Showing posts with label Alabama temperatures. Show all posts
Showing posts with label Alabama temperatures. Show all posts

Wednesday, October 5, 2011

Regional US temperatures, the West, the Middle and the East

The relative changes in the temperature profiles between the Central States and those along the Atlantic Coast, begs the question as to the relative conditions on the West Coast. And so I have combined the state temperatures of California, Oregon and Washington State. As with the other regions the first average that I took was for the homogenized data, and just took an average for the three states, to get a sense of what has occurred over the past hundred and ten years.

Average temperature over time for the Pacific Coast, using the average of state temperatures, and homogenized USHCN reported temperatures

The most obvious immediate conclusion is that the temperature drop from 1950 to 1965 which is so evident on the East Coast does not appear to have happened in the West. And the flat temperatures over the last hundred years found in the Central States don't appear to hold either.

Breaking this down to a review of the initial Time of Observation corrected temperatures, but initially again just averaging the state temperatures, one gets:

Average temperature for the states along the West Coast, averaging the TOBS average state temperatures.

In contrast with the Middle States there does appear to be a steady increase, with time, in the average temperature. Looking at the average of all the stations (a total of 138), the data continues to show that rise:

Average temperature for the states along the West Coast, averaging the TOBS average station temperatures.

The area that each station covers falls from an average of 3,000 sq miles in California, through 2,460 in Oregon to 1,621 in Washington. When one adjusts for area, using this weighting, and adding a trend line to see what the temperature change averages, one gets:

Average temperature for the states along the West Coast, weighting the TOBS averages by area.

There is still no sign of that fall in temperatures in the 1950’s, in fact the temperature goes up. Comparing the results for the three different states:

Average individual West Coast TOBS temperatures over time

The form of the plots are roughly similar with the two more northern states showing almost identical patterns, and a slightly wider range of fluctuation.

When one compares the three regions that I have looked at to date, however, the difference between them is quite clear. Well actually, if I plot the data itself it is a little obscured by the curves as they superimpose.

Average temperatures for different regions of the country over the past century.

The Middle states are clearly warmer than those on the two coasts, which I hadn’t expected, but it is more difficult to separate the two coastal variations. Because of this overlap I have moved the lines a little apart, so that the changes in pattern can be more clearly seen. So, for this plot, ignore the temperature values on the left, since I have, as shown, adjusted the values so that they are separated, and so that the differences in behavior each year can be seen.

Pattern of temperature change for three regions of the United States over the past century.

While there is some congruity between the central and Atlantic values, these are clearly different from the Western states. Perhaps I should see what is happening in the mountains??

I am also curious to see what the effects of El Nino’s have on the land temperatures, but I think that I will add in the mountain states, before putting those effects onto the plot.

I might start here but any other suggestions would be welcome.

Read more!

Tuesday, September 20, 2011

The Temperature fall and the Chickadee

Yesterday (Monday) I was reading the Boston Globe and came across an article by David Abel about a recent report by the Audubon Society on their Breeding Bird Survey, and how it was showing that the changes in the climate was influencing bird ranges and populations. The surveys, which looked at the state of Massachusetts, only went back to 1965, and so, being curious, I wrote to Mr Abel pointing out the drop in temperature (about 4 deg F over 1950 - 1965) for the state that I wrote about a couple of weeks ago. I was, sort of hoping that he might look at the survey data before that date, and see if the bird populations had fallen with that temperature drop. Also I wanted to know if the numbers had returned the the 1950 level. However he merely sent me a note thanking me for my note.

But I was now curious to know the answer, and so I had a look for the data on which the Audubon article and the resulting newspaper story were based. And it turns out that the reason that the data only went back to 1965 is that when the BBS started to be implemented). So where could I go to get earlier data?

The answer turns out to be the Audubon Christmas Bird Count (CBC) But the count is very species specific, and since I have only a limited amount of time I decided that I would just pick one bird (at least initially) and see what the results showed. And if the story was about Massachusetts, then it seemed sort of logical to use the state bird, which turns out to be the Black Capped Chickadee .(And as it happens, though not told until tonight by the Bishop, this is also the Maine State Bird - about which more anon .)

So the first thing to do was to get the basic information on the population with time, looking at the annual counts for the different states along the Atlantic shore from the Audubon site. And that showed the answer to one of my questions. Let me begin by repeating the temperature plot for Massachusetts over the last hundred years from my look at the state. (I am using the USHCN numbers for this particular study) so that you can follow how I got there.

Average temperature as a function of time for the state of Massachussets

You can see the significant drop in average state temperature that occurred in the 1950 - 1965 time frame. So how did the bird population respond?


It has only been in the last few years that the state has returned to those 1950 temperature, so did the bird population change, and has it made it back? The Audubon page allows you to plot a graph:

Black Capped Chickadee population in Massachusetts using the Audubon CBC

The count doesn’t look that valid before 1940, and when I checked other states I had the same problem, so I am going to restrict the rest of this to the years after 1940.

You can see that there was a drop in the count at around the period that the temperature fell, and this was confirmed when I looked at the numbers for the other states sitting close to Massachusetts. But, as you can see, the question that I had asked David Abel as to whether the numbers had returned to the values back before the temperature fell, was answered with a Yes!

However we can now go on from there. Having numbers and a temperature plot, the question can be answered as to whether the population was affected by the temperature, as the two plots above seemed to indicate. So I re-plotted the data as count versus temperature. And this is what I got for the state:

Chickadee count as a function of average annual Massachussets temperature

The plot showed, when I fitted a polynomial, that there was an optimal temperature for the population. I checked this with similar plots for the states along the Atlantic that I covered in the earlier post. Some states didn’t get that warm, and some didn’t get that cold. But for New Jersey, Connecticut, Pennsylvania, Rhode Island and perhaps New York, the data plots showed about the same peak in population at around 51 deg F.

Here is the plot, for example, for Connecticut:

Chickadee count as a function of average annual Connecticut temperature

And similarly for New Jersey

Chickadee count as a function of average annual New Jersey temperature

So with Massachusetts temperature hovering around that level at the moment (as back in 1950) we’ll have to hope for no more change in state temperature, since too much fluctuation either way may cause the population to drop.

So the next question would be whether the population drop is because the birds died off, or moved, if the temperature fell off. I don’t have enough data to answer that part of the question, but I have an indication. If one looks at North Carolina (below which the more Southern states don’t see this species apparently) then when that cold spell hit, further North, the state suddenly saw birds appear. (And having discovered the state they came back when it was colder up North).

Black Capped Chickadee counts for North Carolina (Audubon Society CBC)

So it appears that, not only did I get an answer to my question, I also discovered another temperature index through the birds, and so maybe I’ll do some more correlations from time to time. (I was thinking about doing wild turkeys since they are common around much of the country, but there is the problem that they are harvested also so that the numbers are more artificial).

Oddly the plot of optimal temperatures for the Chickadee did not work for Maine, where the temperatures are cooler. Wonder if the Maine version is turning into a slightly different creature – given that it is the Maine state bird too.

Black Capped Chickadee counts for Maine (Audubon Society CBC)

Read more!

Saturday, June 18, 2011

Alabama combined temperatures

There are fifteen USHCN stations in Alabama that start at Brewton and end at Valley Head. It has two GISS stations on the list, and they are at Mobile and Montgomery.

USHCN stations in Alabama (CDIAC )

The station in Montgomery is yet another where GISS is using a station that does not start its record until 1947. It is interesting to see how many times GISS is using these shorter truncated records when there are a significant number of other stations that have a full record back into the 1880’s. If one were cynical one might think that this was to hide the higher temperatures in the 1930’s.

Average annual temperatures for Montgomery AL, GISS station data

However there wasn’t such a temperature high in the ‘30’s, for Mobile, whose temperatures have remained relatively stable, so we’ll have to see if there was one for the state as a whole.

Average annual temperatures for Mobile AL, GISS station data

When looking at the difference between the average GISS and the average of the homogenized data for the USHCN sites for this state, the GISS sites are, on average 3 deg F warmer, but there is no trend in the data for Alabama.

Difference in average temperature, per year, between the GISS stations and the homogenized values for the USHCN stations in Alabama.

In terms of the temperature trend for the state, as a whole, even the homogenized data from the USHCN sites shows that the state has cooled over the last century.

Average temperature for the state of Alabama with time, using the homogenized temperature data from the USHCN.

When this is compared with the Time of Observation adjusted raw data (TOBS), that also shows a decline. It is interesting that in both there is a drop in the average temperature in 1958, which then seems to re-stabilize around a lower temperature

Average temperature for the state of Alabama with time, using the Time of Observation corrected raw temperature data from the USHCN.

Looking at the populations surrounding the different stations, using the citi-data sites for the communities, This does not work for Gainsville, which has a population of 192 according to HomeTownLocator. Saint Bernard is a little more of a challenge – so I go to Google Earth, and while close in it looks very rural (being in a Benedictine monastery/school) when one zooms out a little it is in the community of Cullman, AL).

Location of the station at Saint Bernard AL, Cullman is over the hill to the left.

Alabama is some 330 miles long and 190 miles wide, stretching from just under 85 deg N to just under 88.5 deg N, and from 30.21 deg W to 35 deg W. The mean latitude is at 32.84 deg N, that of the USHCN stations is 32.8 deg N, and that of the GISS stations is at 31.5 deg N. The elevation ranges from sea-level to 733 m, with a mean elevation of 152.4 m. The mean elevation of the USHCN network is at 122 m, while that of the GISS stations is 40.5 m. The two GISS stations have an average population of 197,648, while that of the USHCN is 14,866. (With a temperature change of 0.024 deg per m, the change in elevation gives a 1.6 degree difference because of elevation changes, and using a temp change that varies with 0.6 x log (pop) would explain another 0.7 degrees of the average 3.04 deg between the two mean temperatures).

Looking at the effects of state geography on temperature change, beginning with Latitude.

Average Alabama temperature as a function of latitude

Alabama is another state where there is a fall in elevation moving West, and so there is a temperature increase with longitude:

Average Alabama temperature as a function of Longitude

While the R^2 value for latitude falls, from 0.86 to 0.71, when the data is homogenized, the value for the longitude correlation increases from 0.14 to 0.28. This again helps explain my concern with the process used to fill out the data.

Looking at the elevation itself,

Change in average Alabama temperature as a function of the elevation of the observing station.

The final correlation is with population, and there is not a really broad distribution of population around the stations., so that when this is included with the other variations, gives a poor correlation in the state with a 5-year average for temperature. (The correlation is, oddly, in this state better over longer periods).

Correlation of Alabama temperatures with the population of the community surrounding the station.

And that leaves the final check, between the raw data and that which has been modified to give the official USHCN numbers. The plot is the official temperature minus the raw temperature averages.


Let me throw in one final fact, just glancing at Chiefio’s site, he has a post up on relative humidity in which he quotes as follows:
As an average, the International Civil Aviation Organization (ICAO) defines an international standard atmosphere (ISA) with a temperature lapse rate of 6.49 K(°C)/1,000 m (3.56 °F or 1.98 K(°C)/1,000 Ft) from sea level to 11 km (36,090 ft). From 11 km (36,090 ft or 6.8 mi) up to 20 km (65,620 ft or 12.4 mi), the constant temperature is −56.5 °C (−69.7 °F), which is the lowest assumed temperature in the ISA.
Since I am using meters for measuring elevation but degrees Fahrenheit for temperature, the gradient is thus officially 11.68 deg/1,000 m or 0.0117 deg/m. The rate that the data fits, as shown above, is 0.024 deg, or about twice the official rate. It will be interesting to test the actual data for the different states against this standard.

E.M. Smith (who writes as Chiefio) has some interesting points to make about the effects of humidity, which I haven’t even thought of looking at, but which might be an interesting place to go. Water vapor is, after all, as he notes, supposed to be a stronger GreenHouse Gas (GHG) than carbon dioxide, and yet more humid places may be cooler than those which are drier. (It just may not feel that way). Read the post, it is interesting!

Read more!