Here is how a BoM report (PDF) following the 1974 flood described the historical context, and offered a prescient warning (emphases added):
Prior to 1900 flooding occurred quite frequently at 1 to 8 year intervals and in one year (1893) four separate floods were recorded. Since 1900 flood rainfall has been much less frequent and the interval between floods has become much longer. Furthermore, dredging and other changes to the hydraulic character of the channel, together with the effect of Somerset Dam have reduced most floods in Brisbane in recent years and have eliminated the smaller floods. . .
. . . flooding is most common in the usual wet season months of January. February and March, and floods are rare from July to December.
The earliest flood recorded was in 1841. Its exact height is uncertain but it was said to be the highest flood known at that time. In 1857 (flood peak 4.42 m) a good deal of land, now the prestige suburb of St Lucia, but then a dense vine scrub, was submerged and in 1864 (peak 4.92 m) flood waters extended from the junction of Oxley Creek and the Brisbane River to the high land at the back of Coopers Plains, a distance of about 11 km. In the 1867 flood the original wooden bridge at the site of the Victoria Bridge was destroyed, and in January 1887 (peak 4.92 m) Bowen Bridge was washed away.
Three floods occurred during February 1893. During the first (peak 9.51 m) the ship Elamang and the gunboat Paluma were carried into and left aground in the Brisbane Botanical Gardens, and the ship Natone was stranded on the Eagle Farm flats. The Indooroopilly railway bridge and the north end of the old Victoria Bridge were washed away. Nine days later a second minor flood was experienced which attained a height of only 3.29 m. However, a week after that there was another major flood (peak 9.24 m) which carried the stranded Elamang, Paluma and Natone back into the Brisbane River!
Prior to January 1974 no flood this century had exceeded 4.5 m at the Brisbane Port Office. The last river flood of any consequence occurred in 1931 (peak 4.48 m), although in recent years there have been several severe floods in the Brisbane metropolitan creeks (in June 1967 and February and April 1972).
Because of changes in the physical characteristics of the river and its catchment, it is very difficult to calculate return periods for flooding in Brisbane. However, four floods well in excess of the 1974 levels have occurred in the past 133 years and, according to the Professor of Economic Geology at the University of Queensland (Professor Sergent), there is geological evidence of water levels 5.5 m higher than the 1974 flood in the Indooroopilly area of Brisbane.
Meteorological studies suggest that rainfalls well in excess of those recorded in the floods of 1893 and 1974 are possible. Therefore it seems certain that unless major flood mitigation schemes, such as the proposed Wivenhoe Dam, are implemented, floods even greater than those of 1974 will again be experienced in Brisbane.
17 comments:
The National Library of Australia have digitised much of Australia's newspapers and magazines dating back to 1804 to about 1982. This is a great research resource.
Access the site here...http://trove.nla.gov.au/ndp/del/home
A search of floods and brisbane yields over 143,000 results. This report on the record 1893 flood one of many...
http://trove.nla.gov.au/ndp/del/article/3555968
Guess it was a bad move to bury the Hydrologists reports and then allow property development on all that land that is best described as a "Flood Plain"
Mother Nature has one thing she wants us peon little humans to remember.
"Life is a series of Lessons, lessons will be repeated until learned"
But I am sure the drumbeat of "Global Warming" did it will drown out Mother Nature's message.
Roger,
I may be reading the chart incorrectly, but it looks to me like the red line is at 3.5 meters. I assume you mean the horizontal red line to be the current flood level, cited in the text at 4.5m. Also, what are the other two lines (blue and green)? Thanks.
-3-Sam
Correct ... blue and green refers to moderate and minor levels of flooding.
Roger ... thanks for posting this. I managed to find this after reading what I presume to be your comment in the Australian. I have been writing on a similar topic from a slightly different perspective.
for instance:
http://cjeastwd.blogspot.com/2011/01/queensland-floods.html
and back in October
http://cjeastwd.blogspot.com/2010/10/how-big-is-current-big-wet.html
I had been wanting to get a copy of that 1974 report but find (happily) it is linked to on your blog. I'll be keeping an eye on this blog too.
Thanks
Sam - Actually, the vertical red line at the far right of the graph represents the current flood level. The horizontal lines represent what have been defined as differing degrees of flooding (minor - blue, moderate - green, and red- major). These are defined based on the areal extent (and comcomitant degree of damage) that will be covered with flood waters at a given river level.
Roger--do you know how Wivenhoe dam or other flood improvements over the last 100 years affect the flood peaks?
-7-Mike
Wivenhoe was built after the 1974 floods so it is likely that it has eliminated minor floods since that time. Once the reservoir is full it has little effect, so I would also venture that it had little effect on the current flood peak. But these are just hypotheses.
The Wivenhoe Dam water level was adjusted to allow for flood mitigation, you might need to test your hypothesis on the effect of these dams. At one time it was at over 200%, hence part of the cause of the flood as the water had to be released. There is also the Somerset Dam further up the river. My hypothesis is that without the dams the level would have been a lot higher. Also the current flood was not cyclone related, as many other floods have been. Perhaps you need to hypothesize on the cyclone effects as well.
-9-MikeAinOz
The effects of the Wivenhoe Dam (and its management strategy) will certainly be the subject of a close look, see:
http://www.theaustralian.com.au/news/opinion/damned-if-they-do-damned-if-they-dont/story-e6frg6zo-1225988018615
While Wivenhoe Dam management, is an issue, my point was to try and get the flood into a historical perspective. I'm familiar with the topography of Brisbane, mainly from the bicycle point of view. The 1.45 million megalitres of water(at least) held back for flood mitigation would have reduced flood levels by a quite a few metres. Brisbane is not on a large flood plain, which is, of course part of the problem. My feeling is that this event is at least equivalent to the 1893 event and one lesson from this is that it flooded three times in that year and this is only the start of the wet season.
I think there needs to be proper hydrological analysis to put the flood in historical context and that your diagram is misleading.
Roger, I just found this which seem to support my gut feel of a couple of metres. http://www.theaustralian.com.au/news/nation/sluices-to-open-to-prevent-possible-second-flood/story-e6frg6nf-1225987410910
Professor Colin Apelt is a retired Brisbane hydrologist an no doubt better informed than either of us on these issues.
-11, 12-MikeAinOz
Thanks for the link, very interesting.
The graph is from the BoM, and is a very standard way to depict streamflow records.
-13-
Here is an engineer who claims that the Brisbane flood peak would have been lower under different management of Wivenhoe:
http://www.theaustralian.com.au/national-affairs/water-releases-before-deluge-too-low-dam-expert/story-fn59niix-1225989066171
Roger I was more concerned about the historical record. Particularly as over at WUWT, your modified graph is being used a a rebuttal of climate change influences in this tragedy. http://wattsupwiththat.com/2011/01/14/bogus-claims-on-australia-and-brazilian-floods-from-abc-and-dr-richard-somerville/#more-31662
Do you support this point of view?
-15-MikeAinOz
Thanks. Let me offer a detailed response.
Climate change, as defined by the IPCC, is a change in the statistics of climate over 30-50 years or longer. Thus it is just nonsensical to ask whether one particular event was influenced by climate change, because it cannot be shown to be the case. I don't expect that such efforts to attribute will go away anytime soon;-)
So long as climate change is defined in statistical terms, then the proper way to look for its signal is statistically.
In the link you point to Watts has the higher scientific ground and Somerville does not (sorry). The graph showing historical flood levels indeed indicates how difficult it is to tease out a causal signal in one event, but also over time. The greater the impact of human interventions in the floodplain (e.g., Wivenhoe) the harder it will be to see any climate signal over the long term, because the statistics are non-stationary.
Whatever impact human influences are having on the climate system (and yes I am convinced there are such influences) it will be impossible to see them in the Brisbane floods, much less demonstrate their impact empirically. People can make up narratives and explanations based on beliefs and assumptions, which is of course what Somereville did that Watts responded to.
The peer reviewed literature is very consistent on this point - a signal of human caused climate change in disasters has not been shown for any phenomena in any location. See this lit review:
http://rogerpielkejr.blogspot.com/2010/08/disaster-losses-and-climate-change.html
Here is a paper that helps to explain why this is the case:
http://rogerpielkejr.blogspot.com/2011/01/crompton-et-al-2011.html
So, if you are interested in winning a battle with Watts, my advice is to suggest to folks like Somerville that they refrain from making scientifically unsupportable claims.
Roger, I think you are wrong in your responce to MikeAinOz's question. Although it is impossible to tell from a statistical analysis whether a single event occurred because of global warming, it is certainly possible to analyze the statistics to show that an individual event is far more probable given global warming than it would have been in its absence.
Assuming no global warming, the 2011 flood is a statistical aberration. We know this because hydrological studies have shown the probable heights that historical floods would have reached if Wivenhoe and Somerset Dams had existed at their time. Most of the major floods cluster around 3.35 meters in height. The highest of them would have been the 1974 flood, at approximately 3.5 meters. In contrast, the 2011 floods reached 4.48 meters despite the effects of Wivenhoe.
http://bybrisbanewaters.blogspot.com/2011/02/2011-and-history-how-big-was-brisbanes.html
This is not by itself enough for attribution, although it is suggestive. However, Dr Sommerville did not rely on mere statistics to make his attribution, rather he relied on known physics. 2010 was an exceptionally warm year, and increased warmth increases water vapour in the atmosphere. This does not bring rain independant of atmospheric circulation, but where atmospheric circulation brings rain, increased relative humidity will result in more rain falling.
In Ausralia's case, the Coral Sea (of the coast of Queensland) was 1 degree warmer than the 20th century mean in December 2010, and 0.2 degrees warmer than in 1974 (which had a similarly strong La Nina). That temperature difference is probably directly attributable to global warming, and is known to be the cause of the increased rainfall relative to 1973/4. The attribution of the flooding events to La Nina conditions, but the unusual intensity and extent of the flooding events to global warming follows straight forwardly.
Post a Comment