Download Earth Surface Temperature Changes above Latitude 45 Degrees

Doi: 10.12982/cmujns.2014.0034
➔ CMU J. Nat. Sci. (2014) Vol. 13(3)
247
Earth Surface Temperature Changes above
Latitude 45 Degrees North from 1973 to 2008
Wandee Wanishsakpong1*, Kehui Luo2 and Phattrawan Tongkumchum1
1
Department of Mathematics and Computer Science, Faculty of Science and
Technology,
Prince of Songkla University, Pattani Campus, Pattani 94000, Thailand.
2
Department of Statistics, Macquarie University, NSW 2109, Australia.
*Corresponding author. E-mail: [email protected]
ABSTRACT
In this study, we examined monthly temperature variation from 1973 to
2008 on grid regions of the earth surface above latitude 45°North, covering the
Arctic Ocean, northern areas of the Atlantic and Pacific Oceans, and the Asian
and European continent. Linear modelling was used to investigate the trends
and patterns of temperature changes and account for auto-correlations of the
temperature changes over time. Factor analysis was then used to model filtered
residuals, i.e., the residuals after removing time trends and auto-correlation,
providing a basis for identifying and classifying regions with similar temperature
change. Twelve large regions, each having similar temperature change patterns,
were identified. Of the 69 sub-regions considered in the study, 64 sub-regions
experienced significant increase in temperature, 2 sub- regions had insufficient
data, and only 3 sub-regions remained unchanged. High temperature increases
(0.200°C to 0.320°C) occurred in the North Pacific Ocean, Alaska and Eastern
Siberia. Moderate temperature increases (0.130°C to 0.199°C) occurred in north
Canada, Greenland, Iceland, Norway, Sweden and Finland. The north of Siberia
and part of the North Atlantic had low increases (0.090°C to 0.129°C) while
northeast Canada and its surrounding seas did not show evidence of warming.
Keywords: Latitude, Climate change, Time series analysis, Correlation, Auto-correlation, Factor analysis.
METEOROLOGICAL APPLICATIONS
Meteorol. Appl. 23: 115–122 (2016)
Published online 25 November 2015 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/met.1536
Modelling of daily maximum temperatures over Australia from 1970 to 2012
Wandee Wanishsakpong* and Nittaya McNeil
Department of Mathematics and Computer Science, Faculty of Science and Technology, Prince of Songkla University, Pattani, Thailand
ABSTRACT: The variations of daily maximum temperatures over consecutive 5 day periods from 1970 to 2012 in Australia
are described. The data were obtained from the Australia Bureau of Meteorology, comprising measurements from 85 weather
stations. Linear regression was used initially to model the seasonally adjusted daily maximum temperatures. Factor analysis
was used to classify the temperatures from the 85 stations into seven factors which correspond to seven geographical regions.
The average maximum annual temperature in these seven regions ranged from 23 to 36 ∘ C. A 6th order polynomial regression
model was then fitted to investigate the trend and pattern of the temperatures in these seven regions. Similar increasing
temperature trends occurred in the central, eastern, southern and southeastern parts of the country.
KEY WORDS
Australia; climate change; time series analysis; linear regression; factor analysis; polynomial regression model
Received 11 June 2014; Revised 18 May 2015; Accepted 12 July 2015
ATMOSPHERIC SCIENCE LETTERS
Atmos. Sci. Let. 17: 378–383 (2016)
Published online 7 June 2016 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/asl.668
Trend and pattern classification of surface air
temperature change in the Arctic region
Wandee Wanishsakpong,1,2 Nittaya McNeil1,2,* and Khairil A. Notodiputro3
1 Department
of Mathematics and Computer Science, Faculty of Science and Technology, Prince of Songkla University, Pattani, Thailand
Centre of Excellence in Mathematics, Commission on Higher Education, Bangkok, Thailand
3
Department of Statistics, Faculty of Mathematics and Science, Bogor Agricultural University, Indonesia
2
*Correspondence to:
N. McNeil, Department of
Mathematics and Computer
Science, Faculty of Science and
Technology, Prince of Songkla
University, Rusamilae, Pattani
Campus, Pattani 94000,
Thailand.
E-mail: [email protected]
Received: 17 July 2015
Revised: 26 April 2016
Accepted: 29 April 2016
Abstract
Monthly seasonally adjusted temperatures above latitude 45∘ N were investigated from January 1973 to November 2013. The study area was divided into 69 sub-regions of similar size
each in the shape of an igloo brick. The data were filtered with a second-order autoregressive process to remove autocorrelation. Two sub-regions did not have sufficient data due to
substantial numbers of missing values. Factor analysis was then applied to the remaining 67
sub-regions and was used to classify regions with similar temperature changes. As a result, 63
sub-regions could be classified based on 12 factors but 4 sub-regions could not be grouped due
to uniqueness. The temperatures for each group of sub-regions were found to increase during 1973–2013. The largest temperature increases of 0.19 ∘ C/decade were found in northern
and southern Siberia and part of the Arctic Ocean. In northern Canada, Alaska, the northern
Pacific Ocean and eastern Siberia the temperatures increased by at least 0.16 ∘ C/decade. In
Iceland, Norway, Sweden and part of the Pacific and Arctic Oceans the temperature increased
by around 0.15 ∘ C/decade. In northeastern Canada, Greenland and its surrounding Atlantic
Ocean and the Arctic Ocean the temperature increased by about 0.15 ∘ C/decade.
Keywords: Arctic; climate change; Time series analysis; autocorrelation; linear regression
model; factor analysis