Climatic windows for human migration out of Africa in the past 300,000 years

The African origin of *Homo sapiens* is well-established, but the timing and routes of their migration into Eurasia remain debated due to limited fossil evidence and ancient DNA. Previous studies have qualitatively explored potential scenarios using sparse empirical climate records or model-based data from limited time slices. Quantitative approaches have attempted to fit demographic rules to match the archaeological or genetic records, but these methods are often based on questionable biological assumptions. The archaeological record itself is sparse, particularly for early periods, and genetic data only reflect successful migrations whose descendants have been sampled. This study aims to identify climatically suitable periods for *Homo sapiens* to leave Africa by using a different approach. It utilizes high-resolution paleoclimate simulations for the last 300,000 years to estimate the minimum rainfall needed for successful migration. This is then combined with anthropological and ecological data to determine the actual climatic tolerances of hunter-gatherers, allowing for the reconstruction of climatic windows for out-of-Africa expansions. The study further assesses the compatibility of these inferred periods with existing archaeological and genetic evidence.

While the African origin of *Homo sapiens* is undisputed, the timing of their expansion into Eurasia is hotly debated. Genetic and fossil evidence suggests a major exit around 65,000 years ago, yet archaeological discoveries in Saudi Arabia (at least 85,000 years ago), Israel (at least 100,000 years ago, possibly as old as 194,000 years ago), and Greece (210,000 years ago) point to earlier excursions. These earlier migrations may have contributed to the genetic makeup of modern populations in Papua New Guinea and might explain evidence of gene flow from *Homo sapiens* into Neanderthals before 130,000 years ago, possibly as early as 250,000 years ago. Previous attempts to use paleoclimate reconstructions to understand these migrations have limitations: either relying on insufficiently detailed empirical data or on model-based data from too few time points. Existing quantitative attempts that link demographic dispersal models with archaeological or genetic data suffer from unrealistic biological assumptions. This study distinguishes itself from previous work by combining high-resolution climate simulations with empirically grounded estimates of hunter-gatherer climate tolerance.

The study uses a recently developed emulator of the HadCM3 global circulation model to generate high-resolution (1,000-year intervals) paleoclimate simulations for the last 300,000 years. These data are downscaled to ~0.5° resolution and bias-corrected to match observed conditions. Simulations of annual paleoclimatic variability further refine temporal resolution to the decadal scale. Two key climate variables—annual precipitation and aridity (using the Köppen aridity index)—are considered. Two migration routes are analyzed: the northern route (Nile-Sinai-Land Bridge) and the southern route (Strait of Bab-el-Mandeb). For each decade, the minimum annual precipitation required for a connected migration path is determined for each route. This is done by considering all possible paths and identifying the lowest precipitation level allowing a continuous path from Africa to Eurasia. An analogous analysis is performed using aridity. The Nile delta is assumed crossable at all times, while the crossability of the Bab-el-Mandeb strait depends on sea level. The results yield the precipitation (and aridity) tolerance required for each decade, indicating the climatic feasibility of migration. To estimate the actual tolerance thresholds of early humans, the authors examine the distribution of contemporary hunter-gatherer populations in an anthropological dataset. Excluding populations near freshwater sources, a precipitation threshold of approximately 90 mm per year is identified, below which no hunter-gatherers are recorded. This threshold aligns with the minimum precipitation needed to sustain a grazer community. By comparing the required precipitation tolerances with this hunter-gatherer threshold, the study identifies climatic windows of opportunity for out-of-Africa expansions along both routes. The analyses consider both the long-term (millennial) and short-term (decadal) variability in climate. The study also compares its model outputs with long-term and high-resolution proxy data (lake levels, leaf waxes, humidity indices, δ¹⁸O records), and pollen-based reconstructions to assess the accuracy and reliability of their climate reconstructions. Additionally, simulated precipitation is compared against the timings of three dated paleolakes on the Arabian Peninsula. The model uses a bisection method to determine the critical precipitation threshold for each route and decade, identifying the minimum rainfall level that allows for a connected path for human migration.

The analysis reveals several climatic windows during the past 300,000 years when migration from Africa to Eurasia was potentially feasible. Before the last interglacial period, the northern route (Nile-Sinai-Land Bridge) was potentially passable at several intervals between 246,000 and 200,000 years ago, with intermittent periods of feasibility until 67,000 years ago. The southern route (Strait of Bab-el-Mandeb) shows extended periods of climatic feasibility, particularly before and during the last interglacial period, assuming the crossing of the strait was possible. The 90 mm annual precipitation threshold for hunter-gatherers reveals a significant climatic window between 65,000 and 30,000 years ago, coinciding with a period of low sea level, that aligns well with both the northern and southern routes. Earlier windows also align with genetic evidence of *Homo sapiens* introgression into Neanderthals between 250,000 and 130,000 years ago and with recent archaeological findings from Israel and Greece. The study finds that precipitation tolerance thresholds above 110 mm per year would have severely restricted migration along the northern route. For thresholds above 130 mm, migration would have been extremely challenging. The southern route, however, shows more feasibility for higher tolerance levels, with opportunities for migration up to 200 mm during the last interglacial period and up to 130 mm between this period and the Holocene. The simulations generally capture key qualitative features of various paleoclimate proxies used for comparison, lending credence to the simulations' reliability.

The results strongly suggest that paleoclimatic conditions significantly influenced the timing and routes of human migrations out of Africa. The identified climatic windows align with both earlier and later proposed migration events supported by archaeological and genetic data. The failure of earlier migration attempts could be due to several factors: challenging environmental conditions between periods of suitable climate, interruptions of demographic influx from Africa, and potential competition with Neanderthals and potentially other hominins. The long period of favorable climate between 65,000 and 30,000 years ago, combined with the potential for seafaring technology along the southern route, may have played a decisive role in the success of the large-scale colonization wave that began around 65,000 years ago. This suggests that favorable climatic conditions facilitated not only the initial migration but also sustained population growth and further expansion into Eurasia.

This study demonstrates the crucial role of paleoclimate in shaping human migration patterns out of Africa. The identification of multiple climatic windows, consistent with both genetic and archaeological evidence, highlights the dynamic interplay between environmental conditions and human dispersal. The study’s limitations include assumptions made regarding the hunter-gatherer precipitation threshold and the potential for seafaring technology. Future research could focus on refining these estimates and incorporating them into more detailed human dispersal models that account for population dynamics and migration behaviors. This integration of data could offer a more comprehensive understanding of human expansions beyond Africa.

The study relies on several assumptions, primarily the applicability of contemporary hunter-gatherer data to those of early modern humans. Ethnographically documented populations are not uniformly distributed globally, and their environmental conditions differ from those of northern Africa and southwest Asia. Technological differences in water storage and transport might also have impacted migration patterns. The study also assumes the crossability of the Bab-el-Mandeb strait, and the effect of short-term climatic variability on human population dynamics remains an area for further investigation. Finally, the study focuses on climatic feasibility and doesn’t explicitly model population dynamics or other non-climatic factors that could affect migration success.