Improved Proxy Network Resolves Apparent Temperature And Precipitation Disagreement During The LIA (AD 1400-1850)

Abstract:

The behaviour of El Nino-Southern Oscillation (ENSO) over the past millennium has been subject to much debate. A period of particular interest is the Little Ice Age (LIA; AD ~1400-1850), which was punctuated by anomalously cold temperatures at least in the Northern Hemisphere. However, there is still dispute over the dominant state of the climate over this period. Climate proxies such as corals, sediment cores and tree rings have been used to extend ENSO reconstructions prior to the instrumental period. Previously published work suggests that there is a discrepancy in the interpretation of precipitation-based and temperature-based ENSO proxies. While several precipitation proxies suggest a more La Nina-like state during the LIA, temperature proxies suggest the opposite. However, this conclusion is largely based on individual records or a combination of just a few.

We show that this apparent disagreement between temperature and precipitation trends over the last millennium can be resolved by the use of an improved, denser proxy network. Two ENSO reconstructions, derived from temperature proxies and from rainfall proxies, are constructed using an empirical orthogonal function (EOF) based method, which is validated using PMIP3 model data. Over the LIA period, the proxy reconstructions display a correlation of 0.55, both tending to a La Nina-like state. This result is corroborated by the model reconstructions, which show high correlations between temperature and rainfall both when using the entire tropical region and when using only data at real proxy locations.

One of the main remaining issues for improving ENSO reconstructions is the lack of tropical proxy records, particularly for temperature. Although the reconstructions here include nearly, if not all, available proxy records, the temperature series is still based on less than 10 records. In order to further strengthen our understanding of ENSO, an effort must be made to collect more data.