Re: Your reply to my 04/28 post

Thanks. Since you were in Ashville, NC last week during your “grand tour”, I decided today to examine one of the staions in that area. I selected Waynesville, NC (319147). The step-wise pattern of NCDC’s quarterly adjustments of temperature data for that station was quite apparent. The quarterly patterns are more interesting than the annual pattern. If NCDC”s adjustment philosophy is to drive delT=0, it was achieved in 1985.

I look forward to your review

I posted the following message on CA (Unthreaded #33) before Steve set up the thread for your NCDC trip. I am posting it here fyi:

HMcCard says:

April 26th, 2008 at 1:51 pm
Steve,

Re: 528 and 529 (Mine)

I think my transmission was interupted because I inadvertently used a “less than sign”. Permit me to start over:

I realize that adjustment of USHCN data has received much attention by CA. However, I am still confused. I told Anthony Watts earlier this week during his visit to NCDC that I echoed Bob Tisdale’s appeal for a simple explanation of NCDC’s adjustments of station temperature data and their significance.

I also told Anthony that a while ago, I became curious about the temporal and spatial variations in surface temperature and chose to examine the average monthly data for the last century (1987 – 2005) from several stations. I became perplexed by the significant differences between the monthly and annual temperature trends for a specific station. The differences in temperature trends between nearby stations was equally perplexing. I lost interest in my project when I realized that NCDC’s adjustment may have confounded the data.

Recently, I revisited the data from one station that I had previously selected (Fort Morgan, CO – 053038). This time, I elected to examine TMAX and TMIN compared to TAVG. I defined TMED=(TMAX+TMIN)/2 and delT=TAVG-TMED. I was surprised to observe a series of nearly noise-free, step-wise changes in delT for both monthly and annual data. It showed the following: 1) 1897-1912: annual delT~+1deg F, 2) 1913-1965: annual delT~-0.4 deg F, 3) 1965-2001: annual delT ~0.0 deg F, and 4) apparently NCDC has not adjusted the data for 2002-2005 yet.

The effect of the 1912 step-wise change varied month-to-month but was significant for all twelve months. The effect of the 1965 step-wise change primarily affected winter months, i.e., DJF. Lesser adjustment in 1987 and 1996 were apparent. I used Excell for analysis and graphical display purposes. I haven’t learned yet how to upload Excel data or graphs to this post.

Yesterday, I examined the Fort Morgan monthly data more closely and observed that step-wise changes in delT occured on a quarterly basis, i.e., DJF, MAM, JJA AND SON. It’s not apparent to me whether TAVG, TMAX, TMIN or all of the above were adjusted. However, it appears to me that NCDC’s adjustment algorithm seeks to drive delT=0. (I can’t think of any reason for that being an end-objective!) In some quarters, detT=0 is achieved earlier with fewer steps. For example, prior to 1913, delT~1.00, 1.16 and 1.40 degF for MAM, respectively; from 1913 to 1987, delT~0.05, 0.09 and 0.02 degF; from 1987 to 2002, delT~0.00.

As I said, it’s not apparent to me whether TAVG, TMAX, TMIN or all of the above were adjusted

Having surmised that NCDC’s adjustment strategy seeks to achieve delT=0, I re-visited five other stations that I examined previously: Boulder, Fort Collins and Wray, CO, as well as, Imperial, NB and Cheyenne, WY. The delT patterns were similar to the patterns for Fort Morgan. In each instance, delT=0 was achieved in three to seven step-wise increments.

As I said, it is not apparent to me whether TAVG, TMAX, TMIN or all of the above were adjusted by NCDC but I suspect that TAVG is the target. If so, TMED may be a better choice for climate trend analysis.

I appologize for the length of this post. If this matter has been discussed previously, will you please direct me to the pertinent archives?

Anthony, If this matter has been discussed previously on your blog, will you please direct me to the pertinent archives?

Thanks

REPLY: I’ll revisit this a bit later, but I have an appointment tonight thanks to a nice fellow from Longmont Colorado.

I tried the “Donate” button route, but did not get a chance to make a comment with the donation. When you see a donation from Longmont CO, that’s for taking your wife to dinner. That might get you two to Chili’s, I hope that others can chip in a little.

REPLY: THANKS AND NOTED

You have given a lot to this project, but there’s one who doesn’t get a lot of credit. That would be your wife (not to mention the kids). Is there a way we can toss a few dedicated bucks into the pot so you can take your wife out to a nice dinner so that she can see how much we appreciate the time that she allows you to spend with us?

REPLY: Thats a fine idea, you can use the dontae button at the right and put a note in appropriately int he form, and I’ll be sure to report back. My wife and kids have indeed lots some quality time due to this project, my wife gets the worts of it since she gets my ups and downs related to some of the issues. Thank you for recognizing this.

“We only get a small feel for that today [distance] when we travel by car rather than plane.”

IMHO, the best way to get a real feel for distance, terrain, and topography is to do it by bicycle. It’s faster than walking, so you can cover a reasonable distance in a day, but boy, you know just how high each hill or mountain is. There are also some interesting effects, like air seems to flow up mountains, not a big surprise on otherwise calm sunny days, but even on windy days I think air will flow up the mountain and keep going up after it passes the ridge and triggers a rolling eddy current that brings air up the other side.

One thing I found bicycling over the Cascades along the Skagit River in Washington was that streams fed by melting snow included chilled air that flowed down with the water. I’d often slow down to let the cool air wash over me. 100 mile uphill but not very steep except by the power dams and when leaving the river to cross Rainy Pass. One of the best legs of the 1974 trip. Much better than the crossing in Oregon in 2003.

Until I got into the comments I had know idea what you were talking about. So these stations track the earth’s temperature? Is it the atmospheric temp? or the crust temp? Sounds like interesting work anyway, best of luck always.

A few years ago after I retired from the Navy I managed to travel 17,844 miles around the U.S. on a Harley Sportster. We follow the road and the miles add up.

REPLY: This is the NOAA surface temperture (air temp) network used for climate monitoring. At 17K+ you have us all beat for road trips.

Maybe ocean based weather bouys would provide a more meaningful result. They are immune to UHI effects. Citing issues should not be a problem. There are no A/C condensers nearby. And the oceans cover 70% of the earth’s surface.

Well it’s just a thought.

I vacation near Southport, NC every summer. Was the station at the Sunny Point Terminal, or actually in Southport? If at Sunny Point, I don’t see how one would ever get access. This is the main terminal for the Army where they load heavy equipment (i.e., tanks) onto ships. It was used heavily in the build-up to the 1st gulf war.

REPLY: It’s the fire station at Sunny Point. Any ideas?

Basil

REPLY: Getting better, thanks

Oklahoma, followed by,

MO
IL
KS
NY
NE
NM
MS
AR
TX,

and

PA
MT
WY
WV
SD
ID
NJ
TN
IA
KY.

It’s interesting that some of these states have a large population (NY, IL), but a dearth of local volunteers.

While schools are still in session, perhaps some methods of attracting science teachers to answer the call to examine a local station as part of the cirriculum could be devised.

REPLY: Thanks for the analysis, I’ll make use of it.