Arora Construction Pty Ltd v Gold Coast City Council
[2012] QPEC 52
•14 September 2012
PLANNING &ENVIRONMENT COURT
OF QUEENSLAND
CITATION: | Arora Construction Pty Ltd & Anor v Gold Coast City Council & Anor [2012] QPEC 052 |
PARTIES: | ARORA CONSTRUCTIONS PTY LTD (ACN 06 275 1820) And JANS CONSTRUCTIONS PTY LTD (ACN 06 962 2297) (Appellants) v GOLD COAST CITY COUNCIL (Respondent) And CHIEF EXECUTIVE, DEPARTMENT OF MAIN ROADS (Co-respondent by election) |
FILE NO: | BD 2355 of 2007 |
PROCEEDING: | Appeal |
ORIGINATING COURT: | Planning and Environment Court, Brisbane |
DELIVERED ON: | 14 September 2012 |
DELIVERED AT: | Brisbane |
HEARING DATE: | 16, 17, 18, 19, 20 April and13 June 2012 |
JUDGE: | Rackemann DCJ |
ORDER: | The further hearing of the appeal is adjourned to enable parties to consider the terms of an approval |
CATCHWORDS: | PLANNING AND ENVIRONMENT – Applicant appeal – Proposed residential unit complex in the Guragunbah flood plain – Where the development is appropriate in terms of land use, density and building heights – Flooding – Where the development site will be immune to flooding up to the probable maximum flood – Where the subject site will not have trafficable road access beyond a Q20 flood event – Whether reliance on early voluntary evacuation followed by ‘shelter in place’ is a viable alternative to evacuation – Whether subject site occupants are exposed to an unreasonable level of risk – Qualitative and quantitative assessment of risk – Whether alternate solutions meet relevant performance criteria – Grounds for approval not withstanding conflict. |
COUNSEL: | Mr C.L. Hughes SC with Mr M.A. Williamson for the appellant Mr R.T. Litster SC with Mr B.D. Job for the respondent |
SOLICITORS: | Thomsons Lawyers for the appellant Minter Ellison Gold Coast for the respondent |
INTRODUCTION
This applicant appeal is against the council’s deemed refusal of an application for a development permit, for a material change of use, to permit the construction of 270 apartments, in seven buildings (ranging from 3 to 7 stories in height), together with a modest 220 m2 restaurant. The site is 10.86 hectares in area and is situated atNerang-Broadbeach RoadCarrara, within the Guragunbah flood plain. It shares a boundary with a major roundabout, providing connection to the North via Nerang-Broadbeach Road, to the East via Gooding Drive, to the South via Robina Parkway and to the West via Gooding Drive. It is located conveniently to a range of urban facilities, including a recently approved Woolworths supermarket-based shopping centre which is on the opposite side of the Nerang-Broadbeach Road and also within the floodplain.
The built form is proposed to be restricted to only 30 per cent of the site, towards its southern end. The balance is to be improved for active and passive recreation and for ephemeral wetlands/flood storage.
The council originally notified a broad range of issues upon which it would rely in resisting the appeal. Most of those have subsequently been resolved as a result of the joint expert meeting process. The town planning experts agree that the development is appropriate in terms of land use, density and building height.[1] The residual issues relate to flooding.
[1]Exhibit 1, Volume 5, pp 1022-3.
The development application requires impact assessment. It is to be assessed in accordance with section 3.5.5 of the Integrated Planning Act 1997 (Qld) (IPA) and decided in accordance with sections 3.5.11 and 3.5.14 of the IPA. An application which conflicts with the planning scheme must be refused unless sufficient “grounds” are identified to overcome the conflict.[2] The expression “grounds” is defined, in schedule 10 of the IPA, to mean matters of public interest. The appellant bears the onus in the appeal.
[2]See Weightman v Gold Coast City Council (2002) 121 LGERA 161; Woolworths Ltd v Maryborough City Council (No. 2) [2006] 1 Qd R 273.
The Guragunbah flood plain is a large flood plain centred on the Merrimac/Carrara locality, generally to the east of the Pacific Motorway. Any development within a flood plain carries some attendant risk. As Senior counsel for the respondent pointed out, the “explanation” to the Natural Hazards Policy in the planning scheme acknowledges that “the introduction of new development into areas subject to natural disasters … such as flood … is unwise”. Although it goes on to say that “it may be possible to minimise risk with appropriate designand thorough environmental management measures. This option can be supported where there is no long term detrimental impact on the environment and the risk to life and property is known and accepted”.
One possible planning approach would be to exclude residential development from flood affected areas within the flood plain, but that would arguably lead to an unnecessary waste of development potential. The planning scheme does not adopt that approach. Rather, the relevant provisions (discussed below) contemplate “a limited mix of urban residential andtourist facilities, within a predominant environment of open space and ecologically sustainablewaterways”.[3] Residential development is contemplated within the flood plain,butthe provisions of the planning scheme require, inter alia, that:
[3]Exhibit 9, p 113.
(a) The hydraulic function of the flood plain is maintained;
(b) People and property in the flood plain are not exposed to an unreasonable level of risk; and
(c) The openness of the flood plain is maintained.
It is uncontroversial that the proposed development would maintain the hydraulic function of the flood plain, would appropriately maintain the openness of the flood plain and would not expose property to damage or risk of damage. The residual issues relate to whether the development would expose people to an unreasonable level of risk, particularly by reason of not having a road access which is trafficable by ordinary car beyond a Q20 flood event.
The safety of access, external to a site, in times of flood has long been recognised as a relevant consideration in development assessment, particularly where the statutory planning authority has incorporated relevant provisions in its planning scheme or in a policy. In Bruigom v RockhamptonCity Council [1997] QPELR 418, for example, a 6-lot residential subdivision was refused on the basis that the access road started to become inundated in a Q12 flood event and was covered by 360 mm of water, at the low point, in a Q20 event. The relevant policy, in that case, required the access to be ‘trafficable’ in a Q100 event.
THE PLANNING SCHEME PROVISIONS
Reference was made to some of the higher-order provisions of the planning scheme, particularly DEO Soc 7 (the location and design of development to minimise the potential risk to life and property from known natural hazards) and the Natural Hazard Mitigation Key Strategy (the first policy relates to ensuring that further development in areas prone to natural disasters is restricted, unless it can be acceptably designed to minimise risk to life and property). It is the relevant Local Area Plan (LAP) and applicable codes,however, which provide the greatest guidance.
The LAPs are intended to implement land use and development controls, and the purposes of the Land Use Themes. Further, “the controls included in each LAP provide guidance for the assessment of a development proposal,and translate the objectives and intent of the relevant Land Use Theme into land use and development provisions”.[4] In particular, the planning scheme states that the Flood Plain Management - Merrimac/Carrara Land Use Theme is implemented directly into the Guragunbah LAP (LAP).[5]
[4]Exhibit 9, p 99.
[5]Exhibit 9, p 100.
The flood plain is the subject of the LAP. The statement of Intent for the LAP includes that:
“This LAP seeks to protect the hydraulic functions of the Guragunbah flood plain, while allowing limited residential, tourism and recreational development. There is also an opportunity to provide an extensive open space resource for the city which complements existing and future urban development.”
The Desired Environmental Outcomes for the LAP relevantly include the following:
“3.1 The achievement of urban development that is of a scale, form and character appropriate to the immediate local area and the intended open character of the flood plain (refer to DEO.Soc.1).
3.2 The optimisation of opportunities for urban development, consistent with the environmental capacity of the land and the continuing functioning of the flood plain (refer to DEO.Soc.7).”
It may be noted that optimisation of opportunities for urban development is, within limits, regarded as a desirable outcome.It has already been noted that the proposal is appropriate in terms of its scale, form and character and would have no adverse impact on the functioning of the flood plain.
In dealing with “Local Area Features”, the LAP states, in part, as follows:
“In all cases, development must be designed to ensure no adverse flooding impacts, having regard to the cumulative effect of all local development within the flood plain. In addition, development will need to ensure that the level of risk to occupants is acceptable during flood events and that appropriate emergency response can be facilitated ...”
The residual issue, between the parties, is whether development ensures the level of risk to occupants is “acceptable” during flood events and that “appropriate” emergency response can be facilitated. There is no concern with floor levels of the buildings. As is discussed later, there is no realistic prospect of inundation of the apartments. The residual concern relates to the extent to which the access to the property, via the Nerang-Broadbeach Road, will be affected by flood.
The LAP has a place code. It is a performance-based code in the familiar performance criteria/acceptable solutions format. Adoption of the acceptable solutions is not mandatory. Compliance can be achieved by an alternative solution. The performance criteria of particular relevance are PC 18 and PC 20 (and their associated acceptable solutions), which provide as follows:
The relevant part of the Table to Acceptable Solution AS20 is as follows:
A maximum flood annual exceedence probability of 1.0 % is equivalent to a
Q100 flood.
The LAP goes on to discuss access to land in more detail as follows:
“Access to land
It is intended that, during a flood emergency, neighbourhood communitieswill continue to function as social entities without adding to the emergency response burden, and people will be able to walk from one dwelling to another, experiencing, at worst, negligible flood hazard conditions. It is hoped that local communities will be virtually self sufficient for the duration of the flood emergency.
Access to land should generally conform to the requirements of the land use served by the road, as indicated in Table to Acceptable Solution AS20, without requiring freeboard. However, a tolerance of 200 mm is permitted for longitudinal road drainage purposes.
Facilities provided will be constructed at a height that will provide immunity from a flood of at least the specified annual exceedance probability, together with an allowance for freeboard, as specified in Table to Acceptable Solution AS20.”
The acceptable solutions contemplate residential building platforms to Q100, habitable rooms to 300 mm above Q100 and road access generally to Q100. The proposal cannot achieve the last. The performance criteria however do not, in terms, necessarily require that level of immunity. The appellants seek approval on the basis of a different solution.
LAP Map 14.7 is a Conceptual Land Use Plan, which includes the site within the following precincts; Residential – Low/Medium Density Precinct (to a small extent in the North-East); Active/Passive Recreation Precinct and Open Space Corridor Precinct. Little assistance is to be gained from this however, because as the map itself states, it illustrates just “one long-term vision” and the LAP otherwise indicates that there is ”considerable flexibility for creative design and land use solutions”.[6]
[6]Exhibit 9, p 114.
The site is also subject to the Flood Affected Areas Constraint Code (FAACC). The stated Purpose of the code is:
“To ensure that, where premises within flood affected areas are to be developed, adequate measures are taken to:
§ ensure that the development does not cause, or have the cumulative potential to cause, real damage (as defined below);
§ provide standards for development in these areas that will ensure that the runoff from land and/or premises does not create any adverse environmental impacts.”
Real damage and cumulative impact are defined as follows:
§ “real damage: whether the development is likely to cause damage that would adversely affect land and/or premises to an extent likely to be actionable;
§ cumulative impact: whether the cumulative impact of development is likely to cause real damage”
It is uncontroversial that the development would not cause, or have cumulative potential to cause, real damage as defined, nor would it create any adverse environmental impacts.
The FAACC also states that development proposals will be ‘fully evaluated” against criteria which include:
§ “flood hazard: whether the development is likely to cause or worsen flood hazard;
§ risks: whether the risks associated with the development are fully known, quantifiable and capable of being dealt with to Council’s satisfaction, without any uncertainties; and
§ flood mitigation: whether flood mitigation works, intended to reduce flood risk, hazard and damage, do so without adversely impacting upon other land and/or premises.”
There is no difficulty with the first and third of those. Debate centred on the risk associated with the access road.
Key objectives of the FAACC include:
“(b) maintaining or improving the City's counter disaster response efforts during a flood emergency;
…
(g) achieving and maintaining a best practice approach to flood plain management;”
The “development requirements” of the FAACC are performance-based. PC5 and PC10, and their associated acceptable solutions, are of relevance to the issues between the parties. They provide as follows:
The table of acceptable solutions to AS10.1 is as follows:
The proposal cannot meet AS10.1 because it must rely on the Broadbeach-Nerang Road as its sole permanent access. That road is below the level of the designated flood and is not defined as “low hazard” for the purposes of the table to AS10.1. The appellants submit however, that the performance criterion is met
THE APPROACH TO ASSESSMENT
It has already been noted that the issues in this appeal arise because of the extent to which the Nerang-Broadbeach Roadis affected by flood. That road, which is a major arterial road within the road network, has, in recent times, been upgraded, but the road authority did not raise the carriageway to achieve Q100 flood immunity, notwithstanding that it is an access route for existing residential development. The road is trafficable by ordinary sedan in events up to Q20 and by heavy trucks to Q100.[7]
[7]T4-4, l 30.
This access issue is not confined to the subject site. It also relates to a good deal of the eastern part of the flood plain. On the western side of the flood plainaccess can be achieved back to the Pacific Motorway, which enjoys Q100 flood immunity. Aerial photography overlaid with the probable extent of flood inundation in a Q100 flood illustrates,[8] as counsel for the appellant pointed out:
“A cursory review of aerial photography for the locality reveals that there is significant development in the Guragunbah flood plain, including residential development and retail development, much of which does not have flood free access (ie access with an ARI 100 flood immunity) and of course substantial development which is inundated in such an event.”
[8]Exhibit 8, p 2.
Mr Molino, who was called by the respondent to give evidence about risk, referred to the floodplain as “high risk”, because of the “thousands of people on the floodplain and the distances they have to travel to get off the floodplain are, in many instances, multiples of kilometres and the roads on which they have to travel are cut by frequent floods”.
The development itself would enjoy far greater flood immunity than is the case for other residential properties shown as inundated on the aerial photography and also includes a number of mitigation measures (discussed later). It was accepted among the risk assessment experts (whose evidence is discussed later) that the flood fatality risk for the proposed development would be less than that for the wider flood-affected population of the Gold Coast. Further, as was pointed out, there is a substantial residential development (known as Emerald Lakes) occurring within the flood plain to the north of the subject site, which also has access only from the Nerang-Broadbeach Road.
The fact that other residential development has worse problems or that residential development has and is occurring in locations with the same issuemight provide a certaincontext or perspective, but does notwarrant approval of the proposal. Each application must be assessed on its own merits in the context of the relevant statutory planning documents. If, assessed on its merits, the proposal is not found to be deserving of approval, then the appeal should be dismissed. That is so even if there is other residential development which has occurred, or is occurring, in the area, the merits of which could also be called in to question for similar reasons.
It has already been noted that the proposal departs from acceptable solutions of relevant performance criteria. In support of its proposal, the appellant asserts that it has alternative solutions which satisfy the performance criteria and/or that there are sufficient grounds to warrant approval notwithstanding any conflict. In order to demonstrate the acceptability of their alternative solutions, the appellants have relied upon various measures which, it was submitted, would substantially reduce, if not completely obviate, reliance on Nerang-Broadbeach Road access in times of floods beyond a Q20 event,and which would reduce residual risk to an reasonable and acceptable level.
The experts called on behalf of the respondentshowed a reluctance to contemplate residential development without road access trafficable by sedan in events up to Q100. Dr Connor, for example, said “Good professional practice in floodplain management would not endorse new developments where access to and from the development is subject to inundation by floods less than the ARI 100 year event”.[9]On the evidence it appeared, at one stage, that this matter was to be a ‘test case’ as to whether, in the context of this floodplain, access via a road without Q100 flood immunity could ever be accepted.
[9]Exhibit 1, Volume 5, p 2094; see also the evidence of Mr Molino at T4-61, ll 1-4.
However, having been given an opportunity to take instructions, Senior counsel for the respondent said that there was no attempt to characterise this case as one which would be determinative of others. He did not urge the court to find that an access road withQ100immunity is a condition precedent to approval. He described his client’s position as “not the simplified position … that we have an unexceptional requirement to provide flood-free access during an ARI year event”.[10]Further, he did not suggest that, in order to be acceptable, the appellants’ alternative solutions must completely remove risk.[11] Rather, he submitted that:
[10]T6-3.
[11]T3-3 to T5-4; T6-18.
“It is a case where your Honour has to determine what is sufficient; what is acceptable; what is reasonable; and whether or not this case has got there ...” [12]
[12]T6-3, ll 39-42.
and,
“what is adequate; what is reasonable; what is acceptable or unacceptable. Those are the questions that are live for your Honour in this case”.[13]
[13]T6-4 ll 9-11.
I have assessed the appellant’s proposalconsistently with that submission.
THE ALTERNATIVE SOLUTIONS
The appellant’s first response to the access road issue was to propose a “voluntary early evacuation” system, so that those on the site could safely evacuate before the Nerang-Broadbeach Roadbecame untrafficable by sedan. The evacuation route would be North along Nerang-Broadbeach Road to Ross Street, then East on Ross Street to a location adjacent to Royal Pines. The length of the route to Royal Pines is 3.6 km.
Such a system is possible because the riverine flood within thiswide floodplain rises more slowly than is the case for areas subject to flash flooding. Floodwaters rise very slowly over more than 2 days, before peak flooding is reached. There is no significant inundation of the site for some 36 hours after the start of the flood event, in which time Bureau of Meteorology flood warnings would have been issued and updated several times. The proposed evacuation route remains trafficable for some 42 hours after the start of an event.[14]
[14]Exhibit 4, p 8.
The voluntary early evacuation system involves an onsite automatic water level monitoring station. Action would be triggered when the water level comes to within 0.5 m of the edge of the pavement of the lowest section of the Nerang-Broadbeach Road. The next ‘trigger’ is when it rises to the road shoulder. The procedure would include a pre-recorded message to advise occupants to safely evacuate (if they wish).The onsite manager/caretaker would also advise residents, particularly the elderly and those with medical conditions or on medication, of the impending flooding. In a Q100 event, there would be 2.5 hours from when site evacuation is initiated until flood waters are lapping at the edge of the road pavement and a further 1.5 hours until the water level over the road reached the upper limit of its trafficabilty by sedan. Once residents are alerted to the prospect of road access being cut off by flood, they would have in the order of four hours within which safely to evacuate by car.
This evacuation is to be managed by the body corporate, rather than burdening emergency services.[15] An Emergency Flood Management Plan is to be settled and adopted.[16] The creation of community emergency plans by developers is encouraged in the Gold Coast City Local Disaster Management Plan.[17]
[15]Some of those who leave may choose to stay at an evacuation centre if they do not have other arrangements. For the reasons documented later however, it is likely that most residents will remain.
[16]This would occur at the time of determining conditions of any approval. Preferably, it would maximise the use of structural/automated mechanisms.
[17]Exhibit 13, p 72.
The time period is not unreasonable. Mr Collins, who was called by the appellant, was comfortable with the opportunity to evacuate. Mr Molino, who was called by the respondent, assessed the reasonableness of that timeframe having regard to the New South Wales SES model for planning evacuations. That model allows one hour for people to accept the warning and another hour to prepare to evacuate. Using the model, Mr Molino predicated that three quarters of an hour would be needed for people to leave the site. He also adopted the SES model’s assumption of allowing one hour for traffic delays due to breakdowns et cetera. Making all of these allowances, Mr Molino predicated that three and three quarter hours would be needed for people to accept the warning, prepare to evacuate, leave and negotiate potential traffic delays. Making a further allowance for poor driving conditions, Mr Molino increased his estimate to four hours for the occupants of the development to be evacuated.[18]He therefore concluded that there would be “barely sufficient” time for all residents to evacuate.
[18]Exhibit 11, para 39.
Mr Collins is of the opinion that the various allowances made by Mr Molino are excessive in the context of a managed evacuation of the kind proposed. The most recent advice from the traffic engineers is that, in the 2.5 hour period immediately after the evacuation is triggered, evacuation of 400 vehicles from the site would take 20 minutes and thereafter, as the road becomes affected by water, and traffic flow slows, vehicles would be able to depart at a rate of about 400 vehicles per hour.[19] Accordingly 400 vehicles per hour could depart even after 2.5 hours of time for warnings to be received and acted upon. I accept that there is sufficient time for the voluntary early evacuation.
[19]Exhibit 1, Volume 5, p 2117.
Mr Molino was of the view that the early evacuation plan should not be relied upon for a full evacuation of the site. Amongst other things, he questioned whether evacuation warnings would be heeded and pointed out that there were some floods which might rise faster than the Q100 design flood, thus reducing the available time.
Senior counsel for the respondent also questioned whether “multiple triggered warnings of flood water levels that do not eventuate in a major event” would lead to a “cry wolf” mentality and associated resident apathy. Given that the warnings would be triggered when water reaches a point in the system consistent with an event falling somewhere between a Q10 and Q20 flood event (i.e. an event which has an average return interval of between 10 and 20 years) it is unlikely that such an effect would materialise.
I am satisfied that the voluntary early warning system provides a substantial opportunity for those who are advised to leave, or otherwise wish to leave, to do so safely by car along the access route. Similarly, there is a substantial opportunity for those who are not at home to return home, if they wish, before the road becomes untrafficable.[20] It is however, unnecessary for me to consider whether that system, by itself, effectively deals with the access issue for the whole of the population, because it is only one component of the response which the appellant ultimately relies upon and it is no longer proposed to use this to evacuate all residents. Indeed residents will generally be encouraged to stay and the majority will likely do so. That only a proportion is likely to choose early evacuation will also somewhat reduce any potential for traffic delays.
[20]In this regard, the early warning system should incorporate text or other messaging alerts to those residents who wish to receive a warning when they are off-site – See evidence of Mr Collins at T4-22.
The appellant altered its proposal on the recommendation of Dr Shorten, a risk assessment expert. Instead of relying upon the early warning system to achieve an evacuation of the site population, the appellant now only relies upon that system as providing an opportunity for those people who wish to leave to do so. Otherwise, it relies upon a range of measures which are designed to convert the proposal into a “safe haven”, so that occupants can comfortably and safely reside on the site until the flood waters have receded to a point where the access roadis again trafficable.In a Q100 flood event, that would be a period of some 30 to 40 hours.[21]
[21]Given present conditions the time would be 30 hours. If global warming is taken into account, and if it is assumed that there is no adjustment to the dam strategy to deal with that, then this time may extend to 40 hours.
The creation of isolated “islands” within a floodplain can be problematic. Mr Granger, an impressive witness who has been involved in emergency and disaster management for more than 20 years, explained:
“In my experience with the SES, it is normal practice for disaster managers to avoid locations for the accommodation of residents that become islands during a flood event. Such an arrangement can give the illusion of a safe haven but in reality may become isolated islands in a sea of water which remain vulnerable to further flood rises. Unfortunately, such locations can be overwhelmed if the expected flood level exceeds the forecast level.”
This topic is also dealt with in the SCARM report, published by the CSIRO, dealing with best practice principles and guidelines for floodplain management in Australia. In relation to the formation of islands, that document states:
“The formation of islands on the floodplain during a flood is always potentially hazardous and is to be avoided. People trapped on the island may be safe for small floods, but are at high risk in extreme floods. The development of land that becomes isolated prior to inundation increases the load on emergency services during flood events. Furthermore, the rescue of people from islands may place rescuers at undue risk.”[22]
[22]Exhibit 1, Volume 6, p 2484.
The document returns to this topic when discussing effective flood access as follows:
“A potentially hazardous situation develops when rising flood waters isolate an area of land, leaving it as an island in a sea of flood water. The degree of hazard depends on the depth, velocity and rate of rise of flood waters between the island and possible places of refuge. Vehicle access may be cut rapidly. Rescue by boat, helicopter or large vehicle may be necessary, so putting the rescuers lives at risk. Although such a situation may not develop for ‘normal’ floods, checks should be made to see whether rare flood events cause islands to develop or even worse, to subsequently be submerged.”[23]
[23]Exhibit 1, Volume 6, p 2496.
These passages reflect a concern for safety where emergency services may be called upon to rescue the population of these “islands”, particularly where those islands become unsafe in times of extreme flood.
Whilst acknowledging those concerns, Mr Granger was supportive of the subject proposal, because of the lengths to which the appellants have gone to ensure that, when the access road becomes untrafficable by ordinary cars, the site acts not as an isolated island at risk from extreme flood, requiring evacuation of the isolated population,but rather as a true “safe haven”, so that residents can “shelter in place” until flood waters subside.
The concept of “shelter in place”, as an alternative to evacuation, is recognised. It is referred to in the Gold Coast City Local Disaster Management Plan, as one of the risk treatments for the Nerang River Catchment,[24] and is reflected in the Queensland Evacuation Guidelines for Disaster Management Groups (developed pursuant to s 63 of the Disaster Management Act 2003 (Qld)),[25]which provides, in part, as follows:-
[24]Exhibit 13, p 46; Mr Granger’s evidence at T2-40.
[25]Exhibit 24A.
“Shelter in place
The best option when evacuation is not necessary is sheltering in a safe and secure structure at home or with family and friends.
Residents, if evacuation is not required, should be encouraged to seek refuge with family and friends who may live in a building structure that meets current wind strength regulations, or is located higher than the expected flood peak and is considered a safer location for the impending hazard.”
It should be noted, in this respect, that evacuation, where available, is not risk free, even if there is a trafficable road. As the consultation draft of the Queensland Evacuation Guidelines acknowledged,[26] evacuation can be a highly destructive activity with its own risks.[27]Reliance on ‘shelter in place’, as an alternative to evacuation, is not necessarily contrary to a best practice approach to flood plain management.
[26]Exhibit 24.
[27]See also T4-103, l 10.
The appellants have taken a multi-faceted approach to ensuring that the proposal would operate as a true “safe haven” for people to “shelter in place”, if they do not take advantage of the voluntary early evacuation process. In that regard:
(i) The proposed development will be built on an engineered development platform, with a minimum ground floor level apartment set at RL 4.9 m AHD. This, without more, will provide greater than ARI 500 year flood immunity for ground floor units, of which there will be 55.[28]
[28]Exhibit 4, para 52.
(ii) All internal access roads in the development are set to be trafficable in the ARI 1000 year event.[29]
[29]Exhibit 4, para 53.
(iii) In addition to providing an engineered development platform, it is proposed to provide bunding and temporary flood barriers to protect not only all of the units but also the car park entrance (and therefore the car parks) in all floods up theProbableMaximum Flood (PMF). This flood event is more extreme than an ARI 10,000 year event.[30] In effect, the development, including the units and car parks, will be flood-proof.[31] Consequently occupants will not only be safe in their units, but will also be able to move safely about the apartment complex even in times of extreme flood.
[30]Exhibit 4, para 52.
[31]Areas of the site beyond the developed portion will become flooded, but that is not unusual for development sites within a floodplain and it does not cause any significant concern.
(iv) The proposed development includes a community room/emergency centre which will serve as a marshalling area for residents during major floods and from where they can be given information/instructions.[32]It is to have, amongst other things, torches, a radio and a first-aid kit. It could also be used for emergency accommodation, but that function would be unnecessary for residents, who would be able to stay in their units. This would be an important resource for community interaction. As Mr Granger said:
[32]It is also intended to offer this space for use by the SES and/or to offer it as “safe haven” for other residents in the flood plain, but I have not assumed that it would be used for those purposes as the evidence of Mr Granger suggests that it is unlikely that the SES would wish to use it and it is likely that evacuations from residential properties in other parts of the floodplain would be to areas beyond the floodplain.
“That sort of resource during a stressful period like a flood is an important resource for community resilience, for the management of the facility to communicate so that people are kept informed of what is happening.”[33]
[33]T2-37.
A question was raised about the size and location of that centre. That can be resolved before any final approval, subject to conditions, is granted.
(v) A sump is to be set below basement level containing adequate submersible pumps capable of operating to remove seepage at all flood levels up to and including the PMF.
(vi) An adequate and managed safe water supply (intended for every day use as well as risk mitigation during flood) is to be provided and maintained. The size will reflect an agreed population of remaining residents during flood (at least 80 per cent of the population for a period of up to three days).
(vii) On-site generated power is to be available to maintain residents’ normal requirements. This power will, amongst other things, be available to run lifts, for pumping potable water from a water-proof basement tank to a roof feeder tank, and for running an on-site sewerage pumping station during severe events or, indeed, whenever there is an outage of mains power.
(viii) Even though residents would continue to have access to their own food supplies in their own units, adequate and safe emergency food rations, in the way of non-perishable supplementary rations, for risk mitigation during flood will also be stored on-site. The amount of this will appropriately reflect an agreed estimated population of remaining residents during flood (at least 80 per cent of the population for a period of up to three days).
(ix) Satellite communications will be available.
(x) An adequate pharmaceutical notification supply scheme (intended for everyday use as well as risk mitigation during a flood) is intended to be provided and maintained through ongoing agreement with a local pharmacy, and a secure on-site dispensary of emergency medical and first aid supplies. The composition of the supplies is intended to appropriately reflect the anticipated profile of the population of remaining residents during flood. It was not explained, in detail, how this could or would operate although reference was made to arrangements used in aged care facilities. It is a desirable concept, and should be implemented if possible, but it is not critical, given that residents with the need for on-going medication will have access to their own supplies, in their own units. It would obviate the slight risk[34] only that a person who, being dependant on medication and having an inadequate supply, chooses to reject the early voluntary evacuation opportunity and then runs out and needs more during the relatively short time that the site remains isolated.
[34] T2-74.
The consequence of these measures is to remove the need for the site population to be evacuated during any flood event and to make provision for the population to shelter safely on-site instead.Mr Collins, who was called by the appellants, described the proposal as “the highest standard development I have ever been involved in”.Obviously, emergency agencies should be advised of the circumstances of the development, so that they are aware that there is no need to direct that the population of the site be evacuated, even in major flood events.
Mr Molino has, in the past, advised that an isolated community would not pose a high risk, where it is protected from inundation in a PMF event. In a report “Pitt Town Flood Risk Management Review”, published in November 2007, and which dealt with the prospect of the residential stock within the town, which becomes isolated by floodwater, increasing by 915 additional lots, over a base of only 447, he said (emphasis added):
“The SES modelling is based on the premise that people will respond to a flood warning evacuation in a timely fashion and the timings of forecasts, flood rises and travel will be in accordance with, or better than, the model. Of course it is possible that evacuation will take longer than modelled or that the available warning time will be less despite the conservative assumptions about these in the model.
What is more likely, is that some residents will choose not to evacuate when advised to do so and will lose the opportunity to leave PittTown before their homes are flooded. These people will effectively be stranded on an island which will shrink in size as the flood waters rise. Because there is an area in PittTown above the PMF level and it is within an easy walk of all of the proposed residential development, failure to evacuate does not pose a high risk to life.”
Confronted with this in cross-examination, he explained that his views had since altered, but he also confirmed that he had not told his former client of that change.
The appellants acknowledge that, notwithstanding the above measures, there is some chance that an emergency evacuation of an individual or individuals might be necessary during times of flood (eg if someone unexpectedly suffers a heart attack or has some other medical emergency). The statistical probability of that occurring, on the conservative assumption that all residents were isolated for two days, is approximately a 50 per cent chance that one evacuation would be required.
The appellants have incorporated means by which an individual evacuation for emergency reasons may be effected during times of flood. They are as follows:
(i) a tractor and trailer (intended for every day use as well as risk mitigation during flood), fitted with emergency equipment and life jackets, and capable of operating in up to 0.9 m of flood water is to be provided and maintained.
(ii) 2 substantial flat–bottomed aluminium boats (intended for restricted use of the body corporate as well as risk mitigation during flood) fitted out with emergency equipment and life jackets, and capable of operating in over 0.9 m of flood water are to be provided and maintained.
(iii) Use of licensed coxswains, trained by the SES, capable of operating the boat in times of flood.
(iv) an adequate and managed safe fuel supplies are to be provided (and set above the PMF), to provide fuel for the tractor and the boat as well as for the operation of the building generators, pumps and lifts during time of flood. Enough fuel is to be stored to provide up to three days supply, for each, without replenishment.
(v) aflood-free helipad is to be provided on-site for high-level emergency evacuation only (in conjunction with requirements of emergency services) without the need for use of winching equipment.
Save for the helicopter rescue option, the necessary measures are to be implemented and operated by the site management.
Dr Shorten thought that the tractor and boats could be used as shuttles, for anyone wishing to come or go during a flood. I accept Mr Granger’s evidence to the effect that generally people should not be crossing the flooded access route by these means. Access/egress should be for emergency purposes only and preferably be by helicopter, if it is available. Mr Granger had some concerns about the use of the tractor/trailer or boats, but was prepared to accept their use as a ‘least risk’ last resort emergency option.[35] The proposals are acceptable if used in that way.
[35]Mr Granger’s evidence at T3-98.
In addition to the above, there is to be education for the residents of the development. This will include resident education and information packs, community management statement requirements for emergency flood management procedures and periodic drills. It would be desirable for those measures to be reinforced by prominent signage within the development.
QUANTITATIVE ASSESSMENT OF RISK
Dr Shorten and Mr Molino sought to quantify the residual risk to life which would be associated with the proposed development occurring on the subject site within the flood plain. There were problems with that exercise. To be fair, they suffered from a relative paucity of data available on which to analyse the history of floods and flood related fatalities on the Gold Coast let alone in the subject flood plain.
Each of them calculated the average risk, expressed in terms of fatalities per million per year (PMPY) across the Australian floodplain for riverine floods. Dr Shorten arrived at a figure of 7.7 PMPY, while Mr Molino adopted a range of 10.4 – 13.9 PMPY. The differences are explained by different sources, assumptions as to the number of affected properties and methodology. Mr Molino, for example, included those flash flooding deaths that did not result from a lack of warning time. Dr Shorten was critical of that approach, because the subject floodplain does not suffer from flash floods and flash flood water has other characteristics that may affect the fatality rate. It is ultimately unnecessary for me to resolve that debate.
Dr Shorten calculated the probable risk of development on the site within the subject floodplain, absent any mitigation measures, to be 0.7 PMPY, or about one tenth of that for the “Australian floodplain”. There are good reasons to suggest that the fatality rate within the subject floodplain may be lower than for many others which contribute to the average figure across the “Australian floodplain”. As Mr Molino acknowledged, the floodwaters within the floodplain are generally slow rising (giving warning of the event), are not high velocity and levels of inundation are relatively low.[36] Mr Molino, on the other hand, pointed to the number of flood-affected residents within the floodplain requiring evacuation and the flood-affected access roads to suggest that this floodplain might have a higher risk than average. Dr Shorten’s defence of his 0.7 PMPY calculation was unconvincing. His reliance on the relative population of this floodplain compared with the population of the “Australian floodplain” (about one tenth) does not logically support a conclusion that the risk, expressed as a rate PMPY, would be proportionally lower. Cross-examination also demonstrated a certain lack of robustness in his attempts to otherwise calculate the risk for this particular floodplain at 0.7 PMPY.
[36]T4-75.
What is plain, however, is that the structural and management measures proposed as part of the subject development will reduce the risk for this development site. That is not something that the generic average across the “Australian floodplain” takes into account. Dr Shorten considered that risk would be reduced by up to 80% by these measures. Without adopting that particular figure, I accept that the reduction in risk would be quite substantial.
I am not persuaded to settle on a specific risk probability figure. I note, however, as Senior counsel for the appellants pointed out, the numbers referred to in the evidence ranged between 0.1 PMPY (in the case of Dr Shorten) - which is similar to the risk of death by lightning strike, and 10.9 to 13.4 PMPY(Mr Molino’s average for the Australian floodplain - which he thought might be exceeded in the subject floodplain) - which is similar to the risk of death by air travel. Theevidence is that it is only risks above 1,000 PMPY which are unacceptable by any definition. The risk associated with this development would not be of that order.
Risks below 1 PMPY (in accordance with Dr Shorten’s opinion) are universally considered to be so extremely small as to be insignificant and negligible. Risk levels of 10.9 – 13.4 PMPY fall at the bottom end of a broad range, of between 10 and 1,000 PMPY, which may be “provisionally accepted”,but where risk should be mitigated to as low as reasonably practicable.The appellants have done all that they could do to mitigate the risk for those who decide to take up residence in the proposed development within the floodplain. A risk between 1 and 10 PMPY (between the figures referred to) falls into a zone where approval is not automatic, but the risk may be accepted.[37]Accordingly, the quantitative assessments suggest that the level of risk is, at worst, something which may still be accepted.
[37]See Mr Molino’s evidence at T4-66, ll 25-40. Dr Shorten described this range as acceptable at T2-45.
The impetus for the quantitative assessment might well have been that part of the purpose of the FAACC, noted earlier, which states that development proposals will be fully evaluated against criteria which include “risk: whether the risks associated with the development are fully known, quantifiable and capable of being dealt with to Council’s satisfaction, without any uncertainties”. Risk assessment intrinsically has a degree of imprecision and uncertainty. Here, the nature of the risk is known and has been discussed in some detail. Similarly, the manner in which those risks are to be dealt with is known. Whilst I have not adopted a precise figure, in terms of fatalities PMPY, to reflect the risk, the evidence is sufficient to conclude that the quantum of the risk is not beyond that which may be accepted as satisfactory. To the extent that some imprecision or lack of certainty remains, it is not such as to warrant refusal of a proposal which, I am satisfied, meritoriously deals with the risk in a way which is satisfactory.
QUALITATIVE ASSESSMENT
It has already been noted that Mr Granger was an impressive witness, who supported the proposal on a qualitative assessment. He expressed the view that the proposal would provide for a “safe and appropriate response in the event of a major flooding event”. He pointed out that the provision of flood immunity to the level of the PMF is most unusual and that the measures to be implemented otherwise provide a true safe haven in the Nerang River flood plain, including an opportunity for a comfortable air conditioned community centre to accommodate people during the period of a flood event, as well as mitigation measures to evacuate people in the event of a medical or other emergency.
Mr Molino accepted that it was difficult to see how the appellants could do much better than what they are prepared to accept as conditions of approval for the purpose of mitigating flood risk. Some concern was raised in relation to reliance on the various measures being maintained in perpetuity. I am not unduly sceptical about that, but I accept that compliance with conditions requiring the provision and maintenance of the range of proposed measures is important. The Council should not be lumbered with overly onerous supervisory obligations. This can be appropriately addressed by conditions which require periodic independent audits and reports, to the Council’s satisfaction, at the expense of the body corporate for the development.
It was submitted, on behalf of the respondent, that the various measures to be implemented do not overcome what was said to be the site’s inappropriate location for a development of this magnitude. It was pointed out that the planning scheme refers to location, as well as design. It was submitted that the location is inappropriate because of the frequency with which the access road will be flooded (ie in events above Q20), the length of road which will be affected and the depth and velocity of the water to which people might fall or be drawn if they were washed off the access road. Further, it was submitted that people could not be relied upon to avoid crossing the flooded access road when it was dangerous to do so. They might do so either in order to get to, or depart from, the subject premises prematurely, before flood waters had sufficiently receded.
It was pointed out that unsuccessful attempts to cross flooded areas, despite general community warnings to the contrary, is a known cause of flood fatalities. Dr Connor put it this way in his testimony:
“Unless you can satisfy the concerns that people want to travel on the road to get to - the home or at times they will want to get to the home and leave. There is nothing in this concept that is stopping that occurring and that is going to be the heightened area of risk, is people and vehicles should traverse flooded water. As we all know what that issue is so therefore you try to avoid that situation if you can. I can’t see that this development is avoiding that situation unless somehow you can just prevent people from movement and, in my view, I don’t think you can.”
This no doubt underscores his reluctance to consider road access with anything other than a Q100 flood immunity.
Whilst I accept that planning should take into account known patterns of behaviour, even if irresponsible, there are a number of matters which are likely to reduce the risk of people trying to come or go while the access road is unsafe:
(i) The site is located in a floodplain where water rises slowly, affording a reasonable opportunity for those who wish to get home or to leave in advance of the access road becoming untrafficable to ordinary sedan. This will be enhanced by the management measures proposed.
(ii) The arterial road is likely to be closed as soon as it is unsafe for travel by ordinary sedan.
(iii) There is a range of measures adopted to ensure that occupants who remain onsite are safe within their units, have adequate food and water for their sustenance, have continued electricity supply (so that they can continue to enjoy the comfort and convenience of lighting, air-conditioning, cooking, electronic entertainment, other electric appliances and the like), and have proper sanitation. Further, residents will not be confined to their units, but will be able to walk about the development and meet in the community/emergency centre,enjoying social interaction and receiving information/instructions. This is likely to reduce the prospect that someone, having decided to stay and shelter in place, would then find sheltering in this development so harsh or inconvenient as to attempt to leave by fording a lengthy flooded access route.
(iv) The time to wait until the flood water hassufficiently receded is limited. In a Q100 flood event, it is only some 30 or 40 hours. Even in the case of a PMF, the period of isolation is only two to three days. This development would not create a population isolated for many days or weeks.
(v) It has already been noted that the proposal includes temporary flood barriers to protect the car park entrance from floods up to the PMF. It is proposed that the barrier be activated when the Nerang-Broadbeach Roadbecomes untrafficable by ordinary sedan, thereby preventing occupants from attempting to leave by car even if they wanted to.[38]Residents familiar with this management measure will also be aware that, should they traverse the flooded road to return home, they will be unable to drive into the flood-proof part of the site, including the car parking facilities and would have to leave their vehicle outside the gate. Further, it is unlikely, in the circumstances, that a person who had decided to stay and enjoying the comfort and convenience of being able to shelter in place for a matter of hours before the access again became trafficable in a Q100 event, would attempt to leave the site by traversing the floodedroute on foot.
[38]Mr Granger’s evidence at T2-42; Mr Collin’s evidence at T4-11 to T4-12.
(vi) I accept that the length of the floodedaccess route would make an ill-advised attempt to traverse it more risky,but it would also logically act as a deterrent to someone who might otherwise be tempted to traverse a much shorter section of flooded road, particularly if that person were proceeding on foot.
(vii) Mr Granger’s evidence was that recent experience of floods isolating communities in St George and Charleville, suggests that once access is cut, there is an acceptance that the community must stay put until the floodwaters ease and access is again safe.[39]
[39]T3-87.
(viii) While it was submitted that the safety and comfort of the proposed development would act as an “attractor”,prompting people to attempt to cross the flooded road to gain access, the development is likely to obviate a number of reasons why people might otherwise be tempted to try to get back to their place of residence. It is foreseeable that people might be tempted to try to cross a flooded road to get to their homes if those homes, or property within them or life (whether that be loved ones or even domestic pets) is in jeopardy. The subject development is however, an apartment development which is to be built to a standard which will see it, including the units, their contents, occupants and the cars in the car park, unequivocally safe from any possible damage or risk of damage by flood waters.
(ix) There is to be a community education and awareness programme for residents.[40] Further, in times of flood, warnings and instructions are to be given. There is also a community/emergency centre within the development. The development is of a nature which lends itself to on-site management via a body corporate. These matters are likely further to reduce the prospect of foolhardy conduct.
(x) Whilst the development will accommodate a significant number of persons, it is also of a kind and scale which can incorporate comprehensive mitigation and management measures, which would not be expected to be found in a residential subdivision of the kind considered in Bruigom v RockhamptonCity Council (supra).
[40]Exhibit 4, p 192.
It may be accepted that one can never rule out the possibility of foolhardy behaviour. I am satisfied however, that in all the circumstances, including the slow rising nature of the floodwaters, the period of inundation and the suite of measures proposed, considered collectively, there will be adequate mitigation of risk such that the solutions proposed are satisfactory and meet the tests postulated by Senior counsel for the respondent, viz sufficient, acceptable and reasonable. In the circumstances the risk associated with locating this development on this site with the management proposed, is acceptable.
COMPLIANCE
For the reasons stated, I am satisfied as follows:
(i) The proposal does not conflict with the higher-order provisions in the planning scheme.
(ii) The proposal meets the intent of the LAP, the relevant DEOs of the LAP, the purpose of the LAP place code and the purpose of the FAACC.
(iii) The proposal meets PC18 of the LAP place code in that it has been designed to ensure that residents, occupants or users are not at unreasonable risk from flooding.
(iv) PC20 of the LAP Place Code is met. The reasonable expectation of evacuation to a place of refuge is logically related to the reasonable necessity to do so. Here evacuation is available until water reaches the Q20 level and is unnecessary beyond that save for emergency purposes. Even assuming conflict existed however, thefact that the proposal is designed to itself be a place of refuge, thereby negating the need for population to be evacuated to another place, provides a sufficient ground to warrant approval notwithstanding any conflict with that performance criteria.
(v) The proposal satisfies PC5 of the FAACC. It will not increase the height or duration of flooding in the floodplain, because it will not adversely affect the hydraulic function of the floodplain. It will not increase the exposure of others within the floodplain. The on-site population would be significant, but will not significantly increase traffic on the evacuation route or significantly affect others using that route, because the majority of residents will “shelter in place” within the development itself. Any new residential development within a floodplain carries some risk and some potential demand on emergency services, because it increases the population within the floodplain but, in this case, the development does not pose a risk for other properties in the floodplain and the risk to its own residents will be both acceptable[41] and less than applies generally across the floodplain. The early managed voluntary evacuation followed by a self-sustaining “shelter in place” strategy will minimise the number of people requiring evacuation and the associated burden on public resources and emergency services.
(vi) The proposal meets PC10 because it has sufficient access or egress available to enable evacuation during a range of floods up to and including the design flood. The proposal has road access by car for a range of floods up to Q20 which affords a reasonable opportunity for the population tocome and go as they please. For the range of floods above that, to the design flood of Q100, it does not have flood free road access, but it does have access or egress which is “sufficient” in the circumstances, since the population which remains on the site will “shelter in place” andaccess will be limited to emergency evacuation, for which there is sufficient access/egress by other means. The performance criterion does not, in terms, require access to be by foot or car along a road (cf Bruigom v RockhamptonCity Council (supra)). As Senior counsel for the respondent acknowledged,[42] it is not a question of whether there is “dry road” access up to and including the 100 year event, but whether there is sufficient access up to such an event. This proposal has that.Alternatively, if it was thought that the lack of flood free road access to the design flood event necessarily creates a point of conflict with PC10, the merits of the proposal in providing a true safe haven for the population to shelter in place without the need for access or egress save for emergency purposes (which is adequately provided for) would deal with the concern underlying the provision and provide sufficient grounds to approve notwithstanding conflict.
[41]Senior counsel for the respondent conceded (T5-3 to T5-4) that read in context and consistently with the settled approach to the construction of planning schemes, the provision should not be read as requiring absolutely no increase in relevant risk or detriment, no matter how trivial. He said that the test was whether the increase is unacceptable (T6-18). All residential development within a flood plain carries some level of risk, even if access roads have Q100 immunity. Evacuation along such roads carries risk and there is a risk of those roads being flooded in more extreme events.
[42]T6-9, l 50.
Quite apart from the aspects of the proposal which address particular areas of conflict, or possible conflict, senior counsel for the appellant pointed out that the proposal is one of significant overall merit, such as to provide a basis to approve notwithstanding conflict. I accept that the proposal has merit in advancing the goal of optimising the opportunities for urban development in a way which would result in a development which is, it was agreed, appropriate in other respect, while minimising flooding concerns. The appellant’s willingness to make the community/emergency centre available to the SES and other residents within the broader floodplain was also relied upon but the evidence does not support a conclusion that such an offer would likely be taken up,[43] and I have not assumed its use for other than the residents of the development.
[43]See the evidence of Mr Granger at T3-84, l 35 to T3-85, l 30.
It is unnecessary for me to reach a concluded view about whether the merits of the proposal overall would outweigh the conflict because, for the reasons stated, I consider that there are sufficient reasons, going to the point of each area of conflict or possible conflict, which justify approval.
CONCLUSION
The appeal will be allowed, subject to the finalisation of an appropriate management plan and the imposition of appropriate conditions. The appeal will be adjourned to give the parties the opportunity to agree on those matters.