COMMISSIONER: This is an appeal under s 8.7 of the Environmental Planning and Assessment Act 1979 (the EPA Act) against the deemed refusal by Wollongong City Council (the Respondent) of Development Application DA 2018/1167 (the DA) for a new two-storey dwelling, proposed by RBFI Pty Limited (the Applicant) who is the owner of a site at 14 Mallangong Close, Figtree.
The Court arranged a mandatory conciliation conference under s 34AA of the Land Environment Court Act 1979 (the LEC Act) on 25 June 2019, which commenced with an onsite view under s 34(1) of the LEC Act between the parties and at which I presided.
However, as agreement was not reached during the conciliation phase, the conciliation conference was terminated and proceeded forthwith to hearing on 25 - 26 June 2019, pursuant to s 34(4) of the LEC Act. The parties did not seek to rely on information disclosed during the conciliation during proceedings.
The site is legally described as Lot 8 in DP1242700, within a newly constructed residential subdivision at Figtree known as 'Cobblers Run', and adjoins a public reserve that follows the line of a watercourse known as American Creek, which is within the Allans Creek catchment.
During the on-site view, the Court was shown the riparian land adjacent to American Creek, and the location of a large culvert through which the river passes, and over which the noise walls and roadway of the M1 motorway could be seen.
The site is zoned R2 Low Density Residential under the Wollongong Local Environmental Plan 2009 (WLEP) which permits dwelling houses with consent. It is common ground that the proposed development is within the maximum height of buildings pursuant to cl 4.3 of the WLEP, and is within the maximum allowable floor space ratio as set out in cl 4.4 of the WLEP.
In essence, this dispute is about the methodology used to derive a flood planning level (FPL) on the site, and the information on which an opinion of satisfaction should be formed pursuant to cl 7.3(3) of the WLEP.
RBFI Pty Limited v Wollongong City Council [2019] NSWLEC 1312 | Case summary | Zoe
Clause 7.3 sets out the objectives, scope and controls that must be satisfied prior to the consent authority, or the Court exercising the functions of a consent authority on appeal, granting consent in the following terms:
7.3 Flood planning
(1) The objectives of this clause are as follows:
(a) to maintain the existing flood regime and flow conveyance capacity,
(b) to enable evacuation from land to which this clause applies,
(c) to avoid significant adverse impacts on flood behaviour,
(d) to avoid significant effects on the environment that would cause avoidable erosion, siltation, destruction of riparian vegetation or a reduction in the stability of river banks or watercourses,
(e) to limit uses to those compatible with flow conveyance function and flood hazard.
(2) This clause applies to:
(a) land identified as "Flood planning area" on the Flood Planning Map, and
(b) other land at or below the flood planning level.
(3) Development consent must not be granted for development on land to which this clause applies unless the consent authority is satisfied in relation to all the following matters:
(a) all habitable floor levels of the development will be above the flood planning level,
(b) the development will not adversely affect flood behaviour resulting in detrimental increases in the potential flood affectation of other development or properties,
(c) the development will not significantly alter flow distributions and velocities to the detriment of other properties or the environment of the floodplain,
(d) the development will not affect evacuation from the land,
(e) the development will not significantly detrimentally affect the floodplain environment or cause avoidable erosion, siltation, destruction of riparian vegetation or a reduction in the stability of river banks or watercourses,
(f) the development will not result in unsustainable social and economic costs to the community as a consequence of flooding,
(g) if located in a floodway area - the development will not be incompatible with the flow conveyance function of, or increase a flood hazard in, the floodway area.
It is commonly held that the site is part of a region that is particularly susceptible to flooding given the behaviour of rainfall and overland flow as would be expected for a low-lying area positioned below an escarpment that is part of the Illawarra Range, and with waterways that are constrained in their downstream flow by culverts passing below the M1 Motorway.
I understand that for some decades, the practice of flood estimation in Australia has been guided by a publication titled "Australian Rainfall & Runoff: a guide to flood estimation". The 1987 edition, which is known as 'ARR1987', is said to have been the culmination of professional inputs and represented what was then state-of-the-art engineering practice.
ARR1987 was superceded in 1997, and the most recent design rainfall data and methodologies for flood estimation are contained in ARR2016 that was published by the Australian government's Geoscience Australia in 2019. An extract is found at Exhibit B.
The Respondent submits that the relevant flood planning level applicable to the site is determined by reference to the dictionary appended to the WLEP, which states that the level is equivalent to 1:100 average recurrent interval flood event plus 500mm for 'freeboard', which is expressed as RL12.30 AHD.
The Respondent completed a Flood study, Floodplain Risk Management Study and Floodplain Risk Management Plan for the Allans Creek Catchment between 2006 - 2008, based on ARR1987.
With the release of ARR2016, Council commenced a review of the study at based on better quality ground level surveys, updated development information and revised culvert blockage factors (Ex 1, Tab 2).
As the results of this study are yet to be settled, the dispute centres on whether there is any reason to depart from the RL12.30 AHD that was applied to the site during an earlier consent for subdivision.
[2]
Respondent's position on the FPL
According to the Respondent, the site is properly defined as being below the flood planning level pursuant to cl 7.3(2) of the WLEP. Furthermore, as cl 7.3 acts as a prohibition on development that does not comply with the clause, the application must be refused.
The Respondent notes the application before the Court is the latest stage in a series of applications that are descended from an original approval granted by the Respondent in 1992 (Ex 1, Tab 7).
The conditions appended to the original grant of consent include an instrument pursuant to s 88B of the Conveyancing Act 1919 (Ex 1, Tab 8) that placed terms of restriction on the site as follows:
"No dwelling house or other structure shall be constructed on the lots hereby burdened with habitable floor levels below the flood planning level of:
+12.30m AHD for Lots 2 to 10
…"
The Respondent notes that Council documents refer to major flood events dating from 1975 to 1999, and that work by Council to update the Allan Creek Flood Study and Floodplain Risk Management Study and Plan has been ongoing, and is subject to wide community consultation in accordance with a process described in the Flood Risk Planning Update Engagement Report (Ex 1, Tab 3).
Furthermore, Council has clearly stated in public communications that the new data contained in ARR2016 is being gradually adopted and that the transition will need to be carefully managed by Councils and consultants.
This is reflected in advice contained in the Floodplain Risk Management Guide, first published by the NSW government Office of Environment and Heritage in 2018 (OEH) (Ex 1, Tab 6) which cautions:
"Where results of studies indicate a reduction in peak flood levels or flood behaviour primarily due to ARR2016 transition, the Council whose area is covered by the study may make a precautionary decision not to change practice or standards (such as minimum floor levels) until they complete further investigations. These further investigations should examine the implications of changes in practices or standards on the full range of flood risk to the community and document any associated decisions in an FRM [floodplain risk management] plan."
The Respondent submits that a properly constituted Floodplain Risk Management Committee has engaged well qualified consultants to undertake an evidence-based review that is yet to be completed and that can be expected to result in a Floodplain Risk Management Study (FRM), a Floodplain Risk Management Plan, and a Plan for implementation as shown at Fig 2.1 of Exhibit 2.
The process followed by the Respondent to develop the FRM Study and Plan, and its implementation plan, involves a range of stakeholders including government agencies, authorities, the private sector and others.
At the completion of the Review, Council could then be expected to adopt the FRM Study and FRM Plan. However until such time, prospective applicants seeking to prepare a DA should rely on the policies, plans and studies previously adopted by the Council as is consistent with any other area of planning policy.
For the Court to grant consent to the applicant seeking to rely on its own methodology, study and plan may set a precedent that is premature given the work currently underway by Council.
[3]
Applicant's position on the FPL
The Applicant submits that it has done everything asked of it by the Respondent in the Wollongong Development Control Plan 2009 (WDCP) in assessing the risk of flooding so as to derive a FPL on its site.
The objectives of Chapter E13 of the WDCP provide the most appropriate guidance to anyone preparing a development application. In particular, (h) is in the following terms:
"Apply a "merit-based approach" to all decisions relating to flood affected development which take account of social, economic and ecological as well as flooding considerations."
As the results of Council's review were not publicly available at the time of preparing the DA, the Applicant undertook a site-specific, detailed flood study that relied on the latest information found in ARR2016 to determine a FPL for the site which is lower than the Respondent's FPL contained in the 1992 consent because while that was correct at the time, it is now acknowledged as out of date.
Chapter E13 of the WDCP provides guidance to an applicant at Part 5 on the key steps to be taken in preparing a DA which includes advice that:
"The assistance of Council staff or an experienced floodplain consultant may be required at various steps in the process to ensure that the requirements of this Plan are fully and satisfactorily addressed."
Part 9 of Chapter E13 of the WDCP advises, in part, the information required to address the relevant part of the DCP in the following terms:
"…
For smaller developments consideration may be given to the use of the existing flood study if available and suitable (e.g. it contains sufficient local detail), or otherwise a flood study prepared in a manner consistent with the "Australian Rainfall and Runoff" publication, Council's Stormwater Design Code and the Floodplain Management Manual, will be required."
The Applicant submits that this is precisely what occurred in preparing the application the subject of the development application, including the engagement of an experienced consultant that adopted an evidence-based, merit-based approach to flooding considerations that was site-specific.
Furthermore, this was a reasonable and correct approach as the existing Allan Creek flood study was superceded in 2016 when Council adopted a revised blockage policy, and so is not suitable for use.
As to the Respondent's preference to rely on ARR1987 as the basis for flood estimation, the Applicant notes numerous references to the 'ARR' in the WDCP are without date reference, including in the definitions of the WDCP itself.
To the Applicant, in the absence of a date that would direct prospective applicant's to a particular and preferred version of the ARR, it is reasonable to rely on the latest and most current version of the 'ARR' to develop flood planning levels and other relevant considerations.
Nevertheless, the Applicant's flood study (Ex A, Tab 8) compared results, in Table 4.1-1, derived from ARR1987 data, ARR2016 with certain data sourced from the Australian Bureau of Meteorology, and from ARR2016 with certain data sourced from OEH.
Table 4.1-1 shows, in essence, that the Applicant's worst case scenario results in a FPL of RL 10.67 AHD. Adding a freeboard of 500mm results in an RL of 11.17 AHD. The Applicant's FPL of RL 11.2 AHD is arrived at by further rounding.
Accordingly, the Applicant considers the site is not at or below the FPL, and so the Court can be satisfied cl 7.3 of the WLEP does not apply to the site.
[4]
Expert Evidence
The Court was assisted by hydrology experts, Mr Anthony Barthelmess, for the Applicant, and Mr Mark Babister, for the Respondent who conferred in the preparation of a joint expert report (Ex 3) and provided evidence at the hearing.
Among a number of areas of agreement, it is of particular relevance that the experts agree:
1. the Allans Creek Floodplain Risk Management Study and Plan 2006-2008 (Allans Creek Study 2006-2008) is out of date and was not relied on by the Respondent when granting the Construction and Subdivision certificates for Cobblers Run subdivision.
2. the site cannot be said to share the same flood behaviour with the wider region. Instead, it is very sensitive to small changes in data inputs given the site is effectively a detention basin that may be subject to flood due to limits in the rate of outflow through the culvert below the M1 motorway.
3. the approach taken in developing ARR2016 draws on a more comprehensive range of datasets than ARR1987. However a key difference lies in the temporal rainfall patterns used. Where ARR1987 relied on one temporal rainfall pattern that was a synthesis of patterns, ARR2016 relies on ten temporal rainfall patterns that are said to be more characteristic of real-world events.
While the experts agree that it is better to have ten temporal patterns to guide flood estimation in preference to just one synthetic model, the experts agree that ARR2016 does not come without its imperfections.
In fact, according to Mr Babister, who is a co-author of the ARR2016 document, some of the datasets in ARR2016 can lead to inaccurate results which suggest that the conditions found in the Wollongong area do not align to the figures published in ARR2016.
For example, Mr Babister noted that the 1998 flood in the area could be considered a 1:1000 year event under the ARR2016, although the rainfall recorded at the time was not consistent with that level.
In his evidence, Mr Barthelemess noted an industry fact sheet prepared by the Council expressing concern at the reliability of ARR2016 and citing four grounds as follows:
1. Rainfall: especially the Intensity-Frequency-Duration (IFDs) published by the Bureau of Meteorology
2. Loss models: the amount of rain 'lost' in to soil in other ways, during a rain event that does not contribute to runoff
3. Pre-burst rainfall: that rain that falls before a main storm event and serves to saturate the ground
4. Temporal rainfall patterns: as temporal rainfall patterns became a particular focus of the expert's evidence, it is discussed at length below.
It is the Applicant's submission that the modelling undertaken by Mr Barthelemess sought, responsibly, to address these four grounds by using rainfall data produced by OEH in 2018 in preference to the data in ARR2016, assuming a zero value in the loss model and a zero value in relation to pre-burst rainfall.
In fact, the Applicant notes that such were their efforts that the rainfall data produced by OEH was initially sourced through an application made pursuant to the Government Information (Public Access) Act 2009.
In relation to the temporal rainfall patterns used in the modelling, Mr Barthelemess undertook what is called sensitivity analysis using patterns that he considered more locally relevant to the site.
This was necessary, according to Mr Barthelemess because, and as agreed by Mr Babister, ARR2016 includes Wollongong in a temporal rainfall pattern comprising a region stretching from the border between Victoria and South Australia, taking in the coastline to the Southern ocean, and the east coast of Victoria, and the southern coast of NSW, as shown in Figure 1 of the Fact Sheet (Ex 3). Wollongong is the northern-most limit of this region.
However Mr Babister describes several physical characteristics that make rainfall and runoff behaviour particularly difficult to estimate in the Wollongong region, including what he terms the significant orographic enhancement caused by the Illawarra escarpment. As a result, the region records some of the highest rainfall in the State but in a very localised area.
It is for these reasons that Mr Barthelemess says he decided to rely on alternative temporal patterns which he regards as more applicable to the site as the data more closely resembles patterns likely to be experienced in the Wollongong area.
Mr Babister considers the Applicant's decision to depart from the temporal patterns said by ARR2016 to apply to the site as understandable, but he did not agree with Mr Barthelemess' choice of temporal pattern, as it also fails to reflect rainfall and runoff behaviour local to the Wollongong area, and he is currently assisting Council to prepare its own dataset that may be incorporated in the Review set out at [14].
During the hearing, the Respondent sought leave to rely on a conference paper presented at the 2019 Floodplain Management Australia National Conference titled 'Cooking Up ARR2016 Storms in Wollongong - the Importance of Using Local Ingredients'. While the Applicant initially objected to its tender, a short adjournment was taken to allow parties to become familiar with the paper.
I acknowledge and accept the Applicant's submission that it is not known if the paper is peer reviewed, however the authors appear to be officers from Wollongong City Council and the NSW Office of Environment and Heritage and so can, in my view, be regarded as intimate to the development of research and policy in this area.
For this reason, I considered the paper may assist the Court to better understand the nature of the challenge faced by the parties in what is broadly described as a 'transition' to ARR2016 and I allowed it to be entered as Exhibit 5.
Table 1 in Exhibit 5 compares the loss values in the ARR2016 datahub, with those derived by consultants in 2018 and 2019. In simple terms, the Applicant asserts that figures in Table 1 validate the assumptions made by Mr Barthelemess to set loss values at zero, which can be regarded as 'conservative' when compared to the ARR2016 datahub which shows a value of 60.5mm.
As I understand it, this means that the model produced by Mr Barthelemess did not allow for any rainfall or runoff to be 'lost' in to the soil, but assumes that the ground is impervious, or so saturated that it is unable to absorb further moisture.
While the parties did not address the Court on the issue, I also note that Figure 7 and supporting text illustrates a point made by the experts in evidence on the composition of data comprising Temporal Rainfall Patterns. Figure 7 appears to show that all but one of the locations from which data is derived for the region said to apply to Wollongong are sourced from Victoria, eight of which appear to be metropolitan Melbourne, which the paper describes as having very different rainfall behaviour from that in Wollongong.
This is said by the authors to be the impetus for Council and OEH to develop an in-house Local Temporal Pattern study, which accords with the evidence of Mr Babister.
The value of locally-derived data is then further explored in Figure 11 which Mr Babister considers to show an under-estimation bias present in the ARR2016 figures as, even when local inputs are used, observed events such as the 1998 and 1999 floods in the region lie outside the model.
According to Mr Barthelemess, this is because Figure 11 does not compare predicted performance with observed and recorded experience, but compares the results of two theoretical models.
Whatever the reason for the disparities evident in Figure 11, the experts agree that in both the 1998 and 1999 floods, the site the subject of the development application, was dry.
[5]
Conclusion
I accept that flood estimation is a complex science that continues to evolve as the quality and detail of data inputs and computational capacity improve.
I regard it as relevant that two eminently qualified and experienced experts agree that during the flood events said to represent some of the most extreme events in the area of the site, in 1998 and 1999, the site was dry.
Furthermore, I regard it as relevant that the Applicant's modelling has assumed that no rainfall will be lost to ground infiltration or the like, and that rounding up has occurred in favour of a more onerous level, and so may be said to incorporate assumptions that are precautionary.
In addition, a freeboard of 500mm required by the WLEP, when added to the Applicant's assumptions, result in the proposed finished floor level of the dwelling being some 590mm above the 1:100 year flood event.
While on first impression the Applicant's derivation of a level that is substantially lower than that previously agreed to apply to the site may be queried, I am satisfied on the basis of all the evidence before me, that the proposed RL of 11.2 AHD is not below the flood planning level.
Consequently, I am satisfied that all habitable floor levels of the development, being RL 11.2 AHD will be above the flood planning level as required by cl 7.3(3)(a).
In written submissions from the applicant, my attention was drawn to s 39(2) of the LEC Act which states:
In addition to any other functions and discretions that the Court has apart from this subsection, the Court shall, for the purposes of hearing and disposing of an appeal, have all the functions and discretions which the person or body whose decision is the subject of the appeal had in respect of the matter the subject of the appeal.
To this end, the Applicant seeks to have the Court order that the instrument issued pursuant to s 88B of the Conveyancing Act 1919 be amended to register a FPL of 11.2 AHD on the site, consistent with my opinion of satisfaction at [66].
The Applicant cites Lloyd J in Pro-Vision Developments Pty Limited v Ku-ring-gai Council (2003) 131 LGERA 108; [2003] NSWLEC 226 where his Honour held that the Court, in having all the functions and discretions as the body whose decisions is the subject of the appeal, had the discretion to release, vary or modify those matters found in an instrument registered under s 88B of the Conveyancing Act 1919 in the course of determining the appeal, if the circumstances so warrant.
As the FPL for Lot 8, the site, is stated on the instrument registered under s 88B of the Conveyancing Act 1919 as being 12.30m AHD, I am satisfied that the Court should exercise its discretion to order that the instrument be amended consistent with the findings at [66].
[6]
Orders
The Court orders that:
1. The Appeal is upheld.
2. Development Application No.DA-2018/1167 for the construction of a two-storey dwelling house on land situated at, and known as 14 Mallangong Close, Figtree NSW, also known as Lot 8 in DP1242700, is approved subject to conditions in Annexure 'A'.
3. The restriction on the use of the land referred to as No.11 in DP1242700 benefitting Wollongong City Council and set out in the supporting s 88B instrument for Lot 53 in DP 838601 is varied as follows:
"No dwelling house or other structure shall be constructed on the lots hereby burdened with habitable floor levels below the flood planning level of:
+11.20m AHD for Lot 8
+12.30m AHD for Lots 2 to 7, and 9 and 10
+12.25m AHD for Lots 1, 11 to 14, and 20 to 22
+12.20m AHD for Lots 15 to 19."
1. All exhibits are returned, except for Exhibits A, 3 and 5.
[7]
DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.
Decision last updated: 04 July 2019