Details for Time series analysis of long-term data sets of atmospheric mercury concentrations

PropertyValue
Name:Time series analysis of long-term data sets of atmospheric mercury concentrations
Description:

Temme C, Ebinghaus R, Einax JW, Steffen A, Schroeder WH. Time series analysis of long-term data sets of atmospheric mercury concentrations. Anal Bioanal Chem. 2004; 380:49350


Abstract


Different aspects and techniques of time series analysis were used to investigate long-term data sets of atmospheric mercury in the Northern Hemisphere. Two perennial time series from different latitudes with different seasonal behaviour were chosen: first, Mace Head on the west coast of Ireland (53°20′N, 9°54′W), representing Northern Hemispherical background conditions in Europe with no indications for so-called atmospheric mercury depletion events (AMDEs); and second, Alert, Canada (82°28′N, 62°30′W), showing strong AMDEs during Arctic springtime. Possible trends were extracted and forecasts were performed by using seasonal decomposition procedures, autoregressive integrated moving average (ARIMA) methods and exponential smoothing (ES) techniques. The application of time series analysis to environmental data is shown in respect of atmospheric long-term data sets, and selected advantages are discussed. Both time series have not shown any statistically significant temporal trend in the gaseous elemental mercury (GEM) concentrations since 1995, representing low Northern Hemispherical background concentrations of 1.72±0.09 ng m−3 (Mace Head) and 1.55±0.18 ng m−3 (Alert), respectively. The annual forecasts for the GEM concentrations in 2001 at Alert by two different techniques were in good agreement with the measured concentrations for this year.

 

Filename:2004-Time series analysis of long-term data sets of atmospheric mercury concentrations.pdf
Filesize: 431.28 kB
Filetype:pdf (Mime Type: application/pdf)
Creator:Unknown
Created On: 08/05/2004 12:14
Viewers:Everybody
Maintained by:Editor
Hits:2052 Hits
Last updated on: 07/23/2009 18:25
Homepage: