Non-linear Dynamic Finite Element

Analysis of Adobe Masonry Structures

with Various Roof Diaphragm
Configurations

Rogiros Illampas, Dimos C. Charmpis and Ioannis Ioannou

Abstract This paper deals with the assessment of adobe structures using non-linear
dynamic Finite Element (FE) analysis. Numerical simulation of adobe masonry
follows a continuum approach and is based on brittle fracture concepts that are
expressed through a damaged plasticity constitutive law. The selection of appro-
priate modelling parameters is hereby discussed. Spatial FE models representing a
traditional single-storey dwelling, before and after the implementation of certain
strengthening interventions which modify the roof diaphragm configuration, are
developed. The FE models are subjected to time history analyses using a real time
accelerogram from a past earthquake. Comparisons among the structures’ seismic
responses are made in terms of the predicted damage distribution, displacement
capacity and overall lateral resistance. Useful conclusions regarding the effect that
different roof diaphragm configurations may pose on the dynamic behavior of
adobe structures are derived. Finally, critical issues that future research should
address in order to enable the efficient computational analysis of earthen structures
are identified.

Keywords Adobe masonry

Time history analysis
Non-linear response



Dynamic behaviour Damaged plasticity

R. Illampas (&)
D.C. Charmpis
I. Ioannou

Department of Civil and Environmental Engineering, University of Cyprus,
75 Kallipoleos Str, P.O. Box 20537, 1678 Nicosia, Cyprus
e-mail: [email protected]
D.C. Charmpis
e-mail: [email protected]
I. Ioannou
e-mail: [email protected]

© Springer International Publishing Switzerland 2015 475

I.N. Psycharis et al. (eds.), Seismic Assessment, Behavior and Retrofit
of Heritage Buildings and Monuments, Computational Methods
in Applied Sciences 37, DOI 10.1007/978-3-319-16130-3_20
476 R. Illampas et al.

1 Introduction

Masonry composed of unfired clay bricks has been used for thousands of years. In
our days, adobes are used mainly in developing countries; nevertheless, a large
stock of historic earthen buildings still survives and constitutes a significant feature
of the international cultural heritage. Furthermore, the use of earth masonry in
contemporary architecture is currently being promoted in the context of sustainable
development.
Experimental [3, 19, 24] and field studies [2, 18, 25] have revealed that unre-
inforced adobe masonry construction exhibits relatively poor performance to
dynamic loading and is prone to seismic damage. This implies that there is an
imminent need for the establishment of appropriate appraisal/design tools, which
can assist in the development and assessment of effective measures to protect such
buildings from the destructive effects of earthquakes.
Considerable efforts are currently in progress to extend the applicability of
existing computational methods to account for the specific characteristics of unfired
earth and to capture the dynamic behaviour of adobe construction with sufficient
accuracy. Noticeable work in this field has been undertaken by Tarque et al.
[21–23]. Using experimental data from cyclic and dynamic tests, these researchers
successfully calibrated continuum FE models. FE models of continua have also
been developed by Che et al. [5] and Gomes et al. [9] for performing linear dynamic
analyses on earth-based structures. More detailed micro-modeling strategies, based
on the distinct element method, have been adopted by Cao and Watanabe [4], who
examined the seismic behaviour of unreinforced and retrofitted adobe structures.
Using a similar approach, Furukawa and Ohta [7] and Furukawa et al. [8] con-
ducted dynamic simulations to investigate the application of casualty estimation
vulnerability functions in the case of earthquake-induced collapse.
This study aims to contribute towards the use of continuum FE models in the
analysis of traditional adobe buildings and the assessment of relevant retrofitting
techniques. An isotropic damaged plasticity model that has been previously cali-
brated using experimental data [12] is adopted for the numerical material modelling
of adobe masonry. A 3D model of a single-storey vernacular dwelling is developed
and subjected to time history analysis before and after the implementation of var-
ious strengthening interventions which modify the roof diaphragm configuration
(i.e. roof-to-wall anchorage, ring beam installation). Comparisons are made
between the predicted modes of damage and the magnitudes of lateral resistance
and displacement capacity. Through the numerical results obtained, the efficiency of
certain retrofitting methods is evaluated. Moreover, key issues that future research
should address are identified.
Non-linear Dynamic Finite Element Analysis … 477

2 Material Models

In this study adobe masonry was numerically handled in the context of a macro-
modelling strategy and was thus idealized as a fictitious homogeneous continuum,
which was discretized into a FE mesh. Damage was treated as a smeared property
spreading over the elements.
For the numerical simulation of adobe masonry, the concrete damaged plasticity
constitutive model implemented in Abaqus/CAE [15, 17, 20] was adopted. This
model assumes two modes of failure: tensile cracking and compressive crushing.
Inelastic behaviour is represented using concepts of isotropic damaged elasticity, in
combination with isotropic tensile and compressive plasticity. The yield surface is
an extension of the classical Drucker-Prager model. The admissible stress field is
controlled by means of hardening variables that are linked to cracking and crushing
strains.
The parameters required for the formulation of the constitutive law were mainly
derived from experimental data. The density of earth masonry composed by adobes
from Cyprus was measured as ρ = 1300 kg/m3. Poisson’s ratio (ν = 0.3) was
evaluated as a transverse-to-axial strain ratio from the deformations recorded during
the compressive strength testing of a stack-bonded prism.
Response to compression was defined in accordance with the polynomial stress-
strain relation developed by Illampas et al. [13] for adobe bricks (Fig. 1a). Com-
pressive strength (fc = 1.2 MPa) and strain at peak compressive stress
(εcu = 0.1 mm/mm) were set equal to the average experimental results obtained by
Illampas et al. [11] through tests on stack-bonded prisms. As elastic response of
soil-based granular materials to compression is generally observed only at rather
low stresses and strains, nonlinearity was assumed at stresses exceeding 5 % of the
maximum bearing capacity. Young’s modulus was estimated from the assigned
compressive stress-strain response as a secant modulus up to the yielding point;
E = 18 MPa.

(a) 1.4 (b) 0.05

1.2
0.04
1.0
Stress (MPa)

Stress (MPa)

0.8 0.03

0.6
0.02
0.4
0.01
0.2

0.0 0.00
0.0 0.1 0.2 0.3 0.4 0.5 0.00 0.01 0.02 0.03
Strain (mm/mm) Strain (mm/mm)

Fig. 1 Compressive (a) and tensile (b) stress-strain response assigned to the homogenized adobe
masonry medium
478 R. Illampas et al.

In tension (Fig. 1b), the medium’s response was considered to be linear up to the
maximum allowable stress. In the post-peak regime, softening behaviour was
defined adopting the exponential function developed by Lourenço [16]:


hft
rt ¼ ft exp  eck : ð1Þ
Gf t

In the above equation, ft is the tensile strength, Gf is the tensile fracture energy, eck
t
is the tensile cracking strain and h is the characteristic crack length.
Uniaxial tensile strength was set as ft = 0.04 MPa in accordance with the outcomes
of diagonal tension tests on an adobe wallette. Tensile fracture energy Gf was esti-
mated based on the results of direct tension tests reported by Almeida [1]. This author
noted that the average energy required for the failure of adobe couplets was 4.5 N/m
for a mean tensile strength of 0.01 MPa. Assuming a linear analogy between the
tensile strength and the fracture energy, ft = 0.04 MPa yields Gf = 18 N/m.
The characteristic crack length h for the first order quadrilateral shell elements
used in the FE simulations hereby presented was computed from the elements’
dimensions hx, hy as [6]:
pffiffiffiffiffiffiffiffiffiffiffi
h¼ 2hx hy : ð2Þ

The elements’ size was selected so as to satisfy the energy criterion:

Gf E
h : ð3Þ
ft2

Following Eq. (3) and taking into account the mechanical properties of the masonry
examined in this study, the elements’ size was kept below 0.2 m in all simulations
conducted. Furthermore, to minimize FE mesh sensitivity of numerical results,
structured meshing was performed.
For the full definition of the yield surface, four plasticity parameters are addi-
tionally required: (i) the dilation angle ψ that controls the amount of plastic volu-
metric strain developed during plastic shearing, (ii) the eccentricity e that controls
the rate at which the hyperbolic flow potential approaches its asymptote, (iii) the
ratio of the initial equibiaxial compressive yield stress to the initial uniaxial com-
pressive yield stress σb0/σc0 and (iv) the parameter Kc that relates the second stress
invariant on the tensile meridian to the equivalent invariant on the compressive
meridian. According to Tarque et al. [23], clays show limited dilatancy; hence,
ψ = 1° was defined. Eccentricity was selected as e = 0.1, implying that the dilation
angle remains almost constant over a wide range of confining pressure stress values.
For the ratio between the initial equibiaxial and the initial uniaxial compressive
yield stresses, the default value suggested in [20] was adopted, σb0/σc0 = 1.16.
Parameter Kc was set as 0.8 in line with the recommendations of Simulia Corp. [20]
for soils modelled with a Drucker-Prager yield function.
Non-linear Dynamic Finite Element Analysis … 479

In addition to load-bearing masonry walls, traditional adobe buildings incorporate

timber structural members, such as lintels and roof rafters. These were modelled to
exhibit isotropic linear elastic behaviour. The material parameters used were drawn
from the literature [10, 14]: density, ρ = 600 kg/m3; Young’s modulus, E = 7000 MPa;
Poisson’s ratio, ν = 0.3.

3 FE Modelling of Adobe Masonry Structures

In order to numerically examine the dynamic behavior of adobe masonry con-

struction, a FE model of a typical earthen building was developed in Abaqus/CAE
and analyzed under seismic excitation (Fig. 2). The FE model prepared represents a
single-roomed rectangular structure composed of four load-bearing walls that carry
the superimposed weight of the roof.
The simulated building has dimensions (height × width × length)
3.5 × 3.5 × 8.5 m3 and its walls are 0.45 m thick. Two doors 2.2 m high and 1.0 m wide
exist on each of the longitudinal walls. Windows 1.0 × 1.0 m2 are formed on the side
walls. Above all openings, timber lintels of cross-section 0.15 × 0.45 m2 and length

X
Y (EAST)
Z

Direction of
imposed
seismic action

Fig. 2 FE model used for examining the dynamic behaviour of a typical adobe structure
480 R. Illampas et al.

1.4 m are installed. Timber rafters 0.10 × 0.14 m2 in cross-section are placed in
parallel at regular intervals of 0.5 m to span the space between the two opposite
longitudinal walls.
Adobe walls were discretized using 4-noded shell elements with reduced inte-
gration formulation (S4R) and 6 degrees of freedom (DOF) per node. The
approximate global size of the elements was set as 0.075 × 0.075 m2 based on the
criterion given in Eq. (3). All wall elements had five Gauss integration points
through the thickness to account for out-of-plane bending. The openings’ lintels
were modelled using the same type of shell elements. Common nodes were
assigned at the areas where the lintels were in contact with the masonry. The roof’s
rafters were represented using 2-noded linear beam elements (B31) linked with the
wall through connector elements (CONN3D2); additional DOFs were generated at
the wall-rafter contact points. The FE mesh contained in total 11,998 shell elements
for the walls and lintels, as well as 360 beam elements and 30 connector elements
for the rafters and their connections, resulting altogether in 77,160 DOFs.
Aiming to obtain comparable results, three structures with different roof con-
figurations were analyzed. The first case considered resembles a building where the
abutments of the roof rafters have been simply set into masonry with the application
of mortar. It was assumed that the timber-mortar interface has limited cohesive and
frictional resistance and thus the imposition of lateral forces can cause sliding of the
rafters. Partial bonding between the walls and the rafters was modelled using rigid
springs in the x and z directions and springs with minimal stiffness (10 N/m) in the
y direction. The second case considered represents a structure where the rafters have
been anchored to the walls to enhance the transfer of forces among the façade and
the rear wall. In this model, bonding between the walls and the rafters was repre-
sented with pinned connectors. Finally, a retrofitted structure incorporating a ring
beam composed of 0.08 × 0.08 m2 timber sections was examined. The ring beam
was modelled using 306 2-noded linear beam elements (B31) and was connected to
the structure’s top via the imposition of tie constraints. The roof rafters were
assumed to be pinned on the ring beam. In total, the FE mesh incorporating the ring
beam results in 78,996 DOFs.
The roof unit weight was set as 130 kg/m2. To account for the swaying roof mass
during the dynamic excitation, the load attributed to each rafter was incorporated
into its mass matrix as an additional weight per unit volume. The total weight of the
structure was estimated as 420 kN. Translational DOFs along the x, y and z axes
were constrained at the walls’ base.

4 Dynamic FE Simulations and Numerical Results

The numerical solution process was completed in two successive steps. At the
initial step, the dead loads were applied using a general static analysis procedure.
Upon the transition from the static to the dynamic step, the boundary conditions at
Non-linear Dynamic Finite Element Analysis … 481

the walls’ base were modified by replacing the translational constraint along the
y axis with an acceleration acting in the same direction.
The total time period of the dynamic step and the amplitude of the acceleration
boundary condition were defined in accordance with the accelerogram of the east-
west component of the 1978 June 20th Thessaloniki Earthquake (Fig. 3). The
duration of the selected seismic event is 30.59 s. Peak ground acceleration is
0.146 g in the west direction and was recorded after 8.83 s. The minimum and
maximum load imposition time increments in the dynamic step were set as
1 × 10−12 and 0.01 s respectively. The effects of geometric nonlinearities were
accounted for in both analysis steps.
The damage pattern yielded by the analysis for each of the three structures
examined is given in Fig. 4. This figure shows a contour representation of the
maximum principal plastic strains computed at the elements’ integration points.
Crack prediction lies on the hypothesis that the direction of the vector normal to the
crack plane is parallel to the direction of the maximum principal plastic strains.
In the structure where sliding of roof rafters was not restrained, the out-of-plane
oscillation of the longitudinal walls led to the development of vertical cracks at their
upper central sections (Fig. 4a). Soon afterwards, cracks were generated at the
window corners and at the vicinity of the lintels due to stress concentrations. As the
intensity of the seismic action progressively increased, the magnitude of axial
tension transferred at the corners of the structure (along the interface of longitudinal
and side walls) exceeded the bearing capacity of the masonry, leading to the
development of vertical detachment cracks. When the peak ground acceleration was
imposed, tensile damage at the two corners of the west wall propagated throughout
the whole height of the structure. An overturning kinematic mechanism was thus
formed and the analysis was terminated due to lack of numerical convergence. It is
worth noting that most researchers [3, 4, 19, 25] point out that out-of-plane collapse
of the walls orthogonal to the direction of the seismic action is quite common in
unreinforced adobe masonry constructions.
Anchoring of the roof rafters to the walls altered the distribution of tensile
damage and changed the structure’s mode of failure (Fig. 4b). Cracking initiated at
the supports of the central and the two outer roof rafters due to stress concentration

1.5
Acceleration (ms-2)

1.0
0.5
0.0
-0.5
-1.0
-1.5
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Time (s)

Fig. 3 Accelerogram of the east-west component of the 1978 June 20th Thessaloniki earthquake
482 R. Illampas et al.

(a)
PEMAG
Multiple section points
(Avg: 75%)
+1.653e+00
+1.515e+00
+1.377e+00
+1.240e+00
+1.102e+00
+9.642e-01
+8.264e-01
+6.887e-01
+5.510e-01
+4.132e-01
+2.755e-01
+1.377e-01
+0.000e+00

(b)

PEMAG
Multiplesection points
(Avg: 75%)
+1.490e+00
+1.366e+00
+1.242e+00
+1.118e+00
+9.935e-01
+8.693e-01
+7.451e-01
+6.209e-01
+4.967e-01
+3.725e-01
+2.484e-01
+1.242e-01
+0.000e+00

(c)

PEMAG
Multiple section points
(Avg: 75%)
+2.067e-01
+1.895e-01
+1.723e-01
+1.551e-01
+1.378e-01
+1.206e-01
+1.034e-01
+8.614e-02
+6.891e-02
+5.168e-02
+3.446e-02
+1.723e-02
+0.000e+00

Fig. 4 Distribution of maximum principal plastic strains at the end of the seismic FE analysis:
a roof rafters assumed to slide freely at their supports, b roof rafters anchored to the underlying
masonry and c roof rafters pinned on a timber ring beam installed at the top of the structure
Non-linear Dynamic Finite Element Analysis … 483

at the rigid pin connectors. Almost coinstantaneously, vertical detachment cracks

developed at the corners of the east wall. Damage propagation at these sections was
rapid and before the peak ground acceleration was reached, tensile plastic strains
had already spread across the whole height of the structure. Out-of-plane deflection
of the detached east wall was transferred to the opposite wall via the rigidly con-
nected roof rafters. This resulted to significant deformation of the west wall and
subsequent crack growth. When tensile damage below the two outer rafters
extended vertically throughout the whole height of the west wall, a rocking
mechanism involving parallel tilting of the two longitudinal walls was formed and
the analysis was again terminated.
A distinctly different response was observed in the case of the retrofitted
structure (Fig. 4c). The installation of a ring beam enabled diaphragmatic function
at the roof level. This provided additional support against lateral deformations and
ensured integral box action of the entire structure. The in-plane loaded side walls
developed cracks that radiated out of the openings’ corners and propagated at a
slightly inclined angle towards the top and base of the structure. The restraint
imposed to the movement of the longitudinal walls decreased the severity of out-of-
plane damage. Indeed, only minor diagonal cracks were noted at the east and west
walls’ upper sides. Despite the fact that the structure suffered considerable cracking
damage, collapse limit state was not reached.
Figure 5 compares the variation of the relative lateral displacement measured at
the west wall’s central upper section over time for the three structures examined.
Relative displacement data were computed by subtracting the translation of the
wall’s central base node along the y axis (caused by the seismic ground accelera-
tion) from the total out-of-plane displacement of the corresponding nodal point at
the top. Similar response is observed during the first 6.8 s of the seismic event,
regardless of the roof configuration. In addition, at 8.83 s when the seismic
Relative displacement (m)

0.3
0.2
0.1
0.0
-0.1
-0.2 Sliding Roof Rafters
-0.3 Pinned Roof Rafters
-0.4 Ring Beam
-0.5
0 4 8 12 16 20 24 28 32
Time (s)

Fig. 5 Time history of the relative displacement measured along the direction of the seismic
action (y) at the west wall’s central upper section for the three structures examined
484 R. Illampas et al.

acceleration attains its peak value, all three analyses predict a relative translation of
−0.098 m. For the structure strengthened with a ring beam, this value coincides
with the maximum deflection generated; from this point onwards, the wall con-
tinues to oscillate around a new equilibrium position at approximately −0.025 m,
indicating residual inelastic deformation [23]. In the case where the roof rafters

(a)
90
Base shear force (kN)

60

30

0
-0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20
-30

-60

-90
Relative displacement (m)
(b)
90
Base shear force (kN)

60

30

0
-0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20
-30

-60

-90
Relative displacement (m)
(c)
90
Base shear force (kN)

60

30

0
-0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20
-30

-60

-90
Relative displacement (m)

Fig. 6 Base shear force versus relative displacement in the y direction at the upper central sections
of the west (a, c) and east (b) walls: a roof rafters assumed to slide freely at their supports, b roof
rafters anchored to the underlying masonry and c roof rafters pinned on a timber ring beam
installed at the top of the structure
Non-linear Dynamic Finite Element Analysis … 485

were assumed to slide freely, load reversals after the peak ground acceleration
caused the detached west wall to excessively deflect out of plane. Analogous
behaviour is noted by the structure with the pinned roof rafters. In this case how-
ever, failure of east wall’s corners preceded damage formation at the west wall.
Therefore, deflection of the east wall is controlled by the parallel out-of-plane
rocking motion of the west wall and the maximum relative displacement predicted
is in the opposite direction.
The variation of the base shear force in relation to the relative displacement
generated at the upper sections of the structures examined is presented in Fig. 6.
The relative displacement data given for the structures with the sliding rafters and
the ring beam (Fig. 6a, c) refer to the relative movement of the west wall’s central
top node, whereas the control nodal point for the structure with the pinned roof
rafters (Fig. 6b) is at the top of the centre of the east wall. Results show that rafter-
to-wall anchorage, in comparison to practically freely sliding roof rafters, increases
the ratio between the maximum base shear force and the structure’s self-weight to
14 % from about 12 % (59.1 instead of 49.3 kN). This is because forces can be
effectively transferred between the walls perpendicular to the direction of the
earthquake and the masonry area resting the seismic actions increases. Considerably
better behaviour can be achieved by the installation of a ring beam. As the dia-
phragm formed enforced equal displacements at the upper sections of the walls,
inertia forces were distributed between the vertical resisting elements as a function
of their stiffness and plane position. Inevitably, the magnitude of the lateral resis-
tance was dictated by the in-plane shear strength of the stiffer side walls that is
significantly higher than the out-of-plane bending capacity. The simulated structure
was able to sustain a maximum horizontal force of 86.0 kN, which accounts for
more than 20 % of its own weight.

5 Conclusions

The numerical results obtained in this study verify that the lack of diaphragmatic
function at the roof level precludes the effective transfer of loads among the load-
bearing walls of masonry structures, adversely affecting their dynamic behaviour.
Unreinforced adobe buildings fail to react as a homogeneous assemblage of
structural elements (i.e. as a structural system) under seismic excitation and out-of-
plane failure due to detachment at the cross-walls and overturning prevails.
Localized interventions such as roof-to-wall anchorage can increase, albeit to a
limited degree, the bearing capacity of an adobe masonry building. However, they
fail to ensure integral box action of the entire structure and may result to more
extended seismic damage due to the introduction of stress concentrations. On the
other hand, retrofitting solutions involving the installation of ring beams can sig-
nificantly improve the poor seismic response of adobe structures, not only by
increasing their overall lateral resistance, but also by producing a confinement
effect, which additionally prevents brittle collapse. Considering the balance
486 R. Illampas et al.

between effectiveness and cost of application, strengthening a structure with the

addition of a perimeter timber beam may be deemed as sufficient. A more holistic
approach for the seismic upgrading of adobe structures may include additional
measures, such as the installation of meshes at the surface of the masonry, the
installation of ties at wall corners, etc.
Although from a qualitative point of view the numerical outcomes hereby pre-
sented are in line with the data reported in other experimental, numerical and field
studies, further research is required to fully capture the dynamic response of adobe
masonry construction through FE simulations. First of all, a thorough and detailed
database of information regarding the properties of adobe materials and masonry
components should be established, in order to facilitate the valid selection of
appropriate modelling parameters. Emphasis should be given on examining their
response to tension, as this appears to strongly influence the overall structural
behaviour of adobe construction. Future work should also focus on investigating
elastic modulus reduction and stiffness recovery effects caused by crack opening-
closing under reversing dynamic loads. Furthermore, the sliding and bonding
strengths at the interfaces among timber and adobe structural members should be
evaluated to enable the correct simulation of the roof-wall interaction mechanism.
Last but not least, large scale dynamic tests on complete adobe structures should be
carried out to enable the accurate calibration of FE models.

References

1. Almeida JAPP (2012) Mechanical characterization of traditional adobe masonry elements.

In: MSc thesis, Department of civil engineering, University of Minho
2. Blondet M, Vargas J, Tarque N (2008) Observed behaviour of earthen structures during the
Pisco earthquake (Peru). In: Proceedings of the 14th world conference on earthquake
engineering, Beijing, China, 12–17 Oct
3. Bossio S, Blondet M, Rihal S (2013) Seismic behavior and shaking direction influence on
adobe wall structures reinforced with geogrid. Earthq Spectra 29(1):59–84
4. Cao Z, Watanabe H (2004) Earthquake response prediction and retrofitting techniques of
adobe structures. In: Proceedings of the 13th world conference on earthquake engineering,
Vancouver, B.C. Canada, 1–6 Aug
5. Che AL, Wu ZJ, Sun JJ, Qi JH (2010) Seismic damage characteristics of rural adobe-wood
building in Gansu Province induced by the Wenchuan great earthquake. In: Proceedings of the
international symposium on geoenvironmental engineering, Hangzhou, China, 8–10 Sept
6. Dhanasekar M (2010) Review of modelling of masonry shear. Int. J Adv Eng Sci Appl Math 2
(3):106–118
7. Furukawa A, Ohta Y (2009) Failure process of masonry buildings during earthquake and
associated casualty risk evaluation. Nat Hazards 49(1):25–51
8. Furukawa A, Spence R, Ohta Y, So E (2009) Analytical study on vulnerability functions for
casualty estimation in the collapse of adobe buildings induced by earthquake. Bull Earthq Eng
8(2):451–479
9. Gomes MI, Lopes M, Brito JD (2011) Seismic resistance of earth construction in Portugal.
Eng Struct 33(3):932–941
Non-linear Dynamic Finite Element Analysis … 487

10. Green DW, Winandy JE, Kretschman DE (1999) Mechanical properties of wood. In: Wood
handbook: wood as an engineering material. U.S. Department of Agriculture, Forest Service,
Forest Products Laboratory, General Technical Report FPL-GTR-113, Madison WI
11. Illampas R, Charmpis DC, Ioannou I (2010) An assessment of the compressive strength of
adobe brick assemblages. In: Proceedings of the 9th HSTAM international Congress on
mechanics, Limassol, Cyprus, 11–14 July
12. Illampas R, Charmpis DC and Ioannou I (2014a) Laboratory testing and finite element
simulation of the structural response of an adobe masonry building under horizontal loading.
Eng Struct 80:362–376
13. Illampas R, Ioannou I, Charmpis DC (2014b) Adobe bricks under compression: experimental
investigation and derivation of stress–strain equation. Constr Build Mater 53:83–90
14. Katsaragakis ES (2000) Timber construction. NTUA Academic Publications, Athens
(in Greek)
15. Lee J, Fenves G (1988) Plastic-damage model for cyclic loading of concrete structures. J Eng
Mech 124(8):892–900
16. Lourenço PB (2000) Anisotropic softening model for masonry plates and shells. J Struct Eng
126(9):1008–1016
17. Lubliner J, Oliver J, Oller S, Oñate E (1989) A plastic-damage model for concrete. Inter J
Solids Struct 25(3):299–326
18. Maheri MR, Naeim F, Mehrain M (2005) Performance of adobe residential buildings in the
2003 Bam, Iran, Earthquake. Earthq Spectra 21(S1):337–344
19. Ottazzi J, Yep J, Blondet, M, Villa-Garcia, G, Ginocchio, JF (1988) Shaking table tests of
improved adobe masonry houses. In: Proceedings of the 9th world conference on earthquake
engineering, Tokyo-Kyoto, Japan, 2–9 Aug
20. Simulia Corp. (2009) Abaqus 6.10 theory manual. Dassault Systèmes, Rising Sun Mills
21. Tarque N, Camata G, Spacone E, Varum H, Blondet M (2012a) Elastic and inelastic
parameters for representing the seismic in-plane behaviour of adobe wall. In: Proceedings of
the XIth international conference on the study and conservation of earthen architectural
heritage terra 2012, Lima, Peru, 22–27 Apr
22. Tarque N, Camata G, Spacone, E, Varum H, Blondet M (2012b) Non-linear dynamic analysis
of an adobe module. In: Proceedings of the XIth international conference on the study and
conservation of earthen architectural heritage terra 2012, Lima, Peru, 22–27 Apr
23. Tarque N, Camata G, Spacone E, Varum H, Blondet M (2014) Non-linear dynamic analysis of
a full-scale unreinforced adobe model. Earthq Spectra 30(4):1643–1661
24. Tolles EL, Krawinkler H (1990) Seismic studies on small-scale models of adobe houses. John
A. Blume Earthquake Engineering Center, Technical Report No. 91. Department of Civil and
Environmental Engineering, Stanford University, Stanford
25. Tolles EL, Webster FA, Crosby A, Kimbro EE (1996) Survey of damage to historic adobe
buildings after the January 1994 Northridge earthquake. The Getty Conservation Institute, Los
Angeles