The Grotto St Pond in Onehunga is an unusual, 100m diameter circular depression with steep walls, located within the One Tree Hill lava flows. It and the adjacent Grotto in Puka St are unique landforms in the Auckland region. The flat-floored depression (512m below the rim) is immersed by water and wetland vegetation. Geophysical studies indicate the presence of up to 20 m of sediment infill beneath the Pond. The upper part of the sediment fill is composed of diatomite (composed of the siliceous skeletons of microscopic algae that once lived in the open water of the Pond). This is the only known diatomite deposit in New Zealand that has accumulated in a pond within a lava flow.


The Pond was well known to early European settlers and was shown on geological maps by Heaphy (1860) and Hochstetter (1864). These two proposed different methods of formation for the depressions and this debate have continued in the geological community through to the present time. The weight of evidence favours Hochstetter’s hypothesis that both the Pond and Grotto were formed by collapse of the roof of a lava cave. I suggest that this occurred while lava was still flowing within the cave and this rafted away the thick roof material leaving behind two, unusually deep, vertically-sided depressions. Later eruptions from nearby volcanoes mantled the area in volcanic ash which possibly half-filled the depressions and blocked the Pond depression’s natural drainage. Thus an open-water pond existed for thousands of years and the algal skeletons slowly accumulated on its floor at a rate of c. 0.21mm/yr. In the 1940s-50s the Pond was drained and some of the surface diatomite was mined for use in a locally marketed abrasive cleaner called “Grotto Maid”. More recently drainage has been impeded and a shallow pond full of wetland vegetation has become established.



The Pond at 36 Grotto St, Onehunga, is a c.100 m diameter, 512m depression within basalt lava flows from One Tree Hill volcano. The flat-floored depression is immersed by water and wetland vegetation with a higher water level in winter than summer. The sides of the depression are mostly steeply sloping (30-50˚) basalt scree that has slumped into the hole from the originally sub-vertical basalt walls. The east side of the depression has the lowest rim, just 45m above the floor, and the west and northern side has a rim 1012m above the floor. 200 m to the north and slightly uphill of the Pond is a second, smaller diameter (c.50 m), 1012m deep, vertically-walled

depression known as the Puka St Grotto. Both the Pond and the Grotto depressions occur within a slightly raised, north-south oriented, surface ridge of lava flow, which drops away by 35m on the eastern side (seen in Heretaunga and Grotto Sts). To the west the lava flow surface is undulating but only slightly lower than in the vicinity of the depressions.


Basalt can only be seen in a few places in the walls of the Pond where it is jointed solid basalt typical of that found in Auckland lava flows. The more extensive exposures in the high walls of the Grotto are also jointed, dense and vesicular basalt in places with small blisters. Several irregular horizons of scoriaceous material in the Grotto’s walls signify the presence of three or more flows or flow pulses within a 10m thickness of lava. Geophysical studies indicate that the lava flow (or flows?) in the vicinity of the Grotto St Pond is c.25 m thick (Cassidy,1985, 1999), which means there is 1015m thickness of lava not exposed below the level of the Pond. We probed and sampled the floor of the Pond on the eastern side. We found there was about 1m of disturbed soil and fill overlying a minimum 2 m thickness of undisturbed, cream, weakly layered, diatomite silt. Geophysical studies suggest that

the floor of the Pond is underlain by relatively soft sediment to a depth of at least 15-20m (Cassidy, 1999). The resistivity study did not detect any evidence of basalt beneath the floor of the Pond at least in the upper 15m.


We were unable to detect positive evidence of any quarrying in the Pond depression, except for the existence of a deep drain on the southeast corner which had been dug back into the scree, presumably to drain the Pond water through joints in the solid lava on the downhill side. We cannot rule out the possibility that some minor quarrying may have occurred or that loose blocks may have been removed, particularly from the lower eastern side.


On top of the lava flow on the northeast rim of the Grotto is a capping of at least 3 m thickness of bedded volcanic ash (tuff), which must have been erupted after the flow was emplaced. This ash could have come from one or more of several surrounding, younger volcanoes – Three Kings, Mt Mangere or Mt Smart. Some of the layers contain scoria up to 2 cm diameter and rare angular basalt bombs up to 6cm diameter. It is unlikely that clasts this size could have been blasted and blown from the first two named centres and thus nearby Mt Smart (1.3 km away) was the likely source.


History of previous geological studies

Three features associated with the One Tree Hill lava flows in western Onehunga were noted by early European settlers. These were the Grotto Street Pond, Puka Street Grotto and Captain Springs freshwater springs. All three were reserved for water, scientific and education purposes by Governor Grey, when the Town Plan for Onehunga was first drawn up in 1855.  Part of Heaphy’s (1860) geological map of Auckland Volcanic Field showing labelled the Grotto St Pond.  The first written geological record of the two unusual depressions in Onehunga was by Heaphy (1860), who showed both the Pond and Grotto on his geological sketch map of the Auckland District, 1857, and labelled the southern one “Pond”. Heaphy would have known of these features prior to the arrival of famous Austrian geologist Ferdinand von Hochstetter in late 1858, and Heaphy probably brought them to Hochstetter’s attention and may have accompanied him during a field visit to the site. Hochstetter clearly visited them himself as he gave an extensive account of the features and drew a cross-section of them in his 1864 book on the Geology of New Zealand.  Hochstetter (1864) states “A phenomenon most intimately connected with the lava caves is the development of deep funnel-shaped or basin-like depressions in the ground, which are frequently found in the lava fields. The Manukau lava field affords the most interesting examples of this, especially the part of the field which seems to have been formed by lava flows from Mt Smart, in the holes east of Onehunga known as “the Pond” and “the Grotto.”


After Hochstetter left New Zealand, Heaphy and Hochstetter had a falling out, with Hochstetter claiming and Heaphy counterclaiming, that each “stole” the other’s ideas and mapping of Auckland’s volcanoes and presented it as his own observations (Mason, 2002). There is clear evidence of this dispute in Hochstetter’s last sentence on the origin of the Onehunga depressions. The argument arose after Heaphy (1860) rushed a map of the Auckland volcanic field off to England where it was presented and published, including the location of the Pond.  It is interesting that even today there is no consensus among geologists on how the depressions were formed. Perhaps this is not surprising as they have attracted very little attention from the geological community since Hochstetter’s description (Fleming, 1959). I can find no mention of the Pond and Grotto in many subsequent geological studies on Auckland’s volcanoes (e.g. Searle, 1964;Kermode, 1992).


In the 1980s and 1990s, Dr John Cassidy of the University of Auckland, undertook two geophysical studies on the lava flow field in this vicinity. The first (Cassidy, 1985) commissioned by Onehunga Borough Council used gravity data to investigate the freshwater aquifer in the base of the lava flows. In the vicinity of the Pond his study indicated a thickness of 25 ± 5 m of basalt lava flow in the vicinity of the Grotto St Pond with a slight thickening attributed to the slightly raised ridge of basalt along the line of the Grotto Pond.


The second study (Cassidy, 1999) was commissioned by the Heritage Division of Auckland City Council who wished to learn “more about the nature and origin” of the Pond and Grotto. A resistivity survey was undertaken in the floor and to the east of the Grotto St Pond during December 1998. From this it was concluded that:

“The floor of the “Pond” depression consists of sediments, some possibly of Pleistocene age, up to a depth of at least 20 m; there is no evidence for basalt material at shallow depth.  Total basalt flow thickness in the vicinity of the “Pond” is about 25 m, thickening slightly towards the “Grotto”.


A collapse origin for the “Pond” and “Grotto” depressions is unlikely and a possible explanation for these features is that they are ‘flow-around’ structures marking the location of ancient topographical knolls.”  The diatomite in the floor of the Pond that was mentioned by Hochstetter (1864) was mined by the owner in the 1940s and early 1950s and used to make a polishing powder called “Grotto Maid”, rather similar to Chemico. It was packaged in flat-topped containers and sold through shops and hawked off locally around Onehunga. Modifications within the Pond depression appear to relate to this period of time, when the owner obtained permits for the use of explosives to assist with his operation. A 1m high concrete wall around the inside perimeter of the floor of the Pond on the north and east side may have been built as a dam to stop water flowing out of the lava flow and into the excavations. A “deep” drain in the southwest corner of the Pond may have also been dug at this time to help dewater the diatomite diggings.



Diatomite is the name of a sedimentary rock composed largely of the siliceous (SiO2), microscopic skeletons of diatom algae. Diatoms live in marine environments, freshwater and soil, some in the plankton and others on the sediment surface. The diatomite in the Pond is composed of freshwater diatoms and the relative purity of

the deposit indicates the absence of wetland vegetation during its accumulation and the presence of a persistent open body of freshwater – a Pond or small lake. High quality diatomite has a number of commercial uses which include as an abrasive cleaner, as a filter medium for wine and as an absorbent kitty litter.


The largest diatomite deposit in New Zealand is Oamaru Diatomite, an Eocene bed of marine origin. Smaller freshwater diatomite deposits are known from the floor of old volcanic crater lakes at Middlemarch (Otago) and Mercer (Waikato) and the floor of a Pond, dammed by a Mt Eden lava flow at Morningside (Auckland).


Inferred origin of the depressions

There are several possible origins for steep-sided depressions within basalt lava flows, such as the Grotto St Pond and Puka St Grotto:

a. Human quarrying activities.

b. Lava flow around a steep-sided topographical knoll and subsequent erosion away of the soft knoll.

c. Small eruption crater.

d. Roof collapse into a lava cave.


a. Quarrying can be excluded as the main cause of the depressions, as the presence of the Grotto and Pond were noted soon after the arrival of Europeans and before there would have been any need to quarry vertical holes into the surface of lava flows in paddocks on the outskirts of Onehunga. The presence of thick deposits of diatomite in the Pond also proves that the feature was not formed by quarrying.  Some limited quarrying around the edges of the Pond could have gone on, but this probably was limited to removal of loose blocks around the foot of the cliffs, especially on the lower eastern side.


b. The mode of origin suggested by Cassidy (1999) is obviously possible but would be highly unusual and requires most unusual circumstances to erode away and remove all visible trace of the soft sedimentary knolls within the depressions. The edges of the knolls would have been baked and hard and one might expect to see some remains, especially on the walls of the Grotto. Having two closely associated knolls would be even more unusual. If the knolls were soft enough to be washed away within the solidified lava flow, it would be most likely that the substantial lava flow itself would have eroded or carried the knolls away itself. I discard this hypothesised origin as highly unlikely and requiring considerably more positive evidence in its favour before it should be adopted.


c. Small eruptions are possible from within or beneath basalt lava flows, although two “craters” so close together and of similar shape, unique for anywhere in the Auckland field, is again an unusual coincidence. These craters are not consistent with lava fountaining eruptions from within the flow or beneath it, which would have built small spatter or scoria cones (e.g. Te Pouhawaiki from beneath Mt Eden lava flows).  An open crater with steep-sided lava flow walls could be produced by a substantial hydrothermal explosive eruption generated beneath the flow. The disrupted basalt blocks from the eruption would have landed all around the crater building up a low surrounding mound and some would have landed back in the crater. There is little evidence for this although they could have been removed in European times. However Hochstetter did not comment upon their presence in 1859

which suggests no mound was present. When a building platform was created in 2006 on the north side of the Grotto, a few large basalt blocks were uncovered, but their origin is unclear. They could have been from the surface of the flow or they could have been from an area that was slightly quarried or had collapsed, as the

northern side does appear to have been disturbed and is not as steep as the other three sides.  There is clearly no evidence of ejected material surrounding the Pond today, that would provide strong support for an eruptive origin and clearly a lot of large blocky material would have been ejected. Again it could have been removed in European times, but Hochstetter would surely have noted it. I conclude that eruption craters are not the likely explanation for these depressions.


d. Current evidence most strongly supports Hochstetter’s original contention that the Pond and Grotto were produced by lava cave roof collapses. The downhill alignment of the two depressions, also in line with a linear surface depression reported by Hochstetter (1864), strongly suggests that their origin was related to an internal

stream of molten lava that flowed from One Tree Hill Volcano and for some time fed lava flows further downslope around west Onehunga’s foreshore.  Most lava cave roof collapses in Auckland have occurred where the thickness of overlying basalt roof was less than 2m, and the presence of debris on the cave floor beneath attests to the fact that the collapse occurred sometime (maybe thousands of years) after the flow had cooled. Many of these roof collapses are small, but where a larger area of roof has collapsed an irregular elongate depression or “trench” has formed (e.g. Edenvale Depression). The sides of these depressions are usually steeply sloping as a result of build-up of collapsed roof and seldom have vertical cliffs like those seen at Onehunga.  I suggest that the unusual “cliffed” walls of the Pond and Grotto are probably a consequence of the much thicker roof of cooled lava that collapsed into the underlying cave. The collapsed roof material did not accumulate against these cliffs.  The evidence that there is no basalt lava rock for at least 20m down beneath the floor of the Pond (Cassidy, 1999) suggests that perhaps the roof collapsed while the internal cave still contained flowing lava and that the roof material was rafted away through the tube.  Once again the Onehunga location is unusual in that it would appear that the lava cave, the roof of which collapsed, was near the base of a succession of lava flows or flow lobes and not just below the surface. This would imply that the lava cave remained open (or active) within the lower flow while additional lava tongues flowed over it, cooled and solidified. Subsequently this lava cave became the conduit for more surges of lava feeding flows further downslope and it was during these later surges that the roof collapsed, and presumably slowly enlarged and was rafted away, as a single 100 m span of roof over a lava cave is untenable.

Although there is anecdotal evidence of a lava cave joining the Grotto and Pond being used by children many decades ago, it is not accessible today. An overhang at the south side of the Grotto (mostly filled with rubbish debris) appears to lead into a dark void that could suggest the entrance to a lava cave. Geophysical evidence suggests that there is a further 1015m thickness of lava flow beneath the present floor levels of the two depressions and thus a substantial lava cave link or its remnants could be present at a lower level.


Inferred origin of the Pond

Basalt lava flows are usually full of numerous cooling joints and fractures which make them good aquifers and thus natural Ponds or lakes of water are not usually formed in depressions within them. In basaltic volcanic fields such as Auckland, lakes are mainly found filling explosion craters (e.g. Lake Pupuke) and sometimes partly filling valleys that have been dammed by lava flows (e.g. Lake Waiatarua).  For a lake to have formed in the Pond depression, the joints in the surrounding lava flow must have been blocked up. The presence of a thick deposit of tuff overlying the flow indicates that a minimum of 3 m of ash must have fallen into each depression and it is highly likely that much more washed in from the surrounding land and possibly washed down through the joints and the presumed lava cave. This water saturated volcanic ash could be expected to have relatively low resistivity consistent with the geophysical measurements of Cassidy (1999).  Cassidy did not consider this option, probably because he was unaware of the tuff cover over the basalt flow.  I hypothesise that the volcanic ash may have half filled the depressions and dammed up any lava cave and eventually the joints in the lava flow. The Grotto,

being uphill, did not become completely choked with ash but obviously the Pond depression did and a standing body of water was created and appears to have existed for thousands of years, as diatomite accumulates extremely slowly.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: