Discontinuous Climate Changes in Japan after 1900

Abstract

Because the statistical hypothesis test (the Lepage test) has successfully shown the regional abrupt changes in precipitation in Japan after 1900, changes of the temperature and sea-level air pressure are studied by means of the same method.The analysis shows that the temperature and sea-level pressure also presented discontinuous changes. Also, it is noted that the changes in the temperature were not fictitious ones caused by changes in observational methods, instrument changes or other factors.Referring to the previous study on the changes in precipitation, the climate changes which occurred in Japan after 1900 can be summarized as follows: (1) Changes detected were discontinuous ones and trends of temperature did not cause any significant changes. (2) Very noticeable abrupt warmings with a magnitude exceeding 1* of the seasonal mean temperature occurred in 1914 and 1949, and the precipitation amount increased in the successive seasons. (3) The seasonal mean temperature also changed in 1957 and 1967. These recent changes were discontinuous ones, but were neither abrupt warmings nor abrupt coolings. The mean temperature for August-October fluctuated more after 1957 and for April-July it fluctuated less after 1967. (4) Cooling trends were noticed during periods between discontinuous changes but they did not cause any definite changes to be detected by the Lepage test.(5) The sea-level air pressure changed three times. Though the quality of the sea-level air pressure records is questionable, it may be worthy to note that one of the changes occurred in 1924 and in 1924 the annual precipitation amount decreased.The discontinuous changes support the idea that the climate system is a chaotic spontaneous dynamic system. Mechanisms of the regional discontinuous changes and the relations between the very noticeable abrupt warmings and the concentration of greenhouse effect gases should be the subjects of future studies.

Detection of Long Term Trend, Cyclic Variation and Step-like Change by the Lepage Test

“…The purpose of this short report is to show the differences in the results of the tests applied to time series that present different changes in types by means of statistical simulation. Therefore, this is also supplementary to the previous papers (Yonetani, 1992a(Yonetani, , 1992b) that showed the domination of discontinuous changes in climate.…”

“…In the papers where the domination of abrupt changes of precipitation and temperature were shown by means of statistical analysis using the Lepage test (Yonetani, 1992a(Yonetani, , 1992b, changes were decided to be abrupt ones by the following facts: One was that the Lepage statistic sharply increased in value and indicated the occurrence of change with high probability around years when abrupt shifts of averaged conditions were recognized. The other fact is that the numbers of the stations where the test detected changes with the 1 % significance level sharply increased around certain years when the sample size was proper.…”

Abrupt changes in regional temperature in the conterminous United States, 1895-1989

“…Recent studies of trends in global temperature suggest a n increase in global annual mean temperature of 0.3 to 0.6OC over the past century (Intergovernmental Panel on Climate Change 1990Change , 1992. On a regional scale, however, this trend is not consistent or even apparent for many regions of the world (Diaz & Quayle 1980, McGuirk 1982, Pittock 1983, Karl & Riebsame 1984, Yamamoto et al 1986, Balling & Idso 1989, Yonetani 1992. The methods used to evaluate trends in temperature, resulting from increasing atmospheric concentrations of greenhouse gases, generally assume that changes in temperature are linear through time (Balling & Idso 1989, Plantico et al 1990).…”

A nonparametric location–scale statistic for detecting a change point

“…To resolve the dilemma, Lepage [15] developed the test statistic that combined the Wilcoxon and the Ansari-Bradley [16] statistics. Yonetani [17][18][19] and Nakazawa and Rajendran [20] studied to detect the changes by using the Lepage test. After the Lepage test was proposed, various Lepagetype statistics have been proposed and discussed over the course of many years, e.g., Büning and Thaewald [21], Neuhäuser [22], Büning [23], Murakami [24].…”