Oxygen isotopic evidence of climate variability in southern England since the Medieval Period.

Late Holocene climatic variations pre-1850 CE are associated with volcanic and solar forcing (Schurer et al., 2013).  Whilst these variations are recorded in speleothems and ice-cores, these archives are often spatially restricted leaving gaps in our knowledge about short-term climate variability in a range of regions. Here, we investigate the potential of using the high-resolution δ¹⁸O analysis of lake carbonates formed within artificially constructed water bodies dating back to the Medieval period. Whilst the isotopic analysis of lake carbonates is a well-established Quaternary palaeoclimate proxy (Leng and Marshall, 2004) it has received less attention as a tool for climate reconstruction over the historic period. In this study we use the δ¹⁸O analysis of winter calcifying ostracod species from lake sediments recovered from Medieval fishponds from the town of Alresford, in southern England, combined with a programme of monitoring within the present-day water body to establish the hydrology and thermal regime of the system. This analysis shows that over the studied interval (the end of the Medieval period through to the 20^th century) the lake system underwent regular inter-annual/decadal isotopic shift of relatively high magnitude (1-2‰).

In order to investigate whether these high magnitude δ¹⁸O fluctuations are explainable by climatic variability or are a result of intra-lake processes we provide a data-model comparison. This approach allows an understanding of the likely mechanistic drivers of climatic change as well as testing if proxy observations are consistent with modelled outputs (Evans et al., 2013). This study compares the δ¹⁸O ostracod record with a synthetic δ¹⁸O_carbonate record derived from the Millennium Data iHadCM3 runs for the period 1200 CE to 1850 CE. The iHadCM3 model generates modelled values for temperature and δ¹⁸O_{precipitation} on an annual and monthly basis. These data were used to produce a synthetic δ¹⁸O_carbonate record on both an annual and seasonal basis using Kim and O’Neil's (1997) equation that describes the relationship between temperature, δ¹⁸O_carbonate and δ¹⁸O_lakewater.

The preliminary outputs of this proxy-model output comparison demonstrate that the magnitude of δ¹⁸O_carbonate variability predicted by the model data is similar to the magnitude of change recorded in the proxy data. This suggests that these variations are real and driven by climatic rather than catchment-specific processes. Ongoing work aims to disentangle primary climate drivers of interannual δ¹⁸O change, at this site, using δ¹⁸O enabled climate model simulations. Our approach of considering what drives interannual δ¹⁸O changes over the last few hundred years, in these lacustrine settings, will help enable more robust palaeoclimatic reconstructions from these records.

References: Evans, M.N. et al., (2013), QSR, 76, pp.16–28.; Kim, S.-T. and O’Neil, J.R. (1997) Geochimica et Cosmochimica Acta, 61(16), pp.3461–3475; Leng, M.J. and Marshall, J.D. (2004) QSR, pp.811–831; Schurer, A.P. et al., (2013) Journal of Climate, 26(18), pp. 6954–6973.