One of our readers asked how it was to read a thermometer. I’ve got a slightly different question: if I can locate daily maximum and minimum readings for Dulan airport on the internet, why can’t NOAA? This is not as trivial a point as it might seem, as it turns out that Dulan airport is one of a very few stations contributing to a lurid blotch of big temperatures on the NOAA gridded March map.
Here is an excerpt from the NOAA MArch 2007 map announcing the xth warmest month since the Eocene or 1960, whatever. The big lurid red dots are 5 deg C positive anomalies and, iin west China, go, on the 42.5N latitude line from 92E to 112E and on the 37N latitude line from 97E to 117E. Dulan, which we’ve been discussing, is in one of these gridcells.
The readme for the NOAA gridded series says:
This data set contains gridded temperature anomalies calculated by the “anomaly method”. Gridpoint temperatures are calculated by averaging the unweighted raw data from all stations within the grid box, and then the anomaly is the difference from 1961-90 mean.
If one then examines the monthly station data in the GHCN compilation of data for the corresponding gridcells, one finds that some of the gridcells contain no data more recent than 1990. However, GHCN has a roster of daily data. I’m not sure how these are integrated in the calculation of the anomalies. In any event, if one then examines the roster of sites with daily data, separately listed here , one finds that there are 6 reporting stations to cover the 10 most lurid red dots, as listed below. Dulan is an old friend, appearing here in a new role.
205 52203 000 42.82 93.52 737 CH HAMI GSN 52203 703
205 53068 000 43.65 111.97 965 CH ERENHOT GSN 53068 707
205 52533 000 39.77 98.48 1477 CH JIUQUAN GSN 52533 776
205 52836 000 36.30 98.10 3191 CH DULAN GSN 52836 776
205 53614 000 38.48 106.22 1111 CH YINCHUAN GSN 53614 778
205 53772 000 37.78 112.55 778 CH TAIYUAN GSN 53772 779
The daily data updated as of April 17, 2007 for Dulan is located here http://www1.ncdc.noaa.gov/pub/data/ghcn/daily/gsn/20500052386.dly. If you look at the March 2007 readings for TMAX and TMIN, you will notice that there are 28 max readings and only 14 min readings. The graphic below shows the number of readings by month during the history of this record (Starting in 1961). There is a hiatus around 1990 – does this represent an inhomogeneity? Inquirigin minds want to know. While measurements were regular as clockwork back in the day, now the number of readings per month is erratic. Indeed, March 2007 set a record low for the number of minimum temperature readings.
But while the NOAA record has become progressively more erratic, with a click of the mouse and a simple google, one can locate daily maximum and minimum temperatures for Dulan airport for March 2007. Indeed, one can get temperatures at 3-hour intervals.
Is there any bias in the NOAA record? I haven’t checked. There are some oddities. The NOAA record shows a high of 1.9 deg C for March 9, 2007, while wunderground.com shows a high of only -2 deg C. I just looked at some values manually and couldn’t figure out how to obtain a digital file of the Dulan airport readings. If someone else can figure this out, it would be a useful exercise to compare the two records. The more puzzling thing is why the NOAA record for Dulan is so incomplete.