The cropmark, that subtle difference in how grass or grain grows over buried ditches and features, was enough to flag the site before construction began, prompting a salvage excavation in 1998 that uncovered something genuinely puzzling: a double-ditched circular enclosure whose origins remain, in a strict archaeological sense, unknown.

The excavation, reported by O'Donnell in 1998 and 2000, revealed a roughly circular space with an internal diameter of 32 metres, defined by a U-shaped fosse, the term used for a ditch forming part of an enclosure or defensive boundary, between 2.35 and 3.66 metres wide and up to just over a metre deep. A gap on the western side of this outer ditch was read as an original entrance. Inside it sat a second, V-shaped fosse, narrower and shallower, enclosing a slightly irregular oval area roughly 25 metres by 22 metres. The two ditches were not concentric, and the space between them varied around the circuit, which gives the whole thing an improvised or perhaps phased quality. No material survived in the ditches that could fix the enclosure to any particular period. It cannot be dated.

What can be said is what happened inside the enclosure after it was already built. Evidence of metal smelting was found post-dating the original construction, including a furnace in the south-eastern quadrant, set about with stake-holes arranged in a semicircle, and a hearth at the centre of the interior surrounded by small pits and further stake-holes. A cluster of shallow pits lay just outside the enclosure to the south-east. The enclosure itself was probably chosen as a ready-made sheltered working space, its original purpose already forgotten by whoever lit the furnace. What that original purpose was, nobody can now say with certainty.