HOURLY TEMPERATURE CHANGE
by Vincent Gray
Expert IPCC Reviewer
“Daily Mean Temperature” from meteorological weather stations is the basis for the compilation known as the Global Mean Temperature Anomaly Record, which is currently regarded as one of the means of assessing the mean temperature trend of the earth’s surface1.
For most of the measurement period since 1850 the only daily measurements of local temperature were the daily maximum and the daily minimum.. There are no published figures for an average temperature for any location on the earth’s surface, despite the current availability of continuous measurement devices.
This paper takes advantage of a series of hourly temperature measurements for 24 weather stations in New Zealand2, for a typical winter’s day (January 1st 2001) and a typical summer’s day (July 1st 2001), intended as a student exercise, to provide plots of hourly temperature change for each station for each day.
The Table below shows the hourly average and the Maximum/Minimum average for each station for both days. The averages for all 24 stations are shown, and in addition, the averages for the “Seven Station Series3” which is used by NIWA as a substitute for an overall average for New Zealand.. . One might note that some of the figures are rounded by the computer calculation.
Since three of the Seven Stations, Masterton, Nelson, and Lincoln did not appear in the 24 station record, the nearest stations, Palmerston North, Blenheim, and Christchurch were used instead for this comparison.
The hourly Maximum and Minimum are not necessarily the same as the actual Maximum and Minimum which were previously measured, so there would be an additional error in assuming they are the same.
It is not intended to imply that these plots are either representative or typical. At least they are both on a single day and from a climate which does not have the extreme temperatures of the larger continents. All the same, some important conclusions may be made
1. None of the individual plots gave a symmetrical curve capable of providing a reliable average or an estimate of variability. Several were highly irregular and the fact that they did not end in the place where they began increased the problem
2 The averaged curves for the 24 stations and the seven stations were much closer to a curve that could provide a plausible average.
|WINTER’S DAY SUMMER’S DAY|
|Station Name||TMax||TMin||AveM/M||Ave Hrly||Diffce||TMax||TMin||AveM/M||Ave Hrly||Dfce|
The most interesting aspect of these results is the difference between the average based on the hourly figures and the average of the maximum and minimum, and the difference between the two series. Winter and summer.
For the winter series this difference varies from -0.2°C and +1.9ºC for individual plots with an average for the 24 series of +0.9ºC and for the Seven Station Series of +0.6ºC
For the summer series the difference varies from -1.1ºC to +0.4ºC with an average of -0.6ºC. with the 24 series and -0.7ºC for the Seven Stations.
This small series shows that there can be discrepancies of almost a degree Celsius either way between an average of a maximum and minimum and an hourly average. This suggests that supposed “trends” of a similar order of magnitude over periods as long as 100 years have a negligible level of statistical significance.
1. Climatic Research Unit, University of East Anglia. (CRU) 2012 http://www.cru.uea.ac.uk/,
Goddard Institute of Space Studies (GISS) http://data.giss.nasa.gov/gistemp/station_data/
2. Mackintosh, L. 2012 National Institute for Water and Air Research, New Zealand (NIWA) “Meteorologist for a day”
3. Gray, V R 2011, The Seven Station Series, Energy and Environment 22 424-434
15th August 2012