Blog 9: Poukawa: The Dig That Keeps the Moa Mystery Alive

Kerry Paul
May 4
5 min read

Updated: Jun 30

JOURNEY 1 - SERIES 4 - BLOG 9 - A PART OF 10 BLOGS IN SERIES 4. - Reading time: 5 Mins 30 Secs

Poukawa Moa Excavation

This blog continues our examination of archaeological features in New Zealand that may preserve evidence older than the last ~800 years. Our case study here is the Poukawa Moa Excavation near Lake Poukawa, Hawke’s Bay—an excavation that has long sat at the centre of debates about how early people may have been present in New Zealand, and what counts as convincing archaeological proof. This discussion adds another dimension to New Zealand’s origin story. These migration patterns contribute to broader debates surrounding Southeast Asian migration to New Zealand and are frequently compared with established interpretations of Polynesian migration to New Zealand. Together, these competing perspectives continue to shape ongoing discussions about New Zealand’s first settlers.

Poukawa matters because it appears to combine three things rarely found together: unusual moa bone deposits, stone tools, and stratigraphic markers in the form of dated volcanic ash layers. If the finds truly lie beneath those ash layers in undisturbed ground, then the site would imply activity that predates well-known eruptions—pushing the timeline earlier than the conventional late-thirteenth-century settlement model.

Location and the 1930 discovery

In 1930, Russell Price, a professional surveyor, was working at Poukawa (about 20 km south of Hastings) on a drainage scheme: a canal outflow for Lake Poukawa intended to provide flood protection and reclaim surrounding land for pasture. During digging, workers encountered a striking deposit: moa leg bones—apparently upright in the soil—with no torsos or heads present.

Richard Price Excavation Site, Poukawa, Hawkes Bay 1965

Investigations (1956–1964)

Price suspected the deposit might reflect human activity—possibly moa hunting or processing—and returned to the site to test that idea. He carried out exploratory digging in 1956, followed by a more detailed excavation in 1962. He reported his findings to the New Zealand Archaeological Association in 1963, with support from scientists associated with the University of Auckland and the DSIR.

The excavations produced broken moa bones and stone tools, including pieces made from black stone and red quartz—materials said to be non-local. The most important claim, however, concerned stratigraphy. Two volcanic tephra (ash) layers were identified in the deposits: an upper ash attributed to the Taupō eruption (c. 186 CE), and a lower ash attributed to the Waimihia eruption (often dated to around c. 1400 BCE). In principle, tephra layers can act like time-stamps: if artefacts and charcoal lie beneath an intact ash layer, they must be older than the eruption—provided the layer has not been disturbed or reworked.

Dating claims and why they were disputed

In 1964, Allan Pullar, a DSIR soil scientist, joined Price. On the basis of charcoal and pollen sampling from deposits reported to lie beneath the tephra layers, Price and Pullar argued that human activity at Poukawa could extend back at least 3,300 years. The claim was controversial and, over time, it was widely treated as unproven or unlikely. The key issue was not simply the presence of charcoal or tools, but whether the site’s layers were securely sealed and undisturbed. In settings like drained wetlands and canal cuts, materials can be moved by water, roots, burrowing, and reworking of sediments—processes that can place older and younger items in misleading positions. For Poukawa to demonstrate very early human presence, the evidence would need to show (1) intact tephra layers, (2) clear, recorded associations between artefacts/charcoal and the layer beneath, and (3) a plausible explanation for how the moa deposit formed.

Even leaving the technical dispute aside, Poukawa connects to a wider puzzle often raised in settlement debates: the speed of moa extinction after the accepted window for first settlement. Several practical observations can be cited:

Estimates of the moa population around c. 1300 CE range widely (figures from tens of thousands to millions are commonly cited in secondary sources).
By about c. 1440 CE (roughly 150 years later), all nine moa species are generally considered extinct.
If initial settlement involved relatively small founding groups arriving in either ships or waka, it raises questions about how quickly people could spread, establish communities across both islands, and exert sustained hunting pressure nationwide.
If moa were eliminated within a short window, one explanation is that substantial human activity was already widespread by the early 1300s—whether through rapid expansion, multiple settlement waves, or earlier, less-visible occupation.

So what might explain a rapid extinction?

The apparent 1300–1440 CE window can be explained in several ways. Poukawa is often cited because it forces us to think carefully about which explanation best fits the evidence.

If moa numbers were still large around 1300 CE and founding populations were initially small, then why did moa disappear so quickly? Possibilities include:

Intensive, localised overhunting of slow-breeding birds, amplified as people rapidly expanded into new territories.
Habitat change (burning and forest clearance) that reduced moa food sources and breeding habitat alongside hunting.
Multiple settlement arrivals or more frequent early coastal movement than the “single arrival” story suggests, increasing pressure sooner.
Earlier, low-density presence that left limited traces—meaning the clock may have started earlier than the best-known sites imply.
Uncertainty in the estimates: moa population sizes, extinction timing, and regional variation are all subject to wide ranges and uneven evidence.

What Poukawa ultimately highlights is the importance of secure excavation processes. At the time of Price’s work, New Zealand had limited archaeological resources, which restricted opportunities to independently validate his stratigraphic observations or to undertake comparable excavations elsewhere that might have provided corroborating evidence. Nevertheless, Poukawa remains a landmark excavation and a valuable single-site case study—showing how debates about New Zealand’s settlement timeline can hinge on the quality of field recording, the security of stratigraphic context, and the ability (or inability) to replicate results across multiple sites.

Key Takeaways

Poukawa is significant because it links tools, moa bone, and volcanic ash layers: if artefacts and charcoal are truly sealed beneath intact tephra, the site could imply human activity earlier than the conventional late-1200s settlement model.
The controversy is mainly about context, not the existence of finds: in wetland and canal settings, reworking and disturbance can place materials in misleading positions, so secure stratigraphy and careful recording are essential. Price worked with professional archaeologists during the excavation, supporting the validity of the methods used.
The site feeds into the wider “speed of moa extinction” debate: the tight window between accepted settlement and moa extinction raises questions about rates of expansion, intensity of hunting/habitat change, and whether earlier or multiple waves of movement are plausible.
The bigger lesson is methodological: Poukawa shows how strongly national timelines can depend on excavation quality, and why replication and corroboration across multiple sites are critical.