Human occupation of northern India spans the Toba super-eruption ~74,000 years ago

The dispersal of *Homo sapiens* out of Africa and into Asia and Oceania is a significant area of research. India's location makes it a crucial region for understanding this migration. The timing of *Homo sapiens*' arrival in India, their material culture, the replacement of archaic populations, and the impact of the Toba super-eruption are all highly debated topics. While the Indian fossil record for this period is limited, mitochondrial DNA analysis suggests India played a vital role in the colonization of Australasia. A key debate centers around whether *Homo sapiens* arrived in India before (~80,000 years ago) or after (~50-60,000 years ago) the Toba super-eruption, utilizing different stone tool technologies. The scarcity of well-dated sites in India from the crucial period between 80,000 and 50,000 years ago hinders definitive conclusions. This study focuses on the Dhaba locality in the Middle Son River Valley to provide more data on this critical time period. The Dhaba site is chronologically positioned between earlier Middle Palaeolithic/Late Acheulean sites (140,000-104,000 years ago) and later blade-based Upper Palaeolithic technologies (previously dated to around 39,000 years ago, although this date is considered problematic). This research aims to use infrared stimulated luminescence (IRSL) dating of sediments to determine the chronological changes in lithic technology at Dhaba and contribute to a broader understanding of the South Asian Palaeolithic and the dispersal of modern humans.

Existing literature presents conflicting hypotheses regarding the timing and technological characteristics of early *Homo sapiens* presence in India. Some studies suggest pre-Toba occupation with an African Middle Stone Age (MSA) toolkit, while others propose a post-Toba arrival with Howiesons Poort microlithic technology. The limited number of reliably dated sites within the critical time frame (80,000-50,000 years ago) contributes to this ongoing debate. Studies have focused on stone tool assemblages, genetic data from modern populations, and rare finds such as engraved ostrich eggshells. Previous work in the Middle Son Valley, such as at sites like Patpara, Nakjhar Khurd, Sihawal, and Bamburi 1, provided insights into the Late Acheulean and early Middle Palaeolithic periods, but a detailed sequence spanning the Toba eruption was lacking. The existing dates for Upper Palaeolithic technologies in the region were also subject to uncertainty. This study therefore addresses the lack of reliable data from this crucial period and seeks to resolve the existing chronological uncertainties.

The Dhaba locality comprises three archaeological excavations (Dhaba 1, 2, and 3) situated on the banks of the Middle Son River. Step trenches were excavated at each locality to expose the stratigraphic sequence. The sediments consist of pedogenically altered alluvial sands, silts, and clays overlying Proterozoic sandstone and shale bedrock. The presence of Youngest Toba Tuff (YTT) deposits nearby allowed for potential correlation of the Dhaba sequence with the Toba eruption. Thirteen sediment samples were dated using the multiple-elevated-temperature post-infrared IRSL (MET-PIRIR) method. This method involves extracting K-feldspar grains, treating them to remove contaminants, and measuring their IRSL signals. Dose rates were determined from field and laboratory measurements of gamma and beta radiation, and estimations of cosmic-ray and internal dose rates. The equivalent dose (D_e) was determined using the MET-PIRIR procedure, which involves measuring IRSL signals at increasing stimulation temperatures and fitting a single saturating exponential function to the dose-response curves. Several criteria (recuperation, recycling ratio, dose recovery, anomalous fading, and residual dose) were used to test the reliability of the method. Stone tools from each stratum were classified into technological categories (core, flake, retouched flake) and further typologically classified. Cores were scanned in 3D to measure key characteristics. Raw material types and flake scar patterns were analyzed to understand technological changes over time.

The IRSL dating revealed that the Dhaba sequence began accumulating just before the YTT event (79.6 ± 3.2 ka and 78.0 ± 2.9 ka in Dhaba 1). Six glass shards consistent with YTT were found in strata dated between 79.6 ± 3.2 ka and 65.2 ± 3.1 ka in Dhaba 1, although contamination cannot be entirely ruled out. The Dhaba sequence continued until close to the Last Glacial Maximum. The stone artefact sequence spans ~55,000 years (80-25 ka), with three major technological phases. The Dhaba 1 assemblage (80-65 ka) predominantly features Levallois technology (cores, flakes, points, blades), primarily using chert, mudstone, and silicified limestone. Red ochre was also found. Levallois technology continued to dominate Dhaba 2 and 3 assemblages (55-47 ka), after which it disappeared. Microlithic technology emerged around 48 ka in Dhaba 2 and 3, co-occurring with Levallois technology in some strata. Analyses of raw material and flake scar patterns showed significant changes over time, supporting the identified technological phases. The study found a strong statistical significance between changes in raw material and flake scar pattern types across the strata.

The Dhaba findings provide crucial chronological and technological information for the Middle Son Valley, filling a gap in the regional sequence. The technological changes observed at Dhaba mirror those at Jwalapuram in southern India, showing a transition from Levallois to multiplatform technology and finally microlithic assemblages. These changes appear gradual, involving shifts in raw material selection, retouch strategies, core reduction techniques, and the introduction of new artefact forms. The Dhaba sequence confirms the presence of MSA-like technologies in India before and after the Toba eruption. The technology evolved from Levallois towards lamellar core reduction systems and, later, microlithic technology. Genetic studies suggest a modern human exit from Africa around 70-52 ka, and fossil evidence supports earlier dispersals. The similarities between the Dhaba lithic industry and those from East Africa, Arabia, and Australia suggest that *Homo sapiens* likely dispersed through South Asia before 65 ka. Dhaba acts as a crucial link between regions with similar archaeological assemblages.

The Dhaba site provides compelling evidence for continuous human occupation in northern India spanning the Toba super-eruption. The consistent lithic technology, resembling MSA assemblages from Africa and Arabia, strongly supports the hypothesis of *Homo sapiens* presence in India prior to 65,000 years ago. The detailed chronological sequence and technological transitions at Dhaba offer valuable insights into human behavioral evolution and migration patterns in South Asia. Future research could focus on further excavations at Dhaba to refine the chronological framework and explore the relationship between technological change and environmental shifts.

While the study provides robust chronological data using the MET-PIRIR method, potential limitations include the possibility of contamination in the YTT glass shards and the limited sample sizes in some strata. The interpretation of technological changes is based on observable trends in the lithic assemblages. Further research could incorporate more sophisticated analyses of tool use-wear and residue analysis to gain a more comprehensive understanding of hominin behavior.