Saturday, September 4, 2010

Maine TOBS temperatures

Here in Maine we have just seen the remnants of Hurricane Earl scurry by. There wasn’t a lot of activity last evening:

Will Earl Park in Maine? It did not!

We woke this morning to about an inch of rain on the flat roof in front of the room, but we’re back to sunny days already. Speaking of which it is time to turn to the review of the TOBS temperature data from Maine and see if it much changes the conclusions that I drew when I first looked at the state back in April.

With the changing list of weather stations that will be used in the future. it appears that more detailed studies of the past might be one of the victims falling by the wayside after the Climategate furor. We are seeing the establishment of new networks so that I am not sure how strongly folk will chase after historic data. Back in April I mentioned the US Climate Reference Network, which will be a set of approximately 100 stations nationwide “selected to capture both the national and regional climate trends and variations for temperature and precipitation.” As I remember, I will look at those sites as well as the ones that I have been following from the US Historical Climatology Network. But having been one of those who sought more data rather than less in setting up experiments, and with all the stations that exist to monitor temperature change under the varying conditions that exist around the country, my study so far has not left me feeling comfortable with such a reduction in input. But for today I am going to input the TOBS data for the 12 stations that are in Maine, compare with the single, partial, station data that GISS uses for Maine, and see what we find.

As with all the raw data sets that I am inputting there are frequent years where there is no data because of a lack of it from the source, but the TOBS data is without homogenization, which allows generation of data from missing months, inter alia, and we will leave that topic until a later discussion.

Looking at the difference between the GISS data and the USHCN, there is an increase, but it should be remembered that this is only since 1947, since the GISS station (at Caribou, one of the coldest places in the contiguous nation) only exists after that time.

In terms of the changing temperature of Maine over the years, this does change when you go from the homogenized data to the TOBS data.

With the homogenized data set the temperature rise was 0.02 deg per year, while with the raw data the temperature rise was only 0.0008 deg per year, a substantial drop in rate.

Looking at how the temperature is affected by geography, in latitude the trend that was clear for the western half of the states examined, continues, if anything more pronounced than before.

And though there is a trend in longitude, this is a little more difficult to assign purely to elevation, since Maine is another state with a strong coastline, and a number of the stations are along the coast.

While not sure yet of the role of the sea surface temperatures (SST) the correlation with elevation continues to be a strong one, though the correlation coefficient has fallen a little. About four or five of the stations are along the coast, which gives the slight variation due to the SST.

Maine was one of the states where I started to correlate local temperatures to a central value for the state, based on latitude correction. I looked at two different correlations with the result. The first is just the state temperature:

There is a stronger increase in temperature with time when I do this, which may be an artifact of the number of stations that are available with full data for each year.

The other correlation I was looking for was with population. When I look at the state initially there is the logarithmic relationship, which popped up in most states out west:

After making the adjustments for location latitudinally, I also checked to see how the population relationship fared.

Slightly better regression coefficient, but I think we have a way to go yet, in order to validate the curve, using states with larger populations.

And the Southern Maine coast was full today (as in "no room at the Inn"), with the large waves coming in after Earl bringing hundreds of folk to stand along the beaches and watch the big waves roll in. Though the water was quite cold - especially up the back of a trouser leg.

Waves at Ogunquit, Sept 4, 2010


  1. This comment has been removed by a blog administrator.

  2. This comment has been removed by a blog administrator.

  3. This comment has been removed by a blog administrator.

  4. This comment has been removed by a blog administrator.