Herding then farming in the Nile Delta

The study addresses long-standing questions about who the earliest settlers in the Nile Delta were, when they arrived, whether their livelihoods centered on herding or farming, and how any transition unfolded. The Nile Delta, fed by rivers originating in humid East African highlands, was pivotal to Egypt’s political and social development and is central to debates on early human migration and agricultural origins across the Mediterranean and SW Asia. Despite prior geoarchaeological work, the drivers and timing of cultural transitions remain unclear. The authors investigate these issues at Sais (Sa el-Hagar) in the west-central delta—one of the earliest Neolithic sites (ca. 6.5–6.2 ka)—by analyzing an 8.5 m sediment core (SH-1) spanning natural and cultural layers, using non-pollen palynomorphs (NPP) and pollen to reconstruct early livelihood patterns and their transitions.

The paper situates its work within debates on early migration and the origins of agriculture in the lower Nile valley and broader Mediterranean/SW Asia. Prior studies noted the importance of the Nile Delta in Predynastic developments and discussed dispersal of Southwest Asian domesticates, but with unresolved questions on timing and subsistence strategies. The authors highlight that few NPP-based geoarchaeological studies exist for the Nile Delta, creating a gap their multidisciplinary approach seeks to fill. They also reference debates over distinguishing cereal versus wild grass pollen based on grain size and note prior regional findings on early domesticates and climate influences on settlement.

Study site and core: At Sais (30°58′05″N, 30°45′56″E), an 8.50 m core (SH-1) was extracted from the site margin. Stratigraphy includes a basal Early Holocene sandy levee (Gezira) below 8.25 m, a natural gray silty clay–yellow fine sand unit (8.25–6.70 m), a thicker muddy cultural layer with pottery/sherds (6.70–1.70 m), and an agriculturally disturbed layer (1.70–0 m; not analyzed). Age model: Twenty-two AMS radiocarbon dates (charcoal and organic mud) were measured (Beta Laboratory, USA; Institute of Earth Environment, CAS, Xi’an). Ages were calibrated at 95.4% confidence using Calib 8.1.0 with IntCal20 datasets. An age–depth model at 10 cm intervals was constructed using the rbacon package in R, results expressed in ka. NPP and pollen analysis: A total of 116 samples at 5 cm spacing were processed. Approximately 10 g dry sediment per sample was treated (10% HCl, 40% HF, 10% KOH), spiked with Lycopodium tablets (27,637 spores/tablet) for concentration calculations, and sieved at 10 µm with ultrasonic treatment. NPP identification followed published descriptions; counts were made with a Nikon Ci-L microscope at 400×. On average, 220 (up to 400) algae and fungal spores were identified per sample. An NPP spectrum was generated using stratigraphically constrained cluster analysis (CONISS). Pollen grain sizes for Poaceae were counted in classes (35–37 µm, 37–40 µm, >40 µm); Poaceae <35 µm and charcoal data (≥100 µm) were from prior work. NPP were grouped into indicator types: humid (e.g., Concentricystes, Mougeotia, fern Pteris sp., HdV-8 fungi), arid (e.g., Pleospora and UG types), erosion (e.g., Glomus, HdV-200), animal dung indicators (Sodaria, Cercophora, Coniochaeta ligniaria), and fiber indicators (Linum and animal hair).

• NPP Zone I (natural layer) before ~7.0 ka shows no evidence of intense land use, consistent with absence of >7.0 ka archaeological sites in the Nile Delta.
• In NPP Zone II (cultural layer), animal dung and hair microfossils rise sharply beginning at ~7.0 ka, indicating the presence of stock animals and systematic pastoralism. Sporadic dung/hair in Zone I suggest indigenous wild animals and possible mixed hunting-husbandry.
• Domesticated cereals appear later: larger Poaceae pollen classes (35–37 µm, 37–40 µm, and >40 µm) emerge abruptly from ~6.7 ka (Zone III). The weak correlation between Poaceae <35 µm (wild grasses) and >35–40 µm classes suggests the larger grains represent domesticated cereals likely imported by migrants, aligning with the existence of domesticates in SW Asia/east Sahara prior to 6.7 ka.
• Timing relationship: herding predates farming at Sais by approximately 300 years (animal dung/hair at ~7.0 ka; cereal pollen at ~6.7 ka).
• Through Zones III–VII, fluctuations in animal microfossils and cereal pollen indicate ongoing husbandry alongside increasing cropping. Proportional changes suggest herding waned as cropping intensified.
• Evidence for fire-enhanced land exploitation and increased soil disturbance after ~6.7 ka includes increasing micro-charcoal (≥100 µm) and NPP erosion indicators (e.g., Glomus) accompanying cereal pollen.
• Linum (flax) and animal hair occur with patterns similar to animal microfossils, indicating fiber use and human activities; sparse Linum before 7.0 ka may reflect local use prior to intensified settlement.
• Paleoenvironmental indicators show wetter conditions prior to 7.0 ka followed by drying, with implications for habitability: high floods likely inhibited early occupation; short-term drying around 7.0 ka coincides with the onset of herding; cropping begins when conditions became wetter again (Zone III).

The high-resolution NPP and pollen record from Sais directly addresses the question of early livelihoods in the Nile Delta. The abundance of dung and hair microfossils around 7.0 ka indicates that the earliest settlers were herders. The subsequent appearance of larger Poaceae pollen (≥35 µm) at ~6.7 ka points to the introduction of domesticated cereals and a transition to mixed herding and farming. The decoupling of cereal-sized pollen from wild grass pollen supports the interpretation that domesticates were imported rather than locally domesticated, in line with broader evidence for Southwest Asian domesticates moving into the lower Nile. Increasing charcoal and erosion signals after 6.7 ka suggest fire-assisted land clearance and intensified agriculture as populations grew and land use intensified. Climatic shifts—from wetter conditions and flood-prone environments before 7.0 ka to short-term drying enabling herding, followed by conditions conducive to cropping—likely modulated the sequence and balance of pastoralism and cultivation. Together, these findings provide a coherent narrative in which pastoralism precedes farming in the west-central Nile Delta and evolves into a mixed subsistence strategy.

This study provides high-resolution biological evidence (NPP and pollen) from the Sais site showing that herding began by ~7.0 ka and farming (domesticated cereals) appeared by ~6.7 ka, indicating herding preceded farming by roughly 300 years. It further documents intensified, fire-assisted land use and soil disturbance accompanying the rise of cropping, and supports hypotheses that domesticates were introduced from SW Asia. Future research directions are not explicitly discussed.

• Few prior NPP-based geoarchaeological studies exist for the Nile Delta, limiting regional comparative context.
• The agricultural surface layer (1.70–0 m) was disturbed and excluded from analysis, potentially omitting later phases of land use.
• The use of Poaceae pollen size as a proxy for domesticated cereals is debated; while applied cautiously here, it introduces uncertainty.
• Interpretation of climatic phases relies partly on proxy inferences and integration with previous studies, which may carry chronological and proxy-resolution uncertainties.