1
$\begingroup$

Nasa's temperature record is showing temperature increase. When I confronted a climate change skeptic with it, he replied that the data has been tempered with, and that there are different data sets which do not show temperature increase.

The exact reply was:

According to RSS, UAH and radiosonde global temp anomaly datasets, there hasn't been a global warming trend in 20 years;

http://www.woodfortrees.org/plot/rss/from:1996.6/to:2015.7/plot/rss/from:1996.6/to:2015.7/trend/plot/esrl-co2/from:1996.6/to:2015.7/normalise/trend/plot/esrl-co2/from:1996.6/to:2015.7/normalise

GISS and HADCRUT4 datasets have added heat to raw temperature data to artificially keep the warming trend going as NOAA freely admits:

https://www.ncdc.noaa.gov/img/climate/research/ushcn/ts.ushcn_anom25_diffs_urb-raw_pg.gif

This sounds reasonable to me.

Can anyone familiar with climate science explain why there are several data-sets, and why they are saying different things, and why NASA's set is the correct one?

$\endgroup$
6
$\begingroup$

There are several climate datasets around the world, two in America - NASA's GISTEMP, and NOAA's MLOST, and one in the UK - HadCRUT. These are the main datasets but there are others around the world, China, Japan etc.

The datasets differ slightly in the starting date, number of met stations included, degree of interpolation, and the way they get round the problem of lack of met stations in oceanic areas.

As to being 'tampered with', this is a subtle malicious implication that the data has been fudged. It is also an insult to the scientific integrity of the thousands of climate scientists involved, whose aim, as in all of science, is to get at the truth. Of course, the data have been 'processed'. This is an essential part of data quality control to interpolate missing data, and to ensure consistency of the data set. But no data has been 'tampered with' in the sense of trying to lead to false conclusions. One of the striking features of all the global datasets is that, despite minor differences, they all show the same basic characteristics of an accelerating global temperature - the 'hockey stick graph'.

There is much dishonesty about, but not with the climate scientists. It is the climate skeptics who are being dishonest (not to mention scientifically incompetent).

$\endgroup$
  • $\begingroup$ Could you please elaborate on the meaning of "data processing " to "ensure consistency of the data set"? What criteria were employed to ensure the consistency? $\endgroup$ – Ale..chenski Sep 10 '16 at 5:40
  • 1
    $\begingroup$ Most raw climatic data is fill of gliches: e.g typos, mixed numeric and alphabetic data entry, confusion between zero and 'no data', transcription errors, discontinuities, systematic errors (for as host of reasons). All of this has to be corrected and kicked into a consistent usable digital format - not so much of a problem in advanced western countries, but very much of a problem everywhere else. Then there is inter-station correlation to check out anomalies. In overseas work I find that more time is spent on all this data pre-processing than on the actual analysis. $\endgroup$ – Gordon Stanger Sep 11 '16 at 9:55
  • $\begingroup$ I understand about consistent digital formats. But the question was, what are the objective criteria by which researchers judge that some data point(s) needs correction? The records are made at least daily, which must follow local weather. What is the radius of "inter-station" correlation, and how it was established? $\endgroup$ – Ale..chenski Sep 12 '16 at 19:00
2
$\begingroup$

Gordon Stanger's reply is the correct answer to this question, but I wanted to point out something else. Saying that there "hasn't been a global warming trend in 20 years" is a tricky statement to begin with. There are a number of papers discussing it, but I think this one is a good summary:
Is the climate warming or cooling?
Essentially, the point is that climate is variable enough that you can get decades of no trend or even slight global cooling; but the longer-term trends always show warming. Where you choose to start and end your regression line to calculate a trend is an easy way to get whatever answer you want. To quote their abstract:

Numerous websites, blogs and articles in the media have claimed that the climate is no longer warming, and is now cooling. Here we show that periods of no trend or even cooling of the globally averaged surface air temperature are found in the last 34 years of the observed record, and in climate model simulations of the 20th and 21st century forced with increasing greenhouse gases. We show that the climate over the 21st century can and likely will produce periods of a decade or two where the globally averaged surface air temperature shows no trend or even slight cooling in the presence of longer-term warming.

$\endgroup$
-3
$\begingroup$

Most estimates of global surface temperature "anomaly" come from NASA GISS network database http://data.giss.nasa.gov/gistemp/station_data/ . There are some stations that show cooling trend over 100-year-long history of observations, as shown here: https://stevemosher.wordpress.com/2010/09/29/needs-chec/ This is considered as a normal "bell-curve" distribution.

However, as it was noted in old discussions, there are several examples where the OPPOSITE trend happens on ADJACENT stations that are only 50 km apart. This means that the spatial distribution of surface stations is insufficient to reconstruct characteristics of temperature field (averages, etc.), because the spatial sampling does not satisfy criteria for Nyquist-Shannon Sampling Theorem.

To get the sampling right, the station site locations must be at least half of 50km, or about 25 km apart around the globe. With Earth surface having 5×108 km2, the 25×25 km grid should have about 800,000 stations. Given 30% dry land, there should be about 240,000 surface stations to have any scientifically reasonable estimation of trends. This is the most conservative estimations for necessary amount of data to collect.The GISS network has less than 6,000 stations world-wide, which means that the data are undersampled by a factor of 40.

If one has a field so vastly undersampled, selective treatment of datasets could result in anything, which explains contradictions in various estimates.

$\endgroup$
  • $\begingroup$ That's why the scientist invented satellites $\endgroup$ – arkaia Sep 14 '16 at 1:46
  • 1
    $\begingroup$ That is why we have a field of science called "remote sensing of environment" in which we TRY to answer those and many other questions. You can question pretty much all the research in: sciencedirect.com/science/journal/00344257 and ieeexplore.ieee.org/xpl/… Good luck in your crusade! $\endgroup$ – arkaia Sep 14 '16 at 12:56
  • 1
    $\begingroup$ My former supervisor on remote sensing used to cite the Bible: Proverbs 23:9: "Do not speak in the hearing of a fool, For he will despise the wisdom of your words." $\endgroup$ – arkaia Sep 14 '16 at 15:54
  • 1
    $\begingroup$ dx.doi.org/10.1175/… dx.doi.org/10.1016/0034-4257(95)00188-3 dx.doi.org/10.1016/0034-4257(95)00036-Z dx.doi.org/10.1016/0034-4257(95)00199-9 dx.doi.org/10.1175/2008JTECHA1052.1 google.com/patents/US5612901 and so many more... They are not my words, rather the words of the experts in the field. Have a conversation with them if you want to learn more. Thanks for the entertaining conversation. $\endgroup$ – arkaia Sep 14 '16 at 17:07
  • 1
    $\begingroup$ My last comment: As somebody that has worked with satellites for a number of year (~15), you demonstrate your ignorance with pretty much all your comments. Claiming that the work of hundreds of experts over more than 60 years of image processing is bogus without any evidence is at least rash, and likely reckless. $\endgroup$ – arkaia Sep 14 '16 at 17:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.