Temperature Extremes in the United States
1953; Taylor & Francis; Volume: 43; Issue: 1 Linguagem: Inglês
10.2307/211537
ISSN1931-0846
Autores Tópico(s)Climate variability and models
ResumoEXTREMES of temperature, like extremes of many other climatic elements, are of continuing interest to the public as well as to meteorologists and geographers. They are often cited uncritically, without regard to accuracy or the probability of recurrence. Yet it is axiomatic that as more and more years of observation accumulate at a place, greater and greater extremes will be observed, so that the magnitude of the recorded extreme depends on the length of the record. Until recently, no valid means for comparing extremes based on different periods has been available, and observed extremes could only be tabulated without analysis. Development in the last decade of the statistical theory of extreme values has provided a means of estimating the most probable extreme value likely in any given period of time, on the basis of all the extremes that have been observed. Application of the theory to rather short series, not necessarily of equal length, for different places yields values that may be compared directly. In this paper the theory has been applied to the highest and lowest air temperatures of each year at ioo places in the United States, obtaining the extremes expected to occur once in a century. These highest and lowest temperatures, estimated on the basis of the annual extremes recorded between I90I and 1930, are shown on two maps, and are compared to the extreme temperatures actually recorded during the entire meteorological history of these and other places. The study is part of a continuing investigation being carried out by the Environmental Protection Section of the Office of The Quartermaster General. The objective is to define the probable extremes of various climatic elements, so that equipment may be designed and tested for satisfactory operation under all conditions likely to be encountered.
Referência(s)