Flood warning system

Johnstone River Flooding

Flood Warning System for the JOHNSTONE RIVER based on information provided to the public by the Bureau of Meteorology

Johnstone River Catchment Map
(courtesy of BOM, Cairns)

Risk

Johnstone River in flood at Innisfail February 1999 (Cyclone Rona) Photo: Brian Cassey

The North and South Johnstone Rivers rise in the tablelands of the north tropical coast and flow through steep narrow gorges to their junction on the coastal plain at Innisfail. The head waters of the catchments are located in high rainfall areas and the rivers are capable of producing severe flooding, especially in the Innisfail area. The North and South Johnstone Rivers have a combined catchment area of about 1,600 square kilometres at Innisfail.

There is a strong rainfall gradient across the catchment with the heaviest rain generally falling along the eastern side of the catchment around Topaz, Crawfords Lookout and Innisfail. In the western area of the catchment, rainfall totals tend to be significantly less. Heavy localised rainfall along the coastal strip to Crawfords Lookout can cause rapid river rises in the lower Johnstone Rivers around Innisfail and Mourilyan, although larger floods tend to be associated with catchment wide heavy rainfalls.

The Johnstone River delta area can be subject to severe flooding with low lying areas being susceptible. Large areas of agricultural areas can be inundated and residential areas affected. Tides can have a significant impact on the smaller floods but little effect in larger floods.

Previous flooding

Severe flooding of the Johnstone River is often associated with tropical cyclones. The two highest floods at Innisfail in recent years occurred in February 1986 and March 1967. The flood records for Innisfail indicate that the highest recorded flood occurred in 1913 and that it was about 1.5 metres higher than the 1986 flood. Historical evidence indicates that the floods in 1893 and 1894 were even higher.

JOHNSTONE RIVER at INNISFAIL WHARF
Highest Annual Flood Peak

Forecasting

The Johnstone Shire Council, in conjunction with the Bureau of Meteorology operates a flood warning system for the Johnstone River catchment. The ALERT network consists of automatic rainfall and river height stations which regularly forward data via radio telemetry to a base station located at the Council offices in Innisfail and the Bureau's Flood Warning Centre in Brisbane. The system provides early warning of heavy rainfall and river rises in the catchment and enables more accurate and timely flood warning and forecasts. The balance of the network consists of volunteer rainfall and river height observers, who forward observations by telephone when the initial flood height has been exceeded at their station. The Department of Natural Resources also operate a number of automatic telephone telemetry stations throughout the catchment.

The Bureau's Flood Warning Centre issues Flood Warnings and River Height Bulletins for the Johnstone River catchment during flood events. Quantitative flood forecasts are issued when moderate flood levels are likely to be exceeded at Innisfail, with an objective to provide between 3 and 9 hours warning of flood levels and forecasts updated every 3 hours during the flood event.

Johnstone River ALERT system

The Johnstone River ALERT flood warning system was completed in 1989 as a co-operative project between the Bureau of Meteorology and the Johnstone Shire Council. The system comprises a network of rainfall and river height field stations located on the Tablelands as well as the coastal plain which report via VHF radio to a base station computer located in the Council office in Innisfail.The field stations send reports for every 1 millimetre of rainfall and every 50 millimetre change in river height.

In consultation with the Johnstone Shire Council, the Bureau issues Flood Warnings for the Johnstone River.

The base station computer located in the Johnstone Shire Council office collects the data and has software that displays it in graphical and tabular form. The data is also received by the Bureau's Flood Warning Centre where it is used in hydrologic models to produce river height predictions.

Flood warnings and bulletins

The Bureau of Meteorology issues Flood Warnings and River Height Bulletins for the Johnstone River catchment regularly during floods. They are sent to radio stations for broadcast, and to local Councils, emergency services and a large number of other agencies involved in managing flood response activities. Flood Warnings and River Height Bulletins are available via :

Radio
Radio stations, particularly the local ABC, and local commercial stations, broadcast Flood Warnings and River Height Bulletins soon after issue.

Local response organisations
These include the Councils, Police, and State Emergency Services in the local area.

Internet/World Wide Web
Flood Warnings, River Height Bulletins and other weather related data is available from the following links:

• The Bureau of Meteorology
  http://www.bom.gov.au/
• The Queensland Flood Warning Centre
  http://www.bom.gov.au/hydro/flood/qld

Telephone weather

Flood Warnings are available through a recorded voice retrieval system, along with a wide range of other weather related and climate information.

Main Directory Tel: 1900 955 360

Flood Warnings Tel: 1300 659 219

Telephone weather services call charges:

1900 numbers: 77c per minute incl. GST; 1300 numbers: Low call cost - around 27.5c incl. GST. (More from international, satellite, mobile or public phones)

Weather by Fax

Flood Warnings and River Height Bulletins are also available through a facsimile information retrieval system, along with a wide range of other weather related and climate information.

Main Directory (FreePoll)	Fax No. 1800 630 100
Flood Warnings	Fax No. 1902 935 065
River Height Bulletins	Fax No. 1902 935 057

Weather by Fax Call Charges: 1902 numbers: 66c per minute, incl. GST; 1800 numbers: Free (More from international, satellite, mobile or public phones)

Interpreting warnings and river heights

Flood Warnings and River Height Bulletins contain observed river heights for a selection of the river height monitoring locations. The time at which the river reading has been taken is given together with its tendency (e.g. rising, falling, steady or at its peak). The Flood Warnings may also contain predictions in the form of minor, moderate or major flooding for a period in the future. River Height Bulletins also give the height above or below the road bridge or causeway for each river station located near a road crossing.

One of the simplest ways of understanding what the actual or predicted river height means is to compare the height given in the Warning or Bulletin with the height of previous floods at that location.

The table below summarises the flood history of the Johnstone River catchment - it contains the flood gauge heights of the highest know floods recorded at selected river height locations, together with heights of recent floods.

River height station	Highest Recorded Flood  (in metres)		Mar 1967	Apr 1982	Feb 1986	Mar 1997	Feb 1999
Nerada	Feb 1999	11.15				10.30	11.15
Tung Oil	Feb 1999	10.81	10.33	10.19	10.17	9.34	10.81
Corsis	Mar 1997	6.68				6.68	6.68
Central Mill	Mar 1967	11.13*	11.13*	8.10	10.84	9.85	9.70
Innisfail	- 1913	8.09*	6.25*	5.81*	6.42*	5.85	6.37

All heights are in metres on flood gauges.
[*] These heights are taken at old gauge sites and may not relate to flood levels from existing gauges sites.

Assessment of flood potential

Major flooding requires a large scale rainfall situation over the Johnstone River catchment. The following can be used as a rough guide to the likelihood of flooding in the catchment:

Average catchment rainfalls of in excess of 200mm in 24 hours may cause stream rises with moderate to major flooding and traffic disabilities to develop, particularly in the lower reaches downstream of Nerada on the North Johnstone River and Corsis on the South Johnstone River extending to the coastal plain around the Innisfail township and the mouth of the Johnstone River.

Average catchment rainfalls of in excess of 300mm in 24 hours may cause significant stream rises with major flooding and traffic disabilities to develop, particularly in the lower reaches downstream of Nerada on the North Johnstone River and Corsis on the South Johnstone River extending to the coastal plain around the Innisfail township and the mouth of the Johnstone River.

Each river height station has a pre-determined flood classification which details heights on gauges at which minor, moderate and major flooding commences. Other flood heights may also be defined which indicate at what height the local road crossing or town becomes affected by floodwaters.

The table below shows the flood classifications for selected river height stations in the Johnstone River catchment.

River Height Station	First Report Height	Crossing Height	Minor Flood Level	Crops & Grazing	Moderate Flood Level	Towns and Houses	Major Flood Level
Nerada			6.0		7.0		8.0
Tung Oil			7.0		8.0		8.5
McAvoy Br		7.1 (B)	5.0		6.0		7.1
Corsis			5.0		5.5		6.0
Central Mill		3.9 (B)	7.5		8.0		8.5
Innisfail	3.5		5.0		5.5	6.0	6.0

Case study background

Key Locations in Flooding study

The Johnstone River system comprises the North Johnstone River and the South Johnstone River with their confluence being at the town of Innisfail. From the confluence, the river flows about 5 km to the ocean. The rivers have a combined catchment of about 1600 km2 with the North Johnstone being the larger of the two with a catchment of about 1030 km2.

The headwaters of the rivers are in the high rainfall area of the Cardwell Ranges. The rivers flow from the range down through gorges to the lower fertile floodplains that are predominantly utilised for agricultural purposes including sugarcane and banana farming. There are a number of townships on the floodplains including the major centre of Innisfail and the smaller townships of South Johnstone, Mourilyan, Wangan and Mundoo.

There is a history of severe flooding on the floodplain with considerable damage to property,agriculture and public infrastructure. Innisfail is most affected being at the confluence of the rivers and with development on flood prone land.

Flooding in and around Innisfail town occurs initially through backup of Saltwater and Sweeneys Creeks and then through overtopping of the banks around Innisfail and further to the north in larger floods.

The construction of floodgates on Sweeneys and Saltwater Creeks has helped to reduce the frequency of flooding in Innisfail, although the floodgates are overtopped in larger floods. The suburbs of Webb, East Innisfail and South Innisfail are affected by overtopping of the river banks and by back up from the Johnstone River through the Ninds Creek catchment. Parts of Innisfail Estate are affected in larger floods through overtopping of the river bank. Mourilyan is affected in larger floods when the South Johnstone River overtops its banks. These floodwaters pass through Mourilyan and into the Ninds Creek catchment before rejoining the Johnstone River at the confluence with Ninds Creek.

Consideration of options to reduce flooding impacts, and planning for future development requires an understanding of the flood behaviour. To develop a greater understanding of flooding, hydrological and hydraulic flood models were developed and calibrated to historical floods. These models were then used to simulate a range of design floods that were the benchmark for assessing both past and future works.
Once flood behaviour is understood, a strategic approach to controlling development on flood prone land, assessing the advantages and disadvantages of flood mitigation options, flood proofing properties and buildings, educating and safeguarding communities and protecting the natural environment can be carried out with confidence. This Study provides such assessments, and actions arising from the Study recommendations will be used to formulate a Floodplain Management Plan.

There have been a number of developments on the floodplain over the last 30 years that have raised concerns within the community as to their impact on flood behaviour. These include a levee on the northern bank of the river downstream of Innisfail known in the community as Carello’s levee, filling of the town swamp and construction of floodgates to protect the town. The flood model was used to quantify these impacts.

Objectives and key stages

The key objectives of the study were as follows:

1. develop a state-of-the-art computer model of the Johnstone Rivers System within the study area to define the nature and extent of the flood hazard;
2. model the effects of existing developments and existing flood mitigation measures to determine their impact on flooding, including community concerns raised during the resident survey and community open sessions;
3. propose, assess and recommend possible flood mitigation measures with consideration given to social, ecological and economic factors;
4. prepare a report detailing the development of the model, the assessment of the effect of existing development and flood mitigation measures, addressing community concerns and detailing proposed flood mitigation measures;
5. prepare a Floodplain Management Plan.

There were seven key stages in the study.

1. Data Collection
2. Flood Model Development and Calibration
3. Design Flood Analysis and Existing Flood Damage Assessment
4. Historical Floodplain Works Assessment
5. Floodplain Management Measures Assessment
6. Recommendation of Management Measures
7. Reporting

Model development and calibration

The flood model comprises a hydrological model and a hydraulic model. The hydrologic model determines the runoff that occurs following a particular rainfall event. The primary output from the hydrologic model is hydrographs at varying locations along the waterways to describe the quantity, rate and timing of stream flow that results from rainfall events. These hydrographs then become a key input into the hydraulic model. The hydraulic model simulates the movement of flood waters through waterway reaches, storage elements, and hydraulic structures. The hydraulic model calculates flood levels and flow patterns and also models the complex effects of backwater, overtopping of embankments, waterway confluences, bridge constrictions and other hydraulic structure behaviour.

The Bureau of Meteorology (BoM) has established and calibrated an URBS hydrologic model of the Johnstone River catchment. This model was reviewed and adopted for the study. Some minor modifications to the model sub-catchments were done by WBM to match the locations of the hydraulic model boundaries.

The complicated nature of the floodplain flow patterns and importance of obtaining community confidence in the process required that state-of-the-art modeling techniques be adopted. For these reasons, TUFLOW, a fully 2D dynamic hydraulic modeling system was adopted. In total, the hydraulic model covers approximately 125 km2 of the rivers and floodplain.

Comparing the model with the actual rainfall at Innisfail

Click the charts to enlarge

Design floods

Design floods are hypothetical floods used for planning and floodplain management investigations. A design flood is defined by its probability of occurrence. It represents a flood which has a particular probability of occurring in any one year. For example, the 1% Annual Exceedence Probability (AEP)
or 1 in 100 Average Recurrence Interval (ARI) flood is a best estimate of a flood which has 1 chance in 100 of occurring in any one year. It should be noted that planning for the 1 in 100 year ARI flood does not guarantee protection for the next 100 years.

Design flood levels, flows and velocities were determined for 100, 50, 20, 10, 5 and 2 year ARI floods.
The design floods were used to make an assessment of the financial losses to residential and commercial properties. These financial losses were then used as a basis to do an economic assessment of potential floodplain management measures. Historical damage to public infrastructure was documented where information was available.

Stream	Station	(a) Highest Gauge Level (m)	(b) Highest Recorded Level (m)	Date Opened	Event	(a)(b) (%)
Nth Johnstone	McAvoy Alert (112908)	Not Gauged	9.80	2000	12/02/1999	-
Nth Johnstone	Tung Oil (112004A)	9.78	10.81	01/10/1966	12/02/1999	78%
Nth Johnstone	Nerada (112905)	Not Gauged	11.35	1989	12/02/1999	-
Nth Johnstone	Innisfail (112900, 112901)	Not Gauged	8.09	1979	30/01/1913	-
Sth Johnstone	Central Mill (12101B)	6.55	10.84	1/10/1974	02/02/1986	34%
Sth Johnstone	Corsis	Not Gauged	8.63	1989	31/01/1994	-

Calibrating the Flood Model example 1999 flood

Flood ID	Recorded Peak Flood Level (m AHD)	Modeled Peak Flood Level (m AHD)	Difference (Modeled Recorded) (mm)
G1	4.23	4.27	40
G2	4.24	4.27	30
G3	4.26	4.27	10
G4	4.25	4.44	190
G5	4.41	4.52	110
G6	4.39	4.53	140
G7	4.27	4.27	0
G8	4.24	4.27	30
G9	4.27	4.27	0
G10	4.24	4.27	30
G11	4.84	4.81	-30
G12	4.21	4.27	60
G13	4.24	4.27	30
G14	4.21	4.27	60
G15	4.73	4.70	-30
G16	4.28	4.27	-10
G17	4.82	4.85	30
G18	4.4	4.38	-20
G19	4.03	3.93	-100
G20	3.87	388	10
G21	4.72	4.49	-230
G22	4.9	4.87	-30
G23	3.94	3.94	0
G24	2.91	3.04	130
G25	2.89	3.04	150
G26	4.25	Not inundated in the model
G27	5.35	5.49	140
G28	3.71	3.73	20
G29	3.33	3.56	230
G30	9.35	9.46	120
G31	4.94	4.88	-60
G32	3.617	3.63	10
G33	4.283	4.27	-10
G34	4.245	4.27	20
G35	4.401	4.52	120
G36	8.4	8.45	50
G37	8.39	8.45	60
G38	6.28	6.34	60
G39	7.45	7.05	-400
G40	8.4	8.44	40
G41	2.1	1.9	-200

Range	Percentage of Calibration Points within Range (%)
-400mm to -200mm	5.0
-200mm to -50mm	7.5
-50mm to +50mm	55.0
50mm to 200mm	30.0
200mm to 400mm	2.5

Overall, good agreement between recorded and hydraulic model flood levels was obtained for the calibration events, especially in the most recent February 1999 flood indicating that the model is reliably predicting the flooding behaviour of the current floodplain. It is recommended that results from the southern part of the hydraulic model be used with caution, as this part of the model was not calibrated.

Design hydrology

Hydraulic Model Calibration

Design floods are hypothetical floods used for planning and floodplain management investigations. A design flood is defined by its probability of occurrence. It represents a flood that has a particular probability of occurring in any one year. For example, the 1% AEP or 1 in 100 ARI flood is a best estimate of a flood which has 1 chance in 100 of occurring in any one year. It is important to acknowledge that the 100 year ARI event may occur more than once in a 100 year period as the definition of the event is that it occurs once, on average, in 100 years. Therefore, planning for the 1 in 100 year ARI flood does not guarantee protection for the next 100 years. Similarly, the 100 year ARI event may not occur at all within a 100 year period for the same reason. The 2 year, 5 year, 10 year, 20 year, 50 year and 100 year ARI were analysed.

Magnitude of flow

There are two main methods of determining the magnitude of the flow for a design event. These are listed below and explained in the following sections:

Method 1. Flood frequency analysis (FFA)

Flood frequency analysis (FFA) enables the magnitude of floods of selected ARI (Average Recurrence Interval) to be estimated by statistical analysis of recorded historical floods.

South Johnstone River

ARI Event (years)	Central Mill + Upstream of Central Mill (BoM Rating) Flows (m3/s)
2	590
5	1020
10	1330
20	1660
25	1770
50	2120
100	2510
200	2920
500	3520

North Johnstone River

ARI Event (years)	Goondi (DNRM Rating) + Tung Oil (BoM Rating) Flows (m3/s)
2	1280
5	2200
10	2910
20	3660
25	3920
50	4760
100	5700
200	6740
500	8280

Method 2. Design rainfalls with the URBS Model

Design flood events are produced using design rainfall events. To determine the intensity and distribution of rainfall that will produce a specified ARI design event, charts developed by the Bureau of Meteorology (BoM) are consulted. These charts are contained in a book called "Australian Rainfall and Runoff"(IEAust, 2001).

General Agreement between the two models

Major Historical Event	ARI based on FFI Results	ARI based on Design Events
1932	13	4
1935	18	5
1967	40	20
1979	12	5
1982	36	18
1986	36	18
1994	26	14
1997	18	9
1999	23	12

Plus historical acedotal evidence

It is really important that whatever model you use, you check it against some real-life flood data. No model is perfect, at best a model is a reasonable approximation of reality. If it does not line up with reak events then the model is of little value.

Year	Estimated Flow* (m3/s)	Comment from Alan Dunne
1878	9000	About 4.5m higher at Innisfail than 1967
1894	5500	About 1.6m higher at Innisfail than 1967
1911	5000	About 0.7m higher at Innisfail than 1967
1913	5800	About 1.7m higher at Innisfail than 1967
1927	4450	About equivalent to 1967

* rough estimate only

Adopted peak flows - human decision

Based on the models and the real data we must now make an informed human decision as to what flow levels constitute a 2, 5, 10, 20, 50, 100 year flood

ARI (Years)	Peak Design Flow (m3/s)
	North Johnstone River	South Johnstone River
2	2100	700
5	3060	1200
10	3630	1600
20	4400	2100
50	5340	2800
100	6140	3330

Comparison between historical and defined flood model

Flood (Year or ARI)	Flood Height at Innisfail Wharf Gauge (m AHD)*	Flood Height at Innisfail Wharf Gauge (m Gauge)
1878	9.0	11.0
1913	6.1	8.1
1894	6.0	8.0
100	5.4	7.4
1911 & 1935	5.1	7.1
50	4.9	6.9
20	4.5	6.5
1999	4.4	6.4
1967 & 1927	4.35	6.35
10	4.1	6.1
1997	3.85	5.85
5	3.8	5.8
2	2.7	4.7

* Some of the early floods are approximate levels only supplied by Alan Dunne (Dunne, 1999 and pers.comm.2003). The 1913 level is considered to be reasonable reliable as the level supplied by the BoM was independently verified by A. Dunne. In the Innisfail CBD the March 1967 flood levels were higher than the February 1999.

Prediction maps

A once in 2 year possible peak level flood	A once in 5 year possible peak level flood
A once in 10 year possible peak level flood	A once in 20 year possible peak level flood
A once in 50 year possible peak level flood	A once in 100 year possible peak level flood
Theoretical maximum possible peak level flood

Possible flood mitigation strategies

1. Constructed Carello’s Channel
2. Webb Levee
3. River Dredging
4. Removal of Carello's Levee
5. Removal of Fill In Town Swamp
6. Lowering the Height of the Bruce Highway
7. Carello's Levee

Carello's channel

The proposed channel would be subject to sufficiently high velocities during flooding to scour its banks. Therefore, it is assumed that the banks will be protected with rock and be sloped.

Sweeneys and Saltwater Creek floodgate levees are raised Based on experience elsewhere, the environmental impacts of the excavation on a wide range of environmental, social and cultural issues would need to be carefully assessed prior to approval being granted. The benefits of the channel would also need to be fully justified.

The analysis indicates that an acceptable return on the initial outlay would be achieved if the channel scoured within 20 to 30 years and no rock protection is required. Without having undertaken specific geotechnical or hydraulic investigations into the likelihood of a similar channel forming naturally, it is difficult to make a definitive statement on the viability of such a proposal. However, based on anecdotal evidence of scouring both prior to and after the construction of Carello's levee, it is considered unlikely that a channel of this size would form naturally over a 20 to 30 year period.

The raising of the levees is not expected to have any environmental impacts given that the floodgates are already in place. The increase in floodgate levee height would bring a positive social benefit by increasing the warning time to residents and business protected by the levees. However, it is likely that the fall of the floodwaters in the town area will be slightly retarded by the increased height on the levees.

Webb levee

It was assumed that the levee would be concrete in front of waterfront properties and earth in other areas.

River dredging

1. dredge to stockpile and sell in Cairns;
2. dredge to stockpile and sell locally;
3. dredge to spoil

However, anecdotal evidence obtained during discussions with long-term observers of the river would suggest that the sediment load in the river is relatively high. If this is the case, then there may be significant costs in maintaining the dredged river profile. It is possible that in a larger flood, the dredged channel may be filled which would then require that the full dredging be undertaken to maintain the flood benefits.

Measure	BCR	Key Issues
Constructed Carello's Channel	0.29	Provides widespread minor reductions in flood levels. BCR does not allow for maintenance, acid sulphate soils or cartage and does not use revised floor level data. High capital cost. Environmental considerations relating to the clearing of mangroves, excavation and disposal of soil.
Scoured Carello's Channel	0.1 to 2.8	BCR strongly dependent on period taken for channel to scour. BCR does not allow for rock protection. BCR used revised floor levels. BCR will be lower if Sweeneys and Saltwater creek floodgate levees are raised. Environmental considerations relating to the clearing of mangroves, excavation and disposal of soil and deposition of scoured material in river system. Further hydraulic and geotechnical investigation recommended to assess the likelihood of success.
Raising of Sweeneys & Saltwater Floodgates	2.6	Significant benefits in areas protected by levees. Some minor increases in flood levels 'outside' of the levee. BCR uses revised floor level data. Additional levee may be required at Scullen Avenue, although it is a high capital cost item with minimal benefit. Further investigation into the significance of a 30 mm increase in the peak 20 year ARI flood levels in the Jones Street area is recommended before including the levees in the scheme. No environmental issues.

Removal of Carello's Levee

Removing Carello's levee would give the river a wider channel through Innisfail

Town Swamp

By removing rubbish and fill that has accumulated in the town /swamp over the years, river flow should be improved and there is a greater low lying area in which to hold surplus water during flooding events.

Bruce Highway

The Bruce Highway is the major and only coastal main north south highway in the region. In order to prevent road transport from being stopped by minor flooding across the highway, it was raised sometime in the past. During major flooding the highway acts as a giant low dam, backing up huge volumes of water and making it harder for the area to quickly drain of flood waters. In backing up there is potential for the waters to flood numerous home and businesses.

Carello's Levee

Modifying the shape of Carello's Levee

This levee protects a residential suburb but also reduces the flow of the main river channel.

Dredging the river channel

By removing accumulated river silt from the main channel, river flow is increased. But dredging lasts only a short time and is expensive. Flooding rivers carry huge volumes of sediment. Sediment load can only be reduced by revegitating river banks upstream.

Impact of Carello's Levee on once in two year flood

The levee protects those on the left bank but puts those on the right bank at greater risk.

Impact of Carello's levee on once in ten year flood

In a once in ten year flood a considerable area od the right bank is inundated to protect a small but populated area on the left bank.

Flood damage assessment

Effects on the local economy

The Lower Johnstone River region is a primary industry based economy serviced by a number of townships, the largest being Innisfail. The region comprises predominantly floodplain lands used for sugar cane, banana and pastoral activities. During flooding under existing conditions, agricultural
activities sustain substantial flood damage, reflecting the location of these activities in the floodplain.

Damages are not limited to the agricultural sector with significant damages also occurring to residential property, businesses and public infrastructure, particularly in larger floods.

Flood damages are classified as tangible or intangible, reflecting the ability to assign monetary values. Intangible damages arise from adverse social and environmental effects caused by flooding, including factors such as loss of life and limb, stress and anxiety.

Tangible damages are monetary losses directly attributable to flooding. They may occur as direct or indirect flood damages. Direct flood damages result from the actions of floodwaters, inundation and flow, on property and
structures. Indirect damages arise from the disruptions to physical and economic activities caused by flooding. Examples are the loss of sales, reduced productivity and the cost of alternative travel if road and rail links are broken.

For the purposes of this assessment, flood damages are classified into the following categories:

Tangible and Intangible.

Tangible damages

Rural damages

• cane farming
• banana farming
• beef grazing

Area of rural land flooded

Flood Event (years ARI)	Total Area Inundated (ha)
100	10,000
50	9,200
20	7,370
10	6,100
5	5,220
2	3,810

 

Flood Event (years ARI)	Total Area Inundated used in Damages Calculation (ha)	Breakdown of Land use Inundation
		Sugar Cane (D>1.2m)(ha)	Banana (D>2.5m)(ha)	Beef (ha)
100	5,950	4,080	410	1,460
50	5,045	3,330	325	1,390
20	3,887	2,365	212	1,310
10	3,058	1,700	148	1,210
5	2,427	1,240	97	1,090
2	1,547	680	27	840

 

	Damages per Landuse ($2002)
Flood Event (years ARI)	Sugar Cane ($)	Banana ($)	Beef ($)	Total ($)
100	1,003,000	7,790,000	350,000	9,143,000
50	819,000	6,175,000	334,000	7,327,000
20	581,000	4,028,000	314,000	4,924,000
10	418,000	2,812,000	290,000	3,520,000
5	305,000	1,843,000	262,000	2,409,000
2	167,000	513,000	202,000	882,000

 

Flood Event (years ARI)	Annual Exceedance Probability	Existing Case ($2002)
		Total Damages	Incremental Area Under Probability-Damage Graph
100	1%	$9,143,000
50	2%	$7,327,000	$82,000
20	5%	$4,924,000	$184,000
10	10%	$3,520,000	$211,000
5	20%	$2,409,000	$296,000
2	50%	$882,000	$494,000
1	99%	$0	$216,000
Average Annual Damage (excl. floods > 100 year ARI)			$1,483,000

Urban damages
(residential, commercial and industrial)

Urban damages in the Johnstone Rivers system are concentrated in the Innisfail region and Mourilyan. However, this analysis also includes damage to residential properties outside of these townships such as smaller communities and farm houses.
The damage to urban areas is principally to property and can be categorised into residential, commercial and industrial sectors.

Flood Event (years ARI)	Annual Exceedance Probability	Existing Case ($2002)
		Total Damages	Incremental Area Under Probability-Damage Graph
PMF*	0%	$210,000,000
100	1%	$83,100,000	$1,460,000
50	2%	$44,800,000	$640,000
20	5%	$3,800,000	$728,000
10	10%	$860,000	$116,000
5	20%	$200,000	$53,300
2	50%	$6,000	$31,200
1	99%	0	$1,500
Average Annual Damage			$3,030,000

* A PMF (probable maximum flood) was not modeled. The total damages estimate for the PMF was calculated assuming a flood level 2 m higher than the 100 year ARI flood level. Neither the damages estimate now the flood level assumption should be quoted.

Flood Event (years ARI)	Annual Exceedance Probability	Existing Case ($2002)
		Commercial	Residential
PMF*	0%	$150,000,000	$60,000,000
100	1%	$65,000,000	$18,000,000
50	2%	$35,000,000	$10,000,000
20	5%	$2,400,000	$1,400,000
10	10%	$585,000	$275,000
5	20%	$133,000	$67,000
2	50%	$4,500	$1,500
1	99%	$0	$0

* A PMF (probable maximum flood) was not modeled. The total damages estimate for the PMF was calculated assuming a flood level 2 m higher than the 100 year ARI flood level. Neither the damages estimate now the flood level assumption should be quoted.

Infrastructure damages

Infrastructure damages includes damages to telephone, electricity, roads, rail, flood structures and other public utilities.

Example road damage

Road	Submergence Damage	Saturation Damage
February 1999
Palmerston Highway	$15,197
Innisfail - Japoon Road	$118,493
March 1999
Palmerston Highway	$20,539
Innisfail - Japoon Road	$58,897
February 2000 & March 2000
Palmerston Highway	$92,515	$99,952
Innisfail - Japoon Road	$104,333
South Johnstone Road	$2,397
November 2000 & February 2001
Palmerston Highway	$4,447
Innisfail - Japoon Road	$34,268

Intangible damages

There are a number of intangible costs of flooding to the community including the following:

• loss of life and limb;
• preparedness (cost of flood warning, planning, community education);
• inconvenience;
• isolation/evacuation;
• stress and anxiety;
• disruption;
• health issues.

These intangible damages are not easily quantifiable and have not been included in the monetary assessment of flood damages.

Cost summary

Summary total cost Including 1 in 100 year floods

Flood Event (years ARI)	Annual Exceedance Probability	Existing Case ($2002)
		Total Damages *	Incremental Area Under Probability-Damage Graph
100	1%	$92,243,000
50	2%	$52,127,000	$721,850
20	5%	$8,724,000	$912,765
10	10%	$4,380,000	$327,600
5	20%	$2,609,000	$349,450
2	50%	$888,000	$524,550
1	99%	$0	$222,000
Average Annual Damage (excl. floods > 100 year ARI)			$3M

* Excluding infrastructure and intangible damages

Summary including possible maximum flood

Flood Event (years ARI)	Annual Exceedance Probability	Existing Case ($2002)
		Total Damages *	Incremental Area Under Probability-Damage Graph
PMF	0%	$210,000,000
100	1%	$92,243,000	$1,511,000
50	2%	$52,127,000	$721,850
20	5%	$8,724,000	$912,765
10	10%	$4,380,000	$327,600
5	20%	$2,609,000	$349,450
2	50%	$888,000	$524,550
1	99%	$0	$222,000
Average Annual Damage (excl. floods > 100 year ARI)			$4.5M

* Excluding infrastructure and intangible damages
* A PMF (probable maximum flood) was not modeled. The total damages estimate for the PMF was calculated assuming a flood level 2 m higher than the 100 year ARI flood level. Neither the damages estimate now the flood level assumption should be quoted.

Economic and environmental considerations

The benefits of the construction of flood modification measures

In general, the benefits of the construction of flood modification measures are as follows:

• increased flood immunity of properties protected by the measure leading to;
• increased flood immunity of roads protected by the measure and thus improved mobility of the community during flooding;
• decreased cost of flood damage to properties protected by the measure;
• decreased potential for loss of life during a flood event within the area protected by the measure;
• decreased emotional, social and psychological trauma experienced by residents in times of
• flooding.

The overall financial viability of an option is initially assessed by calculating the monetary benefit cost ratio (BCR).

A financial project life of 50 years was chosen for this study. This does not imply that the projected structural life of the scheme is only 50 years. In fact, some measures should be effective in reducing the frequency of flooding for centuries to come.
It is not correct to simply multiply a long term average annual benefit by the financial project life of 50 years to derive a total worth of the benefits. To do so would ignore the important point that the benefits from this scheme (ie. reduced flood damages) will occur over time and in the future.

For example, a benefit of $2.3 million to be gained in 10 years time is not worth $2.3 million now but only $1.2 million now. This is because $1.2 million could be invested now and appreciate at say 7 % p.a. over and above inflation for 10 years. This would then be equivalent to $2.3 million in 10 years time.

This is called the Present Worth of the benefit. It is a universally accepted economic theory and used in all major project economic analyses. The adopted rate of 7 % is called the discount rate and is the middle of the range 6 to 8 % recommended by the Queensland Government for assessing public works.

Present worth of annual benefits

Year	Annual Average Benefit ($ million)	Present Worth ($ million)
0	2.3	2.3
1	2.3	2.2
10	2.3	1.2
25	2.3	0.4
50	2.3	0.1

If the present worth benefits for each year are totaled for the 50 years, the total present worth (or total benefit) of the benefits is $ 31.7 million. The calculation of the total benefit can be simplified through the use of a Present Worth Factor.
Rather than calculating the present worth for each year
and summing to calculate the total benefit, a Present Worth Factor can be used when the annual average benefit is identical in each year.

The Present Worth Factor

The Present Worth Factor is calculated using equation (1).

The Present Worth Factor is multiplied by the annual average benefit to calculate the total benefit.
The Present Worth Factor is 13.8 for a 50 year period and a discount rate of 7%.
It is interesting to note that if a longer financial project life of say, 100 years was chosen then the total present worth of the benefits is only $1.1 million more at $32.8 million. This is due to the fact that the present worth of the benefits to be accrued in the second 50 year period is low because of the length of time until the benefits are realised.

The procedure for calculating benefit-cost ratios

The procedure for calculating benefit-cost ratios is outlined below:

1. Calculate the average annual benefit associated with the option (i.e. the reduction in annual average damages) using the method described in Section 6.3,
2. Convert the average annual benefit to a total benefit by multiplying by the present worth factor*;
3. Calculate the total cost of the option.
4. Calculate the monetary benefit-cost ratio:

Benefit + Cost Ratio = Total benefit/total cost

Environmental considerations

Impacts on flood response and evacuations
Impacts on riverbank stability;
Public utility impacts for example, sewer routes may need to be revised.
Visual impacts and blockage of views Levees can have a detrimental impact on the visual aesthetics of an area. They can do this by blocking views or by visually spoiling a formerly attractive area.

• Heritage and archaeology impacts;
• Impacts to traffic routes; and
• Impacts to fauna passage and flora.

Flood hazard categories

It is necessary to divide the floodplain into flood hazard categories that reflect the flood behaviour across the floodplain. CSIRO (2000) refers to the degree of flood hazard as being a function of:

• the size (magnitude) of flooding;
• depth and velocity (speed of flowing water);
• rate of floodwater rise;
• duration of flooding;
• evacuation problems;
• effective flood access;
• size of population at risk;
• land use;
• flood awareness/readiness;
• effective flood warning time.

CSIRO (2000) suggests four degrees of hazard; low, medium, high and extreme.

The categorisation of the floodplain is largely qualitative using the above factors. For example, medium hazard is where adults could wade safely, but children and elderly may have difficulty, evacuation is possible by a sedan, there is ample time for flood warning and evacuation and evacuation routes remain trafficable for at least twice as long for the required evacuation time.

A key factor in the ease of evacuation from an area is the water depth and the velocity along the evacuation route, ie. the stability of pedestrians wading through flood waters or vehicles driving along flooded roads. CSIRO (2000) notes that there are estimation procedures available for stability estimation, but considers that further research is required across a broader range of conditions and so does not recommend a procedure for hazard categorisation on this basis.

Approach

In considering the application of these issues to the specific flood characteristics of the lower Johnstone River floodplain, it is noted that:

• duration of flooding is universally long (in the order of days) across the floodplain;
• warning times can be short (~6 hrs);
• rates of floodwater rise are reasonably fast; and
• flood awareness is generally high and does not vary significantly across the floodplain.

The above four parameters are not significantly variable across the floodplain to warrant specific treatment and are therefore not used to define variations in the flood hazard, but should be included in
development control measures. The flood hazard is therefore defined on the remaining, varying
characteristics of:

• the size of the flood;
• depth and velocity of floodwaters; and
• evacuation and access.

Hazard Category	Base Flood Event	Characteristics
Low	100yr	- Areas that are inundated in a 100yr flood, but the floodwaters are relatively shallow (typically less than 1m deep) and are not flowing with velocity. - Adult can wade.
High - Wading Unsafe	100yr	- The depth and/or velocity are sufficiently high that wading is not possible. - risk of drowning.
High - Depth	100yr	- Areas where the floodwaters are deep (> 1m), but are not flowing with high velocity. - Damage only to building contents, large trucks able to evacuate.
High - Floodway	100yr	- Typically areas where there is deep water flowing with high velocity. - Truck evacuation not possible, structural damage to light framed houses, high risk to life.
Extreme	100yr	- Typically areas where the velocity is > 2m/s. - All buildings likely to be destroyed, high probability of death.

Flood hazard definitions

Local area hazard map

Warning and emergency planning

In the Johnstone region, many people come from families that have resided in the area for several generations. In most cases, these people have either experienced a flood or have heard first hand accounts of floods from family members or friends. Therefore, they are likely to have a high level of flood awareness. However, these people may not be aware that there may be larger floods than those events that they have experienced or heard of. In addition, there are a significant number of new rural and urban residents in the region who may not have the same level of flood awareness. In some instances, these people:

• have not experienced a flood in the area;
• have not heard first hand accounts of previous floods;
• live in houses that are not near the river, but are actually in the floodplain and are subject to flooding; and/or
• are not likely to take flood warnings seriously.

Both groups of people, those who have a low level of flood awareness and those who may not believe that there will be a larger flood than the biggest historical flood, that should be the target of a flood information campaign.

Public information campaign

Some of the initiatives that could be utilised to convey the general messages are listed below:

• Slogan - a simple slogan that could appear on signs, booklets, stickers etc.
• Flood Signs - showing the colour-coded flood bands and the heights of previous floods. These could be erected along the riverbank (e.g. next to bridges) and could include photographs of previous floods at that location.
• Historical Displays – Similar to the flood signs but perhaps more regional in focus. An obvious place in Innisfail would be at the wharf (the display would need to be flood proof or replaced after floods!).
• Totem Poles - showing the colour-coded flood bands (refer to 10.1.5). In order to encourage community acceptance, it is recommended that flood totem poles do not include any information on historical floods nor any signage indicating their relation to flooding to reduce vandalism. This will help to reduce the community feel that they being publicly labeled as “flood prone”. This places a greater reliance on leaflets and advertising to ensure understanding of flood totems significance.
• Flood Awareness Leaflets - containing general flooding information, including an explanation of the colour-coded flood bands. This could be a separate leaflet or an expansion (and renaming) of the current Cyclone Advice Booklet. A leaflet could be sent to homes on a regular basis (e.g. sent out with the rates notice every few months), or the booklet distributed say once a year to every home.
• Flood Awareness Week - a week of the year (preferably at the start of Summer) devoted to promoting flood awareness. Features on flooding, including dramatic photographs of previous floods, could be run in the local newspapers. Local radio stations could hold competitions with a flood theme etc. Flood awareness workshops could be held with flood wardens during this week.
• Guided tours could be run showing historical flood marks, mitigation systems and flood warning systems.
• Flood Education in Schools - provide schools with information kits and activities that are designed to increase flood awareness. This could be coordinated with the Flood Awareness Week.
• Web Site – The JSC web-site should include flood awareness and flood warning information.

Johnstone River

Probable Maximum Flood (PMF)

Estimated probable maximum precipitation

Table 1: Extimated Probable Maximum Precipitation (PMP)

Storm Duration (hrs)	Total Rainfall Depth (mm)
	Combined Johnstone River Catchment
12	780
24	1110
48	1650

Estimate peak probable maximum flows

Table 2: Estimated Peak Probable Maximum Flows (PMF)

Design Flood	Peak Design Flow (m3/s)
	North Johnstone River (McAvoy Bridge: Catchment Area = 971 km2	South Johnstone River (Central Mill: Catchment Area = 480km2
PMF	13,950	5,740
100	6,140	3,330
PMF/Q100 Ratio	2.3	1.7

The peak discharge was within the bounds of rule of thumb calculations for the PMF. For example, the PMF peak discharge is typical 1.5 to 3 times that of the 100 year ARI discharge. The peak PMF discharge in the North Johnstone River at McAvoy Bridge of approximately 14,000 m³/s is about 2.3 times the 100 year ARI discharge at the same location. Another rule of thumb is that for a catchment of about 1000 km² (North Johnstone River at McAvoy Bridge), the ratio of the peak PMF discharge to the catchment area is typically withing 3 to 25. This gives a PMF discharge range of 3,000 m³/s to 25,000 m³/s.

Public Flood Totem

Example flood totem cnr Jodrell and Marjorie streets

Note that the actual totem would only show colours, not flood levels or flood ARI and the change in bands should be based on consequences rather than ARI.

Probable Maximum Flood