Layout 6 ANNALS OF GEOPHYSICS, 56, 6, 2013, S0677; doi:10.4401/ag-6222 S0677 Neotectonics of Graciosa island (Azores): a contribution to seismic hazard assessment of a volcanic area in a complex geodynamic setting Ana Hipólito1,*, José Madeira2, Rita Carmo1,3, João Luís Gaspar1,3 1 Centro de Vulcanologia e Avaliação de Riscos Geológicos da Universidade dos Açores, Açores, Portugal 2 Universidade de Lisboa, Faculdade de Ciências, Departamento de Geologia, and Instituto Dom Luiz (Laboratório Associado)-IDL(LA), Lisboa, Portugal 3 Centro de Informação e Vigilância Sismovulcânica dos Açores, CIVISA, Açores, Portugal ABSTRACT Graciosa is a mid-Pleistocene to Holocene volcanic island that lies in a complex plate boundary between the North American, Eurasian, and Nubian plates. Large fault scarps displace the oldest (Middle Pleis- tocene) volcanic units, but in the younger areas recent volcanism (Holocene to Upper Pleistocene) conceals the surface expression of faulting, limiting neotectonic observations. The large displacement ac- cumulated by the older volcanic units when compared with the younger formations suggests a variability of deformation rates and the possi- bility of alternating periods of higher and lower tectonic deformation rates; this would increase the recurrence interval of surface rupturing earthquakes. Nevertheless, in historical times a few destructive earth- quakes affected the island attesting for its seismic hazard. Regarding the structural data, two main fault systems, incompatible with a sin- gle stress field, were identified at Graciosa Island. Thus, it is proposed that the region is affected by two alternating stress fields. The stress field #1 corresponds to the regional stress regime proposed by several authors for the interplate shear zone that constitutes the Azorean seg- ment of the Eurasia-Nubia plate boundary. It is suggested that the stress field #2 will act when the area under the influence of the regional stress field #1 narrows as a result of variations in the differential spreading rates north and south of Azores. The islands closer to the edge of the sheared region will temporarily come under the influence of a different (external) stress field (stress field #2). Such data support the concept that, in the Azores, the Eurasia-Nubia boundary corre- sponds to a complex and wide deformation zone, variable in time. 1. Geodynamic and volcanic setting The Azores archipelago lies on a complex geody- namic setting on the triple junction between the Eurasian (Eu), North American (NA), and Nubian (Nu) plates (Figure 1). Graciosa Island is located on the west- ern segment of the Eu-Nu boundary, which is charac- terized by diffuse and complex deformation in a dex- tral transtensile tectonic regime [e.g. Lourenço et al. 1998, Madeira and Brum da Silveira 2003, Carmo 2004, Hipólito 2009] (Figure 1). This boundary acts as an ultra-slow (ca. 4.5 mm/a-DeMets et al. 2010) oblique Article history Received October 8, 2012; accepted May 10, 2013. Subject classification: Structural geology, Stress, Plate boundaries, motion and tectonics, Geodynamics, Seismic risk. Figure 1. Main morphotectonic features of Azores region. White lines define approximately the morphological expression of each structure; white shaded area represents the sheared western seg- ment of the Eu-Nu plate boundary, whereas the white shaded area limited by a dotted grey line represents its main structure, the Ter- ceira Rift (TR). Plates: Eu - Eurasia; Nu - Nubia; NA - North Amer- ica; Tectonic structures: MAR - Mid-Atlantic Ridge; EAFZ - East Azores Fracture Zone; NAFZ - North Azores Fracture Zone; GF - Gloria Fault; FFZ - Faial Fracture Zone; AFZ - Açor Fracture Zone; PAFZ - Princesa Alice Fracture Zone; PFZ - Pico Fracture Zone. Is- lands: SMA - Santa Maria; SM - São Miguel; T - Terceira; SJ - São Jorge; P - Pico; F - Faial; FL - Flores; C - Corvo. Azores bathymetry adapted from Lourenço et al. [1997]; World topography and ba- thymetry from GEBCO_08 database [2010]. Datum: WGS 1984. Special Issue: Earthquake geology spreading centre [Vogt and Jung 2004] and as a transfer zone accommodating the differential motion between Eu and Nu plates, due to the higher spreading rates and slightly different spreading direction north of Azores [DeMets et al. 1994, Sella et al. 2002, Altamimi et al. 2002, Fernandes et al. 2003, Kreemer et al. 2003, Calais et al. 2003, DeMets et al. 2010]. The stress regime, the tectonic processes, and the segmentation patterns of the western segment of the Eu-Nu plate boundary are still a matter of debate. This boundary does not corre- spond to a discrete structure but to a wide (~150 km) large-scale interplate shear zone, and the islands within this region present average motions that are interme- diate to those predicted by global plate motion models for Eu and Nu [Fernandes et al. 2006, Trota 2008, Mi- randa et al. 2012, Mendes et al. 2013]. Graciosa and Santa Maria islands, the emerged areas closer to the north and south edges of the plate boundary zone, dis- play kinematic behaviours close to those predicted for stable Eu and Nu, respectively [Fernandes et al. 2006]. Some authors [e.g. Schilling 1975, Cannat et al. 1999, Escartín et al. 2001, Gente et al. 2003, Madureira et al. 2005, Silveira et al. 2006, Yang et al. 2006] consider that the intense magmatism in the region results from a ridge-hotspot interaction. Shifting of the plate bound- ary, possibly associated to migration of the hotspot [e.g. Vogt and Jung 2004], created a topographically dis- turbed area which has been successively attached to the Nu plate as the triple junction jumped northwards [Luis and Miranda 2008]. This region - the Azores Plateau [Needham and Francheteau 1974] - is characterized by an anomalously thick ocean crust [e.g. Hirn et al. 1980, Luís et al. 1998, Miranda et al. 1998, Dias et al. 2007] and shallow depths relatively to the surrounding sea-floor. The neotectonic data [e.g. Madeira and Brum da Silveira 2003] on the islands show a structural pattern indicating three-dimensional strain [Reches 1983, Reches and Dieterich 1983], represented by two main conjugated sets of faults with oblique slip (Figure 2): (1) the main system, trending WNW-ESE to NW-SE, with normal dextral oblique displacement is parallel to the inter-plate shear zone; and (2) the NNW-SSE trend- ing conjugate system, with normal sinistral oblique dis- placement, is oblique to the plate boundary direction. The stress field is characterized by horizontal maxi- mum compressive stress axis (σ1) trending NW-SE in the axial area of the shear zone (which rotates to N-S near Santa Maria island; Madeira [1986], Figure 2), hor- izontal maximum tensile stress axis (σ3) trending NE- SW (rotating to E-W in Santa Maria Island; Madeira [1986], Figure 2), and vertical intermediate compressive stress axis (σ2). Permutations between σ1 and σ2 after stress drop events during transtensile phases, as pro- posed by Reches [1983], account for alternating transtensile and tensile regimes [Madeira 1998]. The changing regimes are deduced both from slickensides indicating events in strike-slip, oblique slip, or normal slip in the same fault, and from different focal mecha- nisms of earthquakes in faults with the same direction (Table 1). In the eastern edge of the deformation zone (eastern part of São Miguel) a distinct set of WNW-ESE to NW-SE normal left-lateral faults conjugated with NE- HIPÓLITO ET AL. 2 Figure 2. Stress pattern and stereographic plots of the main fault systems of the Azores region as inferred from neotectonic data [Madeira 1986, Madeira and Ribeiro 1990, Madeira 1998, Lourenço et al. 1998, Carmo 2004]. The tensile stress field (a) close to the MAR passes lat- erally to compressive (b) as the sea-floor moves away from the ridge. Inside the sheared region (c) stress permutation and deviation of stress trajectories occur. Red dashed lines separate the different stress field domains; White shaded area represents the interplate sheared region. Acronyms as in Figure 1. Bathymetry adapted from Lourenço et al. [1997]. Datum: WGS 1984 (modified from Madeira [1998], Lourenço et al. [1998] with data from Carmo [2004]). 3 NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Table 1. Seismic events with published focal parameters for the Azores region (from 1900 to present). References: CMT: centroid-moment tensor, Havard in Borges et al. [2007, 2008], BUF: Buforn et al. [1988], BOR: Borges et al. [2007]. For the magnitude w and s indicate Mw e Ms, respectively [Borges et al. 2007, 2008]. Event n° Date (d/m/yr) Latitude (°N) Longitude (°E) Depth (km) M Mo (x1017 Nm) Strike Rake Dip ref 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 08/05/1939 06/09/1964 04/07/1966 05/07/1966 20/04/1968 23/11/1973 11/12/1973 01/01/1980 02/12/1981 09/09/1984 16/10/1988 21/11/1988 21/01/1989 26/06/1989 23/09/1989 09/12/1991 20/01/1993 09/03/1996 27/06/1997 27/06/1997 28/06/1997 09/07/1998 01/08/2000 30/11/2002 05/04/2007 04/11/2007 37.40 38.30 37.50 37.60 38.30 38.46 38.74 38.81 38.38 36.93 37.38 38.34 37.92 39.11 39.27 37.22 38.39 37.13 38.33 38.26 38.41 38.65 38.79 39.25 37.45 37.40 -23.90 -26.60 -24.70 -24.70 -26.60 -28.31 -28.67 -27.78 -26.13 -24.60 -25.16 -26.27 -25.92 -28.32 -29.24 -23.61 -29.34 -23.85 -26.68 -26.16 -26.64 -28.63 -29.01 -28.45 -24.62 -24.39 15 15 10 18 15 15 15 7 15 12 15 15 15 15 15 15 15 15 7 15 15 7 15 15 12 12 7.1s 5.1w 5.5w 5.0w 4.6w 5.1s 5.0w 6.8w 5.6w 5.3w 5.3w 5.9w 5.7w 5.8w 5.1w 5.2w 5.4w 5.7w 5.8w 5.2w 5.1w 6.0w 5.1w 5.1w 6.2w 6.0w 199 0.54 1.90 0.41 0.09 2.0 0.34 190 3.20 0.95 0.89 7.10 3.40 5.40 0.44 0.82 1.20 3.80 7.0 0.62 0.58 14 0.51 0.52 41 11 41 185 341 180 117 23 329 149 141 178 303 345 131 105 233 330 132 319 290 284 290 153 97 106 129 133 35 62 49 48 42 90 58 85 42 37 90 29 41 32 45 45 33 28 44 27 44 85 62 45 44 44 -154 3 -42 30 89 -179 -20 -2 -80 -79 180 -37 -87 -110 -90 -90 -59 -106 -114 -147 -114 6 -170 -129 -89 -87 BUF BUF BUF BUF BUF BUF BUF BOR CMT CMT CMT CMT CMT CMT CMT CMT CMT CMT BOR CMT CMT BOR CMT CMT CMT CMT Location Figure 3.Seismicity in Azores region from 1980 to 2012. The red dots mark epicentres of earthquakes. Plot includes earthquakes of all mag- nitudes. Acronyms as in Figure 1. Data from CIVISA (Centro de Informação e Vigilância Sismovulcânica dos Açores), 2013. SW normal dextral faults, is incompatible with the re- gional stress field described above (Carmo [2004], Fig- ure 2). This set of faults is ascribed to a different, previously unknown, stress field, characterized by hor- izontal E-W σ1. Due to this geodynamic setting, the Azores region is subject to active volcanism and frequent seismicity (Figure 3). The seismic activity is characterized by iso- lated seismic events and/or swarms of tectonic and/or volcanic nature, with shallow focal depths and low to moderate magnitudes. Moderate to high magnitude earthquakes, up to M=7, have also occurred in modern times (Table 1). Instrumental data present seismic events with variable focal mechanisms mainly indicating dex- tral (with WNW-ESE and NNE-SSW nodal planes) and sinistral, normal slip (with NW-SE trending nodal planes), oblique-slip, and a few reverse events [McKenzie 1972, Udías et al. 1976, Udías 1980, Hirn et al. 1980, Grimison and Chen 1986, Udías et al. 1986, Buforn et al. 1988, Moreira 1991, Mezcua et al. 1991, Miranda et al. 1998, Lourenço et al. 1998, Matias et al. 2007, Dias et al. 2007, Borges et al. 2007, 2008, Borges and Buforn 2008, Silva et al. 2012].The instrumental seismicity recorded since the 1980’s highlights the Eu-Nu plate boundary (Figure 3). Graciosa Island was affected by a few significant earthquakes since settlement in mid-15th century, caus- ing some deaths and severe damage (Table 2). Based on the historical descriptions of damages, some authors suggest that the epicentre of the January 21st, 1837 earthquake was on land [Madeira 1998, Silva 2005]. In- strumental seismicity in and around Graciosa is pre- sented in Figure 4. Graciosa has a NW-SE trending elliptical shape, 12 km long, 7 km wide and has a smooth relief reaching a maximum elevation of 402 m. As in other volcanic islands of the Azores archipelago, the elongated mor- phology of Graciosa reflects the main tectonic trend of the Azorean segment of the Eu-Nu plate boundary and, more specifically, the NW-SE direction of the western sector of the Terceira Rift [Machado 1959], a tectono-magmatic structure defined by the alignment of alternating volcanic buildings (Graciosa and Ter- ceira Islands, Banco Dom João de Castro seamount, and São Miguel Island) and tectonic basins (West and East Graciosa and North and South Hirondelle). The tectonic structure controlled most of the volcanic ac- tivity responsible for the build-up of the island. That is a common phenomenon that occurs in several oceanic volcanic islands independently of their geodynamic setting, such as the Fuerteventura and Lanzarote is- lands (Canary Islands) in which the elongation clearly HIPÓLITO ET AL. 4 Table 2. Historical seismicity in Graciosa Island (Modified from Gaspar [1996]). References: 1. Moniz [1883], 2. Arquivo dos Açores [1890], 3. Matos [1982],, 4. Pereira [1986], 5. INMG/LNEC [1986], 6. Silva [2005]. Date (d/m/yr) Intensity (EMS-98) Most affected area Damage ref. ?/?/1611/36 ?/?/1717 13/6/1730 ?/3/1787 ?/1/1817 21/1/1837 ?/?/1868 11/12/1883 19/2/1868 - - VII - - VII - - - Santa Cruz (?) Northern part of island (Guadalupe and Santa Cruz villages) South part of island (Luz and Praia villages) - - Central part of the island (Guadalupe and Santa Cruz villages; Caminho das Fontes and Almas) - - - Santa Cruz Church Houses destroyed Total destruction - - Houses destryed; 3 deaths; Probable rupture along the South Serra das Fontes Fault Houses damaged; Landslides - - 4 4,3 6 1 1 6.2 1,5 1 1 5 NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Figure 4.Seismicity in Graciosa region from 2002 to 2012. The red dots mark epicentres of earthquakes and its magnitude. Plot includes earth- quakes of all magnitudes. G - Graciosa Island; SJ - São Jorge Island. Data from CIVISA (Centro de Informação e Vigilância Sismovulcânica dos Açores), 2012. Figure 5. Three-dimensional elevation map of Graciosa Island. Main morphological structures and location of volcanic complexes. Topog- raphy from IGeoE [2001]. reflects the NNE-SSW trend of the East Canary Ridge [e.g. Camacho et al. 2001, Acosta et al. 2003]. The oldest volcano-stratigraphic units occupy the central part of Graciosa. They comprise the Serra das Fontes Volcanic Complex (620 ± 120 ka, Féraud et al. [1980]), a shield volcano composed of basaltic to mugearitic lavas, and the Serra Branca Volcanic Com- plex (350 ± 40 ka, Féraud et al. [1980]) a trachytic (sensu stricto) central volcano [Gaspar and Queiroz 1995, Gaspar 1996] (Figure 5). On the southeast sector a quiescent basaltic to trachytic (sensu lato) polyge- netic volcano with a summit caldera constitutes the Central Volcano Unit (Caldeira Volcano) and in the northwest part of the island a low altitude plateau (the NW Basaltic Platform) is formed by alignments of basaltic cinder cones and the correlative lava flows forming the Vitória Unit [Gaspar 1996] (Figure 5). This fissural volcanic system extends to the southeast mantling the older volcanic complexes and developed simultaneously with the edification of the Caldeira Volcano. The two younger volcanic systems (Caldeira Volcano and NW Basaltic system) are Upper Pleis- tocene to Holocene in age. The most recent volcanic event was a pre-settlement basaltic hawaiian-strom- bolian eruption at about 2 ka B.P. (Walker, unpub- lished data, in Maund [1985]). Currently, the volcanic activity is expressed by secondary manifestations, namely by thermal springs, fumarole fields, and dif- fuse degassing [e.g. Ferreira et al. 1993, Gaspar 1996]. The older volcanic complexes (Serra das Fontes and Serra Branca) are displaced by several NW-SE trending faults morphologically expressed by large fault scarps parallel to the elongation of the island. Those faults define a graben structure that is crossed in its south-eastern portion by a NNE-SSW fault scarp, which separates the older units from the younger volcanic units to the south (Figures 6, 7). The faults can be tracked into the younger volcanic areas where they express as volcanic alignments. The lack of surface ruptures displacing the younger units indi- cates that these structures present low slip-rates dur- ing the Holocene. This work presents the first detailed structural analysis of Graciosa Island. The neotectonic data is used to discuss the relationship between geodynamic processes and the stress fields acting in the Azores re- gion. They also contribute to define neotectonic pa- rameters essential for seismic hazard assessment in order to estimate the seismogenic potential of active faults. HIPÓLITO ET AL. 6 Figure 6. Morphotectonic map of Graciosa Island. Main tectonic structures: SHF - Saúde-Hortelã Fault; SSFF - South Serra das Fontes Fault; NSBF - North Serra Branca Fault; SSBF - South Serra Branca Fault; ESFF - East Serra das Fontes Fault. Topography from IGeoE (2001). 7 2. Data Acquisition In Graciosa neotectonic analysis is limited due to active volcanism that conceals the faults in significant areas of the island, reducing the available outcrops. Major tectonic features correspond to fault scarps sev- eral tens to hundreds of meters tall located in the cen- tral region of the island (Figures 6, 7). These structures extend into the younger volcanic regions as blind faults. The absence of detailed bathymetric surveys on the sur- rounding offshore precludes the reconnaissance of the full length of the faults. The large size of the fault scarps and the absence of recent surface faulting with favourable conditions for trenching preclude paleoseis- mological studies. The neotectonic data was collected mainly in cinder cone quarries (Figure 8) and at sea cliffs (Figure 9). The nature of the younger volcanic deposits, low-cohesion coarse lapilli and lava flows, is not favourable for the generation of slickensides, limiting kinematic analysis. The mapped structures are normal faults or present a normal component. The kinematics were deduced from the displacement of morphologic and strati- graphic markers, and wherever possible, from striated fault planes. Although in most cases it was difficult to recognize a strike-slip component, a few structures pres- ent dextral or sinistral strike-slip components typical of a transtensile tectonic regime. Part of the amount of dip-slip offsets may be apparent and be the result of a strike-slip component displacing inclined surfaces. A preliminary structural mapping was based on vertical air photo-interpretation and DEM (digital ele- vation model) and topographic map analysis, allowing recognition of morphologic features with probable tec- tonic origin (fault scarps) or tectonically controlled (e.g. cone or dome alignments, elongated craters, displaced or linear stream channels, linear sea cliffs). The tectonic nature of the identified alignments was confirmed or in- firmed through detailed analysis of the outcrops. A total of 177 fault planes were measured and analyzed, in- cluding faults of variable importance cropping out in sea cliffs, small scale faults in cinder cones, and map- NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Figure 8. Small faults exposed on wall of a quarry exploiting lapilli from a cinder cone. White arrows specify the dip slip-vector. Figure 7. Photograph of fault scarps in the central region of Graciosa Island viewed from the SE. Continuous white lines indicate fault traces; dashed white lines mark hidden fault traces. Fault legend as in Figure 6. Figure 9. Example of a fault trace exposed in a sea cliff south of Serra Branca. Legend as in Figure 8. scale structures deduced from geological mapping. The structural map was later converted to a digital format, at the 1:25 000 scale (from IGeoE altimetry), using a GIS software. The structural data was included in the AZORIS data base [Gaspar et al. 2004]. 3. Structural data 3.1. Geometric and kinematic fault analysis The geometric analysis shows a main set of faults trending NW-SE to NNW-SSE, dipping to NE and SW or to ENE and WSW, respectively (Figure 10). This set of structures includes the larger structures expressed morphologically as fault scarps. The main NW-SE trending structures are the South Serra das Fontes Fault (SSFF), the Saúde - Hortelã Fault (SHF), the North Serra Branca Fault (NSBF), and the South Serra Branca Fault (SSBF) (Figures 6, 7). The SSFF presents a slightly convex, 4600m-long and 150-200m-high southwest-fac- ing scarp, trending ca. N306º. To the north this struc- ture is marked by an alignment of four basaltic cinder cones. The SHF trace is marked by a 5000m-long, less than 10m-high south-facing scarp, trending ca. N320º from Pico da Hortelã to Pico da Ladeira do Moiro cones, producing 50m of dextral strike-slip displace- ment at the intersection with the East Serra das Fontes Fault (ESFF) scarp. The NSBF presents a 4800m-long northeast-facing scarp, trending N302º, which is aligned with two cones of the NW Basaltic Platform and to the southeast with the long axis of the Caldeira Volcano. The SSBF scarp faces southwest, trends N305º, and may control the location of some basaltic cones to the northwest. These faults define a graben structure in the centre of Graciosa, the Serra Branca - Serra das Fontes Graben, south of which the NSB and SSB faults define a horst (Figures 6, 7). The total length of these faults is unknown because on land they are locally concealed beneath younger lava flows, pyroclastic and/or epi- clastic deposits, and there is no offshore data to assess its extension in the seafloor. A second set of faults trend NNE-SSW to NE-SW and dip to the SE and NW (Fig- ure 10). It is represented by small-scale structures crop- ping out in cinder cones, by cone alignments, and by a major NNE-SSW trending structure, the ESFF (Figures 6, 7). The N20º trending ESFF trace is marked by a 4600m-long and 185m tall southeast facing scarp. Inside the Caldeira Volcano there is a volcanic lineament trending NE-SW (Figure 6). E-W trending faults, al- though less frequent, also occur. In addition to the ob- served faults, some morphological features evidence similar trend, such as two ENE-WSW cinder cones alignments in the NW Basaltic Platform, or a rectilinear sea cliff south of Serra Branca, suggesting an E-W tec- tonic control (Figure 6). Gaspar [1995, 1996] proposes that E-W faults are inherited from the oceanic base- ment on which the islands stand (old transform faults). E-W trending faults were also described by Madeira [1998] in the islands of São Jorge, Faial, and Pico. More- over, the presence of E-W trending structures crossing the Azores Plateau is mentioned by Searle [1980]. The geometric and kinematic fault analyses show that the NW-SE trending, SW-dipping faults, present normal-dextral or dextral-normal oblique slip. Oblique normal-left lateral or left lateral-normal slip was ob- served in NNE-SSW to NE-SW trending, SE to ESE- dipping faults, and normal-dextral or dextral-normal oblique slip in NNE-SSW to NE-SW faults dipping to WNW and NW. Despite the limited number of faults presenting evidence of oblique displacement (strike-slip associated to normal component), the kinematic data obtained from field observations in Graciosa Island sug- gest a transtensive tectonic regime evidencing three-di- mensional strain [Reches 1983, Reches and Dieterich 1983], in agreement with the tectonic pattern described in previous works in other Azorean islands [e.g. Madeira 1986, Madeira and Ribeiro 1990, Madeira 1998, Madeira and Brum da Silveira 2003, Carmo 2004]. Hence, kinematic data from Graciosa indicate the exis- tence of two main fault systems (Figure 11). System A is composed of two sets of conjugated faults, one trending NW-SE, dipping to SW and NE, presenting normal-dextral or dextral-normal oblique slip, and an- other striking NNE-SSW, dipping to ESE and WNW, with oblique normal-left lateral or left lateral-normal slip (Figure 11a). System B includes NNE-SSW to NE- SW trending faults, dipping to WNW to NW or to ESE, presenting normal-dextral or dextral-normal oblique slip (Figure 11b). A family of faults conjugated with these structures was not found. The steep fault dips (the 80º to 90º range dominates) also suggest the presence of a strike-slip component (Figures 10b-11). 3.2. Geometric dyke analysis Several dykes, both of basaltic and trachytic na- ture, were observed at the sea cliffs cutting the Serra das Fontes and Serra Branca volcanic complexes and the Caldeira Volcano. In the first case the dykes’ direc- tions are similar to the main fault systems, with two main trends, NNW-SSE and NE-SW, attesting for the tectonic control of the magmatic phases (Figure 12). The dyke system exposed on the sea cliffs cut on the Caldeira Volcano, represented by NE-SW trending dykes on the northeast flank and NW-SE dykes on the southeast, although scarce, suggests a radial pattern around the volcanic edifice (Figure 13). However, NNW-SSE trending dikes occurring in the southeast HIPÓLITO ET AL. 8 9 coastal cliffs are not in agreement with a radial pattern and may represent en echelon dykes consistent with NW-SE dextral shearing compatible with the kinemat- ics of fault System A, implying a magmatic phase con- NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Figure 10. Geometry of all analyzed faults: a) Circular histogram of unweighted frequencies of fault plane directions; b) fault poles density contour plot (lower hemisphere; Schmidt net); TectonicsFP software® [Ortner et al. 2002]. Figure 11. Stereographic plot of the two main fault systems (lower hemisphere; Schmidt net – β diagram): a) system A - NW-SE to NNW- SSE faults, with normal-dextral oblique slip, conjugate of NNE-SSW to NE-SW normal-left lateral structures; b) system B - NNE-SSW to NE-SW faults with normal-dextral oblique slip. TectonicsFP software® [Ortner et al. 2002]. trolled by ENE-WSW crustal extension. Note that, be- cause of three-dimensional strain, transtensional faults may act as volcanic conduits, so dykes are not necessar- ily normal to the direction of extension (i.e. they may open obliquely to the fracture direction). This is attested by the fact that in many Azorean islands, alignments of monogenetic centres are mostly parallel to the fault sys- tem synthetic with the plate boundary, but alignments along the conjugate direction also occur. 4. Discussion 4.1. Determining paleostress The two main fault systems (A and B) are incom- patible with a single stress field. None of the fault sys- tems is exclusive of a given volcanic system; they affect all stratigraphic units and may occur together. This indi- cates that the different stress fields are neither related to a given period of time nor related to deformation asso- ciated to a particular volcanic centre. The paleostress analysis indicates the presence of two different stress fields. The stress field #1, associated to the fault system A, presents horizontal NNW-SSE σ1 (maximum com- pressive stress axis; Anderson 1951; Angelier 1994), hor- izontal ENE-WSW σ3 (maximum tensile stress axis) and a vertical intermediate compressive stress axis (σ2). Per- mutations between σ1 and σ2 may occur producing al- ternating transtensile and tensile tectonic regimes [Madeira 1998] (Figure 14). The stress field #2, related to the fault system B, is characterized by horizontal WSW-ENE σ1, horizontal NNW-SSE σ3 and vertical σ2. Permutation may also occur between σ2 and σ1 follow- ing events of stress drop during transtensile phases (Fig- ure 15). Because the two different stress fields cannot operate simultaneously, and since they are not local stress fields, they must alternate in time, generating new faults or reactivating pre-existing structures. The spatial-tem- poral relation between the two stress fields is unclear. The stress field #1 agrees with the findings of Madeira and Ribeiro [1990] for the Azores Plateau and with the geodynamic model proposed by Madeira [1998] and Lourenço et al. [1998] for the Azorean segment of the Eu-Nu plate boundary. It considers the occurrence of a rotation to N-S and E-W of σ1 and σ3, respectively, closer the edges of the shear zone, as observed in Santa Maria Island [Madeira 1986], (Figure 2). The maximum tensile stress axis direction is also in agreement with the opening direction according the current kinematic plate models for the relative Eu-Nu motion (e.g. N66º- NUVEL-1A, DeMets et al. 1994; N80º-REVEL, Sella et al. 200; N72º-DEOS2K, Fernandes et al. 2006; N71º- MORVEL, DeMets et al. 2010; Table 3). The stress field #2 is probably external to the shear zone and may temporarily affect the areas at the border of the shear zone, since it is also found in the eastern part of São Miguel Island [Carmo 2004]. The width of the shear zone accommodating the deformation between Eu and Nu may change proportionally to the difference between the NA-Eu and NA-Nu spreading rates, north and south of the triple junction, respectively. Narrowing of the area under the influence of the regional stress field #1 (i.e. the shear zone) would put the margins of that sheared region temporarily under the influence of the outer transtensile stress field #2. Two hypothesis are considered for the origin of the stress field #2: (1) the existence of an intermediate re- gion under transtension, developing parallel to the MAR and establishing a transition zone between the pure ten- sile stress field at the MAR and the distal compressive stress field, established as the sea-floor moves away from the ridge (Figures 16-Ia-16-Ib); (2) the existence of an in- termediate region, with a transtensile regime, defining a narrow band between the interplate region with dextral transtension (stress field #1) and the external compres- HIPÓLITO ET AL. 10 Figure 12. Geometry of all measured dykes: a) Circular histogram of unweighted frequencies of dykes’ plane directions. TectonicsFP software® [Ortner et al. 2002]. Figure 13.Spatial variations of dyke directions around Caldeira Vol- cano (circular histograms of unweighted frequencies of dyke di- rections and dip angles). Topography from IGeoE (2001). 11 NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Figure 14. Stereographic projection of the stress field #1 deduced from structures of fault system A that have kinematic data of both slip components (lower hemisphere; Schmidt net – βdiagram). a) transtensile stress field: NW-SE to NNW-SSE faults with dominant dextral slip component, conjugate of NNE-SSW to NE-SW structures with dominant left-lateral slip component. Horizontal NNW-SSE σ1, horizontal ENE-WSW σ3 and vertical σ2; b) tensile stress field after permutation between σ1 and σ2: NW-SE to NNW-SSE faults, dipping to both sides, with dominant normal slip component conjugate of NNE-SSW to NE-SW faults, dipping to both sides, with dominant normal slip com- ponent. Vertical σ1, horizontal ENE-WSW σ3 and horizontal NNW-SSE σ2. TectonicsFP software® [Ortner et al. 2002]. Figure 15. Stereographic projection of the stress field #2 deduced from structures of fault system B that have kinematic data of both slip components (lower hemisphere of Schmidt net – β diagram). a) transtensile stress field: NNE-SSW to NE-SW faults with dominant dextral slip component. Horizontal ENE-WSW σ1, horizontal NNW-SSE σ3 and vertical σ2; b) tensile stress field after permutation between σ1 and σ2: NNE-SSW to NE-SW faults, dipping to both sides, with dominant normal slip component. Vertical σ1, horizontal NNW-SSE σ3 and hor- izontal ENE-WSW σ2. TectonicsFP software® [Ortner et al. 2002]. HIPÓLITO ET AL. 12 Figure 15. Estimated trajectories of the main stress axis in the western segment of the Eu-Nu plate boundary. Above – hypothesis (1): the tensile stress field (a) close to the MAR passes laterally to transtensive (b) with the establishment of a transition stress field. Eastward, fur- ther away from the ridge, permutation of σ3 with σ2 occurs as a region in compression is entered (c). Inside the sheared region (d) stress permutation and deviation of stress trajectories occur. Ia) Stages of widened shear zone; Ib) Stages of narrowed shear zone: the islands near the edges of the sheared interplate region are subjected to the stress field b. Below – hypothesis (2): the tensile stress field (a) close to the MAR passes laterally to compressive (c) as the sea-floor moves away from the ridge. Inside the sheared region (d) stress permutation and deviation of the stress trajectories occur. During stages of narrowing of the sheared region, permutation between σ2 and σ3 may originate a local in- termediate stress field (b). IIa) Stages of widened shear zone; IIb) Stages of narrowed shear zone: the islands near edges of the sheared in- terplate region are subjected to the stress field b. Red dashed lines separate the different stress field domains; white shaded area represents the interplate sheared region. Trajectories of maximum tensile stress axis near the MAR are based on the spreading direction of NA-Eu ac- cording to the NUVEL-1A (DeMets et al. 1994). Trajectories of the main stresses within the sheared zone are based on neotectonic data from emerged areas [e.g. Madeira and Brum da Silveira 2003, Carmo 2004, Hipólito 2009, and the present work]. Acronyms as in Figure1. Ba- thymetry adapted from Lourenço et al. [1997]. Datum: WGS 1984. Table 3. Relative velocities and slip directions for Eu, Nu, and NA plates in Azores. Global kinematic models: NUVEL-1A [DeMets et al. 1994], REVEL [Sella et al. 2002], DEOS2K [Fernandes et al. 2003], MORVEL [DeMets et al. 2010]. Plate pairs Velocity (mm/a) Azimuth (degrees) Kinematic model Eu-NA(stable) ~23 ~25 ~24 ~23 ~97° ~96° ~94° ~96° NUVEL-1A REVEL DEOS2K MORVEL Nu-NA(stable) ~20 ~19 ~19 ~20 ~103° ~151° ~99° ~149° NUVEL-1A REVEL DEOS2K MORVEL Eu-Nu(stable) ~4 ~6 ~4 ~4.5 ~66° ~72° ~80° ~71° NUVEL-1A REVEL DEOS2K MORVEL 13 sive stress field established as the sea-floor moves away from the ridge. The intermediate region would be es- tablished when the interplate shear stress is weaker (Fig- ures 16, 16,2a, 2b). The E-W trending faults may reflect former deep transform faults cutting the ocean floor generated at the MAR (Gaspar 1995; 1996) that, because of a favourable di- rection, were later reactivated by the current dextral transtensile stress regime, probably as shear fractures (P frac- tures, considering the Riedel shear model; e.g. Petit 1987). 4.2. Insights on the seismogenic potential of Graciosa faults The available data do not provide enough infor- mation to carry out an assessment of seismic hazard associated to the identified structures. Nevertheless, the determination of neotectonic parameters such as the slip-rate is essential to define potential seismogenic sources in a given area, which will be helpful for a fu- ture detailed seismic hazard analysis. Only three tectonic structures allowed slip-rate es- timation: the SHF, the SSFF and the ESFF (Figure 6, 7). However, none of these structures allowed the deter- mination of the total slip (normal slip and strike-slip components) along the fault over the considered time period. In two of them only the normal slip offset was determined, while in the other only the strike-slip could be measured. Given that dated stratigraphical markers only exist for the base of the volcano-strati- graphic units, we consider that the observed deforma- tion accumulated during the 620 ± 120 ka to 31.4 ± 0.3 ka interval. This time interval corresponds to the old- est ages obtained for the Serra das Fontes Volcanic Complex and for the Caldeira Volcano / NW Basaltic Platform Complex, respectively [Féraud et al. 1980, Maund 1985], since the younger volcanic units are not tectonically displaced by those structures. The estimated slip-rates are of the order of tenths of a millimetre to few millimetres per year (Table 4). These values are of the same magnitude as those pro- vided by plate motion kinematic models for the whole archipelago, considering the Eu-Nu relative motion (Table 3). The higher slip-rates could be over-estimated because the age of the younger marker may not rep- resent the end of the deformation period. On the other hand, the beginning of the tectonic deformation af- fecting the older units is unknown and it is probably more recent than the oldest marker used. In addition, the displacements obtained from morphological mark- ers may be underestimated due to infilling of tectoni- cally depressed areas by younger deposits. The deformation accumulated since 350 ± 40 ka BP (or less) in the older units is large (fault scarps as high as 200 m; Table 4). Considering a constant deformation rate it would be expected that the younger volcanic units (> 31.4 ± 0.3 ka, Maund [1985]) would be faulted, NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Table 4. Slip-rates for Graciosa faults. Structure Dip-slip Separation (m) Strike-slip Age of marker (ka BP) Dip-slip Slip-rates (mm/a) Strike-slip SHF SSF ESFF - 200 185 50 - - ca. 620-31.4 ca. 620-31.4 ca. 620-31.4 - 0.3-6.4 0.3-5.9 0.08-1.6 - - Table 5. Maximum expected magnitudes from surface rupture length (SRL) and rupture area (RA) for Graciosa faults. M/SRL - M=5.08+1.16log(SRL); M/RA - M=4.07+0.98log(RA) from Wells and Coppersmith [1994]. Structure SRL (m) Mw (from M/SRL) RA (km 2) Mw (from M/RA) SHF SSFF NSBF SSBF ESFF 5000 4.600 4.800 3.200 4.600 5.9 5.8 5.9 5.7 5.8 75 69 72 48 69 5.9 5.9 5.9 5.7 5.9 with displacements of the order of 1/10 of those ob- served in the older units. Thus, the fault scarps should have surface expression in their projection northwest- wards into the NW Basaltic Platform and southeast- wards into the Caldeira Volcano. This fact suggests slowing down of fault activity in Late Pleistocene and Holocene. The present low seismicity, the large displacement accumulated by the older volcanic units, and the fact that no deformation is observed in the youngest areas of the island suggest that slip-rates must have been higher in the near past. This period with higher slip- rates was followed by a phase dominated by important magmatism, during which Caldeira Volcano was built and the NW Basaltic Platform developed. This de- crease of slip-rates must have increased the recurrence interval between fault rupturing-events. These obser- vations in Graciosa are consistent with the occurrence of variations in the deformation rates in the archipel- ago, with periods with slip-rates higher or lower than the average motion predicted by global plate models, as proposed by Madeira [1998]. In that work the author estimated slip-rate values for a set of faults of the Cen- tral Group of Azores archipelago based on a detailed neotectonic analysis through tectonic mapping, chronostratigraphic correlations, and paleoseismolog- ical studies performed in São Jorge, Pico, and Faial Is- lands [Madeira 1998, Madeira and Brum da Silveira 2003]. However, while in São Jorge, Pico, and Faial, neotectonic data [Madeira 1998, Madeira and Brum da Silveira 2003] indicates recent slip-rates higher than those predicted by global plate kinematic models [DeMets et al. 1994, Sella et al. 2002, Fernandes et al. 2003, DeMets et al. 2010], in Graciosa what is observed is the opposite. The deformation integrated in a longer time frame is larger than what is observed for the last tens of thousands years. Maximum expected magnitudes (MW) were esti- mated for the SHF, SSFF, ESFF, NSBF, and SSBF, using the surface rupture length (SRL)/Magnitude (MW) and the rupture area (RA)/Magnitude (Mw) correla- tions of Wells & Coppersmith [1994], assuming the rupture of the full known length of the faults, a brittle crust thickness of 14 km, based on the depth of the af- tershocks from the January, 1st 1980 and July, 9th 1998 earthquakes (Hirn et al. 1980; Matias et al. 2007), and average fault dips of 70º. The estimated MW values range from 5.7 to 5.9 (Table 5). However, these are minimum values for the expected maximum magni- tudes because the total length of the faults must be longer than the values used since the faults are con- cealed by recent volcanism and the small size of the is- land limits the observation of the full length of the faults that certainly extend to offshore areas. Nonethe- less, the magnitudes estimated from neotectonic stud- ies are within the range of magnitudes of instrumental seismicity in the archipelago (Table 1). 5. Final remarks Several limitations prevent a more detailed neo- tectonic survey of Graciosa. Recent volcanic and epi- clastic deposits mantling the topography conceal the surface expression of major tectonic structures. In most faults, exposed in lapilli quarries, basaltic lapilli deposits do not allow the generation of linear kinematic mark- ers (slickensides). No faults were found presenting scarp heights and lithological conditions (e.g. absence of lava flows) favourable to be trenched for paleoseis- mological studies, thus contributing to seismic hazard assessment. The paleostress analysis suggests that the region is affected by two different stress fields that can alternate in time, which are responsible for the generation of new faults or reactivation of pre-existing structures. The data suggest a diffuse Eu-Nu plate boundary ex- pressed by a complex deformation zone, evidencing variable width in time, probably caused by changes in spreading rate between the Eu-NA and Nu-NA sections of the MAR, north and south of the triple junction. Thus the islands closer to the edge of the shear zone may become under the influence of different (inner and outer) stress fields. The absence of seismic events producing surface rupture since settlement and the current low seismic activity in Graciosa Island, contrast with the youthful aspect of its tectonic morphology. The loss of geo- morphic expression of the fault scarps in the areas cov- ered by the younger units suggests the occurrence of a period of important tectonic activity before ca. 31 ka, with higher slip-rates than those observed in present times, responsible for the tectonic deformation of the central part of the island. Despite the present low seismicity, historical records show that a few significant earthquakes affected Graciosa, since settlement in mid-15th century, causing fatalities and damage. Maximum magnitude values are probably underestimated due to unknown full length of the faults and as result of the island small size. Currently, Graciosa seems to present low tectonic activity. Nevertheless, as the island is located in an im- portant seismogenic zone, seismic hazard cannot be disregarded. The difficulty in developing paleoseismo- logical studies constitutes a significant problem given that the neotectonic information available from tradi- tional structural field analysis comprises a too long time window of the geological record than what is required HIPÓLITO ET AL. 14 15 for seismic hazard assessment. On the other hand, the historical and instrumental seismic data available for the Azores region only portray a very short period of the tectonic history of the analysed faults. Further neo- tectonic work, such as acquisition of bathymetric and seismic reflection surveys offshore Graciosa and re-eval- uating the possibility of a paleoseismology analysis, will contribute to better understanding the tectonic processes and to characterize seismic hazard of emerged areas at the western segment of the Eu-Nu plate boundary. 6. Data and sharing resources Maps were made using ArcGIS 9.3 software, ESRI®. Topographic map of Graciosa Island is from IGeoE - Instituto Geográfico do Exército (2001). St. Cruz da Graciosa (Graciosa-Açores) - Folha 21. In: Carta Militar de Portugal - Série M889 2nd ed. Instituto Geográfico do Exército, Lisboa. World topography and bathymetry from GEBCO_08 database (2010), IHO-UNESCO, General Bathymetry Chart of the Oceans, digital edition at http://www.gebco.net/data and products/gridded ba- thymetry data/. Seismic map of Graciosa region was produced with data from CIVISA Database, Centro de Informação e Vigilância Sismovulcânica dos Açores, Centro de Vul- canologia e Avaliação de Riscos Geológicos da Universi- dade dos Açores. Structural analysis was made using TectonicsFP® software [Ortner et al. 2002]. Tectonic plate motion parameters calculated using the Science product support - Plate Motion Calculator - from UNAVCO, http://www.unavco.org/community sci- ence/science-support/crustal motion/dxdt/model.html. Acknowledgments. Ana Hipólito is supported by a Ph.D. Grant from Fundação para a Ciência e Tecnologia (SFRH/BD/73664/2010). Fieldwork was supported by the Azores Regional Government / Serviço Regional de Protecção Civil e Bombeiros dos Açores through the project “Emergency Planning Studies”, Contract Ref.FGF/2005. We appreciated and acknowledge the reviewers Francesco Mazzarini, João Fonseca, and an anony- mous reviewer, the associate editor Salvatore Barba, and the vol- ume editor Edoardo Del Pezzo for their careful reviews, suggestions, and helpful comments which contributed to signifi- cantly improve this manuscript. References Acosta, J., E. Uchupi, A. Munoz, P. Herranz, C. Palomo, M. Ballesteros and Z.E.E. Working Group (2003). Geologic evolution of the Canarian Islands of Lan- zarote, Fuerteventura, Gran Canaria and La Gomera and comparison of landslides at these is- lands with those at Tenerife, La Palma and El Hi- erro, Mar. Geophys. Res., 24, 1-40. Altamimi, Z., P. Sillard and C. Boucher (2002). ITRF2000: A new release of the International Terrestrial Ref- erence Frame for earth science applications, J. of Geophys. Res., 107 (B10), 2214, 2-1 - 2-19. doi: 10.1029/2001JB000561. Anderson, E. M. (1951). The dynamics of faulting, Oliver and Boyd (eds), Edinburgh, 2, 206 p. Angelier, J. (1994). Paleostress determinations, In: Han- cock, P.L. (Ed.), Continental Deformations. Perga- mon Press, Tarrytown, NY, 53-100. Borges, J.F., M. Bezzeghoud, E. Buforn, C. Pro and A. Fitas (2007). The 1980, 1997 and 1998 Azores earth- quakes and its seismotectonic implications, Tectono- physics, 435, 37-54, doi:10.1016/j.tecto.2007.01.008. Borges, J.F., B. Caldeira, M. Bezzeghoud and E. Buforn (2008). Sismicidade e sismotectónica dos Açores: im- plicações geodinâmicas, In Sousa Oliveira, C., A. Costa, J.C. Nunes (eds.), Sismo 1998 - Açores. Uma década depois. 99-110, ISBN: 978-989-20-1223-0. Borges, J.F. and E. Burfon (2008). Recent seismic activ- ity in the Azores region, In Sousa Oliveira, C., A. Costa, J.C. Nunes (eds.), Proceedings of the Inter- national Seminar on Seismic Risk and Rehabilitation on the 10th Anniversary of the July 9 1998 Azores Earthquake, ISBN 978-989-20-1235-3. Buforn, E., A. Udías and M. Colombas (1988). Seismic- ity, source mechanisms and tectonics of the Azores- Gibraltar plate boundary, Tectonophysics, 152, 89-118. Calais, E., C. DeMets and J.-M. Nocquet (2003). Evidence for a post-3.16-Ma change in Nubia-Eurasia-North America plate motions?, Earth Planet. Sci. Lett. 216, 81-92, doi:10.1016/S0012-821X(03)00482-5. Cannat, M., A. Briais, C. Deplus, J. Escartín, J. L. Geor- gen, S. Mercouriev, C. Meyzen, M. Muller, G. Pouliquen, A. Rabain and P. Silva (1999). Mid-At- lantic Ridge - Azores hotspot interactions: along-axis migration of a hotspot-derived event of enhanced magmatism 10 to 4Ma ago, Earth Planet. Sci. Lett., 173, 257-269. Canto, E. and F. Afonso Chaves (1890). Arquivo dos Açores Online, Digital issue of Arquivo dos Açores (1980 1984), Azores University, XI, 572 p., http://www.uac.pt/~arquivoacores/. Carmo, R. (2004). Geologia estrutural da região Povoação - Nordeste (ilha de São Miguel, Açores), M.Sc. Thesis, Azores University, 133 p. DeMets, C., R. Gordon, D. Argus and S. Stein, (1994). Effect of recent revisions to the geomagnetic rever- sal time scale on estimates of current plate motions, Geophys. Res. Lett., 21(20), 2191-2194. DeMets, C., R.G. Gordon and D.F. Argus (2010). Geo- NEOTECTONICS OF GRACIOSA ISLAND (AZORES) logically current plate motions, Geophy. J. Int., 181, 1-80. Dias, N.A., L. Matias, N. Lourenço, J. Madeira, F. Car- rilho and J.L. Gaspar (2007). Crustal seismic velocity structure near Faial and Pico Islands (Azores), from local earthquake tomography, Tectonophysics, 445, 301-317, doi: 10.1016/j.tecto.2007.09.001. Escartín, J.; M. Cannat, G. Pouliquen and A. Rabain (2001). Crustal thickness of V-shaped ridges south of the Azores: Interaction of the Mid-Atlantic Ridge (36º-39ºN) and the Azores hot spot, J. Geophys. Res., 106 (B10) 21,719-21,735. Féraud, G., I. Kaneoka and C.J. Allégre (1980). K-Ar Ages and Stress Pattern in the Azores - Geodynamic Implications, Earth Planet. Sci Lett., 46(2), 275-286. Fernandes, R. M. S., B. A. C. Ambrosius, R. Noomen, L. Bastos, M.J.R. Wortel, W. Spakman and R. Govers (2003). The relative motion between Africa and Eurasia as derived from ITRF2000 and GPS data, Geophys. Res. Lett., 30 (16), 1828, 1-1 - 1-5, doi: 10.1029/2003GL017089. Fernandes, R.M.S., L. Bastos, J. M. Miranda, N. Lourenço, B. A. C. Ambrosius, R. Noomen and W. Simons (2006). Defining the Plate Boundaries in the Azores Region, J. Volcanol. Geoth. Res., 156, 1-9, doi: 10.1016/j.jvolgeores.2006.03.019. Ferreira, T., J. L. Gaspar and G. Queiroz (1993). Con- siderações sobre as emanações gasosas da Furna do Enxofre (Ilha Graciosa, Açores), Açoreana VII (4), 603-612. Gamacho, G. A., F. G. Montesinos, R. Vieira and J. Arnoso (2001). Modelling of crustal anomalies of Lanzarote (Canary Islands) in light of gravity data, Geophys. J. Int., 147, 403-414. Gaspar, J.L. and G. Queiroz (1995). Carta Vulcanológica dos Açores, ilha Graciosa. Universidade dos Açores e Câmara Municipal de Santa Cruz da Graciosa (Eds). Gaspar, J.L. (1996). Ilha Graciosa (Açores): História Vul- canológica e Avaliação do Hazard, Ph.D. Thesis, Azores University, 361p. Gaspar, J. L., C. Goulart, G. Queiroz, D. Silveira and A. Gomes (2004). Dynamic structure and data sets of a GIS database for geological risk analysis in the Azores volcanic islands, Nat. Hazard Earth Sys. Sci- ences, 4, 233-242. Gente, P., J. Dyment, M. Maia and J. Goslin (2003). In- teraction between the Mid-Atlantic Ridge and the Azores hot spot during the last 85 Myr: Emplace- ment and rifting of the hot spot-derived plateaus, Geochem. Geophys. Geosy., 4 (10), 8514, 23 p., doi: 10.1029/2003GC000527. Grimisson, N. L. and W.-P. Chen (1986). The Azores- Gibraltar plate boundary: focal mechanisms, depths of earthquakes and their tectonic implications, J. of Geophys. Res., 92, (B2), 2029-2047. Hipólito, A. (2009). Geologia Estrutural da ilha Gra- ciosa. Enquadramento no âmbito da junção tripla dos Açores, M.Sc. Thesis, Azores University, 155 p. Hirn, A., H. Haessler, P. Hoang-Trong, P. Wittlinger and L. Mendes-Victor (1980). Aftershock sequence of the January 1st, 1980 earthquake and present-day tecton- ics in the Azores, Geophys. Res. Lett., 7 (7), 501-504. IGeoE - Instituto Geográfico do Exército (2001). St. Cruz da Graciosa (Graciosa-Açores) - Sheet 21, Carta Militar de Portugal - Série M889 2nd ed., In- stituto Geográfico do Exército, Lisboa. INMG and LNEC (1986). Sismicidade histórica e in- strumental do arquipélago dos Açores (Sismos sen- tidos no período 1444-1980), Estudos realizados para o Novo Hospital de Ponta Delgada, INMG/LNEC open-file report, Lisbon, 1. Kreemer, C., W. E. Holt and A. J. Haines (2003). An in- tegrated global model of present-day plate motions and plate boundary deformation, Geophys. J. Int., 154, 8-34. Lourenço, N., J. Luís and J. M. Miranda (1997). Azores triple junction bathymetry, Map edited by Science Faculty, Lisbon University and Algarve University, 1 Sheet. Lourenço, N., J. M. Miranda, J. Luís, A. Ribeiro, L. Mendes-Victor, J. Madeira and H. Needham (1998). Morphotectonic analysis of the Azores Volcanic Plateau from a new bathymetric compilation of the area, Mar. Geophys. Res., 20, 141-156. Luís, J. F., J. M. Miranda, A. Galdeano and P. Patriat (1998). Constraints on the structure of the Azores spreading center from gravity data. Mar. Geophys. Res., 20, 157-170. Luís, J.F. and J. M. Miranda (2008). Reevaluation of mag- netic chrons in the North Atlantic between 35ºN and 47ºN: Implications for the formation of the Azores Triple Junction and associated plateau, J. Geophys. Res., 113, B1o1o5, doi:10.1029/2007Jb005573. Machado, F. (1959). Submarine pits of the Azores Plateau, Bulletin Volcanol. (sér. II), 21, 109-116. Madeira, J. (1986). Geologia estrutural e enquadra- mento geotectónico da ilha de Santa Maria (Açores), M.Sc. Thesis, Lisbon University, 107 p. Madeira, J. (1998). Estudos de neotectónica nas ilhas do Faial, Pico e São Jorge: uma contribuição para o con- hecimento geodinâmico da junção tripla dos Açores, Ph.D. Thesis, Lisbon University, 428 p. Madeira, J. and A. Ribeiro (1990). Geodynamic models for the Azores triple junction: a contribution from tectonics, Tectonophysics, 184, 405-415. Madeira, J. and A. Brum da Silveira (2003). Active Tec- HIPÓLITO ET AL. 16 17 tonics and first paleoseismological results in Faial, Pico and São Jorge Islands (Azores, Portugal), An- nals of Geophysics, 46 (5), 733-761. Madureira, P., M. Moreira, J. Mata and J. C. Allegre (2005). Primitive helium and neon isotopes in Ter- ceira island (Azores archipelago), Earth Planet. Sci. Lett, 233, 429-440, doi: 10.1016/j.epsl.2005.02.030. Matias, L., N. A. Dias, I. Morais, D. Vales, F. Carrilho, J. Madeira, J. L. Gaspar, L. Senos and A. Brum da sil- veira (2007). The 9th of July 1998 Faial Island (Azores, North Atlantic) seismic sequence, J. Seis- mol., 11, 275-298, doi: 10.1007/s10950-007-9052-4. Matos, A. T. (1982). Uma memória setecentista inédita da ilha Graciosa, Boletim do Instituto Histórico da Ilha Terceira, XL, 375-396. Maund, J. (1985). The Volcanic Geology, Petrology and Geochemistry of Caldeira Volcano, Graciosa, Azores, and its bearing on contemporaneous felsic- mafic oceanic island volcanism, Ph.D. Thesis, Uni- versity of Reading, 333 pp. Mckenzie, D. (1972). Active tectonics of the Mediter- ranean region, Geophys. J. R. Astron. Soc., 30, 109- 185. Mendes, V. B., J. Madeira, A. Brum da Silveira, A.Trota, P. Elósegui and J. Pagarete (2013). Contemporary deformation of S. Jorge Island, Azores, from episodic GPS measurements (2001-2011). Adv. Space Res., 51, 1581-1592, doi: 10.1016/j.asr.2012.10.019. Mezcua, J., J. Rueda and J. M. Martínez Solares (1991). Seismicity of the Ibero-Maghrebian region, In: J. Mezcua and A. Udías (Eds.), Seismicity, seismotec- tonics and seismic risk of the Ibero-Maghrebian re- gion, Instituto Geografico Nacional, Madrid, 8: 17-28. Miranda, J., L. Mendes-Victor, J. Simões, J. Luís, L. Ma- tias, H. Shimamura, H. Shiobara, H. Nemoto, H. Mochizukli, A. Hirn and J. Lépine (1998). Tectonic setting of the Azores Plateau deduced from a OBS survey, Mar. Geophys. Res., 20, 171-182. Miranda J.M., A. Navarro, J. Catalão and R.M.S. Fernan- des (2012). Surface displacement field at Terceira is- land deduced from repeated GPS measurements. J. Volcanol. Geoth. Res., 217-218, 1-7, doi:10.1016/j.jvol- geores.2011.10.009. Moniz, A. (1883). Ilha Graciosa (Açores) - Descripção His- torica e Topographica, Angra do Heroísmo, 299 p. Moreira, V.S. (1991). Historical seismicity and seismo- tectonics of the area situated between the Iberian Peninsula, Morocco, Selvagens and Azores islands, In J. Mezcua and A. Udías (Eds.), Seismicity, seis- motectonics and seismic risk of the Ibero-Maghre- bian region, Instituto Geografico Nacional, Madrid, 8: 213-225. Needham, H. and J. Francheteau (1974). Some charac- teristics of the rift valley in the Atlantic Ocean near 36°48' north, Earth Planet. Sci. Lett., 22, 29-43. Ortner, H., F. Reiter and P. Acs (2002). Easy handling of tectonic data: the programs TectonicsVB for Mac and TectonicsFP for Windows, Comput. Geosci., 28, 1193-1200. Pereira, V. C. D. (1986). Igrejas e ermidas da Graciosa, Edição comemorativa do 500º aniversário da ele- vação a vila de Santa Cruz da Graciosa. Direção Re- gional dos Assuntos Culturais/Secretaria Regional Da Educação e Cultura, Angra do Heroísmo, 458 p. Petit, J.-P. (1987). Criteria for the sense of movement on fault surfaces in brittle rocks, J. Struct. Geol., 9, 597- 608. Reches, Z. (1983). Faulting of rocks in three-dimen- sional strain fields. II: Theoretical analysis, Tectono- physics, 95, 133-156. Reches, Z. and J. H. Dieterich (1983). Faulting of rocks in three-dimensional strain fields. I: Failure of rocks in poliaxial, servo control experiments, Tectono- physics, 95, 111-132. Schilling, J-G., 1975. Azores mantle blob: rare-earth ev- idence, Earth Planet. Sci. Lett., 25, 103-115. Sella, G., T. H. Dixon and A. Mao (2002). REVEL: a model for recent plate velocities from space geodesy, J. Geophys. Res., 107 (B4), 2081, 11-1 - 11-12, doi: 10.1029/2000JB000033. Silva, M. (2005). Caracterização da sismicidade histórica dos Açores com base na reinterpretação de dados de macrossísmica: contribuição para a avaliação do risco sísmico nas ilhas do Grupo Central, Ph.D. The- sis, Azores University, 163 p. Silva, R., J. Havskov, C. Bean and N. Wallenstein (2012). Seismic swarms, fault plane solutions, and stress ten- sors for São Miguel Island central region (Azores), J. Seismol., 16(3), 389-407. Silveira, G., E. Stutzmann, A. Davaille, J.-P. Montagner, L. Mendes-Victor and A. Sebai (2006). Azores hotspot signature in the upper mantle, J. Volcanol. Geoth. Res., 156, 23-34, doi:10.1016/j.jvolgeores.2006.03. 022. Trota A. (2008). Crustal deformation studies in S. Miguel and Terceira Islands (Azores). Volcanic un- rest evaluation in Fogo/Congro area (S. Miguel). Ph.D. Thesis, University of Azores, 281 pp. Udías, A., A. Lopez Arroyo and J. Mezcua (1976). Seis- motectonics of the Azores-Alboran region, Tectonophysics, 31, 259-289. Udías, A., A. F. Espinoza, J. Mezcua, E. Buforn, R. Vegas, S. P. Nishenko, J. M. Martínez-Solares and A. Lopez Arroyo (1986). Seismicity and tectonics of the North African-Eurasian plate boundary (Azores-Iberia- Tunisia), U. S. Geol. Survey Open-File Report, 86-626. NEOTECTONICS OF GRACIOSA ISLAND (AZORES) Udías, A. (1980). Seismic stresses in the region Azores- Spain-Western Mediterranean, Rock Mech., suppl. 9, 75-84. Vogt, P.R. and W. Y. Jung (2004). The Terceira Rift as hyper-slow, hotspot dominated oblique spreading axis: A comparison with other slow-spreading plate boundaries, Earth Planet. Sci. Lett, 218, 77-90. doi:10.10116/S0012-821X(03)00627-7. Wells, D. L. and K. J. Coppersmith (1994). New empirical relationships among Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface Displace- ment, Bull. Seism. Soc. Am., 84 (4), 974-1002. Yang, T., Y. Shen, S. Lee, S. C. Solomon and S.-H. Hung (2006). Upper mantle structure beneath the Azores hotspot from finite-frequency seismic tomography, Earth Planet. Sci. Lett., 250, 11-26, doi: 10.1016/j.epsl.2006.07.031. *Corresponding author: Ana Hipólito, Centro de Vulcanologia e Avaliação de Riscos Geológicos da Uni- versidade dos Açores, CVARG, Açores, Portugal; email: ana.rc.hipolito@azores.gov.pt © 2013 by the Istituto Nazionale di Geofisica e Vulcanologia. All rights reserved. HIPÓLITO ET AL. 18