Drought and Rainfall History in Australia

Was 2002 Australia's worst drought ever or worst drought in 100 years ? Read to see how the truth is nothing like the myth.
This site started in 2002 in response to alarmist media reporting of the drought that Australia was then experiencing.  All sorts of hyped up superlatives, "worst drought ever", "worst drought in a hundred years" etc became commonly used  for what was for most of the country, just another damned cyclic El Nino drought. The drought has been blamed on "greenhouse" and we have heard bushfire risk hyped in a similar, unscientific way.

Stunning lack of success by BoM in predicting rainfall 3 months ahead in their "Outlooks"

Following are reviews  of rainfall "outlook" forecast maps against actual results,  from Australian Bureau of Meteorology (BoM) web site. The BoM publish a rainfall outlook every month for the next three months ahead.  So we have these rolling three month forecasts which are in the left hand column.  Each month they also publish the actual rainfall map for the preceding three months so I have matched these side by side.  Any missing periods are due to missing maps on the BoM site.   Newest maps on top now.  The motivation for this comparison is to rub in the point that the best climate scientists can not predict the weather three months ahead yet Australian (and international)  climate scientists purport to predict the climate three decades and more ahead using Greenhouse climate models.  Similar reviews of BoM temperature Outlook maps  to end 2004 are posted in the  Coolwire 8 newsletter found at  http://www.warwickhughes.com/climate/
Temperature Outlooks for 2005 are in Coolwire 14.
From winter 2005 the maps are not clickable for a large version. You can see large maps on the BoM website.

Series of BoM 3 monthly rainfall “Outlooks” or predictions
Rainfall anomalies published by the BoM for same period as “Outlook” in left panel.
March 2006 prediction for April  to June 06, "50:50 chances for above average June quarter rainfall", love that 50:50 as the  BoM opts for a dryer  southern SA and far W Vic but little differentiation anywhere alse
Success for their small area of predicted dry is offset by total failure in north as wet persists and failure to predict intense dry areas in SW WA and eastern seaboard. Small area of success for  Pilbara predicted wet.

February 2006 prediction for March  to May 06, "50:50 chances for a wetter than average autumn", as the  BoM still goes for that dryer NE and this time a wetter Pilbara. Huge failure in far north as dry predict  gets rained out. Large failure in NSW, S Qld and the centre as  rains fail  where predictions were 50 50.  Intensity of SW WA dry not predicted but Pilbara OK.

January 2006 prediction for February  to April 06, "Reduced seasonal falls in north Queensland", as the  BoM opts for a dry NE and wetter S and W. Dry predict for N Qld is total failure. WA pretty good. Predicted wet in SA & Vic a 60% win. Dry result from SE NSW through the centre is a failure as most of those areas were predicted near 50%. Dry SE Qld is a partial win and the wet far north was predicted to be dryer than the norm.

December 2005 prediction for January  to March 06, "Mixed rainfall outlook for early 2006", as the  BoM opts for three definite predictions. A minor win in the WA wheatbelt but larger failure in the wet Pilbara - Gascoyne after predicted less than 45%. Big failure in far N where predicted dry was swamped by reality.  Partial win for small pale blue area in NE NSW but overshadowed by broader failure in surrounding dry areas. Tasmania pretty much dead wrong again.  Should have had some predicted  pale grey or yellow.

November 2005 prediction for December 05 to February 06, "50:50 chances for a wetter than average summer", equivocates the  BoM opting for a wet SE. HUGE area of WA wet was actually predicted sl drier. SE wet predict turned out drier. Large Qld dry area  predicted sl wetter in the 50 to 55% zone. Just so WRONG, everywhere.

October 2005 prediction for November 05 to January 06, "Above average seasonal falls a 50:50 prospect", says the  BoM betting on a wet centre and dry east.  
Wrong by most counts, the dry  E & SE predict turns out average or wet. The HUGE area of WA wet was not predicted at all. SW Qld dry was not predicted.  Partial success with NT wet but offset by being wrong in the centre and 3 corners region.  Tasmania, EXACTLY wrong again.

September 2005 prediction for October to December 05, "Mixed rainfall outlook for the December quarter", says the  BoM betting on a wet Pilbara - Gascoyne. 
BoM have a win in NE Qld. (Cheers) but it ends there. Big wet predict in west turns out average.  Major wet in NT, north WA crept up unseen.  Tasmania  EXACTLY wrong again.

August  2005 prediction for September to November 05, "Drier spring for parts of southern Australia", specifies the BoM Southern Australia was much wetter than the BoM model. Tasmania EXACTLY  wrong again.Another CLANGER as the BoM totally failed to see the major central to southern wet coming.

July 2005 prediction for August to October 05, "Decreased seasonal falls more likely in northeastern Australia", states the pessimsistic BoM Another complete failure for the BoM as NE Australia turned out WETTER than average. The vast majority of Australia turned out much WETTER  than the BoM prediction. Tasmania EXACTLY  wrong again.

June 2005 prediction for July to September 05, "Neutral odds for exceeding average September quarter rain", opines the pessimsistic BoM Some limited success in SW WA, Pilbara, and SA-Vic border. The model completely misread the focus of the standout dry anomaly in the East, which is located in the models main plus 50% zone. Tasmania EXACTLY wrong again. Overall could not score more than abourt 40%..

May 2005 prediction for June to August 05, "Reduced winter rainfall more likely in parts of southern Australia", sez the pessimsistic BoM about our main crop planting season.
Another big failure for the BoM prediction as vast areas of Australia incl Tasmania enjoyed average or over average rain. Their main little win in SW WA  is lessened by the fact that in the mid 1970's rain there declined about 10% and if the more realistic average since 1975 was used, there would be much less red and more blue.

April 2005 predict for May to July 05, "Wetter season favoured in SE Queensland and northern NSW" repeats the BoM.
Good result for the SE Qld / N NSW predict. However west WA, southern Vic, Tasmania and Cape York, not so good but hey, maybe a 55%.
MayJul 05 predict
MayJul 05 actual
March 2005 predict for April to June 05, "Wetter season favoured in SE Queensland and northern NSW".
Maybe a 55% overall result. Despite poor results in SW WA, central WA and  Cape York  and much of Vic. 
AprJun 05 predict
AprJun 05 actual
Feb 2005 predict for March to May 05,  "Little signal in autumn rainfall odds" stumbles the BoM.
Plenty signal here.  The prediction is so wrong everywhere. SW WA wet when predicted dry. Pilbara wet predict less than half correctNorth NT dry marred by too much rain. Stand feature of the result was huge dry anomaly from Snowy Mnts. to E Pilbara, not the slightest hint in the pitiful Outlook.
MarMay 05 preict
MarMay 05 actual
Jan 2005 predict for Feb to April 05, "Drier season more likely in north Queensland" says the BoM. The peak dry event was in SW NT & NW SA. The predicted peak wet area in Pilbara was in fact dry.  The actual wettest area in SW WA was prdicted dry.   The wet predict for southern SA and far west Vic was in fact dry. They totally missed the great central band of rain deficiencies.  Overall, say 35% score.
FebApr 05 predict
FebApr 05 actual
Dec 2004 predict for Jan to March 2005,  “Neutral odds for March quarter rainfall” says the BoM
The broad area of rain deficit sitting east-west from coast to coast was not hinted at in the prediction, so by this major criteria the prediction was a failure. However the 55% contour in SA and Vic turned out a partial success.  Some success can also be claimed for the eastern seaboard 45% prediction. Tasmania EXACTLY wrong.

Nov 2004 predict for Dec 04 to Feb 2005. “Generally neutral outlook for summer rainfall” opined the BoM.
Townsville hinterland dry prediction was 99% dead wrong. Vast areas of the inland and west had a dry summer, not hinted at in the BoM prediction. Victoria’s wet summer also not hinted at in BoM  predict.

Oct 2004 predict for Nov 04 to Jan 2005 “Below average seasonal falls more likely in parts of eastern Australia"  according to the BoM
Dead wrong again, VAST areas of eastern Australia enjoyed average or above average rain.  The BoM dry predict for SE Qld & NE NSW turns out EXACTLY WRONG.  Victorian wet not predicted. Major dry anomaly in centre and Great Sandy Desert not hinted at.

Sep 2004 predict for Oct to Dec 2004, "Increased risk of dry conditions in far southeastern Australia", said the BoM. 
Most of  "..far southeastern Australia." was average or wetter; NOT dryer. Partial success with Pilbara wet predict. West Victorian coast v good but inland Victoria a failure.  North Qld dry predict another furphy.  Vast areas of dry anomalies in centre were predicted in 50 to 55%  zone, exactly wrong.

Aug 2004 predict for Sep to Nov 2004 "Dry Spring more likely in parts of SA, Victoria and NSW", says the BoM
Although the main predict dry area in SA coloured brown was actually sl wet or average, the predict for dry in S SA and W Vic had partial success. The N Qld dry predict was quite good but the SW Qld wet predict turned out average or dry.  The main dry anomaly in the NT was predicted sl wet. Dry anomaly in SW and S of WA carried a 55% wet predict. The NSW coastal wet anomalies are in a predicted sl dry zone.  Overall, broad failures outweigh the few wins.  Tas OK

Jul 2004 predict for Aug to Oct 04 "Seasonal rainfall odds close to 50:50", says the BoM. Taxpayers should be asking what possible use are predictions like this so obviously  lacking any relationship to real world rain patterns.
NT prominent dry areas predicted sl wetter, WA dry anomaly in zone predicted to be sl wetter. Broad sl damp predict for east Aust coastal turns out mainly dryer or average except for NSW coast strip. Area of sl dry predict centred on NE SA turned out on balance av or sl wet.  So most major points of the Outlook turned out to be unreliable. Tas Vic OK.

Jun 2004 predict for July to Sep04  "Increased falls more likely in parts of Qld and WA for September quarter", says the BoM
North Qld predicted wet EXACTLY WRONG.   SW corner WA predicted wet also EXACTLY WRONG.  Pilbara to SA actual wet predicted sl dry.  Tasmania wrong again. I am starting to puzzle that  pure chance does not more often give them a win, anywhere.

May 2004 predict for June to August 2004  "Winter rainfall odds mostly neutral", says the BoM

South Australian wet was actually predicted DRY, how wrong can they be ?  SE Queensland dry actually predicted sl wet. (Sure, I know it rained there on 1st Sept).  Tasmania exactly wrong again.  Even the little predicted dry spot in north WA, exactly wrong and entire predicted dry belt between N WA through NT extending to SA and Vic to Tas just so wrong, actually 90% wetter or average.

April 2004 for May to July 2004 "Generally neutral seasonal rainfall odds", says the BoM
North NSW SE Qld predicted wet exactly wrong. Tas. wrong again. central Aust huge wet area not hinted at in prediction.

Mar 2004 for Apr to Jun 2004 "Mixed June quarter rainfall odds in the east and south"
Large central wet not predicted at all, east NSW predicted wet exactly wrong.  Se Qld dry not predicted.  SA and Vic predicted dry turned out much weaker. West WA  v poor contrary to predicted contours. Tasmania exactly wrong , again.

Feb 2004 for March to May 2004 "Mostly neutral autumn rainfall odds, except in WA", says the BoM
Good result in WA but not covered in glory elsewhere.

January 2004 for February to April 2004 "Neutral seasonal rainfall odds", says the BoM.

What can I say. Model utterly failed to pick up major rainfall patterns,  poor old Tassie wrong again

December 2003 for January to March 2004 "Neutral odds for March quarter rainfall", says the BoM.

Model fails to predict vast areas of northern and WA wet. The centre is best result.   The broad areas of  SA, Vic, NSW and Qld turn out EXACTLY WRONG. For SA Vic the predicted 55% chance of wetter conditions turns out to be mainly drier except for a thin coastal strip.  NSW divides into two zones separated by a NW / SE line, the SW zone was modelled slightly wetter yet turned out drier while the NE zone was modelled slightly drier but turned out wetter. The broad band of better rainfall from New England NW into Qld was actually modelled to be drier than av.  The model also fails to pick up the Cape York rainfall and only the two small pink areas in north Qld could be claimed as success.

For critiques of BoM "Outlooks" prior to January-March 2004

Back to http://www.warwickhughes.com/
Some station rainfall histories to compare.

Bourke         The worst hit in 2002 in data I have sighted, all time dryest year.

Central NSW, two  Lachlan Valley stations, annual rainfall 1876-2001 for Condobolin, Forbes P.O. 

Deniliquin near the NSW Victorian border

For Tamworth we have composite data, Curlewis used 1891-1902 then Carroll to 1997, 1998 was fudged from a neighbour then Tamworth takes over from 1999.   Tamworth

In Western Australia 2003 saw a bumper wheat crop, we now have  Merredin updated through 2003. In vicinity of Newdegate we have Lake Grace, Lake Carmody and Truro all from BoM high quality dataset  indicating rainfall history 1906-2003. Apart from 1918 to 1926 in LG and LC which diverge, the three stations are in fair agreement.

The Perth annual rainfall graph 1876 to 2003 shows the 2001-2002 dry conditions in relationship to previous dry periods and it is obvious there have been several dry spells worse than recent years.

Looking at the Perth graph of monthly rainfall anomalies January 1969 to Dec 2003 we also see several past droughts worse than the moderately dry winter of 2001. Much is made of the fact that January to March 2004 rainfall has been negligible in Perth and I  have not noticed any informed spokespeople, state that there is no correlation between rain in these early months and the annual rainfall, or wet seaon rainfall.  So I am saying it here, in Perth a dry January-March is no guide to a dry year.
Looking at these two graphs of Perth rain data  one would have to wonder why the State Government here is planning to build a desalination plant that would produce water at a cost of three or four times the present supplies from combined dam and underground resources.  There is no shortage of the present resource assuming normal prudent engineering to plan for our expanding population.  The previous Govt. may have neglected water engineering works in the face of growing demand but the current Govt. takes policy into the surreal.  See my Perth Water web site. See new page re cloud seeding.

Back to http://www.warwickhughes.com/

Read how CSIRO Chief made incorrect claims re SW WA rainfall data on ABC TV 7.30 Report

Track all grain and commodity prices

Qld DPI Long Paddock for El Nino Southern Oscillation (ENSO) actual daily values with 30 and 90 day rolling averages.

Site updated April 2005  by Warwick Hughes, email comments etc to wazzau2002@yahoo.com.au

You read it first here.