Post-LGM glacial retreat drives aggradation in the interiors of the Kashmir Himalaya

5 Understanding the response of glaciated catchments to climate change is fundamental 6 for assessing sediment transport from the high-elevation, semi-arid to arid sectors in the 7 Himalaya to the foreland basin. The fluvioglacial sediments stored in the semi-arid Padder 8 valley in the Kashmir Himalaya record valley aggradation during ~19-11 ka. We relate the 9 valley aggradation to increased sediment supply from the deglaciated catchment during the 10 glacial-to-interglacial phase transition. Previously-published bedrock-exposure ages in the 11 upper Chenab valley suggest ~180 km retreat of the valley glacier during ~20-15 ka. 12 Increasing roundness of sand-grains and reducing mean grain-size from the bottom to the top 13 of the valley-fill sequence hint about increasing fluvial transport with time and corroborate 14 with the glacial retreat history. Our result also correlates well with late Pleistocene-early 15 Holocene sediment aggradation observed across most Western Himalayan valleys. It 16 highlights the spatiotemporal synchronicity of sediment transfer from the Himalayas 17 triggered by climate change. 18


21
Understanding the role of past climate change on surface processes is essential to as eccentricity and orbital precision (Milankovich, 1941). While the eccentricity cycles over 36 ~100 ka cause the glacial-interglacial cycles, the ~21-23 ka precision is suggested to be aggradation-incision cycles is yet to be tested.

67
In pursuit of a better understanding of the role of climate change in sediment transport 68 in glaciated catchments, we investigated the aggraded sediments from the Padder valley in 69 the Kashmir Himalaya (cf. Fig.1 for location). In this study, we combined detailed field 70 observations on valley morphology, sedimentology, and sediment chronology to explore how 71 sediment archives can record evidence of glacial retreat.  quartz grains from fine-medium sand layers in the sediment archive is, therefore, a potent 120 option to constrain timings of sediment aggradation. We sampled five samples from the 121 medium sand layers (SD/P01-P05) and one sample from the fine sand layers (SD/P06) for 122 OSL measurement. The sand from the same layers was further used for grain-size and grain-123 shape analysis.

124
All samples were collected in sealed galvanized iron pipes and opened only in 125 subdued red light (wavelength ~650 nm) in the laboratory. The outer ~3 cm of each end of 126 the pipes were discarded to avoid accidental exposure to sunlight during sample procurement.   Fig. 5a). Test doses for samples SD/P01-P05 were set between 40 to 120 Gy (Fig.5b), 134 while the test dose for sample SD/P06 were ranging 8-15 Gy. The aliquots were considered 135 for ED estimation only if: (i) recycling ratio was within 1±0.1, (ii) ED error was less than 136 20%, (iii) test dose error was less than 10%, and (iv) recuperation was below 5% of the  (Table 1). The estimation of moisture content was done using the fractional 142 difference of saturated vs. unsaturated sample weight (Table 1). 145 We sampled the same sand layers which were used for OSL sampling. Samples were 146 dried in a hot-air oven at 50˚C to achieve complete dryness. And then, ~2 kg of each sample 147 were used for sedimentological analysis.  149 Each sample was dry-sieved using 1000 µm, 750 µm, 300 µm, 250 µm, 125 µm and 150 50 µm test sieves. Sediments above 1000 µm (very coarse-gravelly sand) and below 50 µm 151 (silt) were discarded as we wanted to quantify the coarse-grained to very fine-grained fraction 152 of sand (Table 2a) 159 We performed the coning and quartering method several times with the initial mass to 160 finalize 100g of each sample for sediment shape analysis. We separated the quartz grains 161 from the mix by Frantz isodynamic magnetic separator and used quartz as the index grain.

162
Grain-shape was calculated using Powers roundness index (Powers, 1953), where roundness 163 is given by the formula- Here, r = radius of the smallest inscribed circle within the grain and R = radius of the 166 largest inscribed circle within the grain. We made 20 discs of each sample and measured the r 167 and R of at least 20 grains per disc using a scaled Leica microscope. So, the minimum 168 number of counts per sample is 120. The higher the roundness index, the more rounded the 169 grains are. Grain-shape analysis results are provided in Fig.6b. Results of the 170 sedimentological analysis are listed in Table 2.   (Table 2). Samples SD/P01 and SD/P02, collected from the bottom of the log show a high 183 mean grain-size (φ ~0-1); whereas, samples SD/P03 and SD/P04, taken from the middle of 184 the log, yield a lower mean grain-size (φ~2-3) and samples SD/P05 and SD/P06 yield even 185 smaller mean grain-size (φ~3) (Fig.6a). Similarly, the roundness coefficient (according to 186 equation 1, described in section 3.3.2) varies from 0.27±0.08 to 0.60±0.07 (Table 2). The terrace T5 (Fig.3a) records a ~4m thick fine sand layers. The sand layer is devoid of any 217 recognizable laminations, the grain-size is lower and the sorting is higher than fluvioglacial 218 sand samples (Fig.6a). The equivalent dose estimates from sample SD/P-06 are also 219 clustered, having low over-dispersion value (OD ~ 6%, cf. Table 1   Grain-size analysis portrays a fining-upward sequence (Fig.4a), while the average roundness 227 of the grains also increases from the bottom to the top (Fig.4b). Fig.4b illustrates a linear 228 correlation between mean population grain-size and mean roundness co-efficient. It 229 highlights that with time, the grain-size and angularity of grains have reduced simultaneously. 230 We propose that the fluvioglacial sediment sequence recorded more fluvial transport with (see point G1 in Fig.1 and Fig.2). Whereas, in the next ~5 kyr, the valley glacier retreated 241 ~180 km and was at point G4 (~4150 m above msl) ( Fig.1 and Fig.2). We propose that a 242 similar glacial retreat must have been observed in the northern tributaries originating from the 243 arid Zanskar Range (Fig.1)  caused by global as well as a regional temperature change. We acknowledge that the post-

268
LGM deglaciation is coupled with late Pleistocene increased monsoon intensity (e.g.,  LGM to early Holocene period. It is understood that the drainage systems that lie in the           Table 2: (a) Details of grain-size distribution in collected samples. Note that, grains above 1000 µm 599 and below 50 µm are discarded. (b) Mean± standard deviation of roundness co-efficient for sampled 600 sand layers. Minimum number of reading per sample is 120. 601