Things That Make You Go Hmmm Part 3
August 13, 2010Posted by on
In the first two parts we saw that using either the older adjustment method used by GHCN or the new one in USHCN, these adjustments are the only thing that reversed the findings in the Christy et al 2006 paper while using the Hanford and Lemon Cove stations.
In this third part I’m going to look and see if I find the same changing of the results by looking at the Tmax and Tmin trends for Merced and Yosemite stations in the USHCN v2 dataset. You can download the data from the following link:
Couple of things I noticed about the Merced/Yosemite station data, is that there is no Raw data for Yosemite prior to 1911, except for one year and there is no Raw data after 2003. All adjusted data for those years is by the USHCN Filnet computer routine. Merced isn’t much better with very scarce data in the early Raw record but it does happen to have Raw up to 2009. So to make a true comparison for this pair we will have to restrict the data range from 1911 to 2003.
Also take note that both stations have a lot more missing data than the Hanford/Lemon Cove stations in the selected data range.
Now here in Figure 1 we see the Merced (Valley) Tmax raw trend is a warming one of about 2.3° F over the time period. After adjustment the trend becomes a very slight warming trend, maybe .1° to .2° F.
For Yosemite (Sierra) Tmax raw trend is a cooling trend of about -.6° F over the time period. After adjustment the trend becomes a very large cooling trend of -1.75° F.
Now here in Figure 2 we the Merced (Valley) Tmin raw trend is is warming trend of about .5° F. After adjustment the trend is reduced very slightly to about a .4° F warming trend.
Yosemite (Sierra) Tmin raw on the other hand is a eye opener to say the least. You have a very large negative excursion right at the beginning of the graph , followed by a relatively flat period, followed by what looks like a step jump around 1977. Usually those are signs of changing equipment/station move, which means taking a look at the meta data on that station closely. However with all that said the trend is a very large warming one of 7° F. The adjusted trend is still a very strong warming one of about 3.5° F.
So lets see does this match what Dr. Christy had in his 2006 paper of Valley and Sierra showing a lack of warming during daytime (Tmax) but Valley showing a warming during nighttime (Tmin) with Sierra not?
In this pair we saw that the Valley is strongly warming during daytime in the raw data and only slightly in the adjusted, where the Sierra one is showing a cooling in the raw data and is increased via adjustment. On the nighttime side the Valley showed a slight warming trend in both raw and adjusted. The Sierra one on the other hand has a very large warming trend and adjustment cuts it in half and still it’s a very strong warming trend.
Now this goes against the finding of Dr. Christy, but this pair of stations IMHO is a poor pair to use to look to see if irrigation of a desert landscape, changing it into farmland causes a temperature change. The reason for that is if you go into Google Earth yourself and zoom all the way in on the Merced station coords given in USHCN, you find yourself smack dab in the middle of the Merced Airport.
Yes there is farmland to the west of the airport, but there is a city just to the east of the airport and lets not forget the airport itself. To me this doesn’t make a very good station.
Now what about Yosemite? Earlier I pointed out what looked like possible station moves in the record, but that is not why I think it’s a bad choice. IMHO it’s a bad choice because it’s not in the Sierra Foothills in desert like terrain, it is in the Sierras themelves. If you go back and look at the Google Earth Image that shows the stations in relative position to each other, you will see a brown strip between the city of Merced and the Trees and mountains of the Sierras in Yosemite National Park. That brown strip is the desert foothills of the Sierra’s and what we are suppose to be compring to, not the temperatures in the mountains. Matter of fact the Yosemite station can be seen very clearly to be sitting up in the mountains at the Park Ranger’s HQ. IMHO this station should never had been in the 2006 paper, since it is not in the Central Valley, unless it was when the study was done and moved to it’s present location afterward (the USHCN station numbers match between this station and the one in the study), but I can’t find any evidence of that in the USHCN or GHCN records (No multiple stations in GHCN, no flags in USHCN showing a move).
For more on the Merced Station location:
For more on the Yosemite Station location: