Flying Binghi - interesting point, and worth looking into. (regarding an earlier question you asked - yes, work is a rude interruption to this interesting debate
)The BOM makes that dataset available, and you or I can easily look up the daily min-max temperature records for each station. I looked up a couple of dates that were highlighted on her website as having this particular discrepancy and you - and she - are correct in that regard. The minimum daily temperature is recorded as higher than the maximum. Uh oh!
Unfortunately here is where it spears off into entirely uncharted territory and starts drawing conclusions not supported by the evidence. The Joanne Nova website publishes the claim that
This is a blindingly obvious type of error which should not have escaped quality control. It throws serious doubt on the whole ACORN-SAT project. In my opinion, these violations indicate that the entire ACORN-SAT database is suspect, and should be withdrawn for further testing.
whereas the BOM report actually on how the dataset was created specifically looks for these errors, and a number of other error types. It discusses in some detail how these errors may have occurred, their impact on the overall dataset, and the techniques used to mitigate their presence, which include excluding those particular data points if necessary. If the author of the website had read the articles she is attempting to discredit she would have seen this. She makes the specific claim that:
Why are basic checks like these left to unpaid volunteers, while Australian citizens pay $10 billion a year to reduce a warming trend recorded in a data set so poor that it’s not possible to draw any conclusions about the real current trend we are supposedly so concerned about. — Jo
which is factually incorrect. These checks, and a range of others, are done and discussed in great detail in the report produced by the BOM, available here:
http://cawcr.gov.au/publications/tec...ts/CTR_049.pdf which she is attempting to discredit, but it is obvious she hasnt actually read. Unfortunately its errors like this that do a lot of damage to her credibility - she has made statements that are factually incorrect but not at first glance obviously wrong. It then takes a bit of reading and research to find the relevant papers she is referring to, and understand what the papers are actually saying, which in this case at least is entirely different to what she is claiming.
A couple of points that are worth considering here - the dataset consists of approximately 7 million data points, and that particular error was found in just under 1000 points. Thats a large number, but an extremely small proportion of the records, it works out to 0.014% of the records. These errors were typically in the hand-written records derived from manual observations of maximum and minimum recording thermometers, and as you can see on the Joanne Nova website, typically occurred prior to the advent of automated temperature recording methods. This wasnt exclusively the case, and the report also details recording protocols and errors - such as power surges in automated equipment, recordings attributed to preceding days, thunderstorms in tropical areas producing short-term fluctuations in temperature not representative of the entire day that, and so on - that were made that cause these problems in the data. The descriptions are a bit verbose to go into here, but its available in the report linked above. It discusses how the errors occurred, were detected and both the number of changes, exclusions and corrections made, and the direction both positive and negative they were made in.