Arkansas - the homogenized data results

Unfortunately I still do not have access to the CDIAC server showing the information on the USHCN sites in Arkansas, so as I discussed last time I have obtained the information from the surface stations site and then downloaded the various station temperatures from the GISS site. Unfortunately this does not include the TOBS data, only that homogenized before being published. There are some problems with that homogenization, which really requires both sets of data to illustrate, but I’ll leave that for now, just to do an initial review of the Arkansas initial data set, and to look at the population data.

There are only 15 USHCN stations, and one GISS station (Fort Smith) on the list, so that the initial work was not that demanding.

Fort Smith GISS temperatures

The difference between the GISS station and the average of the USHCN curves uses the same data set as in the rest of the series:

Difference between the GISS temperature and the average of the homogenized station data for the USHCN stations in Arkansas

The shape of the curve seems to show an increasing difference through about 1948, and then a decline since that time. Given the switch in some series to only using data after 1948 this is slightly curious.

Turning to the overall change in temperature of the state over the century, and now using the homogenized data (hence the purple) rather than the TOBS raw data, the line still does not show very much of an increase over the past century.

Average temperature change in Arkansas over the past 115 years.

The temperature has been increasing at the rate of some 0.05 deg F per century, which is not a lot.

Arkansas is some 260 miles by 240 miles in size running from roughly 89.6 deg to 94.7 deg W in Longitude, and 33 deg to 36.5 deg N in Latitude. The highest point is at 839 m, and the lowest is at 16.7 m, with the average elevation of the state being at 198 m. The average USHCN station is at 176 m, and the GISS station is at 134 m.

Checking the populations, that of Rohwer is too small for citi-data, so I used zip area code to find 77. Fort Smith (the site of the GISS station) has the largest population of the stations.

Looking at the effects of the geography on the temperatures, beginning with the effect of Latitude.

Effect of latitude on temperature in Arkansas

There is the usual strong correlation, whereas with Longitude:

Effect of longitude on temperature in Arkansas

As has been discussed before, any correlation is likely an artifact reflecting changes in elevation.

Effect of elevation on station temperature in Arkansas

When one looks at the effect of population, using the homogenized data, rather than the TOBS data, then there is sensibly no effect of population, in fact such correlation as exists is negative.

Effect of population on temperature in Arkansas after the data has been homogenized.

Pennsylvania TOBS and homogenized temperatures

Moving East from Ohio the next stop along the line is Pennsylvania, which thus becomes today’s topic. Checking with the US Historical Climatology Network there are 24 stations in the state, so the first job is to download the data from each of them.

Having obtained both the mean temperatures after homogenization of the data, and the TOBS (Time of Observation corrected) data for these 24 stations it is time to find the GISS stations. According to Chiefio the three stations that GISS relies on for information on PA include Philladelphia, Wilkes-Barre and Pittsburgh. So going over to the Goddard Institute for Space Studies (GISS) web page I find Pittsburgh with no problem. For Philadelphia there are two stations available, but since one has only data from 1962 to 1993, I’ll use the other – though it only has data from 1947. It is, after all, listed as Philadelphia USA. And then Wilkes Barre has also only data from 1949.

Putting in the population data there was a little search confusion between Pleasant Mount, (pop 1,022) and Mount Pleasant (pop 4,369). Other than that the entry of data was relatively straightforward, and so on to what it revealed. Looking at the difference between the GISS data and that from the USHCN, when I use the homogenized data, (which I plot in purple) then the difference between the GISS data and the homogenized data reduces with time.



Interestingly, however, while this trend is down, when the original raw data, corrected only for time of observation (The TOBS data) is plotted, the trends are the other way. (TOBS plots use a green line).


Which leads into the next comparison, which is the plot of actual average state temperature, first with the homogenized data set:


Using the homogenized data set (which is the one that feeds into other analyses) this shows that the state temperature has been increasing by roughly 1.5 degF over a hundred years. If, however, I plot the TOBS data then a different rate is seen. The graph shows the much less worrisome rate of 0.5 deg per hundred years.


Turning to the geographical effects on temperature within the state, I will plot only the TOBS data for these. This is partly because the homogenization of the data tends to reduce these effects, as a side effect of the averaging process that they use, but also because both sets are quite close.


With homogenization the R^2 value drops to 0.39, and the gradient (trend) to -2.65.

The correlation with longitude shows relatively poor values in both cases.


In contrast there is the strong correlation with elevation that we have come to expect.


Although, and this is a little surprise, the correlation remains equivalent with the homogenized data, and the trend about the same. (-0.0172).

The final correlation, that of population, has been getting poorer as I have moved East from Missouri, but in the second surprise, it is very strong in PA.


Which, given the correlations found West of Missouri, leaves me wondering what the confounding factors were in Indiana and Ohio?

And having discussed this indirectly in the opening plots of the post, it should come as no surprise when as a final plot, I subtracted the raw TOBS data from the homogenized set, and got a steadily rising change over the course of the time period.


Maybe I should do more of these?

State temperatures - Missouri corrected for observation time

The data that I have been correlating in regard to the temperatures of the different recording stations in the different states of the Union, now comes in three flavors, rather than the one that I started with. The three are the Raw data, data that has been adjusted for the time of day of observation (TOBS), and that which has been homogenized to account for the urban heat island effect, among other things.

However the little data that I have massaged from about 8 states suggests that those massaging the data who ignored town sizes below 10,000 folk were neglecting a sensitivity that the data was beginning to repetitively indicate. Thus it does not seem appropriate to continue with using the “homogenized” data of the set that was the only one available at the U.S. Historical Climate Network (USHCN) when I started . On the other hand, a strong case can be made that the adjustment that the USHCN has made to correct for the varying time of days at which the readings were taken is a legitimate one. If one goes to the USHCN website and selects the time of day corrected data (under the listing TOBS for time of observation) then one can, in the same way as I did last week create a table of information using this simple correction to the Raw data. And so, I will see, this week, whether making that change will affect the trends in the data.

Using the procedures that I have outlined, therefore, I have just done that, and will now continue to work back through the states I have already covered (marking the list with either RAW or TOBS) and beyond using this less manipulated data. For Missouri, you may recall that my original concern was for the difference between the GISS stations and the USHCN. Last week I posted the comparison of the Raw data and it showed that changing the data and the measuring stations both made a difference.

The blue line is for the homogenized data, the red is for the raw.

And if I now look at the difference using the TOBS data I get:


A slight upward trend, but not that significant. As for the temperatures in Missouri, over the past 100 years, with the correction – really there is no trend, it has been relatively stable:


I do note that the highest temperatures were some decades ago.

Actually a little surprise in looking at the changes in standard deviation. They had been reducing over time, with both the raw and homogenized data, but when the TOBS values are used, then there is a slight increase over time.


And for the four parameters that had started to appear to have an impact on relative temperature across the states. First there is latitude:


For which there is not much change. Then there is longitude:


And though the state is relatively flat though it rises toward the west, the change due to elevation:


Which gives a better correlation (and one more logically based) than longitude.

And finally that of population, which I have shown both with log and normal scales, but for tonight – because it emphasizes the effect of towns below 10,000 in size, I’ll use the normal one again.


And so, starting with this as the new baseline of a mid-western state, let’s head off West again, and see what the TOBS data will show.