Late Holocene climate variability in the North Atlantic realm from paleo data and climate simulations

• Main Author: Nilsen, Tine
• Format: Master Thesis
• Language: English
• Published: University of Tromsø 2013
• Subjects: VDP::Matematikk og Naturvitenskap: 400::Matematikk: 410::Anvendt matematikk: 413; VDP::Mathematics and natural science: 400::Mathematics: 410::Applied mathematics: 413; VDP::Matematikk og Naturvitenskap: 400::Fysikk: 430; VDP::Mathematics and natural science: 400::Physics: 430; VDP::Matematikk og Naturvitenskap: 400::Geofag: 450::Andre geofag: 469; VDP::Mathematics and natural science: 400::Geosciences: 450::Other geosciences: 469; EOM-3901; Simu; North Atlantic
• Online Access: https://hdl.handle.net/10037/5337

This thesis presents the results of the comparative analysis of surface temperature series from paleoproxy reconstructions and climate simu- lations. The proxy to model comparison is done using the results from 12 simulations produced by 5 different climate models covering the pe- riod of the late Holocene. Agreement between the model results and the paleoproxy reonstrcuctions is analyzed both on the local/regional scale using the paleoproxy series from two marine sediment cores from the northern North Atlantic, and globally with the available multi- proxy reconstruction of past Northern Hemispheric temperature. Long term temperature trends, long range memory properties, magnitude of variability at different time scales as inferred from series spectral properties quantify the capability of the models to simulate past cli- mate. Persistence on time scales up to a few hundred years is found for the Hemispheric temperature reconstruction under study, and some of the simulated data sets. Further, linear trend estimates in sea surface temperature records from two regions in the North Atlantic suggest an- tiphased linear trends for the reconstructed temperature data, which is only reproduced in one of the model experiments. When studying two known temperature anomalies, the medieval warm period and the little ice age, the anomalies are detected in both paleoproxy and simulated temperature time series.