Another cyclone forecast BOMs out

A characteristic of ABC News is it's lack of inquiry in testing the outcomes of predictions on climatic events made by experts. One would think that the outcome of predictions would be newsworthy given the amount of money the government allocates to our premier weather agency.

The Bureau of Meteorology annually release an outlook of the coming tropical cyclone season in Spring. We have previously  looked at the accuracy of BOM's cyclone forecasts and ABC's reluctance to cover the disparity between forecast and outcomes, but have not had a chance to follow up on recent pronouncements.

Looking at the 2016-2017 season last year ABC reported that "Weather forecasters are expecting an average to above-average number of tropical cyclones to form this season in the region that includes the Northern Territory."

BOM claimed with a "high" level of accuracy :

• The Australian region has a 67% chance of having more tropical cyclones than average, meaning a 33% chance of having fewer tropical cyclones than average. Typically, around four tropical cyclones cross the Australian coastline in a season. Outlook accuracy for the Australian region is high.
• 
  

So what was the outcome? It seems BOM no longer produce the Darwin Tropical Diagnostic Statement that provided a near real time summary of cyclone activity. Perhaps its buried somewhere else in it's labyrinthine web page. However Wikipedia produce a season summary:  

Once again the numbers paint a poor picture of BOM's cyclone forecast skill, but ABC don't think this is newsworthy.

Missing News: climate models stuffed (again)

We know ABC is a deadzone for climate science. The only climate science news its activist reporters cover favours alarm and sensationalism. No surprise then that this new paper by Ross McKitrick and Timothy J. Vogelsang has not made it past the gate keepers:

HAC robust trend comparisons among climate series with possible level shifts
Ross R. McKitrick, and Timothy J. Vogelsang

Environmetrics

Abstract

Comparisons of trends across climatic data sets are complicated by the presence of serial correlation and possible step-changes in the mean. We build on heteroskedasticity and autocorrelation robust methods, specifically the Vogelsang–Franses (VF) nonparametric testing approach, to allow for a step-change in the mean (level shift) at a known or unknown date. The VF method provides a powerful multivariate trend estimator robust to unknown serial correlation up to but not including unit roots. We show that the critical values change when the level shift occurs at a known or unknown date. We derive an asymptotic approximation that can be used to simulate critical values, and we outline a simple bootstrap procedure that generates valid critical values and p-values. Our application builds on the literature comparing simulated and observed trends in the tropical lower troposphere and mid-troposphere since 1958. The method identifies a shift in observations around 1977, coinciding with the Pacific Climate Shift. Allowing for a level shift causes apparently significant observed trends to become statistically insignificant. Model overestimation of warming is significant whether or not we account for a level shift, although null rejections are much stronger when the level shift is included.

For visiting ABC reporters there is a good summary of the paper by McKitrick at Steve McIntyre's Climate Audit.

Bottom Line

Over the 55-years from 1958 to 2012, climate models not only significantly over-predict observed warming in the tropical troposphere, but they represent it in a fundamentally different way than is observed. Models represent the interval as a smooth upward trend with no step-change. The observations, however, assign all the warming to a single step-change in the late 1970s coinciding with a known event (the Pacific Climate Shift), and identify no significant trend before or after. In my opinion the simplest and most likely interpretation of these results is that climate models, on average, fail to replicate whatever process yielded the step-change in the late 1970s and they significantly overstate the overall atmospheric response to rising CO2 levels.

Relentless heatwave continues

BOM's Christmas heatwave continues....

Dramatic photos of Sydney's Christmas heatwave

Warwick Hughes reports on the disparity between BOM's recent Christmas weather forecast and reality. In ABC's report of BOM's forecast this dramatic image forboding of the coming doom:

ABC/BOM: About one-third of the country is expected to be hit by a heatwave over Christmas, with temperatures likely to reach 40 degrees Celsius or more.

This was the heatwave as it passed over Sydney this morning...

Will ABC report on BOM's bomb? 

Extreme extrapolations more common

ABC did not find the time to report on recent news revealing problems with climate models. Instead they have covered a paper that uses climate models to provide a taste of future heat waves. Not surprisingly defective models highly geared to an exaggerated climate sensitivity produce extreme outcomes.

Extreme heat waves to become more common

In their study, Dim Coumou, from the Potsdam Institute for Climate Impact Research, and Alexander Robinson, from Universidad Complutense de Madrid used climate modelling to project extreme heat waves like those that hit Australia in 2009.

They found that extreme heatwaves will by 2020 affect about 10 per cent of total land area -- double today's figure. By 2040, it would have quadrupled.

And just one indication of the reliability of those models?

In contrast to earlier analyses for a ten-year period that indicated consistency between models and observations at the 5% confidence level, we find that the continued warming stagnation over fifteen years, from 1998 -2012, is no longer consistent with model projections even at the 2% confidence level.

In odd contravention of the normal rules of journalism ABC allow this "press release" to be published without question.

Naive journalism: the new standard in news from our public broadcaster.

Update: Some further comments at Judy Curry's Climate Etc blog. Curry one of ABC's missing climate commentators.

from The Atlantic Multidecadal Oscillation in twentieth century climate simulations: uneven progress from CMIP3 to CMIP5

Variability of the AMO in the 10–20/70–80 year ranges is overestimated/underestimated in the models and the variability in the 10–20 year range increases in three of the models from the CMIP3 to the CMIP5 versions. 

Missing News: Climate Models Falsified

Funny how The Science Show and ABC News have not reported on this new paper. Doesn't fit the ABC climate catastrophe Meme.

Can climate models explain the recent stagnation in global warming?

Answer....NO!

In recent years, the increase in near-surface global annual mean temperatures has emerged as considerably smaller than many had expected. We investigate whether this can be explained by contemporary climate change scenarios. In contrast to earlier analyses for a ten-year period that indicated consistency between models and observations at the 5% confidence level, we find that the continued warming stagnation over fifteen years, from 1998 -2012, is no longer consistent with model projections even at the 2% confidence level. Of the possible causes of the inconsistency, the underestimation of internal natural climate variability on decadal time scales is a plausible candidate, but the influence of unaccounted external forcing factors or an overestimation of the model sensitivity to elevated greenhouse gas concentrations cannot be ruled out. The first cause would have little impact of the expectations of longer term anthropogenic climate change, but the second and particularly the third would.

Missing News: More evidence of climate model failure

Roger Pielke Snr reports on another paper that  "documents the lack of skill in multi-decadal global climate models to skillfully predict climate conditions in the coming years. This paper involves the question of accuracy lost when radiation parameterizations are used at time intervals that are long compared to other physical processes in the models. "The paper is

Sun, Z., J. Liu, X. Zeng, and H. Liang (2012), Parameterization of instantaneous global horizontal irradiance at the surface. Part II: Cloudy-sky component, J. Geophys. Res., doi:10.1029/2012JD017557, in press. the full paper is available at the JGR site by clicking PIP PDF.

Look forward to seeing ABC's coverage!

Cyclone forecast a BOM - Part 2

Cyclone tracks in Australia region July 2010-June 2011 (11 tropical cyclones in total)

Abstract
BOM's 2010-2011 tropical cyclone forecast: 20-22 cyclones forecast with a high level of confidence.
BOM Summary: Above average tropical cyclone activity expected for the Australian region  (98% chance!)

Season outcome:  11 actual cyclones. 

We previously looked at the BOM's woeful 2011-2012 tropical cyclone forecast.  ABC provided some coverage of the original forecast (Here and Here)  but didn't think it newsworthy to cover the outcome of that forecast. Remember for the 2011-2012 cyclone season Australia's premier meteorological institute predicted an 80% chance that cyclones would exceed the long term average (12). The actual number of cyclones (8) was 33% below the long term average. Missed it by that much!

Looking to see if previous forecasts were any better, we turned our attention to the 2010-2011 season. This covered the time interval July 2010 to June 2011.

Here was BOM's 2010-2011 forecast (alternate site HERE)

Seasonal Outlook for Tropical Cyclones

Summary: Above average tropical cyclone activity expected for the Australian region

The outlook suggests that the coming tropical cyclone season is likely to have

• a higher than average number of tropical cyclones over the full Australian region,
• a higher than average number of tropical cyclones in the Western region,
• an average to above average number of tropical cyclones in the Northern region,
• a higher than average number of tropical cyclones in the Eastern region.

For the full Australian region, there is a high degree of confidence that the total number of tropical cyclones will be above average (see Table 1). The forecast values from the two models (20-22) are significantly higher than the long-term average value of 12.

Table 1 Forecast values for the 2010/11 Seasonal Outlook for Tropical Cyclones (TCs) for the four main tropical cyclone regions and the northwest WA sub-region (105°E to 130°E).

Region	Chance of more TCs than average	Likely number of TCs (average number)	Confidence (LEPS skill1)
Australian region	98%	20-22 (12)	High (46%)
Western region	93%	11-12 (7)	High (31%)
Northern region	67%	5 (4)	Low (3%)
Eastern region	87%	6-7 (4)	Moderate (21%)
Northwest WA	75%	7-8 (6)	Low (7%)

The likely number of tropical cyclones is indicative only. It is expected that the total number of tropical cyclones will be in the vicinity of the values listed, and not necessarily within the given range. The values are the most likely number of tropical cyclones forecast by two models.

Regional features

• The largest increase in tropical cyclone numbers is expected to occur in the Western region, where 11-12 tropical cyclones are forecast (4-5 more than the average value of 7 tropical cyclones).
• For the Northern region, an average to higher than average number of tropical cyclones is expected with 5 tropical cyclones being forecast (the average value is 4 tropical cyclones). However, as the statistical model used to produce this forecast has low skill in this region, this prediction should be used with some caution.
• The Eastern region is expected to experience a higher than average number of tropical cyclones, with about 6-7 tropical cyclones being forecast (the average value is 4 tropical cyclones).
• The outlook for tropical cyclones in the area from 105°E to 130°E, where tropical cyclones can impact upon coastal WA communities, shows a 75% chance of above average tropical cyclone numbers for the 2010/11 season. However, forecast confidence for this region is relatively low.

This outlook covers the period from July 2010 to June 2011. Most tropical cyclones in the Southern Hemisphere occur from November until April.

So how did they go?
The cyclone tracks for season 2010-2011 are overlain on BOM's regional map at the top of the page. These are derived from reports from the Darwin Regional Specialised Meterological Centre (links to monthly reports below). It seems BOM did worse in 2010-2011. For the 2010-2011 cyclone season Australia's premier meteorological institute predicted a 98% chance that cyclones would exceed the long term average (12). The actual number of cyclones (11) was below the long term average. Missed it by that much! Perhaps the media can exhibit a little more scepticism when the next forecast is made. We look forward to the following questions being posed when the next seasonal outlook is made:
So how did your last two seasonal forecasts go? 

Links to monthly BOM reports.

July-September No cyclones in Australian region 

October 2010 1 tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) Anngrek

November 2010 0 tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) 

December 2010 2 tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) Abele, Tasha

January 2011  5 tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) Vince, Bianca, Yasi, Anthony, Zelia

February 2011 2 tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) Carlos, Dianne

March 2011 0 tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) 

April 2011 1  tropical cyclone in Australia region (5°S-40°S, 90°E-160°E) Errol

May-June, 2011 No cyclones in Australian region 

Total cyclones Australia region July 2010-June 2011: 11


Coverage also at Wikipedia: 2010–11 Australian region cyclone season

Cyclone forecast a BOM

The 2011-2012 cyclone season came to a close in April. In October last year the Bureau of Meteorology issued its forecast for the 2011-2012 season. ABC provided some coverage (Here and Here)  focusing closely on forecasts for NW Australia. Here's BOM's forecast from last year:

2011-12 Australian Tropical Cyclone Season Outlook

Summary: Above average tropical cyclone activity likely for the Australian region this season

The outlook favours the following scenarios for the coming season:

• The whole Australian Region has an 80% chance of having more than the long-term average number of cyclones. The long-term average is twelve.
• The Western Region has a 65% chance of observing above average number of tropical cyclones, where the long-term average is seven.
• The North-western Sub-region has a 60% chance of above average number of tropical cyclones. The long-term average is five.
• The Northern Region has a 60% chance of above average number of tropical cyclones, where the long-term average is two or three cyclones.
• The Eastern Region has a 65% chance of observing above average number of tropical cyclones, where the long-term average is three or four.

Table 1 Forecast values for the 2011/12 Seasonal Outlook for Tropical Cyclones (TCs) for the four main tropical cyclone regions and the north western sub-region.

Region Chance of above average tropical cyclone activity Long-term average Forecast skill Australian region 80% 12 Very high Western region 65% 7 Moderate North-western sub-region 60% 5 Moderate Eastern region 65% 4 Moderate Northern region 60% 3 Low

BOM forecast the number of cyclones in all regions should have been above the long term average. It rated its forecast skill as "Very High". How did it play out???? Here's a map showing tracks of cyclones affecting the Australian region for the forecast period based on reports by the Darwin Regional Specialised Meterological Centre (links to monthly reports below).

Figure 1 Cyclone tracks for 2011-2012 cyclone season

Here's a summary table for Australia as a whole and each region:

Table 2 Actual cyclone numbers for the 2011/12 Tropical Cyclone Season. 

Region	Difference from (Average)	Reported Cyclone Numbers	Forecast skill
Australian region	-4 (12)	8	Poor
Western region	-2 (7)	5	Poor
North-western sub-region	-2 (5)	3	Poor
Eastern region	-2 (4)	2	Poor
Northern region	-2 (3)	1	Poor

Links to monthly BOM reports.

Nov 2011 No Cyclones  in Australian region (5°S-40°S, 90°E-160°E) 

Dec 2012 3 Cyclones in Australian region (5°S-40°S, 90°E-160°E) 0 NW region Alenga, Grant, Fina
Jan 2012 2 Cyclones in Australian region (5°S-40°S, 90°E-160°E) 2 NW region Heidi, Iggy
Feb 2012 1 Cyclone in Australian region (5°S-40°S, 90°E-160°E) 0 NW region Jasmine
March 2012 2 Cyclones  in Australian region (5°S-40°S, 90°E-160°E) 1 NW region Lua, Koji-Joni
April 2012 0 Cyclones  in Australian region (5°S-40°S, 90°E-160°E) 
Total for season: 8 in Australian region (5°S-40°S, 90°E-160°E) 


Seasonal forecasts are news, but the outcome of those forecasts are news also, especially when the forecast skill is so poor. For the 2011-2012 cyclone season Australia's premier meteorological institute predicted an 80% chance that cyclones would exceed the long term average (12). The actual number of cyclones (8) was 33% below the long term average. Missed it by that much!

Wikipedia also have a post HERE with similar numbers. 

Missing news: Problems with climate models

Roger Pielke Snr has an interesting post that will no doubt never be covered by the ABC's Science unit. It's on continuing problems with the skill of climate models, and based on a discussion of the work of Demetris Koutoyiannis.

From Koutsoyiannis et al 2011

“….we tested whether the model outputs are consistent with reality (which reflects the entire variability, due to combined natural and anthropogenic effects). Our results extend Huard’s statements further. Specifically, we show that, climate models are not only unable to predict the variability of climate, but they are also unable to reproduce even the means of temperature and rainfall in the past. For example, as we stated in our paper, “In some [models], the annual mean temperature of the USA is overestimated by about 4–5◦C and the annual precipitation by about 300–400 mm”.

Roger comments:
This Comment/Reply illustrates, in my view, the continued pressure on Editors not to publish papers that conflict with the IPCC perspective of the climate system and the ability of global climate models to provide skillful predictions decades into the future. Instead of showing in a quantifiable manner any flaws in the work by Demetris Koutsoyiannis and colleages, Huard 2011 resorts to semantics and criticisms of the review process. Whenever authors resort to such arguments, it illustrates that they cannot refute the substance of the research study.

See the whole post HERE.

Missing News: No skill in climate modelling

Earlier this year we reported that ABC failed to cover problems with IPCC climate models that predicted both more and less rainfall in Eastern Australia. Now a new study in Hydrological Sciences Journal compares model outputs with observed data finding "that model projections at point scale are poor, results show that the spatially integrated projections are also poor".

See: A comparison of local and aggregated climate model outputs with observed data 

Authors: G. G. Anagnostopoulos^a; D. Koutsoyiannis^a; A. Christofides^a; A. Efstratiadis^a; N. Mamassis^a

Abstract We compare the output of various climate models to temperature and precipitation observations at 55 points around the globe. We also spatially aggregate model output and observations over the contiguous USA using data from 70 stations, and we perform comparison at several temporal scales, including a climatic (30-year) scale. Besides confirming the findings of a previous assessment study that model projections at point scale are poor, results show that the spatially integrated projections are also poor.

Once again ABC provide its audience the climate science it wants you to see.

Climate models: raining on your parade ground

Anonymous writes:
You missed South-east climate changing: CSIRO
The SEACI uses DownScaling to apply the IPCC Global Climate Models to South East Australia. They are actually computer projections of computer projections which as everybody knows result in computer projections! Don't look now fellas but its raining on your parade ground.

With IPCC models predicting more and less rainfall for SE Australia (see our post
Missing News: Climate models are an each way bet) why doesn't the ABC spend some more time actually investigating these claims?

From that post "7 of 15 models predict that mean annual rainfall would decrease by between 2 and 10 per cent, while 8/15 models predict that rainfall would increase by between 2 and 10 per cent by 2030. "

Missing News: Climate models are an each way bet

Updated October 6 2010-see below
The NSW Government are in the process of introducing "Water Sharing plans" across the state. Based on the ABC 's reporting it seems that the plans have been met with some controversy. 

Irrigators await water charge decision (20/5/2010)

About 50 Peel Valley irrigators last night attended their first meeting since the Peel Valley Water Sharing Plan was gazetted. Under the new agreement, which comes into force on July 1, irrigators must make do with only 6.2 gigalitres of water. They are also waiting nervously on a recommendation from the Independent Pricing and Regulatory Tribunal, which will see charges for bulk water rise dramatically. Laurie Pengilly, from the Peel Valley Water Users Association, says water users could be paying more than three times their current rate.

(See also Draft water plan anger sparks talks - 6/11/2009, Call for amended water sharing plans-19/2/2010, and Irrigators hope for future water plan 'reassessment - 23/4/2010).

One aspect of the plans that the ABC does not appear to have reported on is the results of climate modelling contained in the NSW Office of Water's Background document titled "Draft Water Sharing Plan Greater Metropolitan Region unregulated river water sources." dated May 2010. From page 11...

Climate change and variability

The NSW Office of Water has forecast rainfall and runoff across NSW using 15 global climate models for the IPCC SRES A1B climate scenario. The A1B climate scenario indicates a global temperature in 2030 that is 0.90C higher than the global temperature in 1990. For the Greater Metropolitan Region the worst case forecast is a 5-10 per cent reduction in mean annual rainfall by 2030, while the best case is a 5-10 per cent increase in mean annual rainfall. 7 of 15 models predict that mean annual rainfall would decrease by between 2 and 10 per cent, while 8/15 models predict that rainfall would increase by between 2 and 10 per cent by 2030. The worst case prediction for mean summer rainfall is a reduction of between 5 and 10 per cent, while the best case prediction is an increase by between 10 and 20 per cent. 13/15 models predict that summer annual rainfall would increase by between 2 and 20 per cent, while 2/15 models predict a decrease of between 2 and 10 per cent. Worst case winter rainfall is a reduction of between 10 and 20 per cent, while the best case prediction is an increase of between 2 and 10 per cent. 7/15 models predict that annual winter rainfall for the region shall fall by between 2 and 20 per cent, while 8/15 models predict an increase in annual winter rainfall of between 2 and 10 per cent. The Office of Water has recently configured a hydrology model for the Hawkesbury-Nepean River to estimate variation in flows and frequency and duration of licensed access to flows due to water sharing rules. On completion of the hydrology modelling required for the plan, the Office intends to commence modelling the effects of the above rainfall forecasts on flows and water user access across the plan area.

Based on these modelling results the NSW Office of Water could have saved some money and flipped coin or rolled a dice for the same result. No wonder the ABC didn't cover the results - the headline"Climate model results no better than a magic 8 ball" certainly doesn't fit in with the ABC's Groupthink position on climate change. 


Update: We provide the following correspondence between ABC and ANW:
From ABC Received 8 September, 2010:
Thank you for your email.
On review, Audience and Consumer Affairs is satisfied that ABC News reported the Water
Sharing Plan on a newsworthy basis and in keeping with 5.2.2(f) of the ABC Editorial
Policies.   News reported the release of the plans and sought reaction from those
affected.
Nonetheless, please be assured that your comments are noted.  For your reference, the
ABC's Code of Practice is available online at:http://www.abc.net.au/corp/pubs/documents/200806_codeofpractice-revised_2008.pdf
Yours sincerely
Audience & Consumer Affairs


ANW replied 8/9/2010:
ABC Audience and consumer affairs appear to have mis-understood the complaint. I agree
that  the Water Sharing Plan was newsworthy. The complaint outlined lack of questioning
on the part of the ABC such that important information regarding the need for the plan
(ie failure of climate models to provide a degree of certainty over future rainfall
either more or less) was not covered. Are you suggesting that the ABC did not find
problems with the models newsworthy?


ABC responded 6 October 2010
Thank you for your email.  I apologise for the delay in responding.
I do not believe I have misunderstood your complaint; the story in question highlighted newsworthy elements of the Water Sharing Plan and I am satisfied that this was in accordance with 5.2.2 (f) of the Editorial Policies.  It noted that the new agreement was shortly to be introduced and included comment from those most affected; the irrigators.
Nonetheless, please be assured that your comments are noted.
Yours sincerely
Audience and Consumer Affairs


ANC replied 6 October 2010
Thankyou for your reply. I remain unsatisfied with it and in accordance with ABC's complaints process I request you forward it on the ABC Complaints Review Executive for further deliberation. The complaint outlined lack of questioning on the part of the ABC News. Why didn't ABC News provide coverage of the climate modelling results of the NSW Office of Water that show half the models predict more rain, and half predict less rain? Are you suggesting that the ABC NEWS did not find problems with the models newsworthy?


ABC provided this reply October 6, 2010

Thank you for your email.

To clarify, in the context of a short news online story which focussed on the introduction of the new agreement and the concerns of the irrigators, it was not necessary for the item to include reference to the climate models you cite to meet the requirements of 5.2.2(f).

If you would still like your complaint referred to the Complaints Review Executive, I would be happy to do so.

Yours sincerely
Audience and Consumer Affairs

ANC reply 6 October 2010

The concerns of irrigators were the subject of a number of ABC news reports, not just "a short news online story". The omission of important newsworthy points such as the failure of climate models needs to be seen in this broader context, and as such this news should have been part of ABC News' coverage of the issue. In this sense the ABC has also failed to achieve balance in its reporting. 

In light of this, please pass this on to ABC CRE for further consideration.

We await ABC CRE's response.