Discussion
Microstructural analysis of a single pebble from the Indus Formation, within the context of a meso-scale structural framework, has provided useful information for regional and tectonic-scale analysis at this significant location at the Asia/India plate boundary. The inclusion of microstructurally focussed 40Ar/39Ar geochronology on characterised grains within the pebble provides the time component to this history which includes information on both absolute time as well as relative timing of events on the broader scale. The Arrhenius data from this 40Ar/39Ar step-heating experiment undertaken in this study is of such quality that multiple-diffusion domain experiments and modeling normally only done on K-feldspars has been successfully undertaken on white mica. Thus these diffusion experiments have provided information on the relative sizes of different domains and ages preserved on those domains.
More than one gas reservoir (domains of different ages) is preserved in the white mica from the pebble. The smallest domain represents the least retentive sites and is very small. Nevertheless, this domain constrains the timing of cooling below ~200°C at 16.5 ±4.7 Ma. This small domain may have formed during deformation and recrystallization that occurred after the initial cooling of the batholith. In contrast the initial cooling of the batholith is recorded by the main domain with a plateau age of 55.7 ±0.3 Ma (>30% of gas release) and is the most prominent feature in the apparent age spectrum. This plateau age may reflect a cooling age which is comparable to the SHRIMP ages obtained from the Chumathang leucogranite dyke swarm. In addition, the final stages of the apparent age spectrum show an increase in age up to 61.1 ±0.3 Ma, which is consistent with a minimum age estimate for cooling of the Ladakh Batholith. Because the argon system within the pebbles has not been reheated to temperatures or conditions that would have reset the system, the pebble has preserved these distinct gas reservoirs and allowed modelling on the cooling history or any recrystallisation that may have occurred during this tectonic process.
The source of the pebble cannot be known exactly. The most obvious source is from granites that are the closest to the Indus Formation, however the possibility that it may have been derived from a more distal source should also be considered. The Kohistan or Gangdese batholiths did have magmatic activity that was occurring in the early Paleogene during a second stage of plutonism (112-39 Ma), however the actual date of ~56 Ma is not a common age (Heuberger et al. 2007, Wen et al. 2008, Treloar et al. 1989). In addition, the distances, and obstacles associated with transport of eroded pebbles to the Indus Formation make these batholiths an unlikely source. The Ladakh Batholith, immediately adjacent to the Indus Formation, is the most probable source as it preserves ages of 66-64 Ma, 60-56 Ma, ~49-45 Ma and 20-10 Ma (White et al. 2011, cf. Wu et al. 2007), and the petrology is equivalent. Our ~56 Ma pebble is a granite by petrological classification (not a diorite or granodiorite), and we thus consider that it might represent an eroded equivalent of the 56.5 ± 1 Ma Khardung La Granite, or the 58 ± 1 Ma Chang La Granite from the Ladakh Batholith (White et al. 2011) as a source.
Importantly 40Ar/39Ar geochronology on this pebble can be used to determine the relative timing of the regional events of both the Ladakh Batholith and the Indus Formation. For example, 16.5 Ma must be the maximum age of the formation of the molasse. In addition, 16.5 Ma must be the maximum age for the burial and deformation of the Indus Molasse, and the underlying Indus Formation. The sequence of tectonic events needed for this to happen is that the batholith must have been exhumed and eroded with deposits tumbled and abraded and then accreted to the Indus Formation prior to burial and deformation.
The diffusion parameters have been calculated applying the Fundamental Asymmetry Principle to the interpretation of Arrhenius data derived from the 40Ar/39Ar step heating experiment. The following values have been obtained for a spherical geometry: E= 70.7 kcal/mol and D0/r2 = 4.90e+08s-1 rising to 1.14e+06s-1 in the most retentive domains, where E is the activation energy and D0 the frequency factor and r the domain radius. Fractal feathering (as described by Forster and Lister, 2010) can be seen to be a prominent feature in both the Arrhenius and the r/r0 plots. The activation energy would have been substantially underestimated if this FAP had not been observed.
Figure 14. Anisotropic diffusion modelling predicts faster c-perpendicular diffusion compared to the diffusion that would occur c-parallel
a) and b) show how different relative domain sizes of the fractal can emulate the r/r0 plot where:
1. assumes anisotropic diffusion, faster perpendicular to c-axis than parallel to it
2. assumes fractal pattern of very fast pathways (e.g. cleavage and/or stacking faults)
3. perhaps the Ca associated with the margins of such a fractal distribution allows the relative domain size of the fractal to emulate the r/r0 plot.
The diffusion parameters allow estimates as to the temperature during cooling, from which a depth can be inferred, based on assumptions on the temperature gradient. In essence, higher activation energies require higher temperatures to be maintained, and thus reflects the depth at which preserved events have occurred. This in turn affects estimates as to exhumation rates and other aspects of tectonic interpretation. Note that the calculated activation energy for this sample is higher than that estimated from high-pressure experiments (Harrison et al. 2009). Modelling on anisotropic diffusion that predicts faster c-perpendicular diffusion compared to the diffusion that would occur c-parallel can be undertaken on this data (Figure 14). For example where different relative domain sizes of the fractal can emulate the r/r0 plot, different parameters can either assume anisotropic diffusion to be faster perpendicular to c-axis than parallel; or that the fractal pattern of very fast pathways exist (e.g. cleavage and/or stacking faults).
The Ladakh Batholith is reported to have a major pulse of magmatism active at ~56 Ma (White et al. 2011; Weinberg and Dunlap 2000; Upadhyay et al. 2008; Singh et al. 2007; Bhutani et al. 2009), and this is seen in the Chang La and Khardung La granites (White et al. 2011) which correlates well with the ~56 Ma age in the pebble. Metamorphic events recorded within the Ladakh Batholith have not been recorded in the Chang La or Khardung La granites, suggesting that localised activity, either structural or magmatic, has occurred. For example, the Chang La and Khardung La granites do not record evidence for the volcanic magmatism activity at ~47 Ma found else where in the batholith (White et al. 2011; St-Onge et al. 2010; Weinberg and Dunlap 2000: 40Ar/39Ar age of andesitic dyke); Bhutani et al. 2004: 40Ar/39Ar age) and shear zone activity at for example 45 Ma in the Thanglagso Shear zone is localised to that zone (Weinberg and Dunlap 2000). However, Chang La granodiorites record the 20-10 Ma metamorphic event. No trace of a 49-45 Ma event was recorded from our pebble. This is again consistent with the suggestion that the Chang La or Khardung La granites are a possible source.
The Ladakh Batholith also preserves younger metamorphic or cooling ages (White et al. 2011: metamorphic zircon rims 10 - 20 Ma; Bhutani et al. 2003: 40Ar/39Ar age; Kirstein 2011 and Kirstein et al. 2006: apatite+zircon fission track). These younger ages are consistent with the ages of the first steps of the 40Ar/39Ar geochronology step-heating experiment undertaken in this study.
Other geochronometers have been used in this area to draw inferences as to the pattern of tectonic activity. Much of the tectonic interpretations from fission track ages are based on changing tilting directions of the batholith (from north to south) over three distinct phases 49-30 Ma, 30-15 Ma and 15-0 Ma (Kirstein et al. 2009), where the Ladakh Batholith played a passive role in the regional tectonics during shortening associated with the India-Asia collision. This extensive fission track study (Kirstein 2011, Kirstein et al. 2006) suggests that magmatism ceased in the late Eocene with initial denudation driven by erosion due to uplift caused by large-scale shortening, with rapid exhumation and weathering between 35-26 Ma on the southern margin of the Ladakh Batholith. Apatite and zircon fission track and U-Th/He data from the central and northern edge of the Ladakh Batholith showed exhumation commenced after 18 Ma. The northern margin being exhumed at a rate of <0.4 mm/year from ~18 Ma (cf. Kirstein 2011), while exhumation in the central part of the batholith has been suggested to have been rapid and occurred at ~22 Ma (Kirstein 2011). What can be said is that younger ages are being recorded in association with exhumation of this batholith whether they be SHRIMP ages, 40Ar/39Ar geochronology ages or fission track ages,.
Shear zones within, and on the carapace of the Ladakh Batholith suggest that it was not a massive passive block that tilted with regional tectonics but was rather deformed, or was ductiley ripped apart by movement on regional-scale shear zones during its history. The structural features that we observed on the southern margin of the Ladakh Batholith (Fig. 1c) were occasional bounding southwest dipping ductile shear zones. These structures appear to define remnants of a carapace shear zone, as occurs with a metamorphic core complex. For the Indus Molasse to be made from pebbles of the Ladakh Batholith itself, it first must have been exhumed, allowing the batholith to be eroded and debris transported. These pebbles (and cobbles etc.) then accumulated and accreted onto the Indus Formation. According to this hypothesis, parts of the Indus Formation are a molasse that formed as an apron at the margins of the Ladakh core complex, after its Miocene exhumation, once the lower plate was exposed. We suggest that the movement of the late back thrusting of the Indus Formation with the Indus Molasse is thus Miocene or younger.
Microstructural analysis of a pebble within the Indus Molasse in conjunction with 40Ar/39Ar age results from this study indicate that the pebble was eroded from the source, suggested here as the Ladakh Batholith, some time after 16.5 Ma, during its continued extensional exhumation. This produced a molasse that accreted to the Indus Formation. The upright folding observed in the Indus Formation (and Indus Molasse) during inversion of the core complex took place at a later time, i.e. mid-Miocene or younger. The sequence of events that allows a pebble from the Ladakh Batholith to be consolidated and deformed into the overlying material is possible when it is considered that the later inversion of the region resulted in back-thrusting, and recumbent folds (Fig. 1b). These in turn were overprinted by later tight upright folds, during which time a pervasive axial plane slatey cleavage was formed, overprinting earlier formed structures (Fig. 1b).
The way in which the Indus Formation is backthrust over the Indus molasse is similar to tectonic sequences that occurred in Tibet where the Xigaze terrane is thrust back over the Gangrinboche conglomerates (Gansser 1964; Aitchison et al. 2007). In addition, earlier studies on the Tibet uplift or unroofing of correlative rocks (Transhimalayan batholith also known as the Gangdese batholith) suggest that southern Tibet was significantly uplifted and eroded in the early Miocene (Harrison et al. 1992), at about the same time as our data suggests for the Ladakh Batholith and formation of the Indus Molasse in Indus Suture region.