Reasons exist to have serious investigations of the whole of climate change (aka global warming) science.

Global warming inevitably rests on current temperatures setting records in geologic time, or at least since early human civilization.  And this is where the 1998 Nature article by Michael Mann, Raymond Bradley, and Malcom Hughes depicting what has become known as the iconic "hockey stick" graph becomes critical.  The "hockey stick" showed modern temperatures far hotter than in the year 1400.

The hockey stick graph was adopted into the third assessment report (2001) of the United Nations Intergovernmental Panel on Climate Change (IPCC) and contradicted a chart that had appeared only eleven years before in an earlier IPCC assessment report.  The hockey stick eliminated what had traditionally been considered the hottest era, the Medieval Warm Period.

As reasonably accurate thermometers were not developed until well into the 19th century, one would wonder how earlier temperatures were measured.  The answer is the use of proxy data – namely, ice cores, tree rings, bee pollen, ocean and lake sediment.  But a reasonable person would have to wonder by what standards these items are interpreted.

This leaves only modern thermometer data sets, the primary one being that of the Goddard Institute for Space Studies Surface Temperature Analysis (GISTEMP), "an estimate of global surface temperature change ... using current data files from NOAA [National Oceanic and Atmospheric Administration]."  The entire 137-year monthly data set, from 1880 through June 2017, in degrees Celsius anomalies (deviations from the corresponding 1951-1980 means), updated monthly, is available in spreadsheet and text forms.

But the land and seas surface data from which the above is derived suffers serious flaws as far as indicating "global" warming.  First of all, it is not, as implied, indicative of global surface temperatures.  The NOAA website contains in the upper-left-hand corner a small and easily overlooked but important map denoting the location of land-based temperature measurement stations around the world and years of coverage, reproduced below.  Not surprisingly, data for more than about 110 years exist only for the United States; Japan; southeastern Australia; and some areas in Europe, Asia, and India.  Nearly all of Africa, South America, Antarctica, Alaska, northern Canada, Greenland, and Asia contain only a few decades of weather data, from widely dispersed stations.

As NOAA notes, this Global Historical Climatology Network monthly (GHCN-M, version 3) "provides monthly mean temperature data for 7,280 stations from 226 countries and territories, ongoing monthly updates of more than 2,000 stations to support monitoring of current and evolving climate conditions, and homogeneity adjustments to remove non-climatic influences that can bias the observed temperature record. The release of version 3 monthly mean temperature data in 2011 introduced a number of improvements and changes from the previous release that included consolidating 'duplicate' series, updating records from recent decades, and the use of new approaches to homogenization and quality assurance."

The above quote from NOAA itself raises many important issues.  First, while one may at first be impressed that data are collected from "7,280 stations," that is less impressive when compared to the 57.5 million square miles of land surface area in the world, because each station represents approximately 7,900 square miles on average.  Moreover, "ongoing monthly updates" are obtained from only "2,000 stations," each therefore representing 28,750 square miles, represented by a square of about 170 miles per side.  Only a handful of these stations have been around since the 1880s.  Finally, the last sentence should raise concerns about "adjustments" to data, possibly biased toward greater warming, as discussed in a video by Tony Heller, posted July 2, 2017.  Mr. Heller provides his vita at his website.

After describing how satellite data were recently adjusted to reflect hotter land-based temperature data, Heller focuses on how the GISTEMP itself was adjusted in 2016, to show how NASA has doubled the warming anomalies between 1880 and 1999, beginning at the 4:30 mark in his video.  In other words, this is the above mentioned NOAA data and apparently what the above quoted last sentence on the NOAA website refers to.  The adjustments lowered temperature anomalies in earlier years and raised them in later years.  Heller then returns to describing how Carl Mears, at Remote Sensing Systems (RSS), then adjusted satellite data in recent weeks.  Heller points out that the Figure 1 currently on the RSS website is not the one a mere few weeks ago, providing a screenshot of the former Figure 1 at the 6:30 mark!

Ocean temperature data, also collected only since the 1880s, has also been significantly corrupted by adjustment, as described in a March 27 American Thinker article, which extends criticism to the broader underlying global warming theory.  Uncooperative data are periodically adjusted to show more warming, as Rick Moran summarizes a recent study detailing three major such adjustments in the past.