GEOSCAN Search Results: Fastlink


TitleEvaluating periodicities in peat-based climate proxy records
AuthorSwindles, G T; Patterson, R T; Roe, H M; Galloway, J M
SourceQuaternary Science Reviews vol. 41, 2012 p. 94-103,
Alt SeriesEarth Sciences Sector, Contribution Series 20120130
Mediapaper; on-line; digital
File formatpdf
AreaSlieveanorra; County Antrim; Dead Island; County Londonderry; Ireland
Lat/Long WENS -6.5500 -6.2167 55.6833 54.8833
Subjectssurficial geology/geomorphology; Holocene; peat; peatlands; climate effects; climate; spectral analyses; climate change; Cenozoic; Quaternary
Illustrationslocation maps; plots; graphs; tables
Programenvironmental impacts and adaptation in the northern environment, Environmental Geoscience
AbstractProxy records derived from ombrotrophic peatlands provide important insights into climate change over decadal to millennial timescales. We present mid- to late- Holocene humification data and testate amoebae-derived water table records from two peatlands in Northern Ireland. We examine the replication of periodicities in these proxy climate records, which have been precisely linked through tephrochronology. Age-depth models are constructed using a Bayesian piece-wise linear accumulation model and chronological errors are calculated for each profile. A Lomb-Scargle Fourier transform-based spectral analysis is used to test for statistically significant periodicities in the data. Periodicities of c. 130, 180, 260, 540 and 1160 years are present in at least one proxy record at each site. The replication of these periodicities provides persuasive evidence that they are a product of allogenic climate controls, rather than internal peatland dynamics. A technique to estimate the possible level of red-noise in the data is applied and demonstrates that the observed periodicities cannot be explained by a first-order autoregressive model. We review the periodicities in the light of those reported previously from other marine and terrestrial climate proxy archives to consider climate forcing parameters.