Information now available
to calculate flood volumes (APRIL 2011)
The Qld Government’s IQQM computer model
had completed these calculations 
when considering the 66% required for the Ecology at
The Brisbane River mouth. They are done on the basis that no dams or people were
present. It is called pre-development flows. This IQQM computer model is written
into an Act of Parliament and has the force of Law. That chart is not generally available and is
presented here for your inspection. Click to view pre-development inflows
Stream discharge volumes at Savages
Crossing. Savages Crossing is below the Wivenhoe Dam. The Wivenhoe Dam was 
not there in 1974 so the unrestricted flow in megalitres is available. The
recordings commenced in 1909 and are not available for the 1893 floods. This
convenient graph is provided by the Department of Environment and Natural
Resources (DERM). The data that we will inspect is monthly.
The Lockyer Creek discharge is included in
these figures. On catchment size this Creek can produce 25% of the flow through
Savages Crossing. Rainfall figures for 1974, being less than 2011 indicate less
than 25%.
Click to view the 1974 stream
discharges at Savages Crossing was 4,150,00ML
The IQQM computer model and Stream
Discharges are two quite different means of measurement. The IQQM calculations are
based on a daily time-step using rainfall and other associated data. The Stream
Discharges are the actual water that passes through the Crossing unrestricted
except for the Somerset dam.
The Savages Crossing Stream Discharge model
for 1974 shows the volume in megalitres to be 4,150,000ML for the 1974
year. The Pre-Development flows of the IQQM computer model shows 4,300,000ML.
The Savages Crossing graph starts at October 1973 and ends in September 1974.
There is only minor adjustment required to match the the 1974 year. It is
examined fully on this page.
This means that the data
from both sources can be considered reasonably consistent
Seqwater report on the January 2011
flood event 02/03/2011. Report on the Operation of the Somerset Dam and
the Wivenhoe Dam".
Brisbane City Council "Joint
Flood Taskforce Report" (JFTF) of March 2011. The Chair is Professor Colin
Apelt who is also a reviewer of the "Operations Manual" used by
Seqwater in control of the Dams.
Bureau of Meteorology
website has daily rainfall statistics for all rainfall stations in the
catchments. They date from the time each station was first opened.
This website concentrates solely on Flood
proofing Brisbane with mitigation in Ipswich and Gympie and Drought proofing
South East Queensland. It does not seek to
review Seqwater's handling of the 2011 flood event.
However, the pre-development
flows calculated by the IQQM computer model and the Stream Discharge flows at
Savages Crossing suggests caution before acceptance
of the above. The caution relates to the possibility of the floods being higher
than represented above.
A brief overview follows. They are examined
in detail in "Review of flood volumes" which may be skipped to
"How to deal with it".
2011
The volume for 2011 is accurate and accepted
except for the fact that it was two floods similar to 1893(2). In both cases a
minor flood was followed soon after by a major flood. The minor flood was
560,000ML and the major flood 2,090,000ML.
1974. The volume of the 1974
flood put forward by Seqwater for the 1974 flood is 1,410,000ML. They
insist that the 2011 flood of 2,650,000ML was 190 per cent of the 1974 flood.
The evidence that we are about to examine suggests a very different figure.
Savages Crossing records volumes of
2,394,206ML for January 1974 and 682,405ML for February 1974. This totals
3,073,611ML.This includes flows from Lockyer Creek estimated on a rainfall
and catchment basis at around 500,000ML.
The 1974 flood was late in the month of
January and the volume in February, after consideration of rainfall in that
month, was almost entirely related to flood runoff. For example, 170,000ML was
held in the Somerset flood compartment during the flood.
After allowance of inflow from rainfall in
the first 24 days of January, an estimate of the Lockyer Creek flows and a portion of February
2011 there was no less
than 2,200,000ML attributable to the 1974 flood.
This is more fully explored in "Review
of Flood Volumes"
1893.
There were three floods in the month of February. Two were major and one minor
between the two majors. The second flood concatenated (piggy-backed) on the back of the minor flood raising the second major
flood to almost the same
height.
This puts the volumes of both major floods at around 5,200,000ML. The
IQQM computer model suggests that the 1893 year produced 7,500,000ML.
The fourth flood for the
1893 year was in June and was minor. There remains a substantial volume of water
to be accounted for and could affect the volume presented above as the largest
flood.
Further
Review
of flood volumes
The
review's purpose is to determine the largest flows from the Wivenhoe/Somerset
catchments. It is to confirm or otherwise the volumes laid down by Seqwater
which has at its maximum
the 1893(1) at 2,744,000ML
The
final result of this review at the bottom of this section is the adoption of 3,000,000ML as the volume of the
largest flood that we have to deal with.
1893
There were four floods in
1893.
Three occurred in the Month
of February and one in June.
Their respective
heights above "High Water Springs" were Major 
flood
1893(1) 7.25 metres, minor flood of 0.75 metres and Major flood
1893(2) of 6.97 metres. Major flood 1893(2) concatenated (piggy-backed) on the minor flood.
The fourth flood was a minor flood of 1.5 metres in June 1893. Click
to view the February 1893 floods
As I understand it,
"High water springs" has no relation to 
seasons and is when the Sun
and moon have their strongest pull on tides
The Brisbane City Council
Joint task force place the volume of 1893(1) at 2,744,000ML.
Seqwater also places the
volume at 2,744,000ML on page 140 of section 8 of their report. Click
to view the Bureau of Meteorology heights gauge when reviewing the 1974 flood
They
make no comment on the volume of flood 1893(2) and the minor flood combined
total although the height was close to 1893(1).
The IQQM computer model of
pre-development flows has a total of 7,500,000ML for the year.
A reasonable estimate of the
volume of 1893(2) would be 2,500,000ML based on the volume of 1893(1). The total
of the three floods in February 1893 would have been 5,244,000ML. Click
to view rainfall data of February 1893. Below is the 1893 year for Kilcoy
Rainfall station.
The variance to be explained
by the 1893 June flood and other 1893 rainfall is (7,500,000ML - 5,244,000ML )=
2,256,000ML
The IQQM computer
model also tells us that the average pre-development flows at the Brisbane River
mouth for the 111 years 1890 to 2000 was 1,641,331ML including all floods. The
contribution of the Wivenhoe/Somerset was 58% or 951,971ML
This leaves 1,304,029ML
to explain. (2,256,000ML - 951,971ML = 1,304,029ML)
The Kilcoy Bureau
recordings for 1893 are attached. The main Stanley River rainfall stations of Mt
Mee and Peachester were not in operation in 1893. I have singled out these
rainfall stations as the Cromhurst laboratory close to the head water recorded
107 inches for a period of 23 days.
We have seen that the IQQM computer model
had a reasonable match with the Savages Crossing stream flows so it is reasonable to assume that it
is correct bearing in mind it has the force of Law. The recording of
2,744,000ML appears somewhat below the level reached but not significantly.
For the purpose of determining the flood
with the largest volume, I am therefore taking the volume of Flood 1893(1) at 3,000,000ML.
1974
Seqwater has placed this flood at 1,410,000ML.
( Seqwater revised this upwards to 1,724,881ML in the GHD
report December 2011)
The Brisbane City Council Joint Flood Task
Force (JFTF) also places the flood volume at 1,410,000ML.
It is a reasonable assumption that they are
reading from the same information.
It should be pointed out that in the
Seqwater report section 8 
page 140, the 1,410,000ML represents flood water
volume. It indicates that a further 170,000ML was not released from the Somerset
Dam. As we are endeavouring to assess the total inflow, presumably their
recorded
inflow is 1,410,000ML + 170,000ML = 1,580,000ML.
Click to view Seqwater's assessment of all major
flood volumes
While on that schedule, there was no flood
in March 1989. It is most likely a misprint and should have been April 1988. It
was the combination of the April 1988 and April 1989 events that overfilled the
new Wivenhoe Dam. It that condition the Wolfdene Dam, commenced after the
Wivenhoe was completed, was cancelled for political reasons
at that time.
The Savages Crossing Data has a
very different outlook.
The first point that we examine is the
graphs of both the pre-development flows and the graph of the flows through
Savages Crossing.
The first point that we examine is the
graphs of both the pre-development flows and the graph of the flows through
Savages Crossing.
Click to see (1) pre-development flows
(2) Savages Crossing flows and (3) Reconciliation of both of those graphs
Supporting information
The actual data on which Savages
Crossing Stream Flows are based.
The Pre-development flows are calculated by
the Water Resource (Moreton) Plan 2007 Act of Parliament. It is done on the
basis of rainfall and related maters.
Savages Crossing is done of the basis of
stream flows at that site. It does include Lockyer Creek. The flow from that
tributary in 1974 was less than 2011. The 2011 flows from Lockyer Creek appear
to be around 10% of the catchments of the Wivenhoe Somerset
They graphs support each other
and therefore should be judged as reasonably consistent.
The calculation page at Savages Crossing
reveals the the 1974 flood at around
2,200,000ML.
The 1974 flood is also calculated by
comparison with the major flood section of the 2011 flood.
This is best reviewed with the 2011
flood.
2011
Seqwater have confirmed that the 2011 flood
volume was 2,650,000ML. The Joint Flood Task Force (JFTF) of the Brisbane City Council
confirm this volume.
My calculations, made in the same manner as
the the JFTF, came to almost the same conclusion. The exception is that I view
the 2011 flood in the same way that the Bureau viewed the minor flood of 1893 preceding
the 1893(2) flood.
In my view, the 2011 flood was similar
to the second major flood 
in
February 1893 which concatenated (piggy-backed) on the back of a minor
flood.
Click to view Seqwater view of 2011 flood. Yellow additions
are mine. The 1893 graph on the right has been compiled by me.
The volume of the minor flood of 2011 was
calculated from Seqwater flood event, period 6, information with their inflow
volume at 560,000ML. It represents releases of 187,500ML plus held in the dam at
18% being 209,700ML and inflow from rainfall yet to reach the dam of
187,350ML.
That left a volume of 397,050ML (209,700+187,350) not yet
dealt with when overtaken by the major flood.
The volume of the
2011 major flood was 2,090,000ML (2,650,000-560,000). The overall total agrees with Seqwater at
2,650,000ML
This distinction permits us to compare the
1974 flood with the major portion of the 2011 flood. The comparison then becomes
2,090,000ML of 2011 with 1,410,000ML being the stated inflow applied to 1974.
This application reduces both floods to a
three day
event. The hydrologists engaged by the Insurance Council of
Australia seem to recognize this point in their investigations.
For strict comparison, only those rainfall
stations that were in operation in 1974 and 2011 were selected. Those few
outside this rule and in operation in 2011 were examined for statistical
aberration.
The first two aspects to observe are:
* Was the ground saturated?
They were both saturated.
* How does the low rainfall of the minor flood compare with the high
rainfall of
the major flood in production of inflow? This is to confirm saturation.
This chart constructed from Seqwater operational periods for
2011 shows that the rainfall into the dams in the minor flood period produced
the same relative volume of inflow as the major flood. The percentages are in red.
Click to examine
detail
This also means that the catchments were saturated from the
observable prior rainfall. If there was a loss in both events, then it was
consistent.
The rainfall information in the month of January 1974 prior to
the flood commencing on the 25th ensured that the catchments were in a similar
saturated condition.
Click all three to examine detail.
Turning to the catchment rainfall figures from the various
rainfall 
stations, we see that the total of the minor and major floods are
consistent with Seqwater operational periods information.
We now see that the rainfall averages for the 1974 major flood
exceed the major flood of 2011 by 15% in the Somerset and 12% in the Wivenhoe catchment.
With the foregoing in mind it is reasonable to deduce that the
inflow in 1974 to the Wivenhoe was 114% of 2,090,000ML being 2,383,000ML.
The variance from the stated flows of Seqwater and JFTF of
1,410,000ML is approximately 1,000,000ML.
The IQQM computer model of pre-development flows shows inflows
for the 1974 year at 4,300,000ML.
If the calculation of 2,383,000ML is accurate, there remains a
volume of inflow of 1,917,000ML to be accounted for in the balance of January
1974 and the remaining 11 months.
Click to examine the rainfall for 1974.
This remainder of 1,917,000ML is double the
annual flow average of 951,971ML over a period of 111 years. This is in addition
to the 1,000,000ML identified above
