosities in the core due to hydrothermal alterations at
Islands. Secondly, by considering heterogeneous axisym- shallow depths and can also react to the load of the Teide
metric FE models within an advanced nonlinear optimiza- volcano as well as to the deep sagging of the cumulitic
tion procedure, we showed that the surface deformation complex [Merle et al., 2010]. On the other hand, the main
observed in the 1992–2005 period can be explained by a source of the deformation resulting from our models appears
progressive sagging of the denser (less viscous) core of to be broader and is underway at depths up to 10–15 km
the island (cumulitic complex) onto the weaker (but more below sea level. The low viscosity values found in our
5 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Figure 4. Comparison between surface deformation velocities measured via the SBAS‐DInSAR (black
triangles) and those obtained via the FE optimization procedure for homogeneous (green curve), layered
(blue curve), and heterogeneous viscosity models (red curve). The dashed black curve represents the topo-
graphic profile along the considered radial distance. The RMS values of the three models as well as the
accuracy of the deformation velocities retrieved via the SBAS‐DInSAR analysis are reported in the upper
left and bottom right corners, respectively. The RMS values reveal that the heterogeneous model better
fits the observed velocities.
model at these levels might be the result of anomalous of the crust beneath the cumulitic complex. Furthermore,
thermal conditions of the rocks that form the denser core of the sagging effects of the cumulitic complex are markedly
the island, which are probably associated with the upper reduced where a higher value of the upper mantle viscosity
mantle magma plumbing system of Tenerife [Longpré et al., is encountered (about 1 mm yr−1 at a depth of 10–15 km
2008]. versus 4 mm yr−1 at the surface). We speculate that this
[12] Focusing on the details of the velocity field at depth tendency might encourage the driving forces responsible for
(Figure 7), we note that its spatial orientation reveals a lat- the basement flexure observed at a larger spatial scale. If
eral mass migration that accompanies the central subsidence we consider that the Tenerife volcanic complex was largely
Figure 5. The best fit viscosity model obtained via the FE optimization procedure.
6 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Figure 6. Results of the 3D finite element model. (a) Mesh of the 3D FE model. The model includes the
realistic topographic relief of Tenerife Island derived by the SRTM (90 × 90 m) DEM, the 3D density
structure derived from seismic and gravity data, as well as the 3D viscosity model derived by our opti-
mization analysis. The loading and boundary conditions are the same as those defined for the axisymmet-
ric FE model (see the inset in Figure 3). (b) SBAS‐DInSAR velocity map (resampled to a resolution of
500 × 500 m) superimposed on the DEM of Tenerife. (c) Modeled velocities retrieved from the 3D FE
model and projected along the line‐of‐sight of the satellite. (d) Difference between Figures 6b and 6c. P1,
P2, and P4 identify localized residuals related to anthropogenic deformations [see Fernández et al., 2009]
while P3 indicates residuals related to local dynamics of the Las Cañadas caldera and the Teide–Pico
Viejo stratovolcano (see section 3 for details).
built between 11.9 and 3.9 Ma [Longpré et al., 2009, and such timescales are probably affected by nonstationary
references therein], and taking the latter as the start time of processes and that events such as the growth and collapse of
loading of the oceanic crust while considering a stationary portions of the edifice may cause large variations from a
trend of about 1 mm yr−1, we obtain a total flexure of the linear trend. However, it is interesting to remark that starting
lithosphere in the range of 3–4 km. These values are rather from the observation of the deformation processes via the
compatible with independent estimations of the flexure of SBAS‐DInSAR time series over a 10–15 year timescale, our
the lithosphere under Tenerife performed via geophysical modeling strategy provided values rather compatible with
analyses (about 3.5 km; Watts et al. [1997], Collier and the total flexure of the lithosphere beneath the Tenerife
Watts [2001]). We are aware that the deformations on volcanic complex, a process that can be observed at a
7 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
[14] As a final remark, we would like to stress that the
joint analysis of long‐term and short‐term volcanic defor-
mation processes may provide a significant improvement in
the understanding of the behavior of volcanoes. Moreover,
because the compression regime induced by the flexure
of the oceanic crust under the action of volcanic loading
and the sagging of the dense core of the island can affect
the future volcanic activity [van Wyk de Vries and Borgia
1996], the results of this study might also be relevant for
future hazard assessment in active volcanic areas.
[15] Acknowledgments. We thank R. Lanari (IREA‐CNR) for pro-
viding motivation and useful suggestions; V. Acocella (Università Roma
3), M. Battaglia (University of Roma “La Sapienza”) and T. R. Walter
(GFZ Potsdam) for fruitful discussions. The comments of the Associate Edi-
tor and of two anonymous reviewers for their comments that improved the
quality of the manuscript. Research by AGC and JF was supported by the
Spanish MICINN projects GEOMOD (CGL2005‐05500‐C02) and PEL2G
(CGL2008‐06426‐C01‐01/BTE) and was done in the framework of the
Moncloa Campus of International Excellence (UCM‐UPM). The FE models
presented in this work were implemented using the commercial package
COMSOL® Multiphysics.
Figure 7. Distribution of the modeled velocity field in the
upper crust beneath the volcano edifice. The blue arrows References
indicate the field direction and the color scale indicates Ablay, G., and P. Kearey (2000), Gravity constraints on the structure and
the magnitude of the computed field. volcanic evolution of Tenerife, Canary Islands, J. Geophys. Res., 105,
5783–5796, doi:10.1029/1999JB900404.
Amelung, F., S.‐H. Yun, T. R. Walter, and P. Segall (2007), Stress control
Myr timescale. Moreover, the performed two‐scale FDA of deep rift intrusion at Mauna Loa volcano, Hawaii, Science, 316, 1026–
revealed that at Tenerife it is possible to consider a unitary 1030, doi:10.1126/science.1140035.
Ancochea, E., J. M. Fuster, E. Ibarrola, A. Cendrero, J. Coello, F. Henan, J.
physical model to explain both the long‐term deforma- M. Cantagrel, and C. Hamond (1990), Volcanic evolution of the island of
tion recorded in deep geological structures and the cur- Tenerife (Canary Islands) in the light of new K‐Ar data, J. Volcanol.
rently active ground deformation processes. In this context, Geotherm. Res., 44, 231–249.
our short‐term analysis can be considered as a sort of Anguita, F., and F. Hernán (1975), A propagating fracture model versus a
hotspot origin for the Canary Islands, Earth Planet. Sci. Lett., 27, 11–19,
experimental laboratory for the understanding of long‐term doi:10.1016/0012-821X(75)90155-7.
structural evolution starting from recent geodetic observations. Anguita, F., and F. Hernán (2000), The Canary Islands origin: A unifying
[13] Our results might also be very useful for future model, J. Volcanol. Geotherm. Res., 103, 1–26.
Araña, V., and R. Ortíz (1991), The Canary Islands: Tectonics, magmatism,
analyses of deformation processes in other active volcanic and geodynamic framework, in Magmatism in Extensional Structural
areas presenting similar characteristics. For example, Settings: The Phanerozoic African Plate, edited by A. B. Kampunzu
beneath the Hawaiian chain, a broad flexure of the crust and and R. T. Lubala, pp. 209–249, Springer‐Verlag, Berlin.
lithosphere has also been evidenced by geophysical analy- Araña, V., A. G. Camacho, A. Garcia, F. G. Montesinos, I. Blanco,
R. Vieira, and A. Felpeto (2000), The internal structure of Tenerife
ses, in conjunction with idealized elastic models of flexure (Canary Islands) based on gravity, aeromagnetic, and volcanological
[Watts and ten Brink, 1989; Wessel, 1993]. Moreover, data, J. Volcanol. Geotherm. Res., 103, 43–64.
recent analyses have shown that the load of the volcanic Berardino, P., G. Fornaro, R. Lanari, and E. Sansosti (2002), A new algo-
rithm for surface deformation monitoring based on small baseline dif-
complex in the island of Hawaii might control the seismic ferential SAR interferograms, IEEE Trans. Geosci. Remote Sens., 40,
occurrences as well as the ascent of magmas [McGovern, 2375–2383.
2007]. In this context, the availability of spatially and Buckingham, E. (1914), On physically similar systems: Illustrations of the
use of dimensional equations, Phys. Rev., 4, 345–376.
temporally dense surface deformation time series, as those Buckingham, E. (1915), The principle of similitude, Nature, 96, 396–397.
achievable with the SBAS‐DInSAR approach, as well as Camacho, A. G., F. G. Montesinos, and R. Vieira (2000), Gravity inversion
information on the volcanic structures derived from inde- by means of growing bodies, Geophysics, 65, 95–101, doi:10.1190/
pendent geophysical observations, might be useful within a 1.1444729.
Camacho, A. G., F. G. Montesinos, and R. Vieira (2002), A 3D gravity
numerical FDA to study the relationship between long‐term inversion tool based on exploration of model possibilities, Comput.
and short‐term deformation processes of the Hawaiian Geosci., 28, 191–204, doi:10.1016/S0098-3004(01)00039-5.
volcanic complex. However, the deformation signal related Carracedo, J. C., S. Day, H. Guillou, E. Rodríguez, J. A. Canas, and F. J.
Pérez (1998), Hotspot volcanism close to a passive continental margin,
to a potential sagging of the central complex onto the Geol. Mag., 135, 591–604.
oceanic crust might not be as clearly identifiable as in the Carracedo, J. C., E. Rodrìguez Badiola, H. Guillou, M. Paterne, S. Scaillet,
case of the Tenerife volcano. Indeed, earthquakes, inflation/ F. J. Perez Torrado, R. Paris, U. Fra‐Paleo, and A. Hansen (2007),
deflation of shallow magmatic reservoirs, dike intrusions Eruptive and structural history of Teide Volcano and rift zones of Tene-
rife, Canary Islands, Geol. Soc. Am. Bull., 119, 1027–1051, doi: 10.1130/
and/or eruptions, and thus the unrest of volcanoes, such B26087.1.
as Mauna Loa and Kilauea on the island of Hawaii, largely Collier, J. S., and A. B. Watts (2001), Lithospheric response to volcanic
affect the measured short‐term displacements [Amelung loading by the Canary Islands: Constraints from seismic reflection data
in their flexural moats, Geophys. J. Int., 147, 660–676.
et al., 2007; Montgomery‐Brown et al., 2010]. Dañobeitia, J. J., and J. P. Canales (2000), Magmatic underplating in the
Canary Archipelago, J. Volcanol. Geotherm. Res., 103, 27–41.
8 of 9
B12412 TIZZANI ET AL.: DEFORMATION PROCESSES AT TENERIFE B12412
Dañobeitia, J. J., J. P. Canales, and G. A. Dehghani (1994), An estimation Montgomery‐Brown, E. K., D. K. Sinnett, M. Poland, P. Segall, T. Orr,
of the elastic thickness of the lithosphere in the Canary Archipelago using H. Zebker, and A. Miklius (2010), Geodetic evidence for en echelon
admittance function, Geophys. Res. Lett., 21, 2649–2652, doi:10.1029/ dike emplacement and concurrent slow slip during the June 2007 intru-
94GL02552. sion and eruption at Kīlauea volcano, Hawaii, J. Geophys. Res., 115,
Fernández, J., et al. (2009), Gravity‐driven deformation of Tenerife mea- B07405, doi:10.1029/2009JB006658.
sured by InSAR time series analysis, Geophys. Res. Lett., 36, L04306, Pérez, N. M., H. Wakita, S. Nakai, Y. Sano, and S. N. Williams (1994),
doi:10.1029/2008GL036920. 3He/4He isotopic ratios in volcanic hydrothermal discharges from the
Filmer, P. E., and M. K. McNutt (1989), Geoid anomalies over the Canary Canary Islands, Spain: Implications on the origin of the volcanic activity,
Islands group, Mar. Geophys. Res., 11, 77–87. Mineral. Mag., 58A, 709–710.
Gottsmann, J., L. Wooller, J. Martì, J. Fernández, A. G. Camacho, P. J. Perlock, P. A., P. J. González, K. F. Tiampo, G. Rodriguez‐Velasco,
Gonzalez, A. Garcia, and H. Rymer (2006), New evidence for the reawak- S. Samsonov, and J. Fernández (2008), Time evolution of deforma-
ening of Teide volcano, Geophys. Res. Lett., 33, L20311, doi:10.1029/ tion using time series of differential interferograms: Application to La
2006GL027523. Palma Island (Canary Islands), Pure Appl. Geophys., 165, 1531–1554,
Gottsmann, J., A. Camacho, L. Wooller, J. Martì, J. Fernández, A. Garcia, doi:10.1007/s00024-004-0388-7.
and H. Rymer (2008), Shallow structure beneath the central volcanic van Wyk de Vries, B., and A. Borgia (1996), The role of basement in
complex of Tenerife from new gravity data, Phys. Earth. Planet. Inter., volcano deformation, Geol. Soc. Spec. Publ., 110, 95–110, doi:10.1144/
168, 212–230. GSL.SP.1996.110.01.07.
Krastel, S., H.‐U. Schmincke, C. L. Jacobs, R. Rihm, T. P. Le Bas, van Wyk de Vries, B., and R. Matela (1998), Styles of volcano‐induced
and B. Alibés (2001), Submarine landslides around the Canary Islands, deformation: Numerical models of substratum flexure, spreading,
J. Geophys. Res., 106, 3977–3997, doi:10.1029/2000JB900413. and extrusion, J. Volcanol. Geotherm. Res., 81, 1–18, doi:10.1016/
Kennett, B. L. N., and H.‐P. Bunge (2008), Geophysical Continua, Cam- S0377-0273(97)00076-0.
bridge University Press, Cambridge, U. K. Watts, A. B. (1994), Crustal structure, gravity anomalies, and flexure of the
Longpré, M.‐A., V. R. Troll, and T. H. Hansteen (2008), Upper lithosphere in the vicinity of the Canary Islands, Geophys. J. Inter., 119,
mantle magma storage and transport under a Canarian shield‐volcano, 648–666.
Teno, Tenerife (Spain), J. Geophys. Res., 113, B08203, doi:10.1029/ Watts, A. B., and D. G. Masson (2001), New sonar evidence of recent
2007JB005422. catastrophic landslides on the north flank of Tenerife, Canary Islands,
Longpré, M.‐A., V. R. Troll, T. R. Walter, and T. H. Hansteen (2009), Bull. Volcanol., 63, 8–19, doi:10.1007/s004450000119.
Volcanic and geochemical evolution of the Teno massif, Tenerife, Watts, A. B., and U. S. ten Brink (1989), Crustal structure, flexure, and
Canary Islands: Some repercussions of giant landslides on ocean island subsidence history of the Hawaiian Islands, J. Geophys. Res., 94,
magmatism, Geochem. Geophys. Geosyst., 10, Q12017, doi:10.1029/ 10,473–10,500.
2009GC002892. Watts, A. B., and S. Zhong (2000), Observations of flexure and the
Manconi, A., S. Pepe, P. Tizzani, G. Solaro, G. Zeni, and R. Lanari (2010), rheology of the oceanic lithosphere, Geophys. J. Inter., 142, 855–875.
Numerical Inversion of SBAS‐DInSAR time series, Geophys. Res. Watts, A. B., C. Peirce, J. Collier, R. Dalwood, J.‐P. Canales, and T. J.
Abstr., 12, EGU2010‐11877. Henstock (1997), A seismic study of lithospheric flexure at Tenerife,
Martí, J., J. Mitjavila, and V. Araña (1994), Stratigraphy, structure, and Canary Islands, Earth Planet. Sci. Lett., 146, 431–448, doi:10.1016/
geochronology of the Las Cañadas Caldera (Tenerife, Canary Islands), S0012-821X(96)00249-X.
Geol. Mag., 131, 715–727. Wessel, P. (1993), A re‐examination of the flexural deformation beneath
Martí, J., M. Hurlimann, G. J. Ablay, and A. Gudmundsson (1997), Verti- the Hawaiian Islands, J. Geophys. Res., 98, 12,177–12,190.
cal and lateral collapses on Tenerife (Canary Islands) and other volcanic
ocean islands, Geology, 25, 879–882.
A. Camacho and J. Fernández, Instituto de Astronomia y Geodesia,
McGovern, P. J. (2007), Flexural stresses beneath Hawaii: Implications for Consejo Superior de Investigaciones Científicas, Universidad Complutense
the October 15, 2006, earthquakes and magma ascent, Geophys. Res. de Madrid, Ciudad Universitaria, Plaza de Ciencias, 3, ES‐28040, Madrid,
Lett., 34, L23305, doi:10.1029/2007GL031305. Spain.
Merle, O., and A. Borgia (1996), Scaled experiments of volcanic spreading, A. Manconi, IRPI‐CNR, Strada delle Cacce 73, I‐10135 Torino, Italy.
J. Geophys. Res., 101, 13,805–13,817.
M. Manzo, A. Pepe, P. Tizzani, and G. Zeni, Istituto per il Rilevamento
Minshull, T. A., and P. Charvis (2001), Ocean island densities and models Elettromagnetico dell’Ambiente, CNR, Via Diocleziano, 328, IT‐80124,
of lithospheric flexure, Geophys. J. Inter., 145, 731–739, doi:10.1046/ Napoli, Italy. ([email protected])
j.0956-540x.2001.01422.x.
9 of 9
