Page 16 of 18 < 1 2 ... 14 15 16 17 18 >
Topic Options
#1448261 - 10/01/2018 10:47 Re: Observations of climate variation [Re: Surly Bond]
Petros Offline
Meteorological Motor Mouth

Registered: 30/12/2002
Posts: 7128
Loc: Maffra, Central Gippsland, Vi...
"based on 1961-90" - only enhances my potential misconception?

Top
#1448263 - 10/01/2018 11:26 Re: Observations of climate variation [Re: Surly Bond]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 3157
Loc: Buderim
Based on 1961-1990 means that those years have been arbitrarily chosen as the 'average climate' and set to 0. The years 1900-1930 or 1988-2017 could just have easily been chosen to be set to 0. That would change all the numbers on the charts to be higher for 1900-30 or lower for 88-17, but the order of which year is hottest and overall shape would be exactly the same.

Top
#1448304 - 10/01/2018 19:12 Re: Observations of climate variation [Re: Surly Bond]
Petros Offline
Meteorological Motor Mouth

Registered: 30/12/2002
Posts: 7128
Loc: Maffra, Central Gippsland, Vi...
Please describe how 1961-90 can be "arbitary"?

A fair scientific observer would describe 1961-90 as a "good head start - after last cool period" - (or worse if it was "allowed to be said").

Top
#1448305 - 10/01/2018 19:33 Re: Observations of climate variation [Re: Surly Bond]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 3157
Loc: Buderim
What do you think the baseline period should be?

Top
#1448306 - 10/01/2018 19:47 Re: Observations of climate variation [Re: Mike Hauber]
Brett Guy Offline
Meteorological Motor Mouth

Registered: 05/10/2010
Posts: 5091
Loc: Bently Park, Cairns
Originally Posted By: Mike Hauber
What do you think the baseline period should be?


I would think every year we have records for. Why not?


Edited by Brett Guy (10/01/2018 19:48)

Top
#1448311 - 10/01/2018 21:04 Re: Observations of climate variation [Re: Flowin]
Seira Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7493
Loc: Adelaide Hills
Originally Posted By: Flowin
Originally Posted By: Surly Bond
Thanks, Seira.
Rainfall here varies, both in the seasonal bimodality and the bimodality in the total frequency distribution, as well as the independent variation in the four moments of it. I suspect that all of them may relate to the sub-tropical high pressure belt. That is, the relative and absolute effect of its north and south movement from time to time.
But that could be entirely too simplistic.


This topic of bimodality and different characteristics such as skew mean kurtosis etc is complex. A key issue with rainfall data is that analysis of point rainfall such as a specific location or gauge will include a mix of local influences out of sync with a broader region and in sync with a broader region.
That would need an analysis of rainfall over different spatial scale magnitudes of local to perhaps climate zone scale to identify consistency or otherwise. There are data sets such as AWAP and others that could support such analysis ..... But that would be a big effort to do that. I mean a computer with grunt together with clever programming.
Averaging period over 21 years may help bring local to regional scale closer but could also mask some characteristics variation over time.

While I am not greatly familiar with skew-mean kurtosis analysis, it does ring a bell smile . I might look it up in due course, unless others have a more straightforward description of the analysis process than I'd find in the literature...in which case I'm all ears smile .

Top
#1448317 - 10/01/2018 21:59 Re: Observations of climate variation [Re: Surly Bond]
Flowin Offline
Weather Freak

Registered: 15/10/2017
Posts: 481
Loc: Pinjarra Hills, Qld
Seira
Sorry I should have used some commas 😒
Skew and mean and kurtosis as separate stats measures is what I meant not some form of combined skew-mean kurtosis .. which would be unusual or abstract to say the least and I am certainly not in the game of inventing new maths.

The bimodality topic is something that Surly Bond has raised that I have not seen before. I Am familiar with kurtosis or "peakiness" of distributions and it is often more relevant to seasons or months rather than inter-annual. Skew or "longer tails" to one side of the distribution is common in rainfall and much more so in streamflow stats.

My main point though was always be wary of analysis of one gauge and check out further towards regional scale.

Top
#1448318 - 10/01/2018 22:09 Re: Observations of climate variation [Re: Mike Hauber]
Flowin Offline
Weather Freak

Registered: 15/10/2017
Posts: 481
Loc: Pinjarra Hills, Qld
Originally Posted By: Mike Hauber
What do you think the baseline period should be?


You will never find a suitable baseline, and many so called "scientific" studies trying to prove a point can be misleading by arbitrary baseline periods... either unintended or in some deplorable cases it is deliberate.... As has been discussed on forums before.

The best approach I find is test the theory or analysis with reasonably long period of data. Do it with the whole set of data. Repeat it with the first two thirds, and repeat again with the last two thirds. If there are notable differences there are problems with "baseline".

That of all of course assumes reasonable stationary data.... Which really doesn't exist for many variables... A key challenge for that forbidden topic..
_________________________
Models are for estimating and gauges are for knowledge.

Top
#1448323 - 10/01/2018 22:58 Re: Observations of climate variation [Re: Surly Bond]
Seira Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7493
Loc: Adelaide Hills
Thanks for your clarification Flowin.

Top
#1448325 - 10/01/2018 23:13 Re: Observations of climate variation [Re: Petros]
Kino Offline
Weatherzone Addict

Registered: 10/08/2017
Posts: 2312
Loc: Wollongong, NSW, Aus
Originally Posted By: Petros
Please describe how 1961-90 can be "arbitary"?

A fair scientific observer would describe 1961-90 as a "good head start - after last cool period" - (or worse if it was "allowed to be said").


Itís picked because it removes certain years & events which are unsympathetic to the cause.

Top
#1448341 - 11/01/2018 08:47 Re: Observations of climate variation [Re: Surly Bond]
retired weather man Offline
Weatherzone Addict

Registered: 01/07/2007
Posts: 4615
Loc: Wynnum
When I first joined BoM in the 60's, averages were decided on what was known as " standard 30 years ". At that time 1930 to 1960 were used as the standard 30 years. That decision came from WMO at the time.
_________________________
Wyn Nth 2018-Jan12.2(158),Feb264.4(146),Mar217.0(126),Apr65.8(96),May46.0(100),Jun39.8(74),Jul19.6(56.1),Aug 0.4(45),YTD664.2(801.8)

Top
#1448343 - 11/01/2018 09:13 Re: Observations of climate variation [Re: Surly Bond]
retired weather man Offline
Weatherzone Addict

Registered: 01/07/2007
Posts: 4615
Loc: Wynnum
Since I set up the Wynnum North station, my annual temp anomalies are as follows ...
2015 +0.11C
2016 +0.40C
2017 +0.35C.
I would imagine being close to the sea these figures are more tempered by the almost daily sea breeze when compared to inland areas where more extreme ranges exist.

Comparing Brisbane Airport ( 8km away with a similar climate )...

2015 -0.10C
2016 +0.50C.
However 2017 final data not yet available for Bne Apt.
_________________________
Wyn Nth 2018-Jan12.2(158),Feb264.4(146),Mar217.0(126),Apr65.8(96),May46.0(100),Jun39.8(74),Jul19.6(56.1),Aug 0.4(45),YTD664.2(801.8)

Top
#1448345 - 11/01/2018 09:15 Re: Observations of climate variation [Re: Surly Bond]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 3157
Loc: Buderim
Data with baseline of 1910-2017, which is all data as suggested by Brett Guy.



Data with baseline of 1988-2017.



The shape of the chart is exactly the same. The fact that 2017 is the 3rd warmest year is exactly the same.

Top
#1448348 - 11/01/2018 09:25 Re: Observations of climate variation [Re: Surly Bond]
Flowin Offline
Weather Freak

Registered: 15/10/2017
Posts: 481
Loc: Pinjarra Hills, Qld
Yep. Mike what you have shown is for ranks of temperature it is relatively insensitive to selected baseline. Not surprising really because temperature variation is quite limited compared to say other parameters like rainfall.

It is situations like rainfall analysis where selecting a limited or arbitrary baseline where it can go astray to misleading conclusions.

Top
#1448349 - 11/01/2018 09:33 Re: Observations of climate variation [Re: retired weather man]
Petros Offline
Meteorological Motor Mouth

Registered: 30/12/2002
Posts: 7128
Loc: Maffra, Central Gippsland, Vi...
Originally Posted By: retired weather man
When I first joined BoM in the 60's, averages were decided on what was known as " standard 30 years ". At that time 1930 to 1960 were used as the standard 30 years. That decision came from WMO at the time.


For Mike, if a 30 year average is the norm across Met Offices world wide, then I reckon it should be as is was when RWM started, the most recent 30 years.

To me, it should be longer (60+ years), which is possible in other countries I believe, but not for Australia. Here in Australia we have the issue of measurement quality, and especially, the "back adjustment/correction" of historical data that remains highly controversial - but cannot be discussed!

Top
#1448383 - 11/01/2018 17:46 Re: Observations of climate variation [Re: Mike Hauber]
Brett Guy Offline
Meteorological Motor Mouth

Registered: 05/10/2010
Posts: 5091
Loc: Bently Park, Cairns
Originally Posted By: Mike Hauber
Data with baseline of 1910-2017, which is all data as suggested by Brett Guy.



Data with baseline of 1988-2017.



The shape of the chart is exactly the same. The fact that 2017 is the 3rd warmest year is exactly the same.


You would expect the graph to have the same shape. It just alters the apparent anomalies relative to average. Seems to make sense that it should be done that way though.

Top
#1448396 - 11/01/2018 20:50 Re: Observations of climate variation [Re: Surly Bond]
Funkyseefunkydo Offline
Weather Freak

Registered: 30/04/2007
Posts: 629
Loc: East Lake Macquarie
It does have the same shape.

Top
#1448401 - 11/01/2018 21:25 Re: Observations of climate variation [Re: Surly Bond]
Funkyseefunkydo Offline
Weather Freak

Registered: 30/04/2007
Posts: 629
Loc: East Lake Macquarie
Hey kino. Found you on the twitter.

Top
#1448403 - 11/01/2018 21:31 Re: Observations of climate variation [Re: Surly Bond]
Petros Offline
Meteorological Motor Mouth

Registered: 30/12/2002
Posts: 7128
Loc: Maffra, Central Gippsland, Vi...
...the 3rd warmest year.

But to call it the third warmest year "on record". As Mike shows, the "record" is relative to the dataset chosen, for example, the same year shows 2017 to be either 0.5 or 1.0 deg. C above [the dataset chosen]. I acknowledge that the discrepancy appears higher using the longer dataset, the one I'd prefer (if I had faith in the database).

There's got to be a better word than "record" to use when a certain year exceeds a certain chosen dataset range.

I don't think temperature recordings in Australia go back any near far enough for the use of the word "record", for historic temperature comparison. Just my thoughts.

Top
#1448404 - 11/01/2018 21:40 Re: Observations of climate variation [Re: Surly Bond]
Seira Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7493
Loc: Adelaide Hills
Originally Posted By: Petros
I don't think temperature recordings in Australia go back any near far enough for the use of the word "record", for historic temperature comparison. Just my thoughts.

"Record," the word, has a lot of uses. When it is said "3rd warmest year"...on record, it is very clear that's within recorded history, not actual history smile .

Top
Page 16 of 18 < 1 2 ... 14 15 16 17 18 >


Who's Online
6 registered (Raindammit, pete28, Simmo FNQ, kizz, 2 invisible), 63 Guests and 2 Spiders online.
Key: Admin, Global Mod, Mod
Today's Birthdays
Prozerpyne23, Sunday
Forum Stats
29631 Members
32 Forums
23943 Topics
1495682 Posts

Max Online: 2925 @ 02/02/2011 22:23
Satellite Image