Feb 01 2008
Surface Stations Ratings
For the most part, those wishing to deny to existance of global warming have acknowledged the fact that the surface temperature has risen since the industrial revolution. They now claim that it’s all part of a natural cycle. Yet there are still a few that think that the temperature record is contaminated by the placement of the surface weather stations. Previous blog posts about the surface station record: Surface Station Temperature Data, and More GISSTemp and Surface Station Stuff. See links therein to see opinions at other sites.
There is an interesting campaign by the folks at surfacestations.org to survey all the stations in The United States Historical Climatology Network (USHCN). Later, the plan is to survey all stations in the Global Historical Climatology Network (GHCN). The following are two of the goals stated on their website.
- To photographically document good sites that have been well preserved and maintained through their history
- To photographically document poor examples of sites that may introduce biases and errors through faulty siting, encroachments, or maintenance issues, and to identify specific issues when possible
To implement these points, the administrators of the site assign a ranking/rating to each of the sites. A rating of 1 means the sitting is close to perfect, and this is the best score. A rating of 5 indicates there are major problems with the sitting of the station. So today as I was browsing the Internets for fun things to read, I happened across this post at Watts Up With That: How Not to Measure Temperature. The station in question is the one on Mt. Charleston.
There are photos of the station showing that it is very near a parking lot. There is a shed or some sort, and a fire station nearby. A very zoomed-in satellite image shows the general location of the site. The following is provided as a basis for assignment of ratings to the stations.
Climate Reference Network Rating Guide:
- Class 1 - Flat and horizontal ground surrounded by a clear surface with a slope below 1/3 (<19deg). Grass/low vegetation ground cover <10 centimeters high. Sensors located at least 100 meters from artificial heating or reflecting surfaces, such as buildings, concrete surfaces, and parking lots. Far from large bodies of water, except if it is representative of the area, and then located at least 100 meters away. No shading when the sun elevation >3 degrees.
- Class 2 - Same as Class 1 with the following differences. Surrounding Vegetation <25 centimeters. No artificial heating sources within 30m. No shading for a sun elevation >5deg.
- Class 3 (error 1C) - Same as Class 2, except no artificial heating sources within 10 meters.
- Class 4 (error >= 2C) - Artificial heating sources <10 meters.
- Class 5 (error >= 5C) - Temperature sensor located next to/above an artificial heating source, such a building, roof top, parking lot, or concrete surface.
From the images provided, the site would be classified as a 5. Okay, I really don’t have a problem with that rating. It’s not an ideal location. However, there must be some better sites to identify as good. On the same page as that linked directly above, there is a pie chart showing the percentage of stations in each category. There are only 13% in the first 2 categories. Let’s compare.
The site at Berkeley is given a rating of 2. This places it in the top 13% of stations surveyed in the United States. The photographic evidence shows some nice pictures. The station appears to be in an open field, with “natural” grass and wood chips underneath it. There are even close-ups demonstrating the degradation of the shelter.
However, it’s hard to imagine a station much worse than Berkeley. It’s in the middle of a big city! Wasn’t the point of the ratings to document good and poor HCN sites? The temperature of the Berkeley site is going to be influenced in a large part by the urban heat island. It is not a good site based solely on the satellite image of the station.
|
How does the station compare that was criticized as being the wrong way to take a temperature?
|
So this station out in the middle of nowhere, uninfluenced by an urban heat island would have received a 5 rating, and the Berkeley station gets one of the top 13%. That’s backwards. The influence due to humans is much less in the Mt. Charleston case than the Berkeley case, and thus should have a better rating.
(No Ratings Yet)
Loading ...Related Posts:
5 Responses to “Surface Stations Ratings”
To reduce spam, comments are automatically closed 30 days after the last comment. If you would like to comment on any closed thread, please use the contact form at the top of this page.
I think you’ve missed a critical point or two.
The point of the 1-5 ratings system, which was developed by Michel Leroy of Meteo France for their surface network, and then adopted by NOAA for use in defining the new Climate Reference Network (CRN) was to document and rank the microscale issues, not the mesoscale ones. The surfacestations project is not primarily about UHI, it is about microsite. Thus the rankings.
Also please note that the Charleston site doesn’t figure into the study at all since it is not a USHCN station or a GHCN station, it’s simply COOP-A, which goes to the NCDC climate database, but doesn’t get used in other databases like GISS or HadCRUt.
It (photos) was sent in by a blog reader, and it was discussed for it’s merits, but it is not used in the surfacestations study. It is just another example of poor siting for discussion, nothing more.
The classification or urban/rural will figure into the study later when we look at all of the best sites collected, and then determine which ones are truly free from microsite as well as UHI influences.
I agree with you that in the case like Berkeley, the mesoscale UHI factor may very well swamp the microscale issues. But to be true to the rating system as it was defined, I can’t make a value judgement on other factors that aren’t part of the original rating system.
You can see the rating system as defined by NOAA here:
http://www1.ncdc.noaa.gov/pub/data/uscrn/documentation/program/X030FullDocumentD0.pdf
see section 2.2.1
Thank you for your comments.
While microscale issues do play a significant role in determining the temperature measured by each station, I am less convinced that they are important in the temperature trends. A station that is located next to a parking lot, such as Mt. Charleston, is clearly going to have a higher temperature than if it was over natural terrain. Assuming no additional changes are made, shouldn’t the yearly-averaged temperature for that site be the same from year-to-year? Thus, everything else being equal, the temperature trend would be zero?
Dr. Pielke has documented in the literature the effects of microclimate effects on the temperature record of individual stations (e.g. his recent BAMS paper). However, given the exteme similarities between the satellite derived temperature record and the surface records, it seems unlikely that any microsite effects would have a significant impact upon the global average temperature. That is, these microsite effects potentially could have a cooling or a warming influence upon the temperature that seem to average to near zero if the satellite record is correct.
Thank you for the reply.
You asked: “Assuming no additional changes are made, shouldn’t the yearly-averaged temperature for that site be the same from year-to-year? Thus, everything else being equal, the temperature trend would be zero?”
If nothing changed from year to year in that environment, yes that would be true. Essentially you’d have a constant bias. That can sometimes be spotted as a “step change” sometimes not.
But what we are finding is that many of these microscale environments change, erratically and often, and thus the bias would not be constant.
For example, one of the worst sites, the USHCN station inMarysville, CA had two portable electronics buildings added within 10-15 feet of the sensor over a two year span, one the first year, the other the second. Both have electronics and a/c units with exhausts nearby.
This sort of thing is not accounted for in the preparation of data.
In other cases there has been a gradual accumulation around the station of smaller inlfuences which may not be easy to detect.
Thus if you have an accumulating bias (positive or negative) over time, it will result in a trend, and that is not so easy to detect and remove from the record.
I did a post awhile back on this topic, Is it the Heat Island or a Linear Trend. I concluded that it was unlikely to be able to tell the difference if you were not given meta data that included station moves and (not mentioned in the post) the addition or removal of continuous heat sources or sinks.
For what it’s worth, I ran preliminary results based solely on CRN ratings. (With no UHI factor.)
Based on 482 obseved stations (of 1221 in the USHCN), I multiplied the percentage stations in each CRN rating with the associated warming biases and totaled the results.
Using the smallest bias associated with the ratings:
Low Results = 1.98 C warming bias.
It is vital to note that the vast bulk of the violations occurred since the late ’70s as a result of exurban creep (parking lots, shopping malls, etc.) and the fiasco of the switchover from the old Stevenson Stations to the MMTS system.
One also must consider that a heat sink (such as a parking lot) does not produce a mere onetime temperature offset, but continually exaggerates a small temperature increase trend over time (LaDochy, 2007), especially the T-Mins.
This is highly significant because the entire upward shift in measurements from the nadir of 1979 to the acme of 1998 is less than 1 degree C.
My simple calculation is based entirely on the hard work of Anthony Watts, who is making an important scientific contribution and has my deepest respect.