ELSEVIER Tectonophysics 266 (1996) 405-424 TECTONOPHYSICS Determination of present-day stress tensor and neotectonic interval in the Spanish Central System and Madrid Basin, central Spain G. D e V i c e n t e ' , J .L . G i n e r a, A . M u f i o z - M a r t f n a, J .M. G o n z f i l e z - C a s a d o b, R. L i n d o c a Departamento de Geodindmica, E Ciencias Geol6gicas, Univ. Complutense, 28040, Madrid, Spain b Departamento de Geologia y Geoqufmica, E Ciencias, Univ. Aut6noma, 28049, Madrid, Spain c Institut de Physique du Globe, Strasbourg, France Received 27 April 1995; accepted 20 June 1996 Abstract A brittle deformation tectonic analysis was performed in central Spain (Spanish Central System and Madrid Basin) in order to decipher and understand the deformation processes that take place in a typical intracontinental zone. 1174 fault slickensides obtained in materials with ages between Late Cretaceous and Quaternary have been analyzed by means of fault population analysis methods to reconstruct paleostress tensors. Nine earthquake focal mechanisms have been determined, with magnitudes ranging between 3 and 4.1. With regard to regional structural features and sedimentary record data, the characteristics of present-day and neotectonic stress fields have been figured out, which determine the neotectonic period for this region. Thus, we have established that the intraplate zone represented by central Spain has been subjected to a stress field from the Middle Miocene until the present-day with a largest horizontal shortening direction (SHMAX) located between N130E and N160E. Finally, three paleostress maps with the main active structures are presented for: (a) Middle Miocene to Late Miocene, the period when the Spanish Central System was mainly formed, (b) Late Miocene to Quaternary, and (c) the present-day stress field, deduced from earthquake focal mechanisms. Keywords: neotectonics; seismotectonic; fault; focal mechanism; active stresses; stress evolution; Central Spain 1. In troduct ion The concept of 'present-day stress tensor' should be related directly with the term 'active tectonics' , that is, the study of the earth deformation processes that are occurring at present, or could occur in the near future in a specific region (Wallace, 1986). Depending on the author, the term 'neotectonics' could refer to the deformation processes that took * Corresponding author. Fax: +34 1 394-4883. place from the end of the Tertiary (Neogene) or from the first half of the Quaternary to the present. However, we consider that it is more accurate to relate the neotectonic period with the time in which the present stress tensor has been active (Blenkisop, 1982). Thus, the neotectonic deformation phase be- gins at a different time in different places, depending on the tectonic and dynamic regime of the area in question (M6rner, 1990). Therefore, we consider 'neotectonics ' as the com- bined study of active processes and structures, and 0040-1951/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved. PH S 0 0 4 0 - 1 95 1 ( 9 6 ) 0 0 2 0 0 - 4 406 G. De Vicente et al./Tectonophysics 266 (1996) 405 424 under the present-day stress tensor tectonic defor- mations produced by a paleostress tensor of similar characteristics to the current stress tensor. On the one hand, in order to determine the pre- sent-day stress tensor, we could use: (a) geological methods (e.g. fault population analysis), (b) instru- mental methods (e.g. 'in-situ' stress measurements) or (c) geophysical methods (e.g. analysis of earth- quake focal mechanisms). On the other hand, only geological methods can be used when establishing paleostresses. In that case it has been demonstrated that fault population analysis methods give equiva- lent results to those of geophysical methods (Zoback, 1992). With these considerations, it is necessary to mea- sure several fault sets in materials of different ages in order to establish the characteristics of the current and neotectonic stress tensors, as well as the neo- tectonic period. A comparison of these results with those derived from the earthquake focal mechanism analysis is also required. 1.1. Geological setting The Iberian Peninsula shows recent deformation structures in many places (mainly at its southern and eastern limits). The relative motion between the European and African plates has been convergent throughout the Neogene (Dewey et al., 1989). In the Iberian plate, during the N-S to N N W - S S E conver- gence and slip between these plates, stresses from the Pyrenean and Betic borders were transmitted into the Iberian Meseta (a part of the Iberian Massif outcropping in central Iberia). This stress field gave rise to a set of different intraplate chains and basins. Their structural characteristics are influenced by pre- vious crustal structures and discontinuities, and by the thickness of the sedimentary cover. The Iberian Range (Celtiberian Chain) (Fig. 1) separates two different domains: in the eastern part, outcrops of Mesozoic to Paleogene sedimentary cover dominate, while in the western zone, metamor- phic and plutonic Hercynian basement rocks cover almost the whole area. Neogene and Quaternary tectonic activity pro- duced a thick-skin style deformation, within the Iberian Meseta. Where crustal faults were reacti- vated with reverse movements, these faults generated a group of upward and downward moving base- ment blocks that produced Tertiary basins and in- traplate chains. The orientation and distribution of these structures in Iberia follow a marked regular pattern. These structures are the following (Fig. 1), from north to south. Cantabrian Mountains. This chain is the west- ern continuation of the Pyrenees with topographic heights rising up to 2700 m. These mountains have an important topographic relief given their vicinity to the sea. The boundary with the Duero Tertiary Basin consists of a reverse E - W trend fault that dips towards the north (Alonso et al., 1996). Duero Basin. It is filled with Tertiary sediments, that thicken to more than 3000 m at its eastern part (Querol, 1989). Spanish Central System. This range has a N60E trend, and altitudes close to 2500 m. Its boundaries with the surrounding basins consist of two main faults with opposite dips, drawing a large crustal 'pop-up' (Vegas et al., 1990; De Vicente et al., 1994). Madrid Basin. It is filled with fluvial and lacus- trine Tertiary deposits. Next to the northern margin of this basin, the thickness of sediments reaches 3500 m. Sediments are thrust by the basement (Megfas et al., 1983). Toledo Mountains. These mountains are lower than the Central System (1500 m), and have an E - W trend. They thrust northwards over Tertiary sediments of Madrid Basin. Guadiana Depression. This structure does not have a generalized sedimentary cover. Its principal feature is the presence of important Middle Miocene and Pliocene volcanic activity (Ancoechea, 1982). Sierra Morena Mountains. This mountainous range is south bounded by the Tertiary Guadalquivir Fig. 1. Geographical and geological setting of the area of this study. SCS = Spanish Central System: IR = Iberian Range; AR = Altomira Range; TM = Toledo Mountains; GD = Guadiana Depression; CM = Cantabrian Mountain; 1 = Madrid Basin: 2 = Duero Basin; 3 = Ebro Basin; 4 = Guadalquivir Basin. Below, position of microstructural analysis sites. Paleozoic, Mesozoic and Paleogene outcrops are showed in grey, Neogene in white. Line A-B corresponds with the geological cross-section shown in Fig. 2. The coordinates correspond to UTM in kin. G. De Vicente et al. ITectonophysics 266 (1996) 405-424 407 :;.i.i' !/.'2"~.:2' ~ JMAS~S I / - ~ ' ~ ~ ~ ~ ' ~ - , - ~ I I [I PYRENEES ~{~"~ ~ -~ q ;~-~' ~'~-~ ' . ' . ~ xx ~ ~ ~ xx xx xx ~f [ ~ IBERIAN AND CATALAN RANGES ~ ~ i ~ ! ~ ~ [ ~ CENOZOIC BASINS 4650 6 NUMBER (SEE TABLES) <-~-_~-~ I ] \O OLOO.C c.o -SEOT,O. MEASURE SITE , 46°° I DUERO BASIN 4550 4500 4450 ! "~ <>2 ~,~i , SEGOV,A ( , \ ~ "3 9 4 12 - - AVlLA . ¢~ <<>4~ i MADRID BASIN \ 4400 TOLEDO ~B~ 350 400 450 560 5' 9 O3 \ = 550 408 G. De Vicente et al . / Tectonophysics 266 (1990) 405-424 Basin, and the limit between these units has a N60E trend and consists of surficial normal faults, although they have been explained as a large crustal flexure (Van de Beek and Cloetingh, 1992). All these basins and intraplate chains show im- portant tectonic activity during the Neogene, and at present they exhibit a moderate seismic activity. Our study focuses on the Central System, the Madrid Basin and its surroundings. Both features could be considered, in general, as good represen- tatives of Neogene and recent deformation of the Iberian Hercynian basement that outcrops in this part of Spain. The Madrid Basin has a well studied Neogene stratigraphy (Calvo et al., 1989, 1996), that allows us to apply paleostress analysis methods over a rather continuous sedimentary record. Geophysical information about this region, such as gravimetry and seismic reflexion profiles (Surifiach and Vegas, 1988; Querol, 1989; Babfn et al., 1993; Prrez Agudo, 1995), is abundant, therefore the sub-surficial char- acteristics of the Madrid Basin can be determined. There is also accurate information of the date and rate of the Central System uplift, from fission-track studies (Sell et al., 1995). Two research approaches have been employed in order to establish the present and neotectonic stress tensors. On the one hand, fault populations in Meso- zoic and Paleogene materials have been measured at the basin margins, as well as in Neogene and Qua- ternary materials of the basin centre. On the other hand, a set of 9 focal mechanisms of earthquakes with magnitudes between 3 and 4.1 has been studied. Thus, the objective of this study is the estab- lishment of the present-day and neotectonics stress tensor characteristics, as well as the neotectonic pe- riod in an intraplate zone such as the Spanish Central System and the Madrid Basin. 2. Paleostress analysis methods Three independent analysis procedures have been used to infer the current and neotectonics stress tensors. (a) Fault population analyses (FPA) have been used to determine paleostress fields. These proce- dures allow us to establish a reduced stress tensor that activates the fault populations measured in the field. All methods are based on brittle deformation conditions, without volume change. Most stress inversion methods are based on BoWs equation (Bott, 1959): n 2 t a n 0 = /~m[m - ( 1 - n 2 ) R '] (1) This equation relates the shape (R' = (cy=- c ~ ) / ( a ~ . - oi~)) and orientation of the stress ten- sor principal axes (o-y, cr X and az) to the fault plane orientation (1, m, n are the fault directional cosines) and to the slip line (0, the theoretical striation pitch on the fault plane). The stress inversion method (SIM) proposed by Reches (1987) and Reches et al. (1992) has been used here because it assumes the Navier-Coulomb rupture principle as well as Bott's equation, so that cohesion, friction coefficient and other physical vari- ables are taken into account. (b) An analysis procedure, similar to the previous one, is performed in order to calculate the strain ellipsoid from a fault population. In both cases, par- allelism between strain and stress principal axes is assumed. A direct method (De Vicente, 1988; Capote et al., 1991) based on the slip model (SM) (Reches, 1983) was developed, which relates the strain ellip- soid shape (K = e l / e 3 , ei ellipsoid principal axes) with the fault orientation (D, dip) and the striae defining the slip line (B = sin2D • cos2P, being P the slip line's pitch and D, the fault dip) by means of the equation: sin 2 D cos 2 B f = (2) I - sin2D cos2B (c) The method of Rivera and Cistemas (1990) has been applied to determine the regional stress tensor from earthquake focal mechanisms. This pro- cedure uses Bott's equation combined with the radia- tion pattern function for P waves first arrivals. At the same time, it seeks the maximum of the likelihood function, which depends on the radiation function and on the probability of obtaining either a positive or a negative polarity. Since this is a trial-and-error procedure, it makes use of a set of stress tensors and focal mechanisms as starting conditions. Its results are similar and comparable to those of the classi- cal FPA methods (e.g., Etchecopar, 1984; Angelier, 1990). It is remarkable that this method has been used in the determination of the current stress field G. De Vicente et al. / Tectonophysics 266 (1996) 405-424 409 of France (Delouis et al., 1993) with satisfactory results. Once the stress tensor for each site of measure- ment is obtained, the possible deformation age limits are assigned depending on the age of the materials where fault striations were measured, their struc- tural situation, and their position in the sedimentary record of the Madrid Basin. 3. Structure of the Madrid Basin margins The Tertiary evolution of the Madrid Basin and its margins is essential to evaluate and correlate the paleostress data presented further on. Hence, we will start with an accurate description of the structure of the Madrid Basin margins. The Madrid Basin has a triangular geometry. The southern boundary is formed by the Toledo Moun- tains, while the southwestern prolongation of this limit is less well-defined and onlaps with the Gua- diana Depression; the NW boundary is the Central System, and consists of a sharp north-dipping thrust, while the eastern limit with the Iberian Range is rather less defined. Moreover, in the eastern zone, the N-S-trending Altomira Range (Fig. 1), separates the Madrid Basin from the Loranca Basin. 3.1. Spanish Central System This is a straight mountain range, elongated along a principal NE-SW trend. This direction corresponds to several reverse faults that uplifted the basement, causing a double vergence chain with a thick skin tectonic style (Fig. 2). The basement appears clearly involved in this deformation, and there are no observ- able detachment levels in the cover. The shortening related to the NE-SW- and E-W-thrusting is later- ally resolved by strike-slip faults (transfer faults), with N140E (dextral movement) and N10E (sinis- tral movement) trends (De Vicente and Gonz~lez- Casado, 1991) (Fig. 3B). The largest horizontal shortening related with the fault displacement has a N155E trend (Capote et al., 1990). The restored section shows a crustal 'pop-up' structure for the Central System, with an associated shortening of 22% (Fig. 2). In more detail, three minor pop-up structures appear at the southern part, which draw a series of folds with different widths; open anticlines separated by tight synclines. Their wave longitude fluctuates between 5 and 10 km. The northernmost zone has a different structure, with a series of im- bricate thrusts from which a 4 km deep NW-verging detachment level can be inferred. This style differ- ence could be a response to different materials in the basement, granites to the south and metamorphic foliated rocks to the north. The boundaries of the Central System are formed by two reverse faults (the north, NBF, and the south, SBF, border faults). However, it can be observed that, while the southern border (SBF) coincides with the topographic limit (STL), the northern border is displaced towards the south with respect to the NBE This points to a different spatial and tem- poral behaviour of the northern and the southern border faults. This fact is also obvious in the Upper Miocene facies distribution: the lacustrine sedimen- tary deposits (UMLDL) appear next to the NBF, while they are located towards the centre of the Madrid Basin at the south (Figs. 2 and 3B). 3.2. Toledo Mountains These form the southern border of the Madrid Basin. In its northern limit the main structures of this mountain range are several E - W thrusts, dipping to the north. The northernmost thrust produces the overlap of basement rocks over Cenozoic sediments belonging to the Madrid Basin. The shortening direc- tion associated with reverse fault movements shows a N-S trend, similar to that of the western end of the Central System, a region with several features similar to this. Nevertheless, due to the minor topo- graphic expression of this mountainous alignment, it is reasonable to infer that: (a) the alpine deformation intensity was lower than in the Central System; and (b) they were structured first. This latter hypothesis cannot be rejected, given the current knowledge of the sedimentary basin-fill of the Madrid Basin in this area (Calvo et al., 1989). In any case, Toledo Mountains do not close the Madrid Basin, since they finish to the east in a set of N140E striking dextral faults, partly sealed by Pliocene and Neogene sediments. These sediments connect the southern part of the Madrid Basin with the Guadiana Depression (Figs. 1 and 3B). 410 G. De ½"cente et al./Tectonophysics 266 (1996) 405 424 UJ OOI 03 I I ~J3AI~J V N n rv i - - - - '~ _0 ,=, HBAI~I £3WVN3H ---~ -z :~rn >- ~ r ~ o d ~ .=_ = [] =. > m ~_ = ~ P ~ r - q . . G. De Vicente et al./Tectonophysics 266 (1996) 405424 411 3.3. Iberian Range (Celtiberian Chain) 3.4. Altomira Range The Iberian Range is the only boundary of the Madrid Basin with a several thousand metres thick Mesozoic cover, that is locally detached. This is a polyphase chain, structured at least in two episodes: one related to the Oligocene aged Pyrenean stress field, (Sim6n-G6mez, 1986), and the other with the NW-SE-shortening direction developed during the Neogene, and probably related to the Betic pale- ostress field (De De Vicente, 1988). This latter de- formation is more intense in the westernmost sector and it is related to the Central System evolution. In fact, this sector of the Iberian Mountain Range constitutes the main lateral limit of the Central Sys- tem N60E thrusts, given rise to a transpressive zone, with numerous N140E-trending dextral strike-slip faults. This deformation yielded lower topographical heights than the Central System. The microstructural analysis records a last strong extensional process perpendicular to the chain. These stresses reactivated previous N140E faults as normal ones (Sim6n-G6mez, 1986). The important topo- graphic gradients that appear throughout the chain are due to the rapid erosional processes of the fluvial network. The eastern part of Iberian Peninsula has been subjected to stretching due to rifting process, which began in the Middle Miocene (Sim6n-G6mez, 1986; Garcfa-Cuevas et al., 1995). This type of deformation is not recorded in the Madrid Basin, at least until the Late Miocene, although it is more likely that it has a Pliocene age. At the end of the Pliocene, a very characteristic erosional surface was developed (the 'Paramo' surface), that is deformed nowadays, forming a large flexure with an axis parallel to the Iberian Range western sector (NW-SE). This area shows the major erosional processes in the fluvial network. In the Madrid Basin, rivers run southwest, showing a recent tilt in this direction that corresponds to the Iberian flexure's southwest flank. This Pliocene uplifting has been related with the late evolution of the Valencia Trough opening pro- cess (Janssen et al., 1993), although the perpendic- ular disposition between the described structure and the rift (NE-SW) could suggest that the process was not uniaxial. This range is a straight alpine N-S-trending fold- and-thrust-belt that separates the Madrid and Lo- ranca Tertiary basins. This structure was active from the Late Oligocene to the Early Miocene (Dfaz- Molina et al., 1989; Dfaz-Molina and Tortosa, 1996). The study of the sedimentary record and related structures indicates an important decrease in the tec- tonic activity during the Middle to Late Miocene: slightly deformed sedimentary deposits of this age onlap the reliefs of the Altomira Range in some places (Rodrfguez-Aranda, 1995), suggesting a tec- tonically passive phase during this period. The Mesozoic and Cenozoic cover is detached at Triassic Keuper gypsum beds (levels of larger ductility contrast), as can be observed in seismic reflection profiles (Querol, 1989). The western limit of these facies seems to control the formation of these detachments (Van Wees et al., 1995). How- ever, the presence of important basement faults also influences the localization of the thrust and fault propagation folds (Perucha et al., 1995). These base- ment faults also control the lateral termination of the thrust, defining different vergence zones. The maximum horizontal shortening direction as- sociated with the formation of this mountain range is N100E. The genesis of these mountains has been explained as the result of a superposition of two re- gional stress directions, the Pyrenean and the Betic paleostress fields (Mufioz-Martfn et al., 1994), which could have produced a westward escape of the A1- tomira Range, in a general N-S compressive stress field. 4. Establishment of Neogene and Quaternary stress fields In this study 74 measurement sites have been analyzed, with a total of 1174 studied faults (Ta- ble 1), and an average of 16 faults per site. This is a reasonable number for recent materials. A largest SE-trending horizontal shortening (Sn~AX) appears as the main solution at all sites. With regard to the age of the materials and to their structural position, the sites can be divided in two broad groups. The first group consists of the data measured in rocks older than Late Miocene at the Madrid Basin 412 G. De Vicente et al. / Tectonophysics 266 (1996) 405-424 g g g g g ( (~ / / / /1 / / L, 8 o ~ I - E E g g m g g g g g g © .= II =o o . q g E . = = Z ~ z ~ g ~ g ~ - = G. De Vicente et al. / Tectonophysics 266 (1996) 405-424 413 margins (Table 1, 1 to 61 sites). There are some sites in this group that are measured in main faults that, although it is not possible to exclude that these fault populations are not too large, they give very relevant information. Even though it is not possible to exclude a younger fault movement age (Late Miocene or Quaternary), we consider that the main tectonic activity event in the basin margins (Middle Miocene, see next section), must be reflected in the main faults located along these borders. The second group is formed by fault populations measured in the sedimentary basin-fill of the Tajo Basin, in Late Miocene and younger rocks (Table 2, 62 to 74 sites. Station 69 was obtained in Middle Miocene sediments, but due to their structural po- sition it has been considered as belonging to this group). Taking the Upper Miocene and Quaternary outcrop spacial distribution into account, it is diffi- cult to obtain data near the Madrid Basin margins. Anyhow, this group of solutions records the most recent stress field. (1) Middle to Upper Miocene paleostress field. Faults were measured mainly not only in Upper Cre- taceous materials, but also on Paleogene formations (5 sites) and Paleozoic rocks that overlap some Up- per Cretaceous and Paleogene rocks (zone of Toledo, 2 sites), at the Madrid basin margins. In general, all the sites where faults have been analyzed show a N155E largest horizontal shorten- ing, but the easternmost sites (Altomira and Iberian ranges) show other fault population results. This indicates E-W and NNE-SSW largest horizontal shortening directions. The E-W trend is related to the emplacement of the Altomira Range during the Late Oligocene to the Early Miocene (Dfaz- Molina et al., 1989; Dfaz-Molina and Tortosa, 1996; Rodrfguez-Aranda, 1995). Compression patterns ap- pear in the northeastern quadrant of the Iberian Range, that could be related to the paleostress field generated by the Paleogene Pyrenean forma- tion (Mufioz-Martfn et al., 1994). Nevertheless, sites considered in the Central System have only one so- lution, with a largest horizontal shortening towards N155E. If we analyze the stress tensor characteristics related to each fault-slip station, it is possible to find all types of solutions: compressive, extensive and strike-slip stress regimes. However, SHMAX shows always a SE trend (except in the Toledo Mountains, with a N-S trend). The most comprehensive solution, for this group of sites, suggests a strike-slip stress tensor (R = 0.3) (Fig. 5). Fig. 3A shows Sm~ax trends for each station. Re- sults obtained with the 'slip model' (Reches, 1983) and stress inversion method (Reches et al., 1992) are similar. Construction of a SHMAX trajectories map has been carried out by means of an interpolation program, the LISSAGE program (Lee and Angelier, 1994). A regular SHMAX trajectories map is obtained, with a clockwise rotation towards the west due to Toledo Mountains paleostress data. We consider this paleostress field as the cause of the Central System and western branch of the Iberian Range deforma- tion. Taking this paleostress field and the sedimen- tary basin-fill characteristics in the Madrid Basin into account, Middle Miocene active structures are shown in Fig. 3B. (2) Upper Miocene to Quaternary paleostress field. 13 fault-slip sites in the central part of the Madrid Basin have extensional and strike-slip solu- tions (sites 63, 67, 68, 69 and 74) (Fig. 5, Table 2). Fig. 4A shows the SHMAX trajectories map. There is a group of NE-SW-trending normal faults in this zone, with a SE-trending StalIN. Field relationships between this normal fault set and the previously described one, suggest a synchronous movement. Therefore, it is impossible to define two different stress fields. Paleostress data sites showing a SE extension have not been considered in this analy- sis, although they are as abundant as the SE-NW compression sites. The amount of tectonic deformation related to the uplifting of the Central System is smaller during this period than during the Middle Miocene, but this has to be confirmed by other techniques. Sedimentary and structural data show that the Toledo Mountains have not been very active during this period. Fig. 5B shows active structures during the Late Miocene and the Quaternary. 5. Relationship between tectonics and sedimentary infilling of the Madrid Basin The thickness of the Tertiary sediments in the Madrid Basin ranges between 2000 and 3500 m. The Neogene sedimentary sequence overlies dis- 414 Table 1 Strain ellipsoids and Reches et al., 1992) G. De Vicente et al./Tectonophysics 266 (19961 405 424 stress tensors determined by the 'slip model' (Reches, 1983) and by the stress inversion method (Reches, 1987; Stat. N.E Slip model Stress inversion Age Dey S.D. K r (51 a2 0-3 ERR R E.E # 1 31 146 2 20 145 3 24 167 4 23 1 5 27 178 6 12 153 7 10 174 8 7 169 9 10 153 10 12 153 11 9 166 12 15 145 13 8 158 14 19 168 15 14 152 16 25 154 17 8 146 t8 10 152 19 19 163 20 21 162 21 16 151 22 19 160 23 10 155 24 12 169 25 10 138 26 9 152 27 15 171 28 20 142 29 42 t71 30 10 150 31 10 145 32 8 152 33 8 145 34 10 147 35 18 166 36 14 157 37 10 158 38 23 143 39 16 166 40 10 149 4l 17 165 42 20 152 43 13 151 44 9 152 45 23 154 46 6 154 47 6 153 48 11 158 49 50 158 50 14 159 7 1.145 00/320 32/230 57/050 i0 0.09 24 0.6 8 - 1.384 03/316 76/060 13•225 13 0.16 13 1.0 11 0.093 78/347 I 1 /155 02/245 27 0.11 18 1.0 9 13.55 04/002 84/217 03/092 5 0.47 17 1.1 9 - 1.207 00/000 70/092 19/270 8 0.10 18 0.4 9 1.804 60/346 27/145 8/240 9 0.61 8 0.9 11 - 1.292 05/357 34/264 54/095 l 0 0.05 9 0.5 10 - 1.008 03/165 13/075 76/269 6 0.31 7 0.8 6 24.84 35/333 54/144 04/240 1 0.30 8 0.7 11 -1.894 31/155 58/337 00/246 13 0.37 12 0.4 9 16.12 79/298 06/173 08/082 2 0.28 5 0.2 7 -1.014 04/149 02/240 84/000 22 0.58 14 1.0 9 -1.022 14/137 30/038 55/249 6 0.18 7 0.5 9 3.773 01/335 87/114 02/265 9 0.37 12 0.3 9 1.276 04/324 51/061 37/230 9 0.05 9 0.5 7 1.925 64/134 24/334 07/240 4 0.56 18 0.6 6 5.817 12/323 77/169 03/054 2 0.39 7 0.9 7 0.487 07/334 77/217 09/075 12 0.52 8 0.4 5 - 1.035 02/157 07/066 82/268 8 0.3 19 0.6 9 1.148 10/163 70/285 16/070 16 0.21 18 0.6 8 - 1.023 09/142 21/048 66/255 9 0.15 16 0.9 9 18.79 88/185 01/343 00/073 1 0.80 16 0.8 17 0.323 83/305 06/137 01/047 7 0.34 10 0.6 12 36.75 83/009 05/178 02/168 6 0.86 I 1 0.8 5 1.01 15/141 74/328 01/232 9 0.25 8 0.3 7 1.418 00/331 63/240 26/061 5 0.18 6 0.5 8 1.22 12/344 63/228 22/080 8 0.31 I 1 0.3 8 - 1.705 05/322 26/229 62/062 5 0.03 12 0.7 7 -1.314 1 7 / 1 7 1 65/303 17/075 15 0.20 39 0.4 4 0.475 13/332 76/152 00/242 17 0.58 8 0.5 3 - 1.054 05/145 33/052 56/244 I 0 0.13 7 1.1 7 3 . 0 1 28/324 60/164 08/058 5 0.43 6 0.4 10 - I .445 04/128 58/020 31/220 15 0.31 5 0.2 6 0.92 17/335 68/118 11/241 8 0.30 10 0.4 5 0.241 08/169 71/285 16/077 9 0.19 12 0.4 7 0.205 00/336 85/078 04/246 23 0.75 11 0.3 4 0.329 30/343 56/134 13/245 2 0.35 8 0.5 9 0.336 86/347 03/142 01/232 11 0.34 18 0.8 7 3.69 15/159 65/033 19/254 9 0.36 16 0.4 5 -1.32 19/140 61/271 20/043 8 /).17 10 0.6 9 1 .27 08/161 69/272 19/068 7 0.31 12 0.4 6 0.673 00•335 89/130 00/245 13 0.38 18 0.4 4 1.24 08/332 81/134 02/242 11 0.41 12 0.4 4 1 . 4 9 85/357 03/147 02/237 4 0.59 9 0.6 5 0.525 74/146 15/335 02/244 11 0.43 18 0.8 4 3.998 17/326 72/356 02/057 6 0.74 12 0.8 2 I. 19 59/164 30/335 03/068 15 0.34 6 0.6 5 0.472 05/154 72/046 16/245 5 0.24 11 0.4 8 1.234 08/I 59 70/275 17/066 7 0.13 33 0.5 5 10.81 86/033 01/155 02/245 19 0.82 14 0.6 Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Paleogene Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Jurassic Jurassic Jurassic Upper Cretaceous Upper Cretaceous Upper Cretaceous Paleogene Upper Cretaceous Paleozoic Paleozoic Upper Cretaceous Paleogene Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Paleogene Lower Miocene Middle Miocene Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous Upper Cretaceous G. De Vicente et al./Tectonophysics 266 (1996) 405-424 415 Table 1 (continued) Stat. N.F. Slip model Stress inversion Age Dey S.D. K' a l 0"2 a3 ERR R E.E # 51 15 156 5 0.355 76/144 13/334 02/243 5 0.56 11 0.6 Upper Cretaceous 52 9 154 6 0.251 78/143 10/332 01/241 4 0.68 7 0.8 Upper Cretaceous 53 31 154 6 -1.36 00/334 68/243 21/064 12 0.10 24 0.7 Upper Cretaceous 54 17 159 5 0.535 12/343 77/162 00/253 17 0.45 15 0.7 Upper Cretaceous 55 17 149 6 0.8 15/151 71/008 10/244 21 0.30 15 0.6 Upper Cretaceous 56 13 162 4 1 .61 02/160 83/273 06/070 5 0.48 10 0.6 Upper Cretaceous 57 8 161 3 4.67 02/164 79/061 09/255 4 0.17 8 0.3 Upper Cretaceous 58 24 151 5 4.7 04/144 82/015 05/234 6 0.45 22 0.8 Paleogene 59 19 2 11 0.632 06/347 14/079 73/232 18 0.55 16 0.2 Upper Cretaceous 60 16 1 9 0.732 64/322 20/182 15/086 15 0.75 10 0.4 Paleozoic 61 11 35 8 0.688 63/195 22/343 12/078 10 0.12 6 1.3 Paleozoic Stat.: number of measure sites. N.F.: number of faults. Dey: largest horizontal shortening direction. K': ey/ez. ERR: error in degrees in the calculation of SnMAX. R = (a2 -- a3)/(cq -- a3). E.F: number of faults explained by the stress tensor./z: calculated friction coefficient. c o m f o r m a b l y the C r e t a c e o u s a n d / o r P a l e o g e n e for- m a t i o n s at the b a s i n m a r g i n s , w h i l e th i s b o u n d a r y s e e m s to be c o n f o r m a b l e in the cen t r a l par t o f the bas in . D u r i n g m o s t o f the M i o c e n e , the b a s i n was o c c u p i e d b y lakes and a s s o c i a t e d p e r i p h e r a l a l lu- v ia l sy s t ems , f o r m i n g a c en t r i pe t a l d r a i n a g e sys t em. Thus , a c o n c e n t r i c fac ies d i s t r i bu t i on is o b s e r v e d ( C a l v o et al., 1989, 1996). T h r e e m a j o r t e c t o n o - s e d i m e n t a r y un i t s sepa- r a t ed b y u n c o n f o r m i t i e s h a v e b e e n de f i ned for the M i o c e n e : Lower , I n t e r m e d i a t e a n d U p p e r M i o c e n e Uni t s , ( J u n c o and Ca lvo , 1983). A n a l y z i n g the na- ture, e x t e n s i o n and d i s t r i bu t i on o f the a l luv ia l fans a n d the f luv ia l d i s t r i bu t a ry s y s t e m s b e l o n g i n g to the L o w e r and I n t e r m e d i a t e Uni t s , a d i f f e ren t t e c ton i c ac t iv i ty is o b s e r v e d at the b a s i n m a r g i n s d u r i n g t he i r d e p o s i t i o n ( A l o n s o - Z a r z a et al., 1993). T h e a l luv ia l fans o f the L o w e r U n i t are r e l a t ed to the f o r m a t i o n o f the A l t o m i r a R a n g e d u r i n g the L o w e r M i o c e n e a n d are r e p r e s e n t e d b y p r o g r e s s i v e l y f iner and t h i n n e r g r a d e d b e d d i n g c las t ic s e q u e n c e s ( R o d r f g u e z - A r a n d a , 1995). A n oppos i t e s i tua t ion is Table 2 Strain ellipsoids and (1987) and Reches et stress tensors calculated based on the slip model (Reches, 1983) and on the stress inversion method of Reches al. (1992) Stat. N.F. Slip model Stress inversion Age Dey S.D. K' cr I a2 a3 ERR R' E.E /2 62 17 152 5 1.45 73/355 13/152 05/243 5 0.86 12 0.4 Upper Miocene 63 20 166 11 2.48 01/166 81/266 08/076 15 0.38 18 0.3 Upper Miocene 64 4 118 7 0.62 61/315 27/116 07/210 10 0.26 4 0.4 Upper Miocene 65 16 130 10 0.63 76/142 12/304 07/035 25 0.33 8 1.0 Upper Miocene 66 20 132 12 0.61 80/143 08/312 09/044 4 0.4 8 0.4 Upper Miocene 67 22 151 4 - 11.5 08/327 78/104 07/236 2 0.55 19 0.5 Upper Miocene 68 7 144 4 3.37 56/331 33/144 03/236 8 0.05 7 0.4 Quaternary 69 8 145 10 0.445 04/128 58/020 31/220 11 0.31 5 0.2 Middle Miocene 70 13 160 3 -4.637 01/161 88/348 00/251 4 0.23 11 0.7 Quaternary 71 9 129 9 0.095 81/320 08/148 01/058 19 0.19 5 0.3 Quaternary 72 11 140 7 3.72 78/015 06/141 09/232 5 0.2 8 0.6 Upper Miocene 73 17 145 11 0.11 78/046 00/312 11/221 10 0.33 11 0.6 Upper Miocene 74 24 161 10 0.402 20/163 63/300 16/066 15 0.43 12 0.5 Upper Miocene For legend see Table 1. 416 G. De Vicente et al./Tectonophysics 266 (1996) 405-424 -6 .+-A ~ ~ . o o~ _ \ \ ,,o,~ '~'/ # \ \ \ \ \ \ \ \ \ ~<' i ts; <-- p_, e- :_= II g ~4 g © E X Lid ,~ E Z ca ~ e ~ g .~.~ G. De Vicente et al. / Tectonophysics 266 (1996) 405-424 4 1 7 .~ 8 ~ zl ",: I+ + \ - o o / ,,, t ' ~ ~ > . . . . . . . . . . . . . . . % - , + + / ,,, .................... ~; ~ +++++ +++ 8 ° ~d . . . . 03 0 . . . . i----:----;- - : o . . . . , - - - i - - - i - - - ~, ~ Z ~ ~ :::~ ~ >~ . . . . o Z ~ W X W 8 ~ o ' ~ ~ ~ ~ o I:g ~i i i C ~-,+----,---- ,--:! o ,, i J ® . . . . : - - - - - , - - - - 4 ~> . o ¢) b~ Y ~P $ 8 © o o .ff 418 G. De Vicente et al./Tectonophysics 266 (1996) 405-424 noted during the Middle Miocene, when Intermediate Unit lacustrine systems onlap the Altomira Range, suggesting that this border was inactive from a tec- tonic point of view. This unit appears clearly related to the major episode of formation of the Central System, since it represents a main southeastwards prograd- ing alluvial fan system (Calvo et al., 1989). Coarse- grained facies are associated with the Central System south (SBF) and north (NBF) border faults (Figs. 2 and 3B), where they are locally thrust by Cretaceous and Paleozoic rocks. Two minor sequences can be found within this unit, which indicate tectonic ac- tivity of the basin borders: the lower one is related to the uplifted Central System, and the upper one to the Iberian Range (Alonso-Zarza et al., 1990). This indicates a less intense activity of the Central Sys- tem compared to the Iberian Range from the Mid- dle to Late Miocene. It could also suggest a gradual change in the stress regime, shifting to more trans- pressive types associated with the Iberian Range. It seems from the paleostress data that the stress regime changed significantly from the Middle Miocene to the Late Miocene. The two stages have similar SE- trending SHMAX but the average stress regime is strike- slip to compressive for the first stage (transpressive), while it is extensive to strike-slip for the second one (transtensive). Nevertheless, the two groups of data have been obtained in different structural levels. In this way, we cannot assure that stress regime changed between the Middle and Late Miocene. The Upper Miocene Unit consists mainly of flu- vial terrigenous deposits and shallow lacustrine car- bonates. Lacustrine deposits are restricted to the central and eastern parts of the Madrid Basin. Fa- cies distribution is very different to that of the two previous units. It is important to point out that these materials outcrop far away from the south border fault, while they are very close to the north border fault (Figs. 2 and 4B). This could indicate a more important tectonic activity in the south border during the Late Miocene. A succession of uplifts and down- falls appear in the Tertiary sediments from the limit of the lacustrine facies (UMLDL, Fig. 4B) to the southeast, where the Quaternary drainage network settles down. The absence of deep seismic profiles keeps us from establishing this graben system deep structure, although we shall return to this point in view of the result of the seismic analysis. During the Pliocene an erosion-dominated period began, which continues to the present. Materials in this period show a similar distribution to those of the Upper Unit, though there are few materials of this age. Nevertheless, they define a very character- istic erosional surface in this region. At present this surface is structured according to a large-wavelength flexure with an axis parallel to the Iberian Range. The Madrid Basin is located in the southwest flank of this flexure, and is tilted southwest. The Quater- nary drainage network reflects clearly this situation, and forms an exorheic system draining towards the Atlantic. Information related to fission track analysis on some blocks belonging of the inner part of the Cen- tral System indicates an important uplift episode between 7 and 10 Ma (Sell et al., 1995). It could suggest a flexural response to the main tectonic event having occurred before in the Middle Miocene, as can be deduced from flexural analysis to determine the effective elastic thickness (EET) of the area (Van Wees et al., 1995). The Neogene Valencia Trough opening was re- flected clearly in the extensional stress field produced at the eastern part of the Iberian Peninsula (Sim6n- G6mez, 1986). This kind of deformation shifted to the west, and slowed down with time. Due to the Neogene facies distribution, this extensional episode does not appear in the Madrid Basin until, at least, the Late Miocene, and is more obvious in the Pliocene. Therefore, the post-Pliocene block rotation could be related to a generalized uplifting of all the eastern part of the Iberian Peninsula, as a late consequence of the Valencia Trough opening (Janssen et al., 1993). 6. Present-day stress tensor characteristics Nine earthquakes recorded between 1979 and 1994, with magnitudes ranging between 3 and 4.1, and hypocentral depths between 15 and 1 km (Ta- ble 3) have been studied. Seven of them are located inside the Madrid Basin, one (No. 8) is located on the Central System South Border Fault (SBF) and the other (No. 9) towards the easternmost part of the Central System (Fig. 6B). The method of Rivera and Cisternas (1990) has been applied to these earthquakes, with an average of 12 P-wave-first-arrival polarities per earthquake. G. De Vicente et a l . / Tectonophysics 266 (1996) 405-424 419 H r.~ == .,,... H .< --. z~ ,-r, ~z 420 Table 3 Calculated focal mechanisms G. De Vicente el al./Tectonophysics 266 (1996) 405-424 No. Date T i m e Longitude (o) Latitude (o) Depth RMS EH EZ Mag. Int . Plane 1 Plane 2 Site 1 1979-06-30 01-44-35.9 02-31.2 W 40-25.8 N 5 1.4 8 10 4.1 2 1982-02-23 17 59-15.2 02-45.0W 40-38.2N 15 1.3 3 7 4.1 3 1987-10-19 12-54-42.9 03-13.8 W 40-12.9 N 5 0.2 1 I 3.2 4 1988-09-28 12-43-50.7 03 32.1 W 40 05.1 N 1 0.3 2 4 3.0 5 1988-10-04 13 05-10.9 03-33.6 W 40-05.0 N 2 0.3 I 3 3.1 6 1988 10-11 14-15-29.2 03-34.6 W 40-04.3 N 1.5 0.4 2 3 3.1 7 1988 10-24 04-38 51.7 03-14.0 W 40-04.7 N 9 0.4 2 2 3.4 8 1990 07-07 23-30 17.7 03-34.5 W 40-42.5 N 2 0.8 2 2 3.3 9 1994 04-06 03 33 00.0 03-33.0 W 41-33.0 N 4 0.5 1 1 3.4 43/111 62/005 ALCOCER. GU VI 83/313 11/183 DURON. GU 26/233 75/132 V. SALVANES. M 63/340 36/300 ARANJUEZ. M 44/338 35/358 ARANJUEZ. M 29/309 58/122 ARANJUEZ. M lIl 39/114 27/I 14 FUENTIDUEIC,!A. M II 1V 57/245 45/234 S.A.GUADALIX. M V 26/304 13/314 BURGO OSMA. SO Numbers refer to those of Fig. 5. Depth, horizontal (EH) and vertical errors (EZ) in km. Regional strike-slip stress tensor has been obtained (R = 0.0 4- 0.2). Main stress axes are cq = 17/320, cy2 = 68/015 and ey3 = 03/225, with quite good quality parameters (Fig. 7A). Nevertheless, individu- ally obtained focal mechanisms (Fig. 7B) correspond to seven reverse faults with a small strike-slip com- ponent and to two strike-slip-normal faults (No. 3 and 8). Because of the low number of analyzed earthquakes, a division of the population would not be appropriate. The calculated stress tensor presents an R value of 0 because of the presence of reverse and normal faults with same trends (the same as microstructural data). This situation is similar to the fault pattern ob- served in Upper Miocene and Quaternary materials. Thus, one of the characteristics of these stress fields could be a simultaneous reverse and normal fault ac- tivity. Absence of reverse faults at the surface could be an effect of different depths between analyzed faults and earthquake hypocentres. In order to draw the associated SHMAX trajecto- ries map, a projection of P axes (reverse faults) or B axes (normal faults) on the horizontal plane has been used. SHMAX orientation and P axes do not nec- essarily coincide, but in low lateral-slip component focal mechanisms both orientations seem to be very similar (Fig. 6A). 7. Seismicity distribution The instrumental earthquake epicentre distribu- tion from 1979 to 1994 and magnitudes between 3 and 4.1 (Spanish Geographical Institute catalogue) in the studied zone, show the following (Fig. 6B). (a) The Central System southern border fault (SBF) has four epicentres on its trace in the last ten years, so that it could be considered slightly active. The SBF seems to be segmented by several N N E - S S W and NW-SE faults. Attending to the fo- cal mechanism type, obtained on this fault (No. 8), the possibility of a higher seismic activity on these accompanying structures than in the southern border fault (SBF) can not be rejected. In any case, SBF is a seismic boundary that separates two zones: the Central System and the Duero Basin to the north, where the seismic activity is lower, and an area from the SBF to the south, where a great epicentre concentration appears between the UMLDL and the Zfincara Graben System (ZGS). This latter zone con- sists of NE-SW- and E-W-trending graben systems (Fig. 6B). (b) The Tajo-Tajufia Graben System (Fig. 6B, TTGS) is also segmented by N140E-trending faults, that are seismogenic towards the southeast. A normal fault system can be observed on the surface, that in- dicates a clear NW-SE extension. Although there is an extensional focal mechanism (No. 3) in this area, deep faults seem to be mainly reverse, indicating a NW-SE-shortening that agrees with fault-slip data. This situation could be explained by the presence of a crustal flexure compatible with the deduced short- ening which produces normal faults parallel to the flexme axis at the surface. Nevertheless, this assump- tion can not be confirmed because of the absence of published deep seismic profiles. (c) The Zfincara River Fault (Fig. 6B, ZF), south- wards of Altomira Range, shows similar characteris- tics to the previously described graben system, with G. De Vicente et al./Tectonophysics 266 (1996) 405-424 9 3 421 4 5 6 7 R q B S h a p e fac to r : R = 0 .0 +/ - 0 .2 O r i e n t a t i o n : p h i = 201 .7 t he = 23 .0 p s i = 25 .8 Q u a l i t y : L i k e l i h o o d = 0 .893 S c o r e = 0 .817 A Fig. 7. (A) Global solution. (B) Nine individual focal mechanisms obtained by the method of Rivera and Cisternas (1990) (see also Table 3) for the central part of Iberia. a higher prevalence of strike-slip faults. The Toledo Mountains border seems to be inactive. (d) The Iberian Range shows a moderate earth- quake concentration. Specifically, two major earth- quakes of the studied populat ion (numbers 1 and 2) are located at the border between the Madr id Basin and this chain. Number 2 is generated on a N E - S W - trending structure, paral lel to the Central System that may be interpreted as a NW-dipping active thrust. The rest of the earthquakes, located towards an in- ner part of this chain, are upon N W - S E mean trend faults. 422 G. De Vicente el al./Tectonophysics 206 (1990) 405 424 8. Discussion and conclusions According to the results, present-day and neo- tectonic stress tensor characteristics, as well as the neotectonic period for the Madrid Basin and the Central System can be determined. (a) The present stress tensor deduced from focal mechanisms has a N140E largest horizontal short- ening trend, compatible with the one established by MOiler et al. (1992) for Western Europe. This stress tensor activates mainly strike-slip and NE- SW-trending reverse faults. (b) Major seismicity is concentrated on the Cen- tral System south border fault (SBF) and in the Tajo-Tajufia Graben System (TTGS). Both structures trend NE-SW, normal to the main horizontal short- ening direction deduced for the active stress tensor. Nevertheless, numerous faults indicating NW-SE ex- tension have been observed on the surface. (c) Both Upper Cretaceous to Paleogene materials on the Madrid Basin margins and Upper Miocene and Quaternary materials of the central part of the basin record the neotectonic stress tensor. SHMAX trend is N I55E, and all types of faults are active, although the most frequent correspond to strike-slip ones. (d) Thus, the neotectonic period could be defined by the similarity between the present and neotectonic stress tensors with time. This similarity could be established as a function of stress tensor shapes (R), SHMAX trends, and the deformation intensity. According to the stress tensor regime, it seems from the paleostress data that the stress regime changed from the Middle Miocene (transpressive) to the Late Miocene (transtensive). Although, one must consider that there are important differences in depth between the three considered data sources. Depths of earthquake hypocentres are larger than the fault-slip sites located on the basin surface, and perhaps equivalent (the most surficial ones) to some fault-slip data obtained at the basin borders. Thus, it is risky to establish evolutive rules of the stress tensor type with the three data sources. In any case, all described fault sets and the nine established focal mechanisms are compatible with a NW-SE largest horizontal shortening (SHMAX)- On the whole, at the Madrid Basin and the Central System southern border fault, the SHMAX orientation shows a SHMAX trend difference of 10 degrees be- tween two average stress tensors calculated for both of the established paleostress fields (Middle to Up- per Miocene and Upper Miocene to Quaternary) that does not seem to be large enough to separate two different stages, although change in basin fill char- acteristics is also recognized at this time. suggesting the activity of different geodynamic processes. A SHMAX counterclockwise rotation can be ob- served between the Middle Miocene fault-slip data and the present-day stress tensors SHMAX trajectories in the Central System and in the Duero Basin. Nev- ertheless, the latter have been defined only by two focal mechanisms and P axes (not SHMAX orienta- tions), so their reliability depends on more complete studies. In this case, the neotectonic period in these areas began at least in the Late Miocene. Deformation distribution and intensity suggest that active structures migrated from north to south with time: all of the described structures appear as clearly active during the Middle Miocene. The Late Miocene lacustrine facies distribution (UMLDL) makes it evident that there was higher tectonic activ- ity at the south border fault (SBF) than at the north border fault (NBF). Pliocene tilting is more evident in the Madrid Basin than in the Duero Basin. Finally, the present earthquake distribution shows that the seismicity decreases from the Central System south- ern border fault to the north, and concentrates on the central part of the basin, on the Tajo and Tajufia River Graben System. From this point of view, the neotec- tonic period related to the Central System southern border fault (SBF) began in the Middle Miocene, while on the graben system of the central part of the Madrid Basin the neotectonic period could have started during the Pliocene. Acknowledgements The authors thank Dr. Rivera, Dr. Cisternas and Dr. Reches for allowing us to use their programs, and Dr. Lee Ior providing us with a version of his program LISSAGE. The authors appreciate Dr. Delvaux and an anony- mous referee's critical reading and improvement of the manuscript. This work has been funded by the Spanish DGI- CYT grants: PB94-0242 and the Spanish Consejo de Seguridad Nuclear (CSN). G. De Vicente et al. / Tectonophysics 266 (1996) 405~124 423 References Alonso, J.L., Pulgar, J.A. and Garc/a-Ramos, J.C. and Barba, P., 1996. Tertiary basins and Alpine tectonics in the Cantabrian Mountains (NW Spain). In: E Friend and C. Dabrios (Edi- tors), Tertiary Basins of Spain. Cambridge University Press, Cambridge, pp. 214-227. Alonso-Zarza, A.M., Calvo, J.E, Garc/a del Cura, M.A. and Hoyos, M.A., 1990. Los sistemas aluviales Miocenos del borde noreste de la cuenca de Madrid: sector Cifuentes- Las Inviernas (Guadalajara). Rev. Soc. Geol. Esp., 3(1-2): 213-230. Alonso-Zarza, A.M., Calvo, J.E and Garc/a del Cura, M.A., 1993. Paleomorphological controls on the distribution and sedimentary styles of alluvial systems, Neogene of the NE of the Madrid Basin (central Spain). In: M. Mazo and C. Puigdefabregas (Editors), Alluvial Sedimentation. Spec. Publ. Int. Assoc. Sedimentol., 17: 227-292. Ancoechea, E., 1982. Evoluci6n espacial y temporal del vol- canismo reciente de Espafia Central. Tesis Doctoral, Univ. Complutense, Madrid. Angelier, J., 1990. Inversion of field data in fault tectonics to obtain the regional stress, II1. A new rapid direct inversion method by analytical means. Geophys. J. Int., 103: 363-376. Babfn, R., Bergamfn, J.F., Fernfindez, C., Gonz~.lez, J.M., Hern~indez, J.L., Tejero, R. and De Vicente, G., 1993. Modfli- sat/on gravimdtrique de la structure alpine du syst6me central espagnol (secteur NE). Bull. Soc. G6ol. Ft., 164(3): 385-393. Blenkisop, T.G., 1982. Neotectonics. In: EL. Hancock and G.D.. Williams (Editors), J. Geol. Soc. London. Bott, M.H.P., 1959. The mechanism of oblique slip faulting. Geol. Mag., 96:109 117. Calvo. J.P., Alonso, A.M., Garcia del Cura, M.A., Sanz, E., Hoyos, M., Ordofiez, S. and Pozo, M., 1989. Sistemas lacus- tres miocenos de la Cuenca de Madrid. Guia de Campo. V Reuni6n Grupo Espafiol del Terciario, 16, CP-219. Calvo, J.P., Alonso, A.M., Garcfa del Cura, M.A., Sanz, E., Or- dofiez, S., Rodr/guez-Aranda, J.P. and Sanz, M.E., 1996. Sedi- mentary evolution of lake systems through the Miocene of the Madrid Basin: paleoclimatic and paleohydrological constraints. In: P. Friend and C. Dabrio (Editors), Tertiary Basins of Spain. Cambridge University Press, Cambridge, pp. 272-277. Capote, R., De Vicente, G. and Gonz~ilez-Casado, J.M., 1990. Evoluci6n de las deformaciones alpinas en el S/sterna Central Espafiol. Geogaceta, 7: 20-22. Capote, R., De Vicente, G. and Gonzzilez-Casado, J.M., 1991. An application of the slip model of brittle deformation to focal mechanism in three different plate tectonics situations. Tectonophysics, 191: 399-409. Delouis. B., Haessler, H., Cisternas, A. and Rivera, L., 1993. Stress tensor determination in France and neighbouringh re- gions. Tectonophysics, 221 : 413~437. De Vicente, G.. 1988. Anvil/sis Poblacional de Fallas. El sector de enlace S/sterna Central-Cordillera lb6rica. Tesis Doctoral, Univ. Complutense, Madrid, 317 pp. De Vicente, G. and Gonzfilez-Casado, J.M. (Editors), 1991. Las deformaciones alpinas en el S/sterna Central Espafiol. Mem. III Reuni6n de la Comisi6n de Tect6nica (Soc. Geol. Espafia), Vol. I. De Vicente, G., Gonzzilez-Casado, J.M., Mufioz-Mart/n, A., G/net, J.L. and Rodriguez Pascua, M.A,, 1994. Evoluci6n alpina del centro peninsular. Cuad. Lab. Xeol. Laxe, 19: 175- 190. Dewey, J.F., Helman, M., Turco, E, Hutton, D.H.W and Knott, S.D., 1989. Kinematics of the western Mediterranean. In: M.P. Corward, D. Dietrich and R.G. Park (Editors), Alpine Tectonics. Geol. Soc. London, Spec. Publ., 45: 265-283. D/az-Molina, M. and Tortosa, A., 1996. Fluvial fans of the Loranca Basin, Late Oligocene-Early Miocene, central Spain. In: P. Friend and C. Dabrio (Editors), Tertiary Basins of Spain. Cambridge University Press, Cambridge, pp. 292-299. D/az-Molina, M., Arribas, A. and Bustillo, M., 1989. The T6rtola and Villalba de la Sierra fluvial fans: Late Oligocene-Early Miocene, Loranca Basin, central Spain. 4th Int. Conf. Fluvial Sedimentology, Barcelona-Sitges, Field trip 7. Etchecopar, A., 1984. Etude des 6tats de contraintes en tec- tonique cassante et simulations de d6formations plastiques (approche mat6matique). Th6se d'Etat, USTL, Montpellier. Garcfa-Cuevas, C., Gonz~lez-Casado, J.M., Mufioz-Mar6n, A. and De Vicente, G., 1995. Determinaci6n de los tensores de deformaci6n y esfuerzo mediante el estudio comparado de poblaciones de /alias y del maclado mec~inico de la calcita. Geogaceta, 18: 3-6. Janssen, L.N., Torn6, M., Cloetingh, S. and Banda, E., 1993. Pliocene uplift of the eastern Iberian margin: inferences from quantitative modelling of the Valencia Trough. Earth Planet. Sci. Lett., 119:585 597. Junco, F. and Calvo, J.P., 1983. La Cuenca de Madrid. Libro Jubilar J.M. R/os (IGME), II: 534-543. Lee, J-C. and Angelier, J., 1994. Paleostres trajectory maps based on the results of local determinations: the 'lissage' program. Comput. Geosci., 20(2): 161-191. Meg/as, A.G.. Ordofiez, S. and Cairo, J.P., 1983. Nuevas aporta- c/ones al conocimiento geol6gico de la Cuenca de Madrid. Rev. Mat. Proc. Geol., 1: 163-191. M6rner, N., 1990. Neotectonics and structural geology; general introduction. Bull. Int. Quat. Assoc. Neotect. Comm., 13. Mtiller, B., Zoback, M.L., Fuchs, K., Mast/n, L., Gregersen, S.. Pavoni, N., Stephansson, O. and Ljunggren, Ch., 1992. Regional patterns of tectonic stress in Europe. J. Geophys. Res., 97(B8): 11783-11803. Mufioz-Mart/n, A., De Vicente, G. and Gonz~tlez-Casado, J.M., 1994. Atoll/sis tensorial de la deformaci6n superpuesta en el l/mite oriental de la Cuenca de Madrid. Cuad. Lab. Xeol. Laxe, 19: 203-214. P6rez Agudo, M.E., 1995. Interpretaci6n de ondas P y S e n la Meseta lb6rica. Tesis de Licenciatura, Fac. CC Ffsicas, Univ. Complutense, Madrid, 88 pp. Perucha, M.A., Mufioz-Martfn, A,, Tejero, R. and Bergam/n, J.E, 1995. Estudio de una transversal entre la Cuenca de Madrid y la Cordillera Ib6rica a partir de datos estructurales, s/sin/cos y gravim6tricos. Geogaceta, 18:15 18. Querol, R., 1989. Geolog/a del subsuelo de la Cuenca del Tajo. ITGE-ETSIM, Madrid. 424 G. De Vicente et al./Tectonophysics 266 (1996) 405-424 Reches, Z., 1983. Faulting of rocks in three-dimensional strain fields, II. Theoretical analysis. Tectonophysics, 95: 133-156. Reches, Z., 1987. Determination of the tectonic stress tensor from slip along faults that obey the Coulomb yield condition. Tectonics, 7:849-861 Reches, Z., Baer, G. and Hatzor, Y., 1992. Contraints on the strength of the upper crust from Stress Invesion of fault slip data. J. Geophys. Res., 97(B9): 12.481- 12.493. Rivera, L. and Cisternas, A., 1990. Stress tensor and fault plane solutions for a population of earthquakes. Bull. Seimol. Soc. Am., 80(3): 600-614. Rodrfguez-Aranda, J.P., 1995. Sedimentologfa de los sistemas de llanura lutftica-lago salino del Mioceno en la zona oriental de la Cuenca de Madrid (Taranc6n-Aufi6n). Tesis Doctoral, Univ. Complutense, Madrid, 474 pp. Sell, I, Poupeau, G., Casquet, C., Galindo, C. and Gonzfilez- Casado, J.M., 1995. Exhumaci6n alpina del bloque morfo- tect6nico Abantos-Pefialara (Sierra del Guadarrama, SCE): potencialidad de la termocronometrfa por trazas de fisi6n en apatitos. Geogaceta, 18: 23-26. Sim6n-G6mez, J.L., 1986. Analysis of a gradual change in stress regime (example of the eastern Iberian Chain, Spain). Tectonophysics, 124: 37-53. Surifiach, E. and Vegas, R., 1988. Lateral inhomogeneities of the Herynian crust in central Spain. Phys. Earth Planet. Inter., 51: 226-234. Van de Beek, EA. and Cloetingh, S., 1992. Lithospheric flexure and the tectonic evolution of the Betic Cordilleras (SE Spain). Tectonopbysics, 203:325 344. Van Wees, J.D., Cloetingh, S. and De Vicente, G., 1995. The role of pre-existing weak zones in basin evolution: constraints from 2D finite element and 3-D flexure modelling. In: RG. Buchanan and D.A. Nieuwland (Editors), Modern Develop- ments in Structural Interpretation, Validation and Modelling. Geol. Soc. London, Spec. Publ., 99: 297-320. Vegas, R., Vfizquez, J.T., Surifiach, E. and Marcos, A., 1990. Model of distributed deformation, block rotation and crustal thickening for the formation of the Spanish Central System. Tectonopbysics, 184: 367-378. Wallace, R.E., 1986. Active Tectonics. National Academic Press, Whasington, 266 pp. Zoback, M.L., 1992. First- and second-order patterns of stress in the litosphere: The World Stress Map Project. J. Geophys. Res., 97:11703-11728.