Brescia Earthquake (Northern Italy): Seismotectonic Implications

Tectonophysics 476 (2009) 320–335

Contents lists available at ScienceDirect

Tectonophysics

journal homepage: www.elsevier.com/locate/tecto

Active fault-related folding in the epicentral area of the December 25, 1222 (Io=IX MCS) Brescia earthquake (Northern Italy): Seismotectonic implications

Franz A. Livio a,⁎, Andrea Berlusconi a, Alessandro M. Michetti a, Giancanio Sileo a, Andrea Zerboni b, Luca Trombino b, Mauro Cremaschi b, Karl Mueller c, Eutizio Vittori d, Cipriano Carcano e, Sergio Rogledi e a Dipartimento di Scienze Chimiche ed Ambientali, Università dell'Insubria, Como, Italy b Università Statale di Milano, Dipartimento di Scienze della Terra “A. Desio”, Italy c Colorado University at Boulder, CO, USA d ISPRA, Rome, Italy e ENI E&P, Italy article info abstract

Article history: A dense grid of petroleum industry seismic reflection profiles, coupled with field mapping, exploratory Received 9 March 2008 trenching and geomorphic and structural analysis are used to characterize the Quaternary growth history Received in revised form 6 March 2009 of the Capriano del Colle Fault System, one of several inferred active buried thrusts that extend across the Accepted 19 March 2009 Po Plain in northern Italy. Shortening is characterized here by a deeply buried south-vergent forethrust and Available online 5 April 2009 an associated north-vergent backthrust whose upward propagation is expressed by fault-propagation Keywords: folding near the surface. Structural interpretation based on seismic data suggests that strain is fl Active tectonics accommodated at very shallow levels by secondary exural-slip thrusts and reverse faults developed on fl Southern Alps synclinal anks that emanate from active axial surfaces. Analysis of syntectonic growth strata document Fault-propagation folds maximum rates of dip-slip of 3.45±0.66 mm/yr (1.6 Myr–1.2 Myr) and 0.47±0.22 mm/yr during a more Seismic hazard recent time period (0.89 Myr–present). A quarry excavation at Capriano del Colle allows a preliminary paleoseismologic analysis of coseismic surface faulting and liquefaction exposed near the core of an active mid-Pleistocene to Holocene anticline. These features are interpreted to be generated during strong local earthquakes, consistent with the environmental effects and ground motions of an event similar to the December 25, 1222, Brescia earthquake (Io=IX MCS). This indicates, for the first time, that compressive folds and blind thrusts in the Po Plain are currently accommodating slow rates of modern contraction in an active zone of the Southern Alps that extends from Lake Garda to Varese. We thus argue that earthquakes similar to the December 25, 1222 Brescia event are likelytooccurinthisregionandposeadirectthreatto such a densely populated and developed area. © 2009 Elsevier B.V. All rights reserved.

1. Introduction been assessed in a systematic way. As a consequence, historic, destructive seismic events in this area are poorly correlated with The Lombardian Po Plain (Northern Italy) is located at the southern well-defined, causative seismogenic structures. It is still unclear, for fringe of the western Southern Alps, an orogenic retro-wedge where instance, which is the causative tectonic source for local damaging shortening is accommodated by a fold-and-thrust belt (Fig. 1a) mostly seismic events such as the well studied December 25, 1222, Io=IX buried beneath a thick sequence of Plio-Pleistocene sediments. This MCS, Brescia earthquake, an earthquake which has had several region is characterized by low to moderate crustal strain rates and possible epicenters proposed for its source (e.g.; Magri and Molin, moderate to strong, infrequent seismic events, as documented by 1986; Guidoboni, 1986; Serva, 1990; Boschi et al., 2000; CPTI, 2004; historical seismicity catalogues (Boschi et al., 2000; CPTI, 2004; Guidoboni and Comastri, 2005). Guidoboni and Comastri, 2005; Fig. 1b). Even though seismicity Several studies have previously suggested the presence of active catalogues can be regarded as complete for the last ca. 1000 years, the structures in the area we studied. Early work by Desio (1965) identified seismic hazard of this region is still poorly constrained, particularly isolated hills, underlain by Quaternary marine and fluvial deposits because geological evidence for past earthquakes is rare and has not (Castenedolo, Ciliverghe and Capriano del Colle hills, see Fig. 7a for location), whose presence cannot be explained by glacial or fluvioglacial morphogenic processes. These hills were in fact previously ⁎ Corresponding author. Tel.: +39 031 326231; fax: +39 031 326230. interpreted as the culmination of growing anticlines associated with E-mail address: [email protected] (F.A. Livio). underlying thrusts (Desio, 1965; Boni and Peloso, 1982; Baroni and

Fig. 1. (a) Regional tectonic setting of Northern Italy (modiﬁed from Sileo et al., 2007) including the main structural fronts of the Southern Alps and Northern Appennines; barbs indicate the hangingwall; the location of Como (CO), Milano (MI), Bergamo (BG), Brescia (BS), Varese (VA), Triste (TS) and Venezia (VE) is also shown. The inset box locates the study area. (b) Historical and instrumental seismicity map of Northern Italy (Boschi et al., 2000; CPTI, 2004, modiﬁed).

Cremaschi, 1986; Cremaschi, 1987; Castaldini and Panizza, 1991; Curzi et al.,1996; Shaw and Suppe,1996, Mueller and Suppe,1997; Benedetti et al., 1992) but a detailed structural analysis have never been per- et al., 2000; Champion et al., 2001; Shaw et al., 2002; Dolan et al., 2003; formed until now. Ishiyama et al., 2004; Lin and Stein, 2006; Lai et al., 2006; Chen et al., “Blind” thrusts are known to be located near the epicentral area of 2007; Leon et al., 2007; Streig et al., 2007). Kinematic models of fault- the 1222 earthquake; we therefore investigated these structures in related folds have also been developed for both kink band (Suppe, order to determine whether they were potential seismogenic sources. 1983; Medwedeff, 1992; Medwedeff and Suppe, 1997) and trishear Many recent studies point to the importance of active folds as (Erslev, 1991; Hardy and Ford, 1997; Allmendinger, 1998) deformation indicators of coseismic slip on underlying capable faults (Burbank mechanisms. All these models allow restoration of deformed strata 322 F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 based on quantitative geometrical relationships between faults and December 25, 1222, Io IX MCS, Brescia; March 12, 1661, Io VII-VIII MCS, associated fold geometries and provide constraints on deep thrust Montecchio; May 29, 1799, Io VI-VII MCS, Castenedolo; the December geometry based on shallow folding. 5, 1802, Io VIII MCS, Soncino; November 30, 1901, Io VIII MCS, Salò; The architecture of syntectonic strata deposited during growth of November 13, 2002, Ml 4.2, Iseo; November 24, 2004, Ml 5.4 Salò active folds (e.g. growth strata) also provides a powerful tool in earthquake; e.g., Magri and Molin,1986; Guidoboni,1986; Serva,1990; documenting the growth history of compressive structures (e.g. Suppe Burrato et al., 2003; Guidoboni and Comastri, 2005). The distribution et al., 1992; Shaw and Suppe, 1994; Allmendinger and Shaw, 2000). of epicenters, based on historical seismicity catalogues (Boschi et al., Based on these structural methods and fault-related fold theory, we 2000; CPTI, 2004), highlights two main zones of seismicity along this combined an analysis of seismic reflection profiles with field studies sector of the northern Po Plain: a) a WNW–ESE-trending zone of and geomorphologic analysis to illustrate the geometry, kinematics, earthquakes, sub-parallel to the piedmont sector and to the buried Plio-Pleistocene growth history and earthquake potential of active thrusts and b) a NNE–SSW-trending zone, that is readily correlated blind thrusts and secondary faults located at the northern fringe of the with the Giudicarie fault System (Fig. 1b). Evidence for active Po Plain. We then consider these results with respect to a preliminary shortening is provided by geodetic data; these indicate shortening paleoseismological analysis based on trench-scale features observed at rates in the range of 2.2 mm/yr in the Friuli area and ca. 1.1 mm/yr near the Monte Netto site, near Capriano del Colle (Brescia). Finally we Lake Iseo measured with GPS from a fixed point in the Western Alps compare the seismogenic potential of blind thrusts in this region to the along vectors oriented NNE–SSW (Serpelloni et al., 2005). felt effects, macroseismic intensity, and estimated magnitude of the AD 1222 earthquake to determine if they give us a consistent picture, that 3. Stratigraphic setting of the Capriano del Colle region could fit into the local seismic landscape (sensu Michetti et al., 2005). This region of the Lombardian Southern Alps consists of a Mesozoic 2. Regional setting of the Capriano del Colle Fault System: Geological accretionary wedge including Late Triassic to Jurassic carbonatic and seismotectonic framework sediments deposited in the Jura-Cretaceous Lombardian Basin, following the opening of the Tethys Ocean (Winterer and Bosellini, The Capriano del Colle Fault System (CapFS; Fig. 2a) comprises the 1981; Gianotti and Perotti, 1987; Bertotti et al., 1997). The outset of the eastern end of an active fold-and-thrust belt located at the outermost Alpine orogenesis shed sequences of flysch (Cretaceous to Eocene) and buried structural front of the Southern Alps. The study area includes molasse (Gonfolite Lombarda Group, Oligo-Miocene in age) into the Po the piedmont zone of the chain and its foreland, in a sector located Plain Foredeep. The Molasse, outcropping in western Lombardy, between Lake Garda and the Oglio River (Fig. 2). constitutes a Tertiary clastic wedge in the subsurface of the northern The Southern Alps of Lombardy represent a typical example of an Po Plain (e.g.; Gelati et al.,1988; Bernoulli et al.,1989; Gelati et al.,1991; ancient passive continental margin, subsequently shortened into a Bernoulli and Gunzenhauser, 2001; Sciunnach and Tremolada, 2004; compressive orogenic wedge (e.g., Bertotti et al., 1997). The region is Fantoni et al., 2004). characterized by an array of south-vergent thin-skinned thrust sheets, The older stratigraphic sequence shown in the seismic profile in and is bounded to the East by a regional transfer zone, the Giudicarie Fig. 3 has been constrained by deep boreholes drilled by ENI E&P (see deformation system (Fig. 1a; e.g. Castellarin and Cantelli, 2000). The locations in Fig. 2a). Stratigraphic age controls for Quaternary outermost structural front of the Southern Alps is the Val Trompia– sediments are based on in-house stratigraphic, paleontological and Giudicarie belt (Fig. 1a; e.g., Castellarin and Vai, 1986; Castellarin and magnetostratigraphic analysis by ENI E&P and Regione Lombardia Cantelli, 2000; Castellarin et al., 2004; Castellarin et al., 2006), which (Carcano and Piccin, 2002; Scardia et al., 2006) and an extensive mainly developed during Oligocene to Late Miocene time. database of shallow well logs acquired in boreholes drilled for The distribution and segmentation of blind thrusts that comprise groundwater (Piccin, pers. comm.). The oldest strata visible in the the Capriano del Colle Fault System (Fig. 2a) was defined by regional profile belong to the Gonfolite Lombarda Group (GL), a foreland and mapping using an extensive database of ca. 18,000 km of seismic foredeep sequence of terrigenous units that range from Oligocene to profiles provided by ENI E&P that were constrained with hundreds of Miocene in age. The entire Quaternary sequence is deformed by fault- deep boreholes. related folds that form ramps above underlying Mesozoic carbonates; Faults were defined by obvious truncation of strata on seismic profiles these faults offset the overlying Tertiary clastic wedge and fold (e.g. footwall and hangingwall stratigraphic cutoffs) and by construction younger Plio-Pleistocene strata in the basin. Two decollement levels of structure contour maps of folded horizons of various ages in the Plio- can be detected. The deeper one is located at the base of Carbonate Quaternary sequence that infills the Po Plain basin (Fig. 2b). We Units while a shallower rheologically weak level can be identified at characterized segmentation of blind thrusts using several methods (e.g. the Late Eocene Gallare Marls Fm., at the base of Gonfolite Lombarda de Polo et al., 1991; Mirzaei et al., 1999) that included mapping of single Group (Fantoni et al., 2004; Ravaglia et al., 2006). segment faults that terminate at a transverse structure (e.g. a transfer The Gonfolite Lombarda Group has been divided into 3 mega- fault) or where folding decreases to zero at the end of structures as sequences (Fig. 3), based on internal tectono-sedimentary architecture defined by the regional dip of mapped strata. Other folds were identified and sequence stratigraphy. Details are discussed in Section 4.2. The as en-echelon structures that may rupture in multi-segment events. Tertiary sequence is bounded at its top by a regional erosional surface, Structure contour maps clearly show a belt of segmented, 10 to 20 km the Messinian Unconformity, related to an abrupt and long lasting sea long, fault-propagation folds buried beneath the Po Plain (Fig. 2aandb; level drop of the entire Mediterranean (e.g.; Cita and Wright, 1979; e.g. Vittori, 2004). The Capriano del Colle Fault System is composed by a 1982; Cita et al., 1990). Deep incision and associated channel networks south-vergent thrust (CCT) and an associated high angle backthrust are clearly recognizable in the seismic profiles (Fig. 3). At the northern (CCB) located within the Val Trompia–Giudicarie belt (Fig. 2a). edge of the profile a well-imaged channel, ca. 5 km wide, is interpreted Shortening rates accommodated by this thrust belt slows during to record deep Messinian incision of the Paleo–Mella river, cut the Tortonian (“Gonfolite” Deformative Phase: Oligocene–Miocene, longitudinally by the seismic line. Plio-Pleistocene infilling of this mainly Chattian and Burdigalian, and “Valsugana” Deformative Phase: sector of the Po Plain is recorded by a high frequency alternation of fine Serravallian–Tortonian; e.g., Castellarin et al., 2006). Subsequent strain and coarse sediments, mainly influenced by eustatic sea level is however, well expressed by folding of Plio-Pleistocene strata in this oscillations (e.g.; Rossi and Rogledi, 1988; Carcano and Piccin, 2002). sector of the northern Adria (Sileo et al., 2007; Chunga et al., 2007). The sequence defines a general aggradational trend, reaching a phase Evidence for ongoing, modern contraction is also defined by moderate of progradation and offlap, by longitudinal sediment transport, historical and instrumental seismicity recorded in this region (i.e., the between 1.2 and 0.9 Ma (Muttoni et al., 2003). This is recorded in F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 323

Fig. 2. (a) Shaded relief map based on a 20 m DTM showing the locations of buried active thrusts in the northern sector of the Po Plain; barbs indicate the hangingwall of the thrusts. ENI E&P deep boreholes and seismic line location are also represented. Main historical earthquakes are indicated. (Boschi et al., 2000; CPTI, 2004, modified); (b) Structure contour map of the buried 1.6 Myr Blue Surface and fault traces. Note the elongate structures that correspond to active folds developed above blind thrusts. Seismic line trace (Figs. 3 and 4)is also represented. Abbreviations: CapFS, Capriano del Colle Fault System; CastFS, Castenedolo Fault System; CCB, Capriano del Colle backthrust; and CCT, Capriano del Colle thrust. seismic profiles as a thick sequence of eastward prograding clinoforms, borehole data and dated by calcareous nannofossils and a magnetos- recognizable across the entire Po Basin, although this is not strongly tratigraphic analysis based on data from a deep borehole drilled by developed in the study area. Three widespread reflectors in the seismic Regione Lombardia (Muttoni et al., 2003; Scardia et al., 2006). data correlate with sequence stratigraphic surfaces of known age (Carcano and Piccin, 2002), and are recognized in the profile. These 4. Structural model and quaternary fault activity include the Blue, Green and Red Surfaces, respectively dated ca. 1.6,1.2 and 0.89 Myr (Fig. 3). These are penetrated by deep ENI E&P boreholes 4.1. Subsurface structure of the Capriano del Colle Fault System in the central Po Plain and are age dated with calcareous nannofossils (Carcano and Piccin, 2002). The Red surface, first recognized as A seismic reflection profile through the Capriano del Colle Fault outcropping in the Northern Apennines, was correlated with ENI E&P System was interpreted to constrain the geometry of the faults and the 324 ..Lvoe l etnpyis46(09 320 (2009) 476 Tectonophysics / al. et Livio F.A. – 335

Fig. 3. Raw depth-converted seismic proﬁle and stratigraphic interpretation based on correlation with ENI E&P deep boreholes (for the boreholes and seismic line location see Fig. 2a). White arrows indicate structural crest and depocenter points where calculations on structure uplift rates were performed. Lombardian Gonfolite Group was subdivided into 3 tectono-sedimentary sequences (Pre-GS; GS1 and GS2) recording the growth history of CapFS structure. ..Lvoe l etnpyis46(09 320 (2009) 476 Tectonophysics / al. et Livio F.A. – 335

Fig. 4. Depth-converted interpreted seismic profile, across the Capriano del Colle Fault System, constrained by constant-thickness fault-related fold theory. Location is shown on Fig. 2a. There is no vertical exaggeration. (a) Seismic line with structural features used to estimate fault trajectories of the Capriano del Colle Fault System. Numbers in yellow circles indicate principal constraints used for fault recognition: (1) faults cutoff, (2) termination of fold limb or kink band and (3) direct fault-plane reflection. Abbreviations are: CCT, Capriano del Colle thrust; CCB, Capriano del Colle backthrust; CCTFSF and CCBFSF are associated flexural-slip faults. Values above arrows indicate limb widths, measured on the section. Thick dashed lines, active axial surface; dash and dotted thin lines, inactive axial surface, blue dashed lines, erosional features. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 325 326 F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 associated folds. Some well-imaged reflectors, belonging to the GL located ca. 3.5 km below the ground surface, has accumulated a total Unit, were used to reconstruct the geometry of fault-related folds and displacement calculated as ca. 1200±50 m. Total displacement on the then, based on constant-thickness kink-band models (e.g.; Suppe and CCT was estimated using well-imaged reflectors in the Gonfolite Medwedeff, 1990), the deep geometry of fault trajectories, which Lombarda Group (Fig. 4). This value is also in good agreement with the is not imaged by stratigraphic cutoffs in the profile, was constrained. backlimb width of the CCT, as predicted by the model. CCT faces a Fig. 4 shows modeled fault geometry and the criteria upon which fault north-vergent Apenninic thrust, located a few kilometers to the south, trajectories have been defined. Fault geometry has been constrained i.e., the Bordolano thrust (Fig. 4). This thrust, whose tip is located at based on angular fold limbs; this is a reasonable approximation to the 5.5 km in depth, also has a fault-propagation fold in its hangingwall. curved fold hinges seen in the profile. An error analysis of this The Bordolano thrust is defined by a seismic line located near its lateral restoration, based on line length balancing using a well-imaged termination (line trace on Fig. 2), in a sector characterized by low angle horizon within the GL unit (Fig. 4), yields a maximum error of 12% in faulting and limited structural relief. The CCT appears to have the worst case. We also note that analytical solutions for curvilinear propagated southward and overlaps the Bordolano thrust at its fault-bend folding (e.g.; Suppe et al., 1997; Tavani et al., 2005) yield easternmost edge. The CCB is a north-verging backthrust associated solutions comparable with those produced by classical kink-band to the CCT master fault. CCB geometry is constrained through direct models (Tavani et al., 2005). Associated growth strata show fanning fault-plane reflections, the downward termination of the fold limb and limb dips (growth wedge), with progressively less dipping strata in stratigraphic cutoffs across the fault (Fig. 4a). Since the northern younger deposits. This is inconsistent with constant-thickness theory portion of the seismic profile is less well imaged, we cannot exclude: a) unless widely curved fold hinges are invoked (Tavani et al., 2005). the presence of an inner backthrust, as suggested by low-amplitude, Other models, predicting folding by progressive limb rotation and discontinuous reflectors, or b) a slightly different position for the fault growth wedge generation include fixed-axis fault-propagation folding tip. The Plio-Pleistocene sedimentary sequence is clearly folded but (e.g.; Suppe and Medwedeff, 1990), shear fault-bend folding (e.g.; apparently not offset by the CCB fault plane. This could be interpreted Suppe et al., 2004) and trishear fault-propagation fold (e.g.; Allmen- to suggest that CCB tip-line is located within the GL sequence or near dinger,1998). Fixed-axis fault-propagation folding was ruled out based the Messinian Unconformity. Since the front limb of this structure is on limb dips in growth strata (Suppe and Medwedeff, 1990) and a lack not well imaged, we cannot unambiguously restore this fault; in the of evidence for bed thinning or thickening in the forelimb of the fault- present interpretation we consider a tip-line to be located just below propagation fold. Shear fault-bend folding was also ruled out since a the Messinian Unconformity. Total slip is ca. 5700±50 m, measured shear profile (e.g., the difference between the unfolded hangingwall along the CCB backlimb. fault shape and the actual fault shape) calculated on CCT, shows that no The structural style depicted here is therefore characterized by a significant layer parallel simple shear has been accommodated in the triangle zone, or structural wedge, defined by a primary south-vergent hangingwall of the thrust. Moreover the fold-shape itself lacks of forethrust (CCT) and a connected north-vergent backthrust (CCB). typical shear fault-bend fold features (e.g., a significant decrease in dip Secondary flexural-slip faults (CCBFSF e CCTFSF) are associated with between the fault ramp and associated backlimb of the fold) and thus is both these structures. Flexural-slip faults play an important role in the considered to be inconsistent with shear fault-bend folding. Both the accommodation of strain above blind thrusts (Yeats, 1986; Roering CCT and CCB structures are apparently blind, since they do not offset et al., 1997; Nino et al., 1998; Ishiyama et al., 2004). Several models the Plio-Pleistocene sequence (their tips are approximately located and natural examples, in fact, indicate that bedding-parallel strain near the Messinian Unc.), and their upward propagation is expressed tends to consume slip and therefore hinder the upward propagation by an array of fault-related and secondary folds. The CCT, whose tip is of faults and fault-related folds. In flexural-slip faulting only thrust

Fig. 5. Assessment of uplift and in situ sedimentation rates relative to chronologic intervals: a) Sketch of the methodological approach (e.g. Masaferro et al., 2002) applied to calculate uplift rates; b) graphic summary of the results according to the considered time windows. Error bars and U/S ratio value are also indicated. (⁎) indicates minimum uplift rate value, according to a ﬁll-to-the-top growth model. F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 327

Fig. 6. Slip rates for the Capriano del Colle Fault System. Since there is no evidence for significant lateral displacement, only the net dip-slip throw has been considered. (⁎) indicates minimum slip rate value, according to a fill-to-the-top growth model. ramps are usually visible while detachments often remain indis- the CCB structure. This sequence has been therefore interpreted as tinguishable and cannot be resolved (Ishiyama et al., 2004). Thus only CCT piggyback basin deposits whose deposition was contemporary the largest faults, that have accommodated significant displacement, with wedge growth. During this period the southward migration of can be interpreted by analysis based on seismic profiles. We expect the buried Alpine front created a belt of structures directly facing the that many similar structures, at different scales, affect the sedimentary most external (e.g. frontal) thrust sheets in the Apennines belt. GS2 cover over fault-related folds and therefore are partially responsible in (Tortonian) strata record fault/fold growth on structures in the frontal upward consumption of slip. The analysis of axial surfaces recogniz- structures of both the Alps and Appennines, and thus constitute a able in the seismic line is consistent with this interpretation. Some of wedge-top basin. The CCB began to develop, during this period, as an the axial surfaces relative to CCB are in fact crosscut by the CCBFSF out-of-sequence north-verging backthrust. A kinematic solution for fault plane. This is generally diagnostic of break-back imbricate the Capriano del Colle Fault System and a detailed analysis of the Plio- thrusting, where movement of the upper thrust is successive or at Pleistocene syntectonic sedimentary record allows slip rates to be least coeval to the deeper one. Total slip calculated for CCBFSF is ca. determined for the blind thrust beneath the folds. The geometry of 1250±50 m, based on the width of the fault-bend fold backlimb. A growth structures is controlled primarily by the shape of the shallow south-vergent secondary structure, visible in the first 500 m underlying fault and by the relative rates of sedimentation and uplift at the northern edge of the seismic line, cuts through Early Pleistocene (Suppe et al., 1992). The ratio between sedimentation and uplift rate strata, just below the Capriano del Colle hill (Fig. 4). A detailed exerts a significant influence by either enhancing or masking thinning analysis of this secondary fault and its relationship with present-day of strata across a particular structure. The crestal structural relief morphology is discussed in Section 5. (McClay, 1992) at a specific time T has been obtained by subtracting the thickness of all the growth beds deposited on the anticline crest 4.2. Tectono-sedimentary history of the Capriano del Colle Fault System prior to time T from the thickness of the same growth beds in the basin adjacent to the anticline. Rock uplift rates (U) have then been While sedimentation and erosion in the Po Basin is affected largely calculated as the difference in structural relief, relative to the top and by orogenic loading of the foreland of the Alps and Apennines and to the bottom of a specific growth bed (e.g., Suppe et al., 1992; regional climatic and base level change associated with eustacy, local Masaferro et al., 2002; Fig. 5a). Crest and basin depocenter points used stratigraphic architecture reflects growth of Late Quaternary folds. This for calculations are indicated on Fig. 3. In situ accumulation rates (S) allows us to measure rates of rock uplift on these structures, define the were also calculated and compared to relative uplift rates (U/S ratio; kinematic history and style of compressive strain and hence define Fig. 5b). A very high accumulation rate, with respect to uplift rate (very seismic hazard. Syntectonic growth theory (Suppe et al., 1992) low ratio U/S), can in fact, obscure the tectonic signal and appear as an suggests that cumulative displacement will decrease upward within interval apparently not influenced by contemporary uplift. Uncertain- the stratigraphic interval deposited during deformation. In contrac- ties in uplift and sedimentation were evaluated in terms of the tional fault-related folds, syntectonic strata typically thin across folds resolution and variability in seismic reflector character (ca. 25 m), and limbs towards structural highs. We focus on the GL unit, whose errors in age bracketing (ca.±0.1–0.05 Myr, according to the deposition is quite contemporary to the inception of the Val Trompia– considered surface). Errors were calculated following the error Giudicarie structural front. This unit has been divided into three propagation rule for values with unequal standard deviations (Geyh tectono-sedimentary sequences that record the timing, growth and Schleicher, 1990). Results, relative to each structure, are summar- history, and strain rates for the Capriano del Colle Fault System (Fig. 3). ized as follows (Fig. 5). Pre-GS (Chattian–Serravallian p.p.) strata comprise a clastic coarse For the CCT, during the first chronostratigraphic interval (Pliocene – sequence, up to ca. 2 km thick, that is faulted and folded by the CapFS. 1.6 Myr) very low values of uplift and sedimentation rates have been The pre-GS sequence predates tectonic deformation since no changes recorded. Growth beds onlapping the fold limb and thinning as they in sequence thickness and no major onlap–offlap geometries have approach the fold crest predominated during the first part of this been recorded. GS1 (Serravallian p.p.–Early Tortonian) strata record period. Such a low value of average rock uplift rate may be related to contemporary development of the CCT, marked by progressively averaging a relatively short episode of tectonic deformation over a very onlapping reflectors. To the north, thickness remains constant across long time period (duration is ca. 3.6 Myr). The second time window 328 F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 329

(1.6–1.2 Myr) records higher values both in uplift and sedimentation projected locations of hinge zone traces coincide with local structural rates. U=0.51±0.19 mm/yr and S=2.45±0.19 mm/yr. U/S ratio is and topographic relief. The Mella River cuts through the structures 0.21. The following period is characterized by a deactivation of this flowing from north to south and the trace of this river and its thrust, as testified by syntectonic growth strata that maintain the same tributaries appear to be related to mapped structural features. thickness throughout the CCT crest. Since the S value (2.10±0.33 mm/ Anticlinal hinge zone traces run in fact along alignments of river yr) is constant for this period, compared to the previous interval, it is deviations and, as expected in areas of progressively decreasing apparent that the tectonic signal has not been concealed by a gradient (e.g.; Ouchi, 1985), an increase in river sinuosity is observed significant increase in sedimentation rates. The 0.89 Myr to present as each river crosses an anticlinal hinge zone (Fig. 7b). Minor drainage, chronostratigraphic interval records a decrease both in uplift and deep entrenched gullies, affects mostly the southern side of the hill. sedimentation rates. These channels, ca. 3–6 m deep and 2–5 m wide, were excavated For the CCB, during a Pliocene to 1.6 Myr time window, our analysis through the whole sequence of the hill during the fold growth. suggests that this structure initially experienced similar low values of The present channel of the Mella River flows around the western sedimentation and uplift rates. We maintain the same considerations edge of the hill. Three paleo-channels, all attributed to older channels of discussed above for calculations on rock uplift CCT structures to be the Mella River, were recognized both at the level of the modern valid. During the 1.6–1.2 Myr time interval S=1.63±0.19 mm/yr and floodplain and cutting through the hill surface. A 4-stage evolution U=1.35±0.19 mm/yr. In the following period (1.2–0.89 Myr) tectonic model is thus proposed based on Mella River paleo-channel traces deformation is still active (U=1.38±0.33 mm/yr) contemporaneous (numbered circles on Fig. 7b). Relative timing of these channels has been to a significant decrease in sedimentation rates, as highlighted by the inferred from geomorphologic constraints and channel-bed relative high value of U/S ratio. This drastic lowering in sedimentation rate is height. This model implies a progressive increase in the area of rock consequence of the progressive infilling of the Po Plain basin that, for uplift of the Capriano del Colle structure and a contemporary northward this sector, was starting to shift to a continental, alluvial plain migration of the Mella River, which has been progressively deviated to environment. The latest period (0.89 Myr–present) is characterized by the east (stages 1 to 3). During each stage, the Mella River was deviated a decrease in uplift rate (0.22±0.06 mm/yr) and the onset of regional to the east by the growing fold; each paleo-channel therefore records the nondeposition and/or erosional conditions over the entire Po Basin former endpoint of the fault-related fold at the surface. This does not area. The Red Surface is thus not visible throughout the entire CCB actually correspond to the endpoint of the fold as defined by subsurface structural high. Thus, assuming a fill-to-the-top growth model, the mapping, because the structure at the surface is currently in a region of calculated rate has to be considered as a minimum value. Calculated active deposition and sediment onlap. Widening of the area undergoing uplift rates were then used to obtain net dip-slip rates, assuming the rock uplift and related northward migration of the river channels are underlying fault geometry (Fig. 6). consistent with progressive forelimb widening above a kink-band Seismic reflection profile interpretation and calculated uplift and migration model of fault-propagation fold rather than progressive slip rates have constrained the tectono-sedimentary history of CapFS tilting, as would be expected from fixed-axis models. Finally, in stage 4, from Pliocene to Middle Pleistocene time. The calculated slip rates the Mella River deviated to the west, to its present-day position. (Fig. 6), relative to the youngest constrained chronostratigraphic A detailed structural analysis of the seismic line, performed on the interval (0.89 Myr–present), is 0.47±0.21 mm/yr. More recent rock original double-TWTT section, revealed the presence of a shallow uplift above these faults could not be resolved by industrial south-verging structure, cutting through an Early Pleistocene geophysical data, since the available seismic profiles do not contain sequence, recognizable in the northern sector of the seismic line, in reflectors in the upper ca. 150 m and dated surfaces, younger than the upper ca. 500 m (Ccα on Fig. 8b). This structure consists of a multi- 0.89 Myr, are not mapped in this area. bend flexural-slip fault whose ramp sector is pinned, at its base, with the synclinal axial surface on the CCB (α on Fig. 8b). Even though the 5. Landscape response to fold growth: The Capriano del Colle hill front limb of CCB fold is not well imaged on the seismic line, both CCB and Ccα associated folds show narrower hinge zones and a growth The Capriano del Colle hill is located at the northern fringe of the strata architecture typical of constant-thickness model. As the CCB Lombardian Po Plain, just south of the city of Brescia (Fig. 7a). It is an grows, slip is marked by northward migration of the α axial surface. As isolated hill, almost rectangular in shape (ca. 5 km in length and 2 km a consequence, this secondary shallow flexural-slip fault passively in width), trending N 110 and sticking out ca. 30 m relative to the accommodates strain southward. Growth triangles developed above surrounding alluvial plain. The Garda piedmont glacier, coming from the fault hangingwall indicate that fault movement probably started the north, has never reached this sector of the Po Plain, as testified by after 1.2 Myr (Green Surface) and has continued to the present (Fig. 8b the most external terminal moraines of the Garda amphitheater and c). Topography of the central sector of Capriano del Colle hill, (Fig. 7a). The hill exposes a gently tilted, fining-upward sequence of obtained through a total station survey (Fig. 9), corresponds exactly to Pleistocene fluvioglacial and fluvial sediments (braided floodplain the structural relief defined by two outcropping secondary anticlines environment) that is unconformably overlain by loess sequences exposed on a quarry wall that fold and displacing Middle to Late (Fig. 7b). The hill is deeply eroded on its sides. During Quaternary Pleistocene deposits (Fig. 10). These anticlines are located in the time, large glaciers extended along the valleys that now contain Lake crestal sector of Ccα associated fold (Fig. 9). Seismic reflection data are Garda and Lake Iseo; these are located 15 and 10 km to the east and to not resolved well enough to define the architecture of growth strata the northwest of Capriano del Colle hill, respectively. The morphology on this structure. of the piedmont belt in this area, except where it is covered by moraines, is essentially constituted by a huge outwash fan, developed 6. Surface faulting from folding: The Monte Netto site during the last glacial maximum by a meltwater channel network (e.g., Baroni and Cremaschi, 1986; Marchetti, 1996) that has been In order to define the most recent evidence for compressive subsequently incised by local drainages. The surface projection of the tectonics in the CapFS area, a detailed geological and geomorphological CCB fault is located just north of Capriano del Colle hill (Fig. 7b). The field survey was conducted. Investigation sites were selected based on

Fig. 7. (a) 3D view of northern Po Plain and Capriano del Colle area. Vertical exaggeration: 5×; (b) geological map of Capriano del Colle area: fold hinges, paleo-channels and secondary folds are indicated. Numbered circles indicate progressive northward migration and deviations of Mella River around the hill. Note that present-day river deviations and pattern changes coincide with their passage across the folds. Mella River trace was derived from a 1968 topographic map. 330 ..Lvoe l etnpyis46(09 320 (2009) 476 Tectonophysics / al. et Livio F.A. – 335

Fig. 8. Multi-scale seismic reflection images and interpretations of the CCB fault and Capriano del Colle topographic profile. Topographic vertical scale is strongly exaggerated in order to highlight subtle geomorphic features. Light grey lines indicate line drawing; black continuous lines, modeled horizons; thick dashed lines, active axial surface; dash and dotted thin lines, inactive axial surface; α indicates CCB synclinal active axial surface and Ccα an associated passive flexural-slip fault, see text for details. F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 331

Fig. 9. Modeled fault/fold geometry of Ccα structure: topographic profile, vertically exaggerated, was obtained through a total station survey, central white-shadowed sector indicates lack of data; see figure for outcropping secondary anticlines photographs. the most promising geomorphological evidence of recent tectonic clearly displaced. The maximum throw on the exposed normal fault is activity and available outcrops of Quaternary deposits. A quarry ca. 2 m, measured at the top of the central graben (Fig. 10d). Offset excavation at the top of the Capriano del Colle hill, hereinafter referred across the faults decreases downward, confirming their origin as to as the Monte Netto site, allowed us to conduct a preliminary secondary structures that form in response to folding. This fault cuts at palaeoseismological analysis, supported by new trenching and soil least the lower part of the Fg2 sequence, without evidence of colluvial stratigraphy data, on ca. 150 m long, 7 m high, north–south and east– re-deposition or strata thickening thus it clearly postdates the west trending trench walls. The detailed paleoseismic analysis of the displaced sequence. Furthermore, the upper part of the paleosol Monte Netto site is beyond the scope of this paper and will be the sequence is characterized by the occurrence of sub-vertical parallel subject of a companion paper; here we provide the main preliminary fissures, regarded as secondary brittle deformations successively results which are relevant for the understanding of surface slip affected by pedogenetic processes with illuvial clay infilling. The propagation from the buried CCB fault, and related seismotectonic exposed section also contains a liquefaction feature (Fig. 10e) implications. Two secondary anticlines, tens of meters in size (ca. 4 to consisting of a sand and gravel dike cutting through ca. 60 cm of the 9 m of exposed amplitude) and deforming a sequence of Quaternary sequence and whose presence is interpreted to be related to the strong continental strata (gravel, sand and clay), have been recognized (Fig. ground motions produced by a local earthquake, possibly generated 10a and b). The core of the anticlines is composed of a sequence of by the underlying CCB structure. fluvioglacial sediments. These units are overlaid by a loess–paleosols sequence, laterally thickening up to 6 m (Fig. 10b). The stratigraphy of 7. Discussion this sequence has been described in detail by Cremaschi (1974, 1987) based on observations from an older quarry exposure located nearby. Reconstruction of CapFS fault geometry and kinematics allow us to The paleosols sequence includes three different paleosol-parent test the consistency between fault parameters and the seismotectonic materials, which are related to different episodes of pedogenesis. The framework depicted by historical seismicity. Available seismic data, oldest paleosol, red colored, was developed on the fluvioglacial although of good quality and resolution, do not allow us to derive a sequence (Middle Pleistocene (?); Fg2), whereas the other two soils single deterministic model for maximum expected Magnitude for were developed on the loess sequence (Upper Pleistocene–Holocene structures that are deforming at the low strain rate common in the (?); LL). Radiometric and OSL dating is in progress at this site, however Southern Alps. A better estimate can, however, be derived from an some age constraints are provided from the development of paleosols analysis of several scenarios for fault derived Mw, including and from paleo-ethnological studies. (e.g., Cremaschi, 1974). Recent uncertainties related to fault geometry. Table 1 summarizes the flint artifacts findings give us an age range of 300 to 30 kyr B.P. for the maximum and minimum values of for fault length, width, fault area, whole loess sequence (Berlusconi et al., 2007). net dip-slip rates and range of derived Mw for each of the studied The loess was deposited onto the topographic surface after folding structures. Scalar relationships between fault area and maximum of the fluvioglacial sequence and ongoing strain caused the develop- expected Magnitude (e.g. Wells and Coppersmith, 1994), indicate that ment of brittle deformation. Deformation is characterized by exten- both CCT and CCB, if considered active seismogenic sources, could be sional bending-moment faults, formed on the culmination of the able to produce Mw=5.9–6.5 earthquakes. northern anticline, that offset the entire exposed sequence (Fig. 10f). A The maximum recorded historical earthquake in the region is the reverse fault was exposed by trench excavation at the southern December 25, 1222 Brescia earthquake Io IX MCS), one of the largest termination of the northern anticline (Fig. 10b and c). This fault offsets events in the entire Po Plain, whose epicentral area has been located the sequence up to 3 m and reveals a southern vergence for the just south of Brescia city (Fig. 11). This event, which characterizes the northern anticline. seismotectonic framework of the Lombardian Southern Alps (e.g., Bending moment faults are interpreted to result from coseismic Magri and Molin, 1986; Serva, 1990), produced significant damage stretching of the top of the anticline where the uppermost soil is across much of Northern Italy and generated ground effects in the 332 ..Lvoe l etnpyis46(09 320 (2009) 476 Tectonophysics / al. et Livio F.A. – 335

Fig. 10. (a) Western quarry wall on the top of Capriano del Colle hill. Location is shown on Fig. 7b; (b) detail on the sector between northern and southern anticline, location of the trench represented in (c) is also shown; (c) eastern trench wall excavated at the quarry floor: a high angle reverse fault, also indicated in (b), has been documented; (d) detail of the bending-moment normal faults affecting the upper part of the sequence: note the progressively downward decrease in offset of normal faults; (e) detail on the identified liquefaction feature: a sand and gravel dike, ca. 80 cm high, and an associated flame-like secondary deformation affecting the overlying light brown clays; (f) structural data of the northern anticline outcrop: fold limb, fractures and faults are indicated. Calculated paleo-stress (after Tucker, 1951) documents a σ1 principal stress almost vertical and a σ3 stress, directed N190. Abbreviation are: Fg2, Upper Fluvioglacial sequence (Middle Pleistocene ?), LL, Loess deposits (Middle to Late Pleistocene?). F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 333

Table 1 The tectonic source of this earthquake still remains questionable but Geometric and seismotectonic characteristics of the studied faults; minimum and some important observations can be summarized as follows: maximum values are indicated.

Fault name W (Km) L (Km) Area (km2) Slip (mm/yr)a Mwb – The CCB structure is located beneath the epicentral area of the CCT 3.5–10.5 20–23 70–242 0.04 5.9–6.4 earthquake (Fig. 10b) as well as the CastFS north-verging thrust CCB 8–11.6 20–30 160–348 0.43 6.2–6.6 (Fig. 2a); they both belong to the same structural front and show aAverage slip rate is considered over the most recent time interval (0.89 Myr–present, similar structural architecture. see Fig. 6). – The Macroseismic ﬁeld shows an almost ENE–WSW major axis, b Derived from empirical regression between Subsurface rupture area and Mw (Wells consistent with the average trend of the buried structures (Fig. 11). and Coppersmith, 1994). – Intensity derived Magnitude for the Dec. 25, 1222 event is Me=6.2 (Guidoboni and Comastri, 2005). This estimate is also in good epicentral area such as sudden river deviations and ground cracks agreement with the distribution of Magnitude values for each Intensity (Guidoboni, 1986; Boschi et al., 2000; Guidoboni and Comastri, 2005). class, based on an extensive catalogue in the Mediterranean region No direct association can presently be demonstrated between CapFS (D'Amico et al., 1999). Mean value for Intensity (MCS) IX is Ms 6.2, in a structures and the Christmas 1222 event, notwithstanding the high range between 5.8 and 6.7. All these estimates are comparable with quality subsurface information available and analyzed in this study. CCB seismogenic source characteristics (ca. Mw=6.2–6.6, see Table 1).

Fig. 11. (a) Macroseismic map of the December 25, 1222 Brescia earthquake derived from point data published by Guidoboni and Comastri (2005). Intensity data are indicated as light grey dots; (b) detail of the macroseismic ﬁeld compared with a digital elevation model of the area and the surface projection of main buried structures. 334 F.A. Livio et al. / Tectonophysics 476 (2009) 320–335

– Surface environmental effects for such strong earthquakes usually anonymous reviewer for their constructive suggestions and criticisms, include fracturing, surface displacement and liquefaction features which greatly improved the manuscript. This research has benefited (Serva, 1994; Guerrieri and Vittori, 2007). Features observed in from funding provided by the Italian Presidenza del Consiglio dei Capriano del Colle site are consistent with such effects and with Ministri – Dipartimento della Protezione Civile (DPC). Scientific those quoted for the Christmas 1222 event (Guidoboni and papers funded by DPC do not represent its official opinion and policies. Comastri, 2005). In particular surface deformation has been interpreted as due to secondary flexural-slip faulting, as observed References for the CCB structure. – CapFS and adjacent structures (i.e. CastFS, to the east, see Fig. 2a) Allmendinger, R.W., 1998. Inverse and forward numerical modelling of trishear fault- – are characterized by similar fault geometry and Quaternary uplift propagation folds. Tectonics 17, 640 656. Allmendinger, R.W., Shaw, H., 2000. Fault nucleation and propagation from fold shape: rates (Livio et al., 2009). Deformation rates in the order of potential implications for earthquake seismicity. Geology (28), 1099–1102. fractions of a mm per year are consistent with the earthquake Baroni, C., Cremaschi, M., 1986. Geologia e pedostratigrafia della Collina di Ciliverghe – size, slip per event and recurrence interval displayed by Christmas (Brescia). Nat. Brescia., Ann. Mus. Civ. Sci. Nat. 23, 55 78. Basili, R., Valensise, G., Vannoli, P., Burrato, P., Fracassi, U., Mariano, S., Tiberti, M.M., 1222-like earthquakes, regarded as the maximum historical Boschi, E., 2008. The Database of Individual Seismogenic Sources (DISS), version 3: earthquake experienced by the northern Po Plain (e.g., Serva, summarizing 20 years of research on Italy's earthquake geology. Tectonophysics 1990; see the empirical relations presented in Slemmons and 453 (1–4), 20–43 16. Benedetti, L., Tapponier, P., King, C.P.G., Meyer, B., Manighetti, I., 2000. Growth folding DePolo, 1986). and active thrusting in the Montello region, Veneto, Northern Italy. J. Geophys. Res. – Repeated 1222-like earthquakes occurred during the Late Qua- 105 (B1), 739–766. ternary in this area would have produced a cumulative observable Berlusconi, A., Livio, F., Sileo, G., Zerboni, A., Michetti, A.M., Cremaschi, M., Trombino, L., fi 2007. Evidenze paleosismiche nell'area epicentrale del terremoto del 25.12.1222: effect of coseismic surface deformation and faulting, i.e., a speci c risultati preliminari. Gruppo Nazionale di Geofisica della Terra Solida – 26 Congresso seismic landscape (Michetti et al., 2005). The observations at Nazionale – 13–15 Novembre 2007. – Roma, pp. 238–241. Capriano del Colle site are consistent with this hypothesis. Bernoulli, D., Bertotti, G., Zingg, A., 1989. Northward thrusting of the Gonfolite Lombarda (“South-Alpine Molasse”). Eclogae Geol. Helv. 82, 841–856. Bernoulli, D., Gunzenhauser, B., 2001. A dolomitized diatomite in an Oligocene–Miocene 8. Conclusions deep-sea fan. Sediment. Geol. 139 (1), 71–91. Bertotti, G., Capozzi, R., Picotti, V., 1997. Extensions controls, Quaternary tectonics, This study defines the growth and seismic hazards posed by blind geomorphology and sedimentation of the N-Appennines foothills and adjacent Po Plain (Italy). Tectonophysics (282), 291–301. thrusts in the Po Plain using interpretation and structural analysis of Boni, P., Peloso, G.F., 1982. Dati sulla neotettonica dei fogli 34 qBrenoq,47qBresciaq edi seismic data, geomorphological analysis of rivers and trench-scale parte dei fogli 35 qRivaq e48qPeschieraq. Vol. pubbl. 506 – P.F. Geodinamica, in excavations of active folds and faults. A structural solution and Contributi conclusivi per la realizzazione della Carta Neotettonica d'Italia., di CNR, pp. 189 – 208. geomorphic evidence are consistent with a deformation model that Boschi, E., Guidoboni, E., Ferrari, G., Mariotti, D., Valensise, G., Gasperini, P., 2000. Catalogue implies progressive kink-band widening above faults characterized by of strong Italian earthquakes from 461 B.C. to 1997. Ann. Geofis. 43, 609–868. curved fold hinges. Bedding parallel faults, formed near the surface, Burbank, D., Meigs, A., Brozovic, N., 1996. Interactions of growing folds and coeval – fi depositional systems. Basin Res. (8), 199 223. play an important role in accommodating folding and induce sur cial Burrato, P., Ciucci, F., Velensise, G., 2003. An inventory of river anomalies in the Po Plain, deformation, where no primary surface faulting has to be expected. Northern Italy: evidence for active blind thrust faulting. Ann. Geophys. 46 (5), 865–882. This finding has primary importance for the identification of capable Carcano, C., Piccin, A. (a cura di), 2002. Geologia degli acquiferi Padani della Regione Lombardia. Firenze: S.EL.CA. faults (sensu IAEA, 2002; Michetti et al., 2005) in active compressional Castaldini, D., Panizza, M., 1991. Inventario delle faglie attive tra i fiumi Po e Piave e il tectonic environments and for surface faulting hazard evaluation in Lago di Como (Italia settentrionale). Il Quaternario 4 (2), 333–410. the Po Plain, particularly as related to siting and seismic design of Castellarin, A., Vai, G.B., 1986. Southalpine versus Po Plain Apeninic Arcs. In: Weizel, C. (Ed.), The origin of Arcs. Elsevier Science Publisher B.V., Amsterdam, pp. 253–280. critical facilities. The reconstruction of Quaternary tectono-sedimen- Castellarin, A., Cantelli, L., 2000. Neo-Alpine evolution of the Southern Eastern Alps. tary history of CapFS structures, based on geophysical data, demon- J. Geodyn. 30, 251–274. strates that both the CCB and CCT are active at least up to Middle Castellarin, A., Transalp Working Group, 2004. Structural syntesis of the Eastern Alps: a fi Pleistocene time. Geomorphological data and our preliminary analysis collisional orogenic chain. In: d'Of zi, F., Merlino, S., Sacchi, L., Crescenti, V. (Eds.), Geology of Italy – Special Volume of the Italian Geological Society for the IGC 32 on outcrops of folded strata probably record active strain during Late Florence 2004, pp. 3–13. Pleistocene time. Work is in progress to obtain a complete dating of Castellarin, A., Vai, G.B., Cantelli, L., 2006. The Alpin evolution of the Southern Alps the Capriano del Colle quarry sequence through OSL and AMS around the Giudicarie faults: a Late Crataceous to Early Eocene transfer zone. Tectonophysics 414, 203–223. analyses. Earthquake scenarios depicted for CapFS faults seismic Champion, J., Mueller, K., Tate, A., Guccione, M., 2001. Geometry, numerical models and potential (Mw=5.9–6.6) are consistent with the known northern Po revised slip rate for the Reelfoot fault and trishear fault-propagation fold, New Plain seismotectonic framework and historical seismicity. From a Madrid seismic zone. Eng. Geol. 62, 31–49. Chen, Y.G., Lai, K.Y., Lee, Y.H., Suppe, J., Chen, W.S., Lin, Y.N.N., Wang, Y., Hung, J.H., Kuo, Y.T., deterministic point of view, an earthquake of similar energy could 2007. Coseismic fold scarps and their kinematic behavior in the 1999 Chi-Chi represent a serious hazard and cause significant losses in such a earthquake Taiwan. J. Geophys. Res. 112, B03S02. doi:10.1029/2006JB004388 2007. densely populated area as the Northern Italy. Chunga, K., Livio, F., Michetti, A.M., Serva, L., 2007. Synsedimentary deformation of Pleistocene glaciolacustrine deposits in the Albese con Cassano Area (Southern Nevertheless in recent seismic hazard analyses and seismogenic Alps, Northern Italy), and possible implications for paleoseismicity. Sedimentary fault compilations (Basili et al., 2008; Meletti et al., 2008) these Geology 196 (1–4), 59–80. structures have not been adequately taken into account and their Cita, M.B., 1982. The Messinian salinity in the Mediterranean: a review. In: Berkhemer, E., Hsu, K.J. (Eds.), Alpine-Mediterranean Geodynamics. Geodynamics Series, vol. 7. meaning in terms of seismic potential haven't been considered. This Cita, M.B., Wright, R.C., 1979. Geodinamic and Byodinamic Effects of the Messinian Crisis underlines the primary role that geological and paleoseismological of Salinity. Paleogeogr., Paleoclimat., Paleoecol., vol. 29. Spec. Issue. data should play in the understanding of earthquake hazard, Cita, M.B., Bini, A., Corselli, C., 1990. Superfici di erosione messiniana: un'ipotesi especially in regions characterized by moderate seismicity and low sull'origine dei laghi sudalpini. Il lago Maggiore dalla ricerca alla gestione. Documento dell'Istituto Italiano di Idrobiologia, vol. N.22, pp. 33–54. strain rates. CPTI, 2004. Catalogo Parametrico dei Terremoti Italiani, versione 2004 (CPTI04). INGV, Bologna. http://emidius.mi.ingv.it/CPTI/. – Acknowledgments Cremaschi, M., 1974. Manufatti del Paleolitico medio inferiore, provenienti da Monte Netto di Brescia e loro rapporti con i depositi quaternari del Colle. Nat. Brescia. – Ann. Mus. Civ. Stor. Nat. Brescia. 11, 41–57. We thank the municipality of Capriano del Colle, Brescia Province Cremaschi, M., 1987. Paleosol and Vetusols in the Central Po Plain (Northern Italy) a administration and in particular Fornaci Laterizi Danesi S.p.A. for study in Quaternar geology and soil development. Edizioni Unicopli, Milano. Curzi, P.V., Castellarin, A., Ciabatti, M., Badalini, G., 1992. Caratteri morfostrutturali, permission to site access and trenching. Also F.lli Arici S.r.l. is thanked sedimentologici e genetici del Lago di Garda. Boll. Soc. Torricelliana di Scienze e for logistic support. Special thanks to F. Storti, G. Roberts and an Lettere 43, 3–111. F.A. Livio et al. / Tectonophysics 476 (2009) 320–335 335

D'Amico, V., Albarello, D., Mantovani, E., 1999. A distribution-Free analysis of magnitude– Mueller, K., Suppe, J., 1997. Growth of Wheeler Ridge anticline, California: geomorphic intensity relationships; an application to the Mediterranean region. Phys. Chem. Earth evidence for fault-bend folding behavior during earthquakes. J. Struct. Geol. (19), Part A: Solid Earth Geod. 24 (6), 517–521. 383–396. de Polo, C., Clark, D., Slemmons, D., A.R., R., 1991. Historical surface faulting in the Basin Muttoni, G., Carcano, C., Garzanti, E., Ghielmi, M., Piccin, A., Pini, R., Rogledi, S., Sciunnach, and Range province, western North America: implications for fault segmentation. D., 2003. Onset of major Pleistocene galciations in the Alps. Geology 31 (11), J. Struct. Geol. 13, 123–136. 989–992. Desio, A., 1965. I rilievi isolati della Pianura Lombarda ed i movimenti tettonici del Nino, F., Philip, H., Chery, J., 1998. Role of bed-parallel slip in th formation of blinf thrust Quaternario. Rend. Ist. Lomb. Acc. Sci. Lett., Sez. A 99, 881–894. faults. J. Struct. Geol. (20), 503–516. Dolan, J.F., Christofferson, S.A., Shaw, J.H., 2003. Recognition of Paleoearthquakes on the Ouchi, S., 1985. Response of alluvial rivers to slow active tectonic movements. Geol. Soc. PuenteHills Blind Thrust Fault, California. Science 300, 115. Am. Bull. 96, 504–515. Erslev, E.A., 1991. Trishear fault-propagation folding. Geology 19, 617–620. Ravaglia, A., Seno, S., Toscani, G., Fantoni, R., 2006. Mesozoic extension controlling the Fantoni, R., Bersezio, R., Forcella, F., 2004. Alpine structure and deformation chronology Southern Alps thrust front geometry under the Po Plain, Italy: Insights from sandbox at the Southern Alps–Po Plain border in Lombardy. Boll. Soc. Geol. Ital.123, 463–476. models. J. Struct. Geol. 28, 2048–2096. Gelati, R., Napolitano, A., Valdisturlo, A., 1988. La gonfolite lombarda; stratigrafiae Roering, J.J., Cooke, M.L., Pollard, D.D., 1997. Why blind thrust faults do not propagate to significato nell' evoluzione del margine Sudalpino Lombardo. Riv. Ital. Paleontol. the Earth's surface: numerical modeling of coseismic deformation associated with Stratigr. 94 (2), 285–332. thrsuts related anticlines. J. Geophys. Res. 102, 11901–11912. Gelati, R., Napolitano, A., Valdisturlo, A., 1991. Results of studies one the Meso-Cenozoic Rossi, M., Rogledi, S., 1988. Relative sea-level changes, local tectonic settings and basin succession in the Monte Olimpino 2 tunnel. The tectono-sedimentary significance margin sedimentation in the interference zone between two orogenic belts: sesmic of the “Gonfolite Lombarda”. Riv. Ital. Paleontol. Stratigr. III–IV (97), 565–598. stratigraphy examples from Padan foreland basin, northern Italy. In: Nemec, W., Steel, R. Geyh, A., Schleicher, H., 1990. Absolute Age Determination. Springer-Verlag, Berlin. (Eds.), Fan Deltas: Sedimentology and Tectonic Settings. Blackie & Sons, pp. 368–384. Gianotti, R., Perotti, C.R., 1987. Introduzione alla tetonica e all'evoluzione strutturale Scardia, G., Muttoni, G., Sciunnach, D., 2006. Subsurface magnetostratigraphy of delle Alpi Lariane. Mem. Soc. Geol. Ital. (32), 67–99. Pleistocene sediments from the Po Plain (Italy): constraints on rates of sedimentation Guerrieri, L., & Vittori, E.(Editors), 2007. Intensity Scale ESI 2007. Mem. Descr. Carta and rock uplift. GSA Bull. 118 (11/12), 1299–1312. Geologica d'Italia, 74, Servizio Geologico d'Italia, Dipartimento di Defesa del Suolo, Sciunnach, D., Tremolada, F., 2004. The Lombardian Gonfolite Group in the central APAT, Rome, Italy. Brianza (Como and Milnao Provinces, Italy): calcarous nannofossil biostratigraphy Guidoboni, E., 1986. The earthquake of December 25th, 1222: analysis of a myth. Geol. and sedimentary record of neo-alpine tectonics. Eclogae Geol. Helv. 1 (97), 119–131. Appl. Idrogeol. 21, 413–424. Serpelloni, E., Anzidei, M., Baldi, P., Casula, G., Galvani, A., 2005. Crustal velocity and Guidoboni, E., Comastri, A. (editors), 2005. Catalogue of earthquakes and tsunamis in strain rate fields in Italy and surrounding regions; new results from analysis of the Mediterrean area from the 11th to the 15th century. vol. II, INGV Istituto permanent and non-permanent GPS networks. Gephys. J. Int. 161 (3), 861–880. Nazionale di Geofisica e Vulcanologia & SGA Storia Geofisica e Ambiente S.r.l., Serva, L.,1990. Il ruolo delle Scienze della terra nelle analisi di sicurezza di un sito per alcune Bologna, 950 p. tipologie di impianti industriali; il terremoto di riferimento per il sito di Viadana (MN). Hardy, S., Ford, M., 1997. Numerical modeling of trishear fault-propagation folding and Boll. Soc. Geol. Ital. 109, 375–411. associated growth strata. Tectonics 16, 841–854. Serva, L., 1994. Ground effects in the intensity scales. Terra Nova 6, 414,416. IAEA, 2002. Evaluation of Seismic Hazards for Nuclear Power Plants. IAEA Safety Shaw, J.H., Suppe, J., 1994. Active faulting and growth folding in the eastern Santa Standard Series, Vienna. No. NS-G-3.3. Barbara channel, California. Bull. Geol. Soc. Am. 106, 607–626. Ishiyama, T., Mueller, K., Togo, M., Okada, A., Takemura, K., 2004. Geomorphology, Shaw, J.H., Suppe, J., 1996. Earthquake hazards of active blind thrust faults under the kinematic history, and earthquake bahavior of the active Kuwana wedge thrust Central Los Angels basin, California. J. Geophys. Res. (101), 8623–8642. anticline, central Japan. J. Geophys. Res. (109), B12408. Shaw, J.H., Plesch, A., Dolan, J.F., Pratt, T.L., Fiore, P., 2002. Puente Hills blind-thrust Lai, K., Chen, Y., Hung, J., Suppe, J., Yue, L., Ya-Wen Chen, Y., 2006. Surface deformation systems, Los Angeles, California. Bull. Seismol. Soc. Am. (92), 2946–2960. related to kink-folding above an active fault: evidence from geomorphic features Sileo, G., Giardina, F., Livio, F., Michetti, A.M., Mueller, K., Vittori, E., 2007. Remarks on the and co-seismic slips. Quat. Int. 147, 44–54. Quaternary tectonics of the Insubria Region (Lombardia, NW Italy, and Ticino, SE Leon, L.A., Christofferson, S.A., Dolan, J.F., Shaw, J.H., Pratt, T.L., 2007. Earthquake-by- Switzerland). It. J. Geosc. Boll. Soc. Geol. It. 126 (2), 411–425. earthquake fold growth above the Puente Hills blind thrust fault, Los Angeles, Slemmons, D.B., DePolo, C.M., 1986. Evaluation of active faulting and associated hazard. California: implications for fold kinematics and seismic hazard. J. Geophys. Res. 112, In: Wallace, R.E. (Ed.), Active Tectonics. National Academy Press, Washington D.C., B03S03. doi:10.1029/2006JB004461. pp. 45–62. Lin, J., Stein, R.S., 2006. Seismic constraints and Coulomb stress changes of a blind thrust Streig, A.R., Rubin, C.M., Chen, W.S., Chen, Y.G., Lee, L.S., Thompson, S.C., Madden, C., Lu, S.T., fault system. 1: Coalinga and Kettleman Hills, California, USGS Open-File Report 2007. Evidence for prehistoric coseismic folding along the Tsaotun segment of the 2006-1149, 17 P. Chelungpu fault near Nan-Tou, Taiwan. J. Geophys. Res. 112, B03S06. doi:10.1029/ Livio, F., Michetti, A.M., Sileo, G., Carcano, C., Mueller, K., Rogledi, S., Serva, L., Vittori, E., 2006JB004493. Berlusconi, A., 2009. Quaternary capable folds and seismic hazard in Lombardia Suppe, J., 1983. Geometry and kinematics of fault bend folding. Am. J. Sci. 283, 684–721. (Northern Italy): the Castenedolo structure near Brescia. It. Journ. Geosc. Boll. Soc. Suppe, J., Medwedeff, D.A., 1990. Geometry and kinematics of fault-propagation folding. Geol. It. 128, 191–200. Eclogae Geol. Helv. 83, 409–454. Magri, G., Molin, D., 1986. I terremoti del 3 Gennaio 1117 e del 25 Dicembre 1222. ENEA- Suppe, J., Chou, G.T., Hook, S.C.,1992. Rates of folding anf faulting determined from growth RTI-PAS-ASPGEOL-LO, 2. strata. In: McClay, K. (Ed.), Thrust Tectonics. Chapman & Hall, London, pp. 105–120. Marchetti, M., 1996. Variazioni idrodinamiche dei corsi d'acqua della Pianura Padana Suppe, J., Sabat, F., Munoz, A., Poblet, J., Roca, E., Verges, J., 1997. Bed-by-bed growth by centrale connesse con la deglaciazione. Il Quaternario 9, 465–472. kink band migration: Sant Llorenc de Morunys, Eastern Pyrenees. J. Struct. Geol. 19, Masaferro, J.L., Bulnes, M., Poblet, J., Eberli, G.P., 2002. Episodic folding inferred from 443–461. syntectonic carbonate sedimentation: the Santaren anticline, Bahamas foreland. Suppe, J., Connors, C., Zhang, Y., 2004. Shear fault-bend folding. In K. Mc Clay, Thrust Sediment. Geol. 146, 11–24. tectonics and hydrocarbon systems. AAPG Mem. 82, 303–323. McClay, K., 1992. Glossary of thrust tectonics terms.» In Thrust tectonics, di K. McClay, Tavani, S., Storti, F., Salvini, F., 2005. Rounding hinges to fault.bend folding: geometric pp.419–433. Chapman & Hall, London. and kinematic implications. J. Struct. Geol. 27, 3–22.Vittori, E., 2004. The map of Medwedeff, D.A.,1992. Geometry and kinematics of an active, laterally propagating wedge capable faults of Italy. Int. Geol. Cong. Florence vol. 32, 20–28 August, 2004. Firenze, thrust, Wheeler Ridge, California. In: Fisher, S., Mitra, G.W. (Eds.), Structural Geology of 2004. Fold and Thrust Belts. Jhon Hopkins University Press, Baltimore, Md., pp. 3–28. Vittori, E., 2004. The map of capable faults of Italy [Atti di convegno] // 32th Int. Geol. Medwedeff, D.A., Suppe, J., 1997. Multibend fault-propagation folding. J. Struct. Geol. 19, Cong. Florence, 20–28 August, 2004. 279–292. Wells, D.L., Coppersmith, K.J., 1994. New empirical relationship among Magnitude, Meletti, C., Galadini, F., Valensise, G., Stucchi, M., Basili, R., Barba, S., Vannucci, G., Boschi, E., rupture length, rupture width, rupture area and surface displacement. Bull. Seismol. 2008. A seismic source zone model for the seismic hazard assessment of the Italian Soc. Am. 84, 974–1002. territory. Tectonophysics 450 (1–4), 85–108 1. Winterer, E.L., Bosellini, A., 1981. Subsidence and sedimentation on a Jurassic passive Michetti, A.M., Audemard, F.A., Marco, S., 2005. Future trends in paleoseismology: continental margin (Southern Alps, Italy). A.A.P.G. Bull. 65, 394–421. Integrated study of the seismic landscape as a vital tool in seismic hazard analysis. Yeats, R.S., 1986. Active faults related to folding. In: Wallace, R.E. (Ed.), Active Tectonics. Tectonophysics 408, 3–21. National Academy Press, Washington, pp. 63–79. Mirzaei, N., Gao, M., Chen, Y., 1999. Delineation of potential seismic sources for seismic zoning of Iran. J. Seismol. (32), 17–39.