The structure of the Alpujarride Complex and its division into tectonic units is a subject of continuous updates and scientific debates. Several authors over recent decades have worked on the division of the tectonic units in the Alpujarride Complex of the Internal Zone of the Betic Cordillera (Elorza et al. 1978; Simancas and Campos, 1993; Sanz de Galdeano, 1997; Alonso-Chaves & Orozco, 2007; etc.). The division proposed in each case depends not only on the geological data gathered through fieldwork but also on the tectonic model that had been applied by each author. In some cases, these models lead to inaccurate interpretations. In this sense, studies in related sciences such as hydrogeology may reach conclusions or interpretations that are not compatible with some of the preexisting geological premises in the study area. Also, an accurate control of the stratigraphic sequences is necessary to avoid, how it can be observed, the multiplication of tectonic units, something now signaled in several sectors of the Betic Cordillera (Martín-Martín et al., 2006; Sanz de Galdeano & López Garrido, 2003; Sanz de Galdeano, 2017).

In the northern area of Sierra Tejeda (S Spain), different aspects are discussed regarding the possible contradictions raised by the previous interpretations of units, in some cases separating continuous lithological sequences or, on the contrary, joining different lithological sequences at present separated by tectonic contacts, based on proposals that are incompatible with hydrogeological field data and observations. Consequently, the aim of this work is to propose a differentiation of tectonic units in the study area consistent with several geological data, including those of hydrogeological and stratigraphic features. To achieve this purpose the successions of deformations affecting the area are also described, integrating them in a wider regional frame. Moreover, the carbonate sequence to the N of Alcaucín is presented, something not done so far. In the description made, the study area was divided into two separate sectors: Alcaicería and Majadas. The former, the main sector, is located in the south of Alhama de Granada (Figs. 1 and 2), and the latter is situated W of the first one, to the N of Alcaucín and S of Zafarraya (Figs. 1 and 3).

According to the field observations, in both studied sectors, only two tectonic units could be identified, i.e. Almijara and Guájares. The name of the Guájares unit actually refers to a group of units all with similar tectonic and lithologic character, which –without disregarding their relations—can be called differently in each area. In the present work, it will be named as Robledal unit (Table 1), referring to several farmhouses of this area and seeking to avoid confusion with that of Alcaicería used by Alonso-Chaves et al. (1993) or by Alonso-Chaves & Orozco (2007) to denominate a subdivision of the Charcón unit of Elorza et al. (1978).

The marbles of the Sierra Tejeda constitute an aquifer whose northern limit is formed by the contact with metapelitic rocks of low permeability (of the Robledal unit). According to the previous geological works, the outcrop of marbles of Rodaderos had never been considered specifically as part of the Sierra Tejeda aquifer, since it was considered to have an upper tectonic position, and therefore hydraulic connection was deemed unlikely. The first reference about the Rodaderos aquifer, from hydrogeological point of view, was carried out by ITGE-DG (1990) as part of the hydrogeological atlas of the Granada province, where they define it as a small independent aquifer since they did not evaluate its water resources (2 hm³/year) jointly with the aquifer of Sierra Tejeda. In the other hand, the study of the origin of the waters drained by a thermal spring (Baños de Alhama), located at north to the study area, reveal that the most probable recharge area of this outlet must be Sierra Tejeda aquifer because the isotopic δ¹⁸O value (-8.58‰) imply a high altitude of recharge (López-Chicano, 1992; López-Chicano and Pulido-Bosch, 1996). In these works, the tectonic position of the Rodaderos aquifer is not addressed directly, but the conclusions suggest the existence of an underground flow from Sierra Tejeda to the north through the marbles of Rodaderos.

The water table has been measured at a well (1040 m a.s.l) in Sierra Tejeda (S. Tejeda in Fig. 7) and at four points in Rodaderos area (Fig. 7): a quarry (1012 m a.s.l), Parrillas well (1005 m a.s.l), Rodaderos well (999 m a.s.l), and Rodaderos spring (951 m a.s.l). The water table elevations observed at the study site progressively decrease from Sierra Tejeda to Alhama River. The major groundwater flow path in the Rodaderos sector goes predominantly towards the NE, reaching the main karst outlets to 945 m a.s.l (Alhama River; Fig. 7). The Parrillas well (Fig. 7) is exploited for water supply agricultural irrigation, but in the same location has been documented a sporadic spring active during high waters periods (ITGE-DG 1990; López-Chicano, 1992) and flowing toward the polje of Zafarraya (NO Rodaderos). The occasional functioning of this outlet as a consequence of the piezometric level rise, in response to significant pluviometric episodes, seems to indicate that this groundwater source is a trop-plein or overflow spring.

The groundwater drained by Alhama River, the source of which is in Los Nacimientos sector in Sierra Tejeda aquifer (Alhama spring; Fig. 7), was measured upstream (Q=179 l/s as average value) and downstream (Q=383 l/s) of the calcareous outcrop of Rodaderos in several flow-measurement campaigns (Tab. 2). During the execution of these hydrodynamic records, no significant water runoff was detected through tributary streams, which does not mean that in high waters it may be not exist. Table 2 shows that the Alhama River increases the flow more of 200 l/s (114% more) as it passes through Rodaderos outcrop.

The marbles of Rodaderos occupy an area of 7.74 km² and annual rainfall for the 2016/2017 water year was 682 mm. If we take into account an efficient infiltration of 42% calculated with the APLIS method (Andreo et al. 2018) for the marbles of Sierra Almijara (SE of the study area) in the same tectonic unit, the average resources of the Rodaderos aquifer are 2.21 hm³/year, similar to the value obtained by ITGE-DG (1990).

Rodaderos aquifer would require an effective infiltration of 6.4 hm³/year to justify an average discharge of 204 l/s (Table 2), without taking into account the resources extracted for irrigation. In a preliminary interpretation, the source of the increase of flow in the Alhama River cannot come only from with the recharge occurring directly on Rodaderos marbles, because this area is too small to account for this. Additionally, from the isotopic point of view, the average values of δ¹⁸O of the water drained by Rodaderos (-8.41 ‰) and Alhama (-8.30 ‰) springs were similar to those obtained by López-Chicano and Pulido-Bosch (1996) in Baños de Alhama spring, where estimated a recharge height above of 1500 m a.s.l. The elevation of Rodaderos outcrop (1203 m a.s.l) seems to be insufficient if we assume that the height of recharge of the groundwaters that drains is similar to the obtained in the study of López-Chicano and Pulido-Bosch (1996), which would be consistent with the hypothesis of the existence of the hydraulic connection between Rodaderos and Sierra Tejeda aquifers.

The Robledal tectonic unit thrusts the marbles of Rodaderos with a contact visible on both sides, N and S, of this relief. The surface of this contact coalesces laterally, something particularly visible to the E (Fig. 2). This structure indicates that only a single unit thrusts the marbles of Rodaderos and that a division into two or three tectonic units has no justification. This thrust surface is affected in some places by later faults, as can be seen in the southern part of the cross-section B of Figure 5.

The relationship between the dark and light schists of the Robledal unit is well preserved in some places, indicating a former stratigraphic contact, as occurs to the SE of Rodaderos, while in other places this relation has been tectonically affected. This last feature occurs N of Rodaderos where a large part of the lithological sequence underwent a clear extension, or near the farmhouse of Robledal Alto, where the light schists have been practically laminated, putting the dark schists and the marbles of this unit close together (Fig. 2).

On the southern edge of the Rodaderos anticline, particularly in its SW part, the contact between the marbles of Rodaderos and the dark schists was interpreted by Elorza et al. (1978) and Alonso-Chaves & Orozco (2007, among others, as a stratigraphic contact. However, the dark schists are situated generally over the marbles, this being clearly visible in many places where different levels of marbles (not a single layer) are under the dark schists and, moreover, part of this contact is affected by a more recent E-W fault (Fig. 6B), whose surface is well exposed and presents minor structures, corresponding to dextral and normal dextral displacement.

The previous division into tectonic units had a key point in this contact. If the contact between the marbles of Rodaderos and the dark schist were a single stratigraphic contact, then these marbles and the dark schists would form the Venta de la Palma unit of Elorza et al. (1978). However, this continuity is contradicted by the features described in the preceding paragraph. A major tectonic contact is present, changing the previous interpretation—that is, the Venta de la Palma unit does not exist as defined, and furthermore the dark schists, with the gneisses at the bottom, and the light schists and other marbles above (forming the Robledal unit), must be separated from the Rodaderos marbles.

This new interpretation agrees with the preliminary hydrogeological data shown. The spatial distribution of groundwater levels and the increasing discharge (around 100%) of Alhama River through the Rodaderos carbonate outcrop suggest the hydrogeological connection with Sierra Tejeda aquifer. Rodaderos outcrop is too small to be responsible for the increasing discharge of Alhama River. Therefore, the tectonic and the hydrogeological data clearly point to the connection of the marbles of Rodaderos with those of Almijara unit. According to these hydrogeological data and their interpretation, the marbles of Sierra Tejeda and Rodaderos outcrop would constitute the same aquifer, so that they must be geologically located in the same tectonic unit.

The extensional structures previously described by Alonso-Chaves & Orozco (2007, among other articles) were interpreted as contacts separating tectonic units (Venta de Palma, Bentomiz and Alcaicería). However, these deformations, of fragile character in the study area, affect the former continuous lithologic sequence, without changing its original order—that is, there are extensional deformations but this does not necessarily means the formation of new tectonic contacts separating new units.

In the southern sector of the Alcaicería area, north of Sierra Tejeda, Alonso-Chaves & Orozco (2007) indicated the existence of more extensional structures in places corresponding to E-W faults in which minor structures cannot be seen or there is no tectonic contacts. Moreover, these authors assigned lower and middle Miocene ages to these structures, although this area lacks any sediments of these ages. Apparently, the authors followed the age attribution of extensional structures in other places of the Cordillera and applied them to the area of the NW of Sierra Tejeda. These types of interpretations correspond to the application of previous models, but in the study area, it is not possible to maintain these age attributions.

The E-W faults correspond to the system of E-W dextral strike-slip faults affecting the Betic Internal Zone (Sanz de Galdeano et al., 1985; Sanz de Galdeano, 1989, 1990, 2008). According to the interpretation of Sanz de Galdeano (1989), a major fault lies N of Sierra Tejeda, but the new data show that this, in fact, corresponds to a wide fault zone. This interpretation is applicable not only to the Alcaicería sector but also to the northern border of the Majadas sector, where the contact between the Alpujarride and Malaguide complexes as well as with the Campo de Gibraltar units and the Subbetic leads to the same conclusion.

The Majadas sector preserves a large part of the formations of marbles of the Robledal unit. In the NW of Sierra Tejeda, a local depressed area was formed owing to the coincidence in this sector of the aforementioned E-W faults, the normal sinistral faults also cited previously, and the NW end of the Frigiliana fault (Sanz de Galdeano, 1989).

In what concerns to the elongated body of phyllites situated within the Malaguide Carboniferous rocks, for its present position cannot correspond to the Robledal unit or other lower Alpujarride units. Its origin is to be found in some of the Casares-Federico units (Didon et al., 1973), also called Imbrications of Benarrabá (Balanyá, 1991) and Intermediate units (Sanz de Galdeano et al., 2001). This may be because during the double process of the thrusting of units and the immediate E-W drift of the Internal Zone, part of these units were dragged together with the Malaguide Complex.

On a regional scale, the Almijara unit corresponds to an Alpujarride unit situated in a middle position within this complex, if the division that considers three main groups of units is used (Aldaya et al., 1979), although these authors situated it in the upper position of units. The Robledal unit, as indicated previously, forms part of the Guájares nappe, which can be directly correlated with the Jubrique-Los Reales unit (Sanz de Galdeano, 2017), belonging to the upper Alpujarride units. This correlation is based in its tectonic position and lithological sequence, particularly the presence of gneisses in equivalent positions.

A brief summary of the geological evolution of the study sectors within the Internal Zone is the following: there is a preliminary stage related to the presence of gneisses, presumably formed by the Paleozoic emplacement of the Ronda Peridotites under the Jubrique-Los Reales (Guájares) unit (Sanz de Galdeano, 2017). This aspect is still under discussion and does not directly concern the present study. The present structuring of tectonic units studied here occurred during the Alpine Orogeny. This gave rise to the superposition of the Robledal unit over the Almijara unit. Later, the Internal Zone was drifted westwards, then forming the major E-W faults.

In this drift process, the pile of units underwent extensional deformations, affecting in an irregular shape the previous structures. In this aspect, Simancas (2018) discusses the significance of the extension and thinks that it is compatible with a moderate orthogonal shortening. In any case, in the study area the importance of this extension, even noticeable in some places, does not justify the previous proposed divisions of the Alpujarride Complex into tectonic units based in its possible effects.

In the study area, only two Alpujarride tectonic units have been identified: Almijara and Robledal units (in ascending order). This interpretation is consistent with the hydrogeological data. The piezometric levels in the aquifer of the Rodaderos shows a potential flow path towards the Alhama River, which increases its flow around 100% as it passes through this marble outcrop. The direct recharge from a possible isolated Rodaderos aquifer would not be sufficient to justify this remarkable increase in flow, so it would require a hydraulic connection with the aquifer of Sierra Tejeda. For this reason, and according to field observations, it can be concluded that the Rodaderos marbles do not belong to an upper unit but rather, to the contrary, they occupy the same position of the marbles of Sierra Tejeda, which is also coherent with the isotopic composition of groundwater. That is, they correspond to the marbles of the Almijara tectonic unit and they are hydrogeologically connected. However, it is necessary to continue investigating the hydrochemical and isotopic characteristics of the water drained by the springs of this sector of Sierra Tejeda aquifer and the groundwater of Rodaderos, in order to gather new data to test the conclusions of the present work.

The presence of gneisses in the lower position of the Robledal unit permits its correlation with other equivalent units of the Guájares-Jubrique/Los Reales group of units.

Over the previous structures, there is a superposition of extensional deformations, but unlike previous interpretations, there is no evidence of the enormous importance attributed in previous models, at least in this area, where new units cannot be differentiated for its cause.

The new E-W faults identified offer a better understanding of the westward displacement of the Betic Internal Zone.