Holocene climate history of British Columbia using pollen-based climate reconstruction techniques

Using pollen data to reconstruct past climate requires testing to understand the strengths and weaknesses of methods, and to develop accurate interpretations. This thesis presents the development of a British Columbia (BC) modern pollen dataset (BCMPD) for Holocene paleoclimate reconstruction. The B...

Full description

Bibliographic Details
Main Author: Goring, Simon James Allan Elliott
Format: Thesis
Language:unknown
Published: 2011
Subjects:
Online Access:http://summit.sfu.ca/item/12178
Description
Summary:Using pollen data to reconstruct past climate requires testing to understand the strengths and weaknesses of methods, and to develop accurate interpretations. This thesis presents the development of a British Columbia (BC) modern pollen dataset (BCMPD) for Holocene paleoclimate reconstruction. The BCMPD used to test seven climate reconstruction techniques including the new Non-Metric Multidimensional Scaling / Generalized Additive Model method. The BCMPD includes assemblages from lacustrine, bog and terrestrial depositional environments. Depositional site characteristics may affect the ability of pollen-based climate models to predict climate accurately. Results show that combining samples from multiple depositional environments can increase, decrease or have no effect on model error, independent of sample size. The effects vary among model types and across climate variables. Given the variable effects of depositional environments, researchers are recommended to test the effect of mixed depositional environments prior to model calibration when using large datasets. The accuracy of pollen datasets for climate reconstruction depends on turnover rates in source vegetation. Correlations to weather station records from the lower mainland of BC show that aerial pollen, pollen from a lacustrine sediment core and the BCMPD all show different pollen-climate relationships. These differences are likely the result of interplay between individual-level physiological responses to climate and changes in regional vegetation as a response to either succession or climatic change. Pollen may be a weak predictor of rapid climate change in forests dominated by long-lived species. Climate reconstruction in BC during the past 10,000 years indicates changes in temperature and precipitation that compare well to other proxy observations. These changes can be explained by interactions between changes in solar insolation in addition to dominant weather systems such as the Aleutian Low and Pacific High. Vegetation reconstructions using pollen-based climate models show that coastal and northern boreal vegetation zones declined in area while interior-type vegetation increased in area during the last 10,000 years.