NOTICE!

The Weatherzone forum has now closed and is in read-only mode until the 1st of November when it will close permanently. We would like to thank everyone who has contributed over the past 18 years.

If you would like to continue the discussion you can follow us on Facebook and Twitter or participate in discussions at AusWeather or Ski.com.au forums.

Page 8 of 19 < 1 2 ... 6 7 8 9 10 ... 18 19 >
Topic Options
#1003530 - 30/07/2011 20:31 Re: Observations of climate variation [Re: ROM]
ROM Offline
Meteorological Motor Mouth

Registered: 29/01/2007
Posts: 6628

Top
#1003619 - 31/07/2011 17:24 Re: Observations of climate variation [Re: ROM]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
July climate change at Manilla, NSW: 13 years

July is usually the cold month at Manilla, on the North-West Slopes in NSW. During the last thirteen years the weather or climate, as shown in the first graph, has varied quite a lot.
To clarify the kinds of change, I have divided the data into two groups: "drought indicators" and "heat indicators", shown in the second and third graphs.

Drought Indicators
Drought conditions are indicated not only by low monthly rainfall, but also by low humidity, sunny days, and wide range of daily temperature. This graph puts drought conditions at the bottom, and "deluge" or (D. McKellar's) "flooding rains" at the top. The right side scale is for monthly rainfall in millimetres and for percentage of cloudy mornings greater than four octas. The degree scale on the left is for humidity, expressed as early morning Dew Point. It also shows daily temperature range, but indirectly: the mean daily temperature range for July (Decade 1999 to 2008) is subtracted, to give the temperature range anomaly, then the sign of the anomaly is changed so as to plot in the same sense as the other variables.
There is little pattern or trend in July drought indicators in this period, except in two months: July 2002 and July 2010.
The year 2002 is recognised as one of extreme drought. At Manilla, July 2002 had extreme values of low rainfall (1.0 mm), low Dew Point (-1.4 degrees), few cloudy days (23%) and high daily temperature range (18.5 degrees).
Wet climate prevailed in late 2010. July 2010 had extreme values of high rainfall (79.8 mm), high Dew Point (+4.2 degrees), many cloudy days (61%) and low daily temperature range (12.3 degrees).
Some other July months have very high or low drought indications on one or two of these variables, but not on all four. In particular, July this year had very low rainfall and humidity, but did not have many sunny days or a high daily temperature range.
Cloudiness, as discussed earlier , increased dramatically in August or September 2007. July months before that averaged 33% cloudy mornings: those after that averaged 51% cloudy mornings.
I started a thread on "Cloudiness" , but very few were interested.

Heat Indicators
Indicators of climate heat or warmth include mean daily maximum temperature, mean daily mean temperature, mean daily minimum temperature, and mean subsoil temperature. To show variations clearly, I have subtracted the mean values for July (Decade 1999 to 2008) to plot anomaly values.
In general, daily maxima ("days") do not track daily minima ("nights"). Very warm July days came in 1999 and very cold ones in 2007. Very warm July nights came in 2010 and very cold ones in 2002. Mean daily temperature varied much less. So did subsoil temperature, but subsoil temperature did not track mean daily (air) temperature. July this year had normal temperatures, except that the soil was very warm.
In this period at Manilla, the linear trend of July daily maximum temperature sloped down at 0.83 degrees per decade, and that of daily minimum temperature sloped up at 0.93 degrees per decade.
In the words of Mark Twain:
"Therefore, any calm person, who is not blind or idiotic, can see that" ... Manilla's July daily maximum temperature, starting at 17.0 degrees in 2004, and daily minimum temperature, starting at 2.2 degrees, if they converge at 1.76 degrees per decade, will reach the same value (10.0 degrees) in the year 2088. After that, nights will be warmer than days. evillaugh
_________________________
Data are cheap; information is expensive!

Top
#1003660 - 31/07/2011 21:50 Re: Observations of climate variation [Re: Keith]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
Originally Posted By: Keith
Surly, that link to Drosdowski in your penultimate post is most useful. I'm going to have a go at reproducing the process it describes, in relation to rainfall and the SSTs...although I've not yet worked out whether I can take it further and if so, how.


This looks like it will have to go through to the keeper in the way the paper deals with it. I think the authors accessed various corrections that were applied or used to fill gaps.

In any case it's now 10 years or more since that paper was written. SST data have since been reconstructed and although I have these, there are gaps all over the place. Also I noticed that some of these data were adjusted down sharply from 1950. Surely this wasn't a natural process; I've been suspicious of the whole process ever since noticing it.

Nevertheless, despite all of the foregoing, I will see what I can come up with based on whatever looks reliable. In any case, rainfalls in NSW are unlikely to be influenced directly by SSTs more than 20° either side of the Equator.

Top
#1003684 - 01/08/2011 08:28 Re: Observations of climate variation [Re: Keith]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
Just as a follow up to my last post..

I looked at the link that ROM posted in one of the GW threads to the GISS data, and how they appear to have done a lot of 'adjustments'.

If this is what's happening to SST data in general, and if it's a regular process, then I would say any scientific endeavours to ascertain the effects of SSTs on rainfall and temperature, have been seriously compromised.

Top
#1004079 - 03/08/2011 19:26 Re: Observations of climate variation [Re: Keith]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Dry air and little rain
Manilla Smoothed Monthly Anomalies of Climate Variables
Parametric Plots
Update for July 2011

These graphs have data points for the last 36 months. In that time, the climate at Manilla, NSW has experienced a cycle from near-normal (Aug-08), through very warm and dry (Nov-09), then very cool and wet (Oct-10) , and back to near normal (Jul-11). For most variables, the raw anomaly values of July 2011 are near the smoothed anomaly values for August 2008.
Three variables have not returned to normal. Rainfall and Dew Point are far below normal, as typical in drought, and Subsoil temperature is high.
Cloudiness is "normal" in the new cloudy regimen in place since September 2007.


Note:
New data for July 2011 allow updating with more smoothing applied to all months back to January 2011, which is now fully smoothed. Fully smoothed data - gaussian smoothing with half-width 6 months - are plotted in red, partly smoothed data uncoloured, and unsmoothed data for the last data point in orange. January data points are marked by squares.
_________________________
Data are cheap; information is expensive!

Top
#1004245 - 04/08/2011 15:29 Re: Observations of climate variation [Re: Keith]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
Originally Posted By: Keith
Originally Posted By: Keith
Surly, that link to Drosdowski in your penultimate post is most useful. I'm going to have a go at reproducing the process it describes, in relation to rainfall and the SSTs...although I've not yet worked out whether I can take it further and if so, how.


This looks like it will have to go through to the keeper in the way the paper deals with it. I think the authors accessed various corrections that were applied or used to fill gaps.

In any case it's now 10 years or more since that paper was written. SST data have since been reconstructed and although I have these, there are gaps all over the place. Also I noticed that some of these data were adjusted down sharply from 1950. Surely this wasn't a natural process; I've been suspicious of the whole process ever since noticing it.

Nevertheless, despite all of the foregoing, I will see what I can come up with based on whatever looks reliable. In any case, rainfalls in NSW are unlikely to be influenced directly by SSTs more than 20° either side of the Equator.


So I've had a bit of a go at this, although the paper under reference was more aimed at working out the relationship between large areas of the mainland and the SSTs.

I have run some simple linear regressions between the Manilla Post Office rainfalls from 1950, and a whole batch of SSTs. These were extracted for 40° to lon 280°E and 0°S to 40°S and were divided up into blocks of 10°.

The SSTs explain the highest percentage of the variance in the rainfall during El Nino years and the lowest in neutral years.
In winter in an El Nino year this percentage is 68.3% but drops to zero in neutral years. I'm at a loss to explain why.

For all seasons the total percentage for El Nino is 33.1% which is also odd seeing that individual seasons are higher, so it must have something to do with the way the months are divided up or there may be individual months with much weaker correlations.

There is, however, a sting in the tail. When, for good measure, I threw into the mix for winter and neutral ENSO years, a whole lot of other likely influences on the climate, such as specific and relative humidity, various other Nino indexes and so on, 4 of the previous SSTs that had no significance (refer to above) assumed better than 95% significance (that is, we could be 95% certain that the influences weren't just random).

Taken in isolation, for any independent variable the regressions will show how much of the variance in the rainfall might be explained by it. To get a better idea of things it's necessary to add more 'ingredients' into the 'pudding'. In the example of the previous paragraph, the total explained variance came up to 78%.

The above results should be read in conjunction with my earlier posts about SST adjustments.

Top
#1004250 - 04/08/2011 15:47 Re: Observations of climate variation [Re: Keith]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Thanks for the interest, Keith.
My impulse would be, having got a believable correlation, to log the actual local data together with the model data derived from regression to see whether the patterns agree, at least for some of the time.
I don't have the sequence of El Nino and non-El Nino years in my head. Is there an agreed tabulation?
_________________________
Data are cheap; information is expensive!

Top
#1004254 - 04/08/2011 15:58 Re: Observations of climate variation [Re: Surly Bond]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
I will look into that Surly and post back. It's easy to generate what the model predicts and what rainfall actually fell. I can't tabulate data in the threads themselves, only a horrible sloppy-looking spaced out series of numbers, but I'll put up a link on my site with a couple of charts.

Top
#1004424 - 05/08/2011 09:09 Re: Observations of climate variation [Re: Keith]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
This is a chart of actual versus modelled rainfall:



The rainfall values are after using a power transformation.
The green bar chart shows the phases of El Nino; the negative readings are for El Nino years, the positive readings, La Nina. Neutral ENSO years are depicted by the gaps between the other 2 traces.

Top
#1004496 - 05/08/2011 16:26 Re: Observations of climate variation [Re: Keith]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
Here's the same plot with 2 forecasts tacked onto the end:



The neural network (NN) plot hasn't really done things any justice although it's based on the same independent variables. The regression plot looks more convincing. I had also worked with the approach taken in the Drosdowski paper regarding the reduction of the data dimensions by principal components but this resulted in a very poor model. This requires more experimenting.

Correlations between the original data and both forecast models is around 73% to 75%.

I also forgot to mention before that I've only modelled the data from 1950; the charts are for a more recent time frame to conform better to forum graphics size requirements and also for better clarity.

Top
#1004873 - 07/08/2011 17:10 Re: Observations of climate variation [Re: Keith]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Great stuff, Keith!
Thanks for using Manilla data as a demonstration, and thanks for the graphical plot of the "ENSO code".
_________________________
Data are cheap; information is expensive!

Top
#1006802 - 18/08/2011 21:05 Re: Observations of climate variation [Re: Surly Bond]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Manilla NSW in Global Warming Context
Up-to -date data on global temperature change can easily be down-loaded from Ole Humlum's website "climate4you". (Thank-you ROM for the link.)
Humlum favours sampling windows 37 months wide. I have always used windows about six months wide, which show up Australia's vigorous Quasi-biennial oscillations of climate, seen here.
I tried Humlum's 37-month window, with quite startling results, as shown in the graph below.
Humlum presents three records since 1979 of global monthly air surface temperature anomalies:
* HadCRUT3: by the (UK Met Office) Hadley Centre for Climate Prediction and Research, and the University of East Anglia's Climatic Research Unit (CRU).
* NCDC: National Climatic Data Centre (USA).
* GISS: Goddard Institute for Space Studies, Columbia University, New York.
When smoothed by a 37-month running average, these data sets give very similar results. I use the GISS data because it matches my data best.

The match is very good, particularly in the sharp fall from the maximum in April 2006 to the minimum in September 2007. Where my data begins in September 2000, both curves rise steeply from low values, but mine peaks in August 2001, more than a year before a corresponding peak in global temperature (September 2002). After that, there is a plateau, where the graphs rise together to the highest peak (April 2006).
The other global data sets, HadCRUT and NCDC, have temperature falling or steady along the 2002-2006 plateau.
There are two reasons for plotting my data on a separate axis (on the right). First, the reference periods are different: GISS uses 1951-1980, while I use 1999-2008. Second, temperature varies much more at a single station than in the average of many stations around the world. I use a scale six times larger.
It turns out that the cold time in Manilla in late 2007, seen here and here, was a cold time world-wide.

I am over the moon bounce at getting agreement between data from my home-made thermometer screen and the best that world climatologists can do. It makes me inclined to believe some of the things they say. wink


Edited by Surly Bond (18/08/2011 21:12)
_________________________
Data are cheap; information is expensive!

Top
#1006870 - 19/08/2011 09:43 Re: Observations of climate variation [Re: Surly Bond]
Locke Offline
Weatherzone Addict

Registered: 27/12/2007
Posts: 4553
Loc: Brisbane
Interesting. For me the obvious question is why the big disparity on the data prior to 2002.
_________________________
This post and any other post by Locke is NOT an official forecast & should not be used as such. It's just my opinion & may or may not be backed by sound meteorological data. For official information, refer to Australian Bureau of Meteorology products.

Top
#1006892 - 19/08/2011 12:35 Re: Observations of climate variation [Re: Locke]
Keith Offline
Meteorological Motor Mouth

Registered: 16/12/2001
Posts: 6453
Loc: Kings Langley, NSW
I don't want to spoil the party but I find I can't win with the climatologists' data. They are constantly 'reanalysing' it and they redefine periods for which anomalies are calculated (I don't know how often but the latest SST anomalies are now measured against the 1980-2010 period (used to be 1960-1990 and, I suppose, 1970-2000). Some of the charts I have plotted from their data show a sudden nose-dive from around 1950. Is it only to suit a preconceived global warming hypothesis? I think this is a valid point to raise.

As usual I smell a rat when this happens and it makes me very angry. May as well fiddle the accounts of a commercial organisation. But in this day and age, skepticism is the unforgiveable sin...let alone holding any minority view that counters the mainstream's agenda of avoiding accountability.

Top
#1006896 - 19/08/2011 13:21 Re: Observations of climate variation [Re: Keith]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Locke, we should not expect agreement in every detail. There is much more agreement in the pattern, especially in the timing of events, than I thought possible.

Differences arise due to the difference in vertical scale. The 100-year linear trend of mean temperature in New South Wales is +0.08 degrees per decade. Thus, the upward slope on the 2002-2006 "plateau" in the GISS global mean air temperature data, smoothed using a 37-month average, is parallel to that NSW trend: temperatures remained uniformly above the trend. The upward slope on the Manilla data in the same interval was actually six times steeper. Then, the decline in Manilla data to the September 2007 trough was also six times greater than in the world figures.
_________________________
Data are cheap; information is expensive!

Top
#1010633 - 07/09/2011 19:01 Re: Observations of climate variation [Re: Surly Bond]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Summer 2010-11 showed warming and drying
Manilla Smoothed Monthly Anomalies of Climate Variables
Parametric Plots
Update for August 2011

New fully-smoothed data for summer 2010-11
Daily maximum temperature anomalies (x-axis, all graphs) are more clearly shown on the top right graph. They rose at 0.3 degrees per month: about the same rate as they had fallen in the winter. From the extreme cold (anomaly -1.96 degrees) of October 2010, the February value was half-way back to normal.

Monthly rainfall anomalies (y-axis, top left graph) decreased through summer, exactly reversing the winter trend.

Cloudiness (top right graph) also reversed the trend seen in winter but, for a given temperature value, there was now more cloud.

Dew Point anomaly (centre left) fell much more rapidly than it had risen in winter. It was near normal by February.

Temperature range anomaly (centre right) rose much more rapidly during summer than that of maximum temperature, and reached higher values. As has appeared persistently in this thread, temperature range is the first variable to reach a positive or negative peak anomaly value. Here, its minimum value was in September, a month before the minimum value of daily maximum temperature. On these axes, the curve loops clock-wise as a result.

Daily minimum temperature anomaly (bottom left) fell steadily towards normal through spring and summer, completing a broad 30-month loop with daily maximum temperature anomaly.

Subsoil temperature anomaly (bottom right) fell along with daily maximum temperature anomaly.

Partially smoothed and unsmoothed data since February
Early winter 2011 had very low rainfall, very low Dew Points, and warm subsoil. Raw anomaly values for August suggest that the climate has since returned to normal, if somewhat warm and dry.

Note:
New data for August 2011 allow updating with more smoothing applied to all months back to February 2011, which is now fully smoothed. Fully smoothed data - gaussian smoothing with half-width 6 months - are plotted in red, partly smoothed data uncoloured, and unsmoothed data for the last data point in orange. January data points are marked by squares
_________________________
Data are cheap; information is expensive!

Top
#1012339 - 15/09/2011 13:35 Re: Observations of climate variation [Re: Surly Bond]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
All days of the year trended cooler: nearly all nights of the year trended warmer
Latest Knappenberger graphs for Manilla, NSW

Graphs of this type, pioneered by Knappenberger et al. 2001, show, for a chosen sequence of years, how the trends of temperature differ between the coldest days (or nights) of the year, the hottest ones, and all those in between.
I have chosen to analyse data in nine-year blocks. In Post #3875 I compared the two nine-year blocks ending August 2008 and August 2009; in Post #884837 I compared the two nine-year blocks ending August 2009 and August 2010.


Graph No. 1 shows trends in daily maximum temperature (“Days”).
In the earlier period (blue) there was a trend to cooler days in the cooler half of the year, and trends to warmer days were grouped at the boundary between summer and spring or autumn. In the later period (red) ALL days of the year, without exception, trended cooler. Summer days cooled at about -0.15 degrees per year; winter days cooled at about -0.05 degrees per year.
This graph is less "interesting" than earlier graphs: temperature trends for the hottest days and the coldest days are not much different from those of the mean, which would appear even in the simplest analyses.

Graph No. 2 shows trends in daily minimum temperature (“Nights”).
As in last year's graphs, the new curve (red) is almost the same shape as the earlier one (blue), but each trend has moved: in the negative direction this time. Earlier, the move was in the positive direction. In fact, the trends for the nine-year period ending August 2011 are close to those for the period ending August 2009. The middle period, ending August 2010, was a time of maximum warming of nights at Manilla.
Throughout the periods considered (beginning 1999), the hottest summer nights have consistently trended cooler than other nights of the year. The cooling trend of mid-summer nights reached an extreme of -0.28 degrees per year in the nine-year period to August 2009. The following nine-year period to August 2010 was the period when most nights of the year were warming at their extreme rate (about +0.15 degrees per year). In that period mid-summer nights were still cooling, but at only -0.12 degrees per year.

These graphs have no predictive value. One could find out if the observed trends have affected peoples lives. Cooler summer days might have reduced sun-screen sales and office air-conditioner load: cooler winter days have surely reduced solar gain in solar-passive houses. Warmer winter nights may have reduced heating costs; cooler mid-summer nights may have required less air-conditioning in bedrooms.
_________________________
Data are cheap; information is expensive!

Top
#1012825 - 18/09/2011 11:55 Re: Observations of climate variation [Re: Surly Bond]
Seina Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7770
Loc: Adelaide Hills
Originally Posted By: Surly Bond
All days of the year trended cooler: nearly all nights of the year trended warmer
Latest Knappenberger graphs for Manilla, NSW

Very interesting post heading smile ... brings to mind a few ideas.

Top
#1012835 - 18/09/2011 12:40 Re: Observations of climate variation [Re: Seina]
Seina Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7770
Loc: Adelaide Hills
Originally Posted By: -Cosmic- (naz)
Originally Posted By: Surly Bond
All days of the year trended cooler: nearly all nights of the year trended warmer
Latest Knappenberger graphs for Manilla, NSW

Very interesting post heading smile ... brings to mind a few ideas.

The suggestion or idea is that cooler daytime/warmer nighttime implies a trend towards more "even" temperatures...

In the case of Southern Australia, becoming more humid sub-tropical.

It might not necessarily apply for these obs, but it's a thought nonetheless.

Top
#1012996 - 19/09/2011 10:51 Re: Observations of climate variation [Re: Seina]
Surly Bond Offline
Weatherzone Addict

Registered: 23/08/2003
Posts: 2193
Loc: Manilla, near Tamworth NSW
Yes, naz, it is a trend toward more even temperatures.
I see it as a trend from continental towards maritime.
This trend appears in world climates and in Australian climate over the last century.
No-one pays any attention to it, which is puzzling.
_________________________
Data are cheap; information is expensive!

Top
Page 8 of 19 < 1 2 ... 6 7 8 9 10 ... 18 19 >


Who's Online
1 registered (Foehn Correspondent), 47 Guests and 3 Spiders online.
Key: Admin, Global Mod, Mod
Today's Birthdays
droughtboy, stormchaser88
Forum Stats
29947 Members
32 Forums
24194 Topics
1529243 Posts

Max Online: 2985 @ 26/01/2019 12:05
Satellite Image