Abstract

Variability of the North Atlantic region is a key factor in the workings of global climate. Changes related to orbital factors, and to the thermohaline circulation (THC), occur on timescales ranging from 1000 to 100,000 years, whereas subdecadal, decadal and multidecadal trends are associated with changes in atmospheric circulation, sea surface temperature (SST), and salinity. Long-term changes are known only from proxies, whereas shorter periodicities are identified in instrumental and other observational data. Between lies a range of less well known centennial periodicities, too long to be recorded observationally, and too short for all but the highest resolution proxy records, such as tree rings, atmospheric Δ 14C, and varved lake sediments. Numerous investigations of such periodicities involve comparisons between the proxy and the observational record. However, results depend on the length of the datasets being analysed, the relative strength of the periodicities present, whether data are stationary or non-stationary, the degree to which they have been smoothed in order to remove particular fluctuations, and the methods of analysis employed. Ideally, comparisons should be between datasets of equal length and of similar properties, but these criteria are not always easy to fulfill. A search for periodicities in a range of well known multidecadal datasets reveals that whilst some frequencies previously identified in the Central England Temperature (CET) record, and in the Koch sea ice Index, are statistically significant, those found by several authors in the Index of the North Atlantic Oscillation (NAO) are not, as the index approximates to ‘white noise’. Instead, decadal periodicity of the NAO described by several authors appears to be result of smoothing of the raw data. Periodicities of 6.1, 33.9 and 213 years identified in a proxy record from laminated sediments from Loch Ness, Scotland appear, respectively to be related to (a) subdecadal variations in SST and salinity in the North Atlantic ocean, (b) multidecadal interactions between the NAO and the North Atlantic THC, and (c) a well known centennial solar periodicity.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call