Leucogranites in the Garhwal Himalaya: The story according to zircon Charlie Oldman1, Clare Warren1, Christopher Spencer2, Tom Argles1, Nigel Harris1, and Sam Hammond1

Contact author: Charlie Oldman ([email protected]) 1 The Open University, Milton Keynes (UK) 2 Queen’s University, Kingston (Canada) Motivation for this study • Partial melting of metapelitic rocks has occurred throughout the high-structural levels of the Himalayan orogen A typical - tourmaline • The presence of melt impairs local mechanical leucogranite strength and may be the driving force behind the mid- to deep-crustal exhumation

• Analysis of chemistry in structurally related rocks of various melt fraction, we can gain a better understanding of this process

• In this study, we use a suite of geochemical signatures in zircon from leucogranites, , and host metapelites to inform us of specific interactions occurring at the time of crystallisation A diatexite Geology of the Garhwal • Sampling locations were mostly contained to the Rishi Ganga (Badrinath) and Alaknanda (Mana) valleys in the Garhwal region of the Indian Himalaya

• Valleys sampled are part of the Greater Himalayan Sequence (GHS) of crystalline metamorphic rocks, marked in blue, found right across the orogen

• Line A-B transects the GHS and is shown in cross-section below

• Extent of migmatization is greatest towards the north- east, with larger granitic bodies below the South Tibetan Detachment (STD) B

A • Sampling targeted the Badrinath Formation After Weinberg 2016

A B

Spencer et al., 2012 Analysis • Analytical spots targeted the dark (high-U) outer growth rings typical of Himalayan-aged zircon

• Oxygen isotope ratio (δ18O), analysed with SIMS, is unaltered by anatexis and is inherited from the source rock • Measured at the Guangzhou Institute of Geochemistry

• U-Th-Pb analysis, using LA-ICP-MS, dates the crystallisation event • Measured at Curtin University

• Trace element analysis (LA-ICP-MS) allow CL imaging δ18O U-Th-Pb Trace for the U-Pb dates to be linked to geological processes • Measured at Curtin University

Source Date Reactions 100 µm n = 105 Probability Density Plots Leucogranites • On the right are probability density plots of zircon U-Pb dates for both leucogranite and migmatite samples

• Leucogranite zircon crystallisation is concentrated between 12 and 24 Ma, with a prominent peak at 16.5 Ma

• Migmatite zircon dates can largely be split into two groups – those occurring Date (Ma) before and those contemporaneous n = 36 with the main body of leucogranite Migmatites zircon

• Within this relative small area of the Himalaya, melt would have been present and repeatedly generated across a 20 Ma timespan

Date (Ma) Leucogranite Date vs δ18O

• Here is a subset of leucogranites analysed, with date plotted against δ18O

• δ18O varies by up to 2‰, which suggests a high degree of heterogeneity in the early melt

• All spot analyses fall within the expected δ18O range of sedimentary origin

• Samples show periods of both continuous and punctuated zircon crystallisation REE profiles – BAD 01d (Leucogranite)

• Chondrite-normalised REE profiles of zircon from sample BAD 01d (grt-tur-ms-bt leucogranite), coloured with respect to U-Pb date (left) and δ18O (right)

• Elements Sm-Lu show consistent trend of increasing concentrations in grains with younger crystallisation dates

• Profiles are heterogeneous with respect to δ18O REE profiles – BAD 03 (Leucogranite)

• Chondrite-normalised REE profiles of zircon from sample BAD 03(tur-bt-ms leucogranite), coloured with respect to U-Pb date (left) and δ18O (right)

• Concentrations are not correlated to changes in grain date, with two distinct age populations (15.1- 16.5 and 19.4-22.2 Ma)

• Decreasing δ18O may correspond with increasing REE concentrations • However, total δ18O variation (0.95‰) is close to overlapping error (±0.39‰) REE profiles – BAD f02 (Leucogranite)

• Chondrite-normalised REE profiles of zircon from sample BAD f02 (tur-ms-bt leucogranite), coloured with respect to U-Pb date (left) and δ18O (right)

• Profiles show general trend of decreasing REE concentration in younger grains (the opposite seen in sample BAD 01d)

• Again, decreasing δ18O may correspond with increasing REE concentrations • Total variation is 1.75‰, with average error of ±0.41‰ Migmatites Date vs δ18O

• Here is a subset of migmatites analysed, with date plotted against δ18O

• For many migmatites samples the extent of new zircon growth is low, resulting in a reduced dataset

• Again, all δ18O values are within the sedimentary-derived range, with strong heterogeneity in both date and oxygen isotopes (BAD07c) or surprisingly homogeneity (BAD02b) • The later is likely due to the small sample size of zircons suitable for analysis REE profiles – BAD 04a (Migmatite)

• Chondrite-normalised REE profiles of zircon from sample BAD 04a (sil-bt diatexite migmatite), coloured with respect to U-Pb date (left) and δ18O (right)

• Decreasing LREEs concentrations in younger grains, with narrowing variation of HREEs

• Possible trend of increased REE concentrations with decreased δ18O • Grey profile does not have oxygen data Preliminary Observations • Zircon crystallisation events in leucogranites and migmatites are offset, with populations of migmatite- hosted zircon crystallising before and during leucogranite formation

• Leucogranites show a variety of zircon populations: • Younger dates with restricted δ18O, shifting δ18O ranges with increasing crystallisation, etc. • Continuous and punctuated episodes of crystallisation, with potentially varying sources

• Migmatite data is sparse, with huge variety between samples • Less zircon has crystallised in the migmatites than in the leucogranites, result skewed towards larger grains

• REE profiles show variable trends (decreased conc. with date, increased conc. with light δ18O, etc.) • Oxygen isotope values have are a complimentary analysis to U-Th-Pb and REEs, showing relationships not otherwise visible

References Spencer, C. J., Harris, R. A., and Dorais, M. J. ( 2012), The and exhumation of the Himalayan metamorphic core, eastern Garhwal region, India, Tectonics, 31, TC1007, doi:10.1029/2010TC002853. Weinberg, R.F. (2016), Himalayan leucogranites and migmatites: nature, timing and duration of anatexis. J. Metamorph. Geol., 34: 821-843. doi:10.1111/jmg.12204