Developing Fossil Beetle Analysis for Past Environmental and Climate Reconstruction in New Zealand

Abstract

This thesis presents the first attempt in New Zealand to use beetle fossils as indicators for past temperatures and vegetation histories. Host specific taxa provided precise information for vegetation reconstructions, including Hypotagia Lewisi, which showed that Nothofagus was growing in the lower Awatere Valley during the last glacial maximum (LGM). This discovery of a LGM forest patch is significant because until now the eastern South Island was considered largely treeless. The Awatere Valley Holocene fossils indicate Nothofagus forest persisted at the site from the LGM to the early Holocene then was replaced by a podocarp forest in the mid-late Holocene. The late Holocene is characterised by low diversity and the absence of forest species and a loss of beetle biodiversity. The last interglaciation fossils from the southern Wairarapa identify swamp forest environments, and host specific beetle species indicate that Cordyline and Phormium species were major components of these forests. At the Banks Peninsula site, the beetle fossils identify an estuary surrounded by montane forest. The host specific beetle Pachyurinus metallicus indicates Podocarpus totara or Podocarpus halli was part of the forest. Climate reconstructions: LGM temperature estimates for the Awatere Valley (c. 21,000 - 19,000 yrs BP) show February mean temperature was about 3.5 - 4°C cooler, and mean daily minimum temperature in July was about 4 - 5 °C cooler than present day. This is consistent with LGM temperature estimates of 4 - 7 °C from other climate proxy indicators. However, earlier in the LGM (22,800 years BP), estimates for the mid-Canterbury Ranges site show mean February temperature was between 0 and 2°C cooler than present and mean daily minimum temperature in July was between 0 and 2.6°C cooler. These results are consistent with plant macrofossils data from the site. Temperature estimates for the last interglaciation are provided for southern Wairarapa. They are inferred from the modern distribution of species and indicate summer (January) temperatures were 1.6 - 2.5 °C warmer and 2.3-3.2 °C warmer in the winter (July) than present day conditions. The assemblage is characterized by eight species with modern day ranges up to 700 km north of the study site. This is the first quantification of the last interglaciation climate in New Zealand from terrestrial data and is consistent with the thermal optimum of MIS5 and with warmer conditions inferred from vegetation fossils. Temperature estimates for the penultimate glaciation are also derived from the modern distributions of species from Banks Peninsula and indicate temperature at this stage in MIS 6 was 2-3 °C cooler than present day. These estimates are consistent with pollen reconstructions from the site that suggest interstadial conditions. As well as local paleoecological and paleoclimatological reconstructions, this thesis contributes to the general development of beetle analyses by: 1) Introducing a trophic approach to paleoenvironmental reconstruction. Applied at the Awatere site, it provided a measure of stand stability indicating changing conditions at the LGM and early Holocene but a mature ecosystem in the late Holocene. 2) A 'new' model and statistical approach is presented for quantifying the New Zealand beetle data. It is a maximum likelihood and a sine distribution model, which is an appropriate approach when quantifying past climates from taxa with finite temperature distributions and when relying on the small data sets that characterise the New Zealand beetle data. In summary, this thesis demonstrates that beetle studies are not only possible in a New Zealand context, but represents one of the best avenues for paleoecological and paleoclimatological work in New Zealand.