Susceptibility and radiometry data used for stratigraphic correlations: case study on Upper Triassic beds in Turkey

Abstract A Julian/Tuvalian (=Lower/Upper Carnian) substage boundary within the Kasımlar Formation, recently detected at Aşağiyaylabel (Taurus Mountains, Turkey) by facies analyses and biostratigraphic ammonoid investigations, was additionally detected by magnetic susceptibility (MS) and radiometry data. The Aşağiyaylabel sequence, a key section concerning environmental changes during Early to Late Carnian time, represents a deepening sequence from platform carbonates to pelagic limestones and marls. The Julian/Tuvalian boundary strata can be correlated over wide areas due to a positive shift in MS values from 11–105×10−6 SI (range AS I, KA IV) at the Lower Carnian sediments to 62–458×10−6 SI (range AS I, KA IV) at the Upper Carnian sediments as well as in radiometry values from 21.35–83.27 nGy h−1 (range AS I, KA IV) at the Lower Carnian sediments to 38.43–130.24 nGy h−1 (range AS I, KA IV) at the Upper Carnian sediments. A second, smaller positive shift in MS occurs at the transition from shallow-water carbonates of the Kartoz Formation (−11 to 71×10−6 SI; range AS I–KA II) into deeper-water carbonates from the base of the Kasımlar Formation (3–108×10−6 SI; range AS I–KA II). This study presents the first MS and radiometry data from Lower to Upper Carnian sediments. They reliably reflect lithological changes and display a direct function of enhanced terrigenous input into marine sediment systems, most probably caused by variations in climatic conditions.

The sudden deepening at AS I was apparently triggered by synsedimentary tectonics, which led to a new basin geometry along a deeper ramp setting, due to down-faulting and tilting of the platform (Lukeneder et al. 2012). Accordingly, the depositional environment at the northern Tethyan passive margin changed from a mid ramp over an outer ramp into a basinal position. Shallow-water carbonates from the Kartoz Formation were replaced by deeper-water limestones and marlstones of the Kasımlar Formation. A low-energy, bottom environment was interpreted based on the autochthonous fine-grained deeper-water limestones. Allochthonous biogenic material, for example, thick-shelled bivalves, gastropods and plasticlasts within the pelagic limestones, indicate episodic input of thin (several millimetres-thick) coarse-grained sediment layers from different source areas (proved by the microfacies; see Table 1) of the foreslope or shallow marine ramp (Lukeneder et al. 2012). The general deepening of the carbonate ramp at AS I is demonstrated by both the ammonoid biostratigraphy (Lukeneder & Lukeneder 2014) and the facies analysis (Lukeneder et al. 2012).
The Julian/Tuvalian boundary (¼Lower/Upper Carnian boundary) is also clearly observable at AS I, as demonstrated by detailed facies analyses and ammonoid-biostratigraphic investigations (Lukeneder et al. 2012, Lukeneder & Lukeneder 2014. Correlatable sections in the vicinity of AS I represent similar sediment sequences, with somewhat different thickness. Due to the similarities in lithological units (e.g. shallow-water limestones, ammonoid beds, Julian/Tuvalian boundary) the distinct sections can be well correlated. These interpretations, based on facies and biostratigraphy, are strengthened by the magnetic susceptibility (MS) and radiometry values as exemplified in the present paper. MS is explained as the induced magnetization which arises at minerals within a certain rock sample, brought into a low-strength magnetic field (Kodama 2012). It is one of numerous methods for detecting the properties of magnetic rocks. Such features are important in certain environments (Jovane et al. 2013). Since Robinson (1990), MS has become a standard technique for logging sediment cores. A number of additional authors have described a close relationship between the MS signature and the facies of marine environments (Devleeschouwer 1999;Ellwood et al. 2000;Zhang et al. 2000;Racki et al. 2002;Da Silva & Boulvain 2006;Babek et al. 2010;Lukeneder et al. 2010). This relationship is the result of erosion and weathering, caused by sea-level changes (regression/transgression; Crick et al. 1997, Ellwood et al. 1999 or by climatic changes (Crick et al. , 2002. Whilst most studies concentrate on climate-driven cyclostratigraphic aspects (magnetosusceptibility event and cyclostratigraphy (MSEC); e.g. Crick et al. 1997;Ellwood et al. 2001Ellwood et al. , 2013aEllwood et al. , 2013bWhalen & Day 2010), regional events, caused by tempestites or storm events, can be also important for correlation and interpretation at a local scale. Furthermore, many studies measure MS by detailed rock sampling and subsequent highly sensitive laboratory measurements (e.g. Hladil et al. 2006).
In contrast to such highly sensitive measurements, the present study represents an 'averaging' method (Hladil et al. 2006) of MS and radiometry measurements. It compares the MS and radiometry data of the Upper Triassic section AS I from Aşagiyaylabel from the Taurus Mountains of Turkey (Anatolia) with four time-equivalent sections near Aşagiyaylabel (AS IV) and Karapinar (KA I -II & IV). The MS and radiometry values or peaks depict mineralogical and sedimentological variations. These variations can be directly used for straightforward correlations of time-equivalent sequences on a regional scale. MS and radiometry results should be interpreted with caution as trend-data. Nonetheless, the characteristic peaks at the Julian/ Tuvalian boundary as well as at the lithological boundary between the shallow-water carbonates (Kartoz Formation) and deeper-water carbonates (Kasımlar Formation) help to localize these particular sequences at surrounding localities. These peaks form the basis of the detailed facies and biostratigraphic investigations undertaken in this study.

The facies changes at Aşagiyaylabel (AS I and AS IV) and Karapinar (KA I -II & IV)
Aşagiyaylabel I (AS I). The main section at Aşagiyaylabel (AS I) consists of the Kartoz Formation and the Kasımlar Formation, embracing the Lower and Upper Carnian. The Kartoz Formation is only marginally represented by the lowermost 1.7 m of the log; this reflects a shallow palaeowater environment (massive limestone-bearing in situ reefs, megalodont bivalves, gastropods, sponges). The Kasımlar Formation, in turn, is represented by at least 33 m, bearing the Carbonate member (Julian 2), the Marlstone member and the Shale member (Tuvalian 1; Lukeneder et al. 2012, Lukeneder & Lukeneder 2014Fig. 2). Based on its different facies description, the Carbonate member is subdivided into Carbonate member A (well-bedded dark-grey limestones yielding Kasimlarceltites krystyni in masses), B (Cipit-like boulders ¼ mixed carbonate blocks embedded within reworked material) and C (well-bedded dark-grey limestones bearing millimetre-thick spherical bioturbations). The overlying Marlstone member consists of ochre, marly limestones which fade at the top into tectonically laminated shales (Fig. 2). The whole 33 m of the Kasımlar log represents a maximum time interval of 1.89 -5.5 Ma (Ogg 2012;Lukeneder & Lukeneder 2014). This big difference in absolute values reflects an extreme divergence of two different chronostratigraphic models used for the absolute durations of the Upper Triassic (Ogg 2012). Whilst one model suggests a longer duration of Tuvalian time, the other suggests a longer Rhaetian duration. The deposits of the section AS I show an inclination of c. 508 towards NE. The whole facies description of AS I (Lukeneder et al. 2012) is summarized in Table 1.
Aşagiyaylabel IV (AS IV). The sequence starts with 14.5 m of massive shallow-water limestone (equivalent to the Kartoz Formation at AS I), yielding coral debris, sponges and bivalves (articulated and disarticulated). This is followed by 4 m of darkgrey, bedded ammonoid limestones (equivalent to Unit A at AS I; Fig. 2), bearing abundant K. krystyni. Event beds, which bear masses of K. krystyni, alternate with beds exhibiting only rare specimens of K. krystyni. The bedded limestones are overlain by 5.2 m of reworked material (equivalent to Unit B of AS I with Cipit boulders) and are sealed by a calcite layer; this may indicate tectonic faulting of undefined dimension. This is tectonically overlain by 3 m of carbonate reef-limestones yielding crinoid debris, and is topped by 0.5 m of coral limestone bearing coral debris, sponges and articulated bivalves (Figs 2-4, Table 1).
Karapinar I (KA I). The section Karapinar I (KA I) has one of the shortest but most characteristic sequences, represented by only 0.5 m of shallowwater limestones followed by a 5.5 m-thick bedded Kasimlarceltites-limestone sequence (Fig. 2). The shallow-water limestone and deeper-water limestone layers are tectonically overturned. The lithological transition from the shallow-water carbonates of the Kartoz Formation into the ammonoid beds (K. krystyni mass occurrence) of the lowermost Kasımlar Formation is the main feature within the transitional part in all known sections (Figs 2-4, Table 1).
Karapinar II (KA II). KA II, situated close to KA I (100 m SE), yields a very similar sediment sequence. The main difference is the extent of the ammonoid beds with their mass occurrence of K. krystyni. Shallow-water carbonates (which are equivalent to the Kartoz Formation of AS I) are replaced after 2 m by 17 m of black ammonoid beds (K. krystyni mass occurrence), equivalent to the Kasimlarceltites beds at the lowermost part of the Kasımlar Formation at AS I. This section seems to be more complete at the lower part of the Kasımlar Formation, causing an increased thickness of the ammonoid beds in comparison with AS I, AS IV and KA I. The top is formed by 2 m of dolomitized limestone. The shallow-water limestone and deeperwater limestone layers are characterized by a tectonically inverse stratification (Figs 2 -4, Table 1).
Karapinar IV (KA IV). The section KA IV, about 100 m to the north of KA II, represents the shortest, but probably the most important, sequence. It comprises the Lower Carnian (Julian) to Upper Carnian (Tuvalian) facies change with its typical boundary strata, ranging from marly limestones into marlstones. The section is equivalent to the corresponding part at the main section at Aşagiyaylabel (AS I), where the Carbonate member (Unit C, Julian 2) is followed by the Tuvalian Marlstone member above. The abrupt change from the 1 m-thick darkgrey carbonate sequence to ochre marlstones is obvious and is a clear marker level in that area (Fig. 2). Two metres of marlstones are directly visible, but an unknown thickness of marlstones  Table 1).

Methodology
Magnetic Susceptibility (MS) was measured directly in the field with a hand-held, highly sensitive (1 × 10 27 , measured in SI units; results are indicated throughout as 'value × 10 26 SI'), lowlevel SM-30 MS meter (GF Instruments) bearing a 50-mm pick-up coil. Bed-by-bed measurements were conducted at distinct beds (e.g. Carbonate member Units A and C). This approach was replaced by a measuring interval of between 5 and 50 cm where distinct beds were missing (e.g. Carbonate member Unit C and Marlstone member);186 measurements of MS were performed in sediments of the Kartoz and Kasımlar Formations at AS I; 171 at AS IV; 85 at KA I; 156 at KA II; and 38 at KA IV. Each single bed was measured at least three times and the average calculated. To prevent measurement errors due to the sensitivity of the hand-held instruments (uneven terrain and therefore variable distance between coil and rock surface), a fresh smooth surface was prepared for each single point measured.
The radiometry was also conducted directly in the field using a hand-held scintillation gamma radiometer (GF Instruments), bearing a 51 × 51 mm (6 in 3 ) NaI (Tl) detector. Measurements were taken with an acquisition time of 1 s and a sampling interval between 5 and 50 cm through the whole log. We applied this method as an additional correlation tool and therefore concentrated on the dose rate (nGy h 21 ) without testing the spectral results (K, U or Th): such detailed information is not supported by the instruments used. The resulting values were transposed to curves (using Microsoft Excel) and correlated with the log (Fig.  2). Furthermore we tested the correlation between the MS and radiometry signal for each sequence. The measurements were carried out during several field trips from October 2010 to October 2012 by the Natural History Museum Vienna (NHMV). In contrast to most studies, which focus on cyclostratigraphic results and therefore use highly sensitive laboratory measurements along with highresolution measuring intervals, this study focused only on the MS and radiometry trends in order to correlate different logs of a regional area.
Accordingly, data were measured directly in the field at the main locality (AS I) as well as at additional surrounding sites (AS IV, KA I-II & IV) with a maximum measuring interval of 50 cm. Samples did not receive any special preceding preparation; 152 radiometry measurements were performed in sediments of the Kartoz and Kasımlar Formations at AS I; 155 at AS IV; 81 at KA I; 132 at KA II; and 33 at KA IV. Data were subsequently compared with facies analyses and biostratigraphic data of AS I (Lukeneder et al. 2012;Lukeneder & Lukeneder 2014) as well as with sections from nearby localities (AS IV, KA I -II & IV).

Results of MS & radiometry measurements
General trends in MS in comparison with the CaCO 3 content A general increase in the MS values from shallowwater platform limestones over bedded limestones up to the marlstones is evident within all five sections and logs at Aşagiyaylabel and Karapinar (Fig. 2, Table 2). The most drastic difference can be recognized within both logs from AS I and KA IV in the median values between the bedded limestones (58 × 10 26 SI and 56 × 10 26 SI for AS I and KA IV, respectively) and the marlstones (171 × 10 26 SI and 142 × 10 26 SI for AS I and KA IV, respectively). This corresponds to the Julian/Tuvalian boundary (i.e. Lower /Upper Carnian boundary). At this boundary the CaCO 3 content at AS I decreases from about 70% at the Julian Carbonate member to 60% and lower at the Tuvalian Marlstone member. The same trend is present at KA IV: the CaCO 3 content decreases from . 55% at the Julian Carbonate member to ,30% from the Tuvalian Marlstone member onward. Although their MS range overlaps (11 to 105 × 10 26 SI v. 62 to 458 × 10 26 SI), a general increasing trend is clearly recognizable ( Table 2). The trends and positive shift in the MS curves can be precisely detected within the logs of AS and KA (Fig. 2).
Shallow-water carbonates: Kartoz Formation. The shallow-water carbonates of the Kartoz Formation are generally related to diamagnetic substances (Dearing 1999) and therefore show, as expected, negative MS values. The shallow-water platform carbonates, which partly bear in situ corals (Figs 2 & 3a -d), show very similar MS values as well as similar CaCO 3 contents (92.9 -95.8%). Hence a differentiation based on MS signals or CaCO 3 content is not possible in these shallow-water carbonates from the Kartoz Formation. Nevertheless, the lowest MS values are clearly much lower within carbonate deposits having in situ corals (minimum 211 × 10 26 SI) than within deposits of shallow-water limestones without in situ corals (minimum 23 × 10 26 SI). Furthermore a generally higher trend of MS values is observable at the shallow-water limestones without in situ corals (211 to 49 × 10 26 SI v. 23 to 71 × 10 26 SI).
Kasimlarceltites limestones: Kasımlar Formation (Carbonate member Unit A). Well-bedded Kasimlarceltites limestones occur in almost every log, except for KA IV due to its different stratigraphic position . The MS values of these black ammonoid beds range between 3 and 93 × 10 26 SI at AS IV, between 33 and 108 × 10 26 SI at AS I, between 16 and 74 × 10 26 SI at KA I, and between 4 and 72 × 10 26 SI at KA II. Their values are generally slightly higher than those of the shallow-water limestone deposits (211 to 71 × 10 26 SI; Table 2). While their median values are not always higher than the median values of the shallow-water limestones, their maxima are higher by far within all sections (108 × 10 26 v. 36 × 10 26 SI at AS I; 93 × 10 26 v. 71 × 10 26 SI at AS IV; 74 × 10 26 v. 31 × 10 26 SI at KA I; 72 × 10 26 v. 11 × 10 26 SI at KA II; Fig. 2, Table 2). The carbonate content in all sections (average 90.6%) is significantly lower than in the shallow-water carbonates (94.1%). This decrease corresponds well to the increase in terrigenous material.
Cipit-like boulders: Kasımlar Formation (Carbonate member Unit B). An interval with Cipit-like boulders (these being metre-to tens-of-metre-sized isolated blocks embedded in different kinds of sediments; see Richthofen (1860) and Lukeneder et al. (2012)) was detected in the Carbonate member (Unit B, Kasımlar Formation) within both sections of Aşagiyaylabel (AS I and AS IV; Fig. 2). Cipitlike boulders represent a mixture of allochthonous material from different source areas (e.g. shallow-water environments and deeper deposits), making it impossible to distinguish such sediments    Table 2). The distinct sandstone layer, breccia layer and pyrite layer are detected exclusively within the Marlstone member at the AS I section (Fig. 2). We therefore assume that the range of peaks in the MS curve at the Marlstone member can be assigned to such characteristic event beds or layers and are indicative for marginal breccias, glauconite or pyrite layers, which are hardly detectable without geophysical measurements (Figs 2 & 4h-j). MS measurements at Aşagiyaylabel and Karapinar generally reflect the influence of a terrigenous input. Higher susceptibility values argue for an increased entry of terrigenous material, additionally reflected by a decrease in the carbonate content.

The radiometry measurements
The radiometry variation of a studied section reflects the presence of detrital clays, feldspars, heavy minerals and other minerals arriving by riverine, aeolian or other inputs (Serra et al. 1980;Serra 1984;Rider 1990). Higher radiometry may reflect higher clay content. Measurements of radiometry response (nGy h 21 ) are a powerful tool for interpreting lithological changes within an outcrop (Lukeneder et al. 2010).
The radiometry values confirm the increasing trend and positive shift already described by the MS values at the Julian/Tuvalian boundary at AS I and KA IV (Fig. 2, Table 3). Shallow-water limestones (Kartoz Formation) and well-bedded ammonoid limestones (Kasımlar Formation, Carbonate member, Unit A) are not well distinguishable by the radiometry data (Fig. 2, Table 3). A significant increase is recognizable again at the Lower/Upper Carnian boundary beds (Julian 2/Tuvalian 1) at AS I and KA IV. The positive shift in radiometry values marks the change from well-bedded carbonates of the Carbonate member (Unit C) to the calcareous marls and marlstones of the Marlstone member, both within the Kasımlar Formation (Fig. 2).

Kartoz Formation and Kasımlar Formation with
Carbonate member (Units A -C). The whole sequences of the Kartoz Formation (shallow-water limestones partly with in situ corals) and the Carbonate member of the Kasımlar Formation (wellbedded ammonoid limestones Unit A, Cipit interval Unit B, well-bedded limestones Unit C) are hardly distinguishable from each other by means of their radiometry values (Figs 2 & 3g, Table 3). Their general radiometry ranges are very similar, except  Fig. 4g, Table  3). The positive shift in radiometry values is most distinct at section KA IV (Fig. 2)

Radiometry and MS data correlation: time-series analysis
For comparing the MS and radiometry curves (Fig. 5), we only used data from beds at which both measurements were available. The time alignment of both signals (MS and radiometry) was tested by using cross correlation of two time series with the correlation coefficient r (Davis 1986;Hammer & Harper 2006). Cross correlation is used for optimal alignment of two series (Hammer et al. 2008). To do this, one time series is kept in a fixed position, whilst the second time series is slid past, and for each position, the correlation coefficient (r) computed (Fig. 5, black line). The correlation coefficient (r; black line) is then plotted as a function of alignment position (Hammer & Harper 2006). The horizontal axis (x) represents the displacement of the radiometry data with respect to the MS data, whilst the vertical axis (y) shows the correlation of both time series for a given displacement. Therefore, the position with the highest correlation coefficient indicates optimal adjustment. Significance is shown by plotting the p-values ( Fig. 5; grey line). Correlation coefficients (r; black line), which show higher values than the corresponding p-values (grey line), indicate a significant correlation. For a more detailed description of time-series analyses see Hammer & Harper (2006) and Hammer et al. (2008).
The results (Fig. 5) clearly reveal that both curves of the section AS I and KA IV correlate significantly. A significant peak positive correlation occurs at AS I in the region of two samples delay of the MS signal with respect to the radiometry signal (see Fig. 5). This is in contrast to the sections AS IV and KA I -II, where no significant correlation between the two signals was observed. Moreover, certain sequences of AS I do not show positive correlations between the MS and radiometry signals. Examples are the Kasimlarceltites beds, or from the top of the Marlstone member onward, where the homogeneous marlstone beds start to alternate with sandstone, breccia and pyrite layers; this might affect the MS signal more significantly than the radiometry signal.

MS and radiometry data used for the correlation of time-equivalent sequences
The investigated MS and radiometry data from the area around Aşagiyaylabel and Karapinar are the first to have been measured from Lower to Upper Carnian carbonate sequences. At section AS I, the Julian/Tuvalian boundary could be fixed by ammonoid biostratigraphy (Lukeneder & Lukeneder 2014) and facies analyses (Lukeneder et al. 2012).
In the present study, the characteristic lithological change exactly at the Julian/Tuvalian boundary from well-bedded limestones (Unit C) into calcareous marls and marlstones (i.e. Marlstone member at AS I and KA IV) is demonstrated by a distinct MS increase from 58 × 10 26 SI at AS I resp. 56 × 10 26 SI at KA IV to 171 × 10 26 SI at AS I resp. 142 × 10 26 SI at KA IV. A similar increase in radiometry values is observed from 44.84 nGy h 21 at AS I resp. 51.24 nGy h 21 at KA IV to 66.19 nGy h 21 at AS I resp. 106.75 nGy h 21 at KA IV (Figs 2 & 5, Tables 2 & 3). MS and radiometry signals from section AS I were then compared with those from additional sections at Aşagiyaylabel (AS IV) and Karapinar (KA I -II & IV). The intention was to localize a specific part of the sequence at these sections, namely that part which possibly also bears an equivalent of the Julian/Tuvalian boundary. The opportunity to detect such important and characteristic biostratigraphic and lithological boundaries by geophysical measurements (e.g. MS and radiometry) is an additional, preliminary tool for correlation in the field. The measurements can be supported by subsequent measurements in laboratories (e.g. Kappabridge, Karl-Franzens-University, Graz), and ultimately proved by palaeontological and geological investigations. The MS studies, conducted here with hand-held instruments, can also easily be used to find characteristic features in sequences as sequence-boundaries, marked by lithological changes, even at sections where no macrofossil markers are detected in the field.
Sections AS IV and KA I -II lack this characteristic boundary. Section KA IV, however, which shows a similar lithological change from wellbedded calcareous limestones to marly limestones, shows an MS (up to 217 × 10 26 SI) and radiometry peak (up to 130.24 nGy h 21 ) comparable to that found at the Julian/Tuvalian boundary at AS I (Fig. 5, Tables 2 & 3).
Radiometry and MS values studied by a timeseries analysis show that in both cases (at sections AS I and KA IV) the curves are mostly significantly positively correlated (Fig. 5). Accordingly, the primary signal seems to be preserved (except for some parts discussed below).
Both signals show an abrupt increase at the Julian/Tuvalian boundary, which can be interpreted as an increased influx of detrital minerals into the system. The preceding demise of the carbonate platform and the decrease of the CaCO 3 content from the Julian to the Tuvalian sediments clearly reflect this siliciclastic entry. This entry probably affected fluviatile systems due to the more humid conditions, which are described in several studies dealing with the carbonate platform demise during Carnian time (Lukeneder et al. 2012). The possible trigger mechanisms for this extraordinary ecological incision are still under discussion and therefore the subject of a series of studies. One potential reason is an uplift of extended areas (e.g. the Fennoscandian High of the Cimmerian orogeny), which could have been a trigger for a monsoonal climate (Hornung & Brandner 2005;Preto et al. 2010). The resulting new landmasses could have been the source areas for the increased terrigenous input into the northwestern part of the Tethyan Sea. This influx might have led to the well-known carbonate platform demise within the Tethyan realm (Tollmann 1976;Aigner & Bachmann 1992;Hornung & Brandner 2005;Preto et al. 2010). Another plausible scenario is magmatic or volcanic activities (e.g. the volcanic series of the Wrangellia terrane or Eastern Mediterranean alkaline volcanics), which could have influenced oceanic circulation, resulting in global climate change (Hornung et al. 2007;Preto et al. 2010). Robinson (1973) and Parrish (1993) suggest that the wet climate during Carnian time might have been the peak of the monsoonal character that already existed during the transition from the Permian to the Triassic, and therefore can be traced back to the maximum aggregation of the landmasses at that time (Preto et al. 2010). The result was a hot and dry climate within the continent interior, alternating wet and dry seasons at the coast of eastern Laurasia and Gondwana (Parrish & Peterson 1988;Dubiel et al. 1991;Mutti & Weissert 1995), and a generally wet climate at higher latitudes (Robinson 1973;Preto et al. 2010). Furthermore, a sealevel drop, followed by submarine clathratedestabilization and methane degassing (Hornung 2008), are also part of a large compendium of discussed scenarios along with dramatic sea-level rise (Arche & López-Gómez 2014), sediments representing anoxic events beneath siliciclastic sediments (Jerz 1966;Schuler 1968;Hagemeister 1988) and rapid variations in seawater chemistry.
Although the radiometry and MS signals generally correlate positively in both sequences (AS I and KA IV) at which the Julian/Tuvalian boundary was detected, certain intervals of AS I do not support this general correlation trend. This for instance is the case of the well-bedded limestones bearing K. krystyni, the top of the Marlstone member, as well as the whole Shale member of AS I, which might be caused, to some extent, by diagenetic overprint. The stylolitic structures (Table 1) observed within these beds would support this hypothesis. Nonetheless, the MS signals between these Kasimlarceltites beds of the different localities are rather similar (61 × 10 26 SI at AS I; 24 × 10 26 SI at AS IV; 43 × 10 26 SI at KA I; 36 × 10 26 SI at KA II), perhaps reflecting a similar diagenetic history of those beds.
These characteristic MS and radiometric peaks at the Julian/Tuvalian boundary can be used as additional stratigraphic correlation tools. Moreover, the other lithological parts (below and above) at AS I also show values similar to those of the surrounding sequences. Although the primary signals might be diagenetically overprinted (because the MS and radiometry signals do not show significant correlation), some comparisons between similar lithologies are possible and are therefore discussed: Whilst at both AS I and KA II the carbonate platform sediments are represented by only 1.8-m-thick and 2.3-m-thick coral-limestones, respectively, the carbonate platform sediments at section AS IV is 14.5 m thick (Fig. 2). KA IV lacks the carbonate platform sediments, due to its higher stratigraphic position (Julian/Tuvalian boundary). The corresponding MS values of the shallow-water limestone deposits (AS I, AS IV, KA I, KA II; Fig. 2) are very low to negative (211 to 71 × 10 26 SI), as already estimated and expected for the diamagnetic mineral calcite (Algeo et al. 2007a). Ellwood & Ledbetter (1977) have already mentioned that MS values are inversely proportional to the CaCO 3 content of unconsolidated marine sediments. This fits well with the picture and data shown from Aşagiyaylabel and Karapinar: low MS values, from 211 to 71 × 10 26 SI, occur in shallow-water carbonates (Kartoz Formation), higher values, from 3 to 108 × 10 26 SI, are found in well-bedded limestones (Carbonate member, Kasımlar Formation), and much higher values, from 62 to 458 × 10 26 SI, occur in calcareous marls and marlstones (e.g. Tuvalian Marlstone member, Kasımlar Formation). Algeo et al. (2007b) have described atypical positive instead of negative correlations between MS values and carbonate beds from the Permo-Triassic Guryul Ravine section in Kashmir (India). This Indian section should appear as diamagnetic due to its high calcite content (Algeo et al. 2007b). Algeo et al. (2007b) explained this positive correlation with the theory that the carbonate fossil concentrations might be produced by winnowing, yielding incorporated paramagnetic heavy minerals which potentially cause the higher MS signal (Ellwood et al. 2000;Algeo et al. 2007b). The Indian example shows that MS values should always be used cautiously and interpreted solely in combination with facies analyses and different available proxies.
In contrast to the shallow-water carbonates with almost pure limestones, the overlying dark-grey, well-bedded limestones, which bear masses of K. krystyni, occur in nearly every section (Unit A  at AS I,  The sudden demise of the carbonate platform at the main section AS I, represented by the shift from shallow-water limestones bearing in situ corals over shallow-water limestones to bedded limestones and calcareous marls and marlstones, is interpreted as being caused by an increased input of siliciclastic material from the hinterland, triggered by more humid conditions (Lukeneder et al. 2012). Due to similar deposits and the similar MS and radiometry trend of the correlated sequences (AS IV, KA I, KA II), the local deepening can be interpreted at least as a regional event that affected a recent area of at least 2.8 km diameter (Aşagiyaylabel to Karapinar).
As mentioned above, we used the MS and radiometry results with caution and interpreted them solely as a trend. This is because we measured them directly in the field without highly sensitive laboratory methods and without special preceding sample preparation. Nonetheless, specific trends and shifts in MS and radiometry values can be detected and are used in combination with other analyses (e.g. geochemistry, facies changes, biostratigraphy). This interdisciplinary approach was particularly successful with regard to the characteristic peaks at the Julian/Tuvalian boundary. Additional distinct peaks along the MS curve, detected within the Marlstone member (AS I), are significant and are well correlated to a single sandstone layer, as well as to a distinct pyrite layer. Although pyrite is a paramagnetic mineral and generally shows low positive susceptibility values, the above-mentioned pyrite layer appears with values of 200 × 10 26 SI. Therefore it is remarkable that such high susceptibility values can be associated with pyrite. Nonetheless, other minerals exhibit stronger paramagnetism than pyrite, such as biotites or certain iron silicates (Dunlop & Ö zdemir 1997). Moreover, this could hint at partly oxidized pyrite which is probably associated with some magnetite or maghemite, and might be caused by the formation of magnetite or goethite (Lazaroff 1997). Such high values therefore indicate the occurrence of a range of distinct and thin event layers as sandstone and pyrite layers (e.g. AS I, top of marlstone beds). These are lithologically hardly observable in the field but easily detectable by geophysical measurements (e.g. hand-held MS). This shows that MS data and radiometry measurements, along with a trendline and shift interpretation undertaken by comparing and proving already-analysed lithological parameters, are valid methods for a better understanding of stratigraphic and palaeontological datasets. Geophysical measurements, although performed in the field by hand-held instruments, can hint at characteristic parts or layers within sections that are different and yet can be correlated. They can serve as a preliminary tool for subsequent stratigraphic investigations. MS and radiometry data are thus significant proxies for the amount of siliciclastic input, which in turn might reflect some climatic conditions.

Conclusions
This study presents the first magnetic susceptibility (MS) and radiometry signature of Lower to Upper Carnian sediments. The results from the MS and radiometry logs of the sections are therefore only comparable among themselves (AS I, AS IV  and KA I -II & IV). The investigated section comprises the important Julian/Tuvalian substage boundary (¼Lower/Upper Carnian boundary) within the Kasımlar Formation, recently detected at Aşagiyaylabel (Taurus Mountains, Turkey). The data are interpreted solely as a trend. Nonetheless, an indicative shift of the MS values (11 to 105 × 10 26 SI up to 62 to 458 × 10 26 SI) and an increase of the radiometry values (21.35 -83.27 nGy h 21 up to 38.43-130.24 nGy h 21 ) perfectly reflect the Julian/Tuvalian (Lower/Upper Carnian) boundary at the sections AS I and KA IV (Fig. 2, Tables 2  & 3). The Julian/Tuvalian boundary is of special interest due to its key position at the beginning of the Carnian Crisis within this area. The boundary is characterized by the change from a carbonate system into a marl-dominated system, triggered by severe climate change during Lower to Upper Carnian time. The Julian/Tuvalian boundary was already indicated by biostratigraphic investigations and facies analyses (Lukeneder et al. 2012;Lukeneder & Lukeneder 2014), strengthened herein by the MS and radiometry data from several sections. A second, smaller positive shift in MS and radiometry values is observed at Aşagiyaylabel and Karapinar at the transition from shallow-water carbonates of the Kartoz Formation to deeper-water carbonates from the base of the Kasımlar Formation at its ammonoid beds (AS I, AS IV, KA I, KA II). The latter shift is typical for a transition from pure carbonates to more marly limestones. This seems to have been caused by an increasing amount of siliciclastic material, which accumulated within the northwestern Tethyan shelves and basins due to the onset of increased river runoff, probably triggered by humid conditions. The situation is prolonged and more pronounced at the Julian/Tuvalian boundary when the Carnian Crisis (¼Carnian Pluvial Event) started at the western end of the Cimmerian Terranes, as postulated by numerous studies (e.g. Tollmann 1976;Simms & Ruffell 1989;Aigner & Bachmann 1992;Hornung & Brandner 2005;Kozur & Bachmann 2010;Lukeneder et al. 2012). This more humid climate phase, with increased terrigenous and siliciclastic input, occurs somewhat later in the Taurus Mountains than in other sections of various European areas (Lukeneder et al. 2012).
This study shows that MS and radiometry data, used with caution and in combination with facies analyses and biostratigraphic investigations, constitute a robust tool to detect lithological changes by quantitative trendlines. Additionally, very thin and distinct layers in the otherwise-uniform marls can be detected by the MS measurements. Such event beds are formed by brief but high input of siliciclastic material such as breccia, pyrite accumulation and sandstones. This trendline method, performed by hand-held MS and radiometry instruments, is straightforward, fast, inexpensive and nondestructive. We therefore suggest applying these geophysical methods as additional tools for comparing geological sections and logs. These geophysical approaches (using MS and radiometry devices) are more sensitive than biostratigraphic methods or facies analyses, making them useful tools to detect possible sequences for which more detailed biostratigraphic investigations should be conducted. They are also suitable for comparing logs at which proper biostratigraphic markers are not preserved. More exact data call for the additional use of laboratory equipment (e.g. Kappabridge, Karl-Franzens-University, Graz). This approach supports common facies analyses and palaeontological studies with quantitative data, enhancing the interpretation of past environmental conditions.
The study is part of project P 22109-B17, funded by the Austrian Science Fund (FWF). It was supported by the Turkish Geological Survey (MTA), which was responsible for approving digging permission within the area investigated. The authors appreciate the suggestions and comments of two anonymous reviewers which improved the quality of the manuscript. We thank the volume editor Simo Spassov (Royal Meteorological Institute of Belgium) for his positive advice. The authors thank Oleg Mandic (Natural History Museum Vienna) for useful comments concerning the MS and radiometry analyses, as well as Anton Englert and Franz Topka (both at the Natural History Museum Vienna) for sampling assistance and thin-section preparation.