Easton Agriculture Ltd v Manawatu-Wanganui Regional Council HC Palmerston North Civ-2008-454-31

[2011] NZHC 1005

For a Court ready (fee required) version please follow this link

IN THE HIGH COURT OF NEW ZEALAND PALMERSTON NORTH REGISTRY

CIV-2008-454-31

BETWEEN  EASTON AGRICULTURE LIMITED First Plaintiff

ANDEVELEIGH FARMING COMPANY LIMITED (IN RECEIVERSHIP) Second Plaintiff

ANDMANAWATU-WANGANUI REGIONAL COUNCIL

Defendant

Hearing:         30-31 May, 1-3 June, 7-10 June 2011

Counsel:         J O Upton QC with M S Dobson for Plaintiffs

D J Heaney SC with S H Macky for Defendant

Judgment:      7 September 2011

JUDGMENT OF THE HON JUSTICE KÓS

Introduction  [1] Background  [5] The February 2004 flood  [36] Parties  [67] Pleadings  [76] Issues  [93] Issue 1:  Can the Council be liable other than in negligence?  [94] Issue 2:  Did the Council owe the plaintiffs a duty of care in its monitoring

and maintenance of the stopbank?  [122] Issue 3: Was the Council negligent?  [143] Issue 4:  Did the Council’s negligence cause the plaintiffs' loss?  [195] Conclusion  [224] Disposition  [225]

EASTON AGRICULTURE LIMITED v MANAWATU-WANGANUI REGIONAL COUNCIL HC PMN CIV-

2008-454-31 7 September 2011

Introduction

[1]      A floodway stopbank fails. The floodway diverts part of a river’s flow during flood conditions.   Together the floodway and river can cope with a 1 in 100 year flood.  But the flood that leads to the stopbank failure is greater still. A 1 in 110 year flood.  The floodway is crossed by a highway bridge.  The bridge cuts through the top of the stopbank on the southern side of the floodway.  During the flood the water reaches and laps the bottom of the bridge.  Yet the flood should still be able to be contained within the floodway stopbanks.

[2]      But the southern stopbank fails.  First, just upstream of the bridge.  Then, a few hours later, downstream of the bridge.  Ultimately a gap of 40 metres opens up about the bridge.   The exact cause of the failure is in dispute. Nearby croplands belonging to the plaintiffs are flooded.   The stopbank is the responsibility of a regional council.   Must the council compensate the plaintiffs for the loss of their crops?

[3]      The main questions in this case are whether the Council was negligent in its monitoring and  maintenance of the stopbank,  and  whether  any such  negligence caused the plaintiffs’ loss.  The plaintiffs expressly do not claim the construction of the bank or bridge was negligent.1    Any such claims would long since have been time barred.

[4]      Evidence as to quantum was received by the Court.  By agreement the issue of quantum was reserved for further argument, if need be.  This judgment is confined to whether the Council is liable to the plaintiffs.

Background

Manawatu River

[5]      The Manawatu River is about 160 kilometres long.   It rises on the eastern side of the Ruahine Ranges.  Fifty kilometres later it passes through the Manawatu

1      This was made clear in the plaintiff ’s closing submissions: see [81] below.

Gorge, and then flows on past the city of Palmerston North.  From there to the small settlement of Opiki, the river is steep enough to transport gravel when the river is in flood, and the riverbed is gravelled.  Downstream of Opiki the gradient is flat.  The riverbed is silty.   Flood velocities here are typically about 1.5 metres per second, compared to 3 to 4 metres per second in the steeper Palmerston North section.

[6]      River level and flow records have been held longer for the Manawatu than any other river in  New  Zealand.   An automatic recorder has  been  operating at Palmerston North since 1929.   Manual information is available back to the 19th century.   The February 2004 flood, with which we are concerned, was the third

largest flow recorded on the river:

Year	Discharge (cumecs)2
1880	4000
1982	3800
2004	3500

[7]      At Opiki, where the river gradient flattens out, the Oroua River joins the Manawatu.   The Oroua is the Manawatu’s major tributary.   Below the confluence with the Orua, at Opiki, the Manawatu wends its way on towards the sea past flat lands called the Makerua and Moutoa basins.   The plaintiffs farm in the Moutoa basin.

[8]      Substantial swamp drainage works were undertaken in these areas in the 19th century.   Flax was planted in the newly drained land.  The flax fibre was used to make rope for sailing vessels and lashings for wool bales.  Much of it was exported to Australia.  Maori had exploited this resource for generations by the time pakeha settlers established ropewalks along the Manawatu river banks in the 1840s.3   When the flax industry declined in the late 19th century the land was drained and converted to farming.  First Makerua, and later Moutoa, the area with which we are concerned

in this case.

2      Measured at Palmerston North.

3      T Buick Old Manawatu (Buick & Young, Palmerston North, 1903) 142; A J Dreaver

Horowhenua County & Its People (Dunmore Press, Palmerston North, 1984) 141-145.

[9]      The first flood protection works were constructed between 1923 and 1925. The Makerua Drainage Board constructed stopbanks to protect the Makerua basin. In the 1930s and 1940s more stopbanks were built, by the Manawatu-Oroua River Board and the Palmerston North River Board.

Lower Manawatu Flood Control Scheme

[10]     The Lower Manawatu Flood Control Scheme (LMS) was undertaken by the Manawatu Catchment Board between 1959 and 1965.  The LMS protects 320 square kilometres of land from flooding.  But for these flood protection works, the city of Palmerston North, the town of Fielding, and significant areas of the region, in particular the Taonui, Makerua and Moutoa basins, would be prone to flooding in severe weather events.

[11]     The LMS relies primarily on stopbanks to contain floodwaters. A key part of the scheme, as we shall see, is the Moutoa floodway.   The LMS is designed to contain a “1 in 100 year flood”, meaning that such a flood is expected (measured over a very long period of time) to recur on average once every 100 years.  To put it another way, there is a 1 per cent statistical probability of that flood size being equalled or exceeded in any given year.   The LMS stopbanks were built with a design objective of a freeboard of 450 millimetres.  That is, at all points the stopbank crests were to be 450 millimetres higher than the calculated 1 in 100 year flood level.

[12]     At the time of the February 2004 flood, the 1 in 100 year flood flow was set at 3,450 cumecs (measured at Palmerston North).  At 3,500 cumecs, the February

2004 flow exceeded that level slightly.  It is common ground that it was a “1 in 110 year” flood.

Moutoa floodway

[13]     Downstream of the settlement of Opiki, and the confluence of the Manawatu and Oroua Rivers, there is a long oxbow bend where the Manawatu doubles back on itself.   On the apex of that bend vast sluice gates have been built.   These are the Moutoa sluice gates.   The sluice gates were built in the early 1960s.   They are

regarded as a major engineering feat.  When open, they allow the river flow to be split.  The major part of the water is sent down the Moutoa floodway.  The existing river channel will take the rest.  The sluice gates serve as a critical safety valve.  The change in gradient at Opiki means the river flow is far slower here than higher up the river.  But for the ability to open the sluice gates, and divert part of the river down the floodway, floodwaters would bank up and overwhelm the stopbanks.

[14]     The floodway passes directly across low-lying farmland in the Moutoa basin. It runs from the oxbow bend to a point near the Manawatu river mouth.   It is 10 kilometres long, and bypasses 30 kilometres of winding river channel. The floodway averages 600 metres in width.  Its presence means that in all but floods exceeding the

1 in 100 year level, the Manawatu River water can be contained wholly within the existing river structure and the Moutoa floodway.   The part of the Moutoa basin lying between the floodway and river channel should remain free from flood.  It is that land the plaintiffs farm.

[15]     The flood levels in February 2004, a 1 in 110 year flood, were always going

to challenge the floodway and river’s capacity to cope.

Trestle bridge

[16]     Running at right angles to the floodway is State Highway 1.  Just south of Foxton it crosses the floodway using a trestle bridge.  The trestle bridge was built in the 1930s.  The floodway, in the 1960s.  The trestle bridge is lower than the adjacent floodway stopbanks.  So, in effect, it cuts through them.  Either side of the bridge, the stopbank crest runs up against the outer fascia of the bridge, at about road level. Beneath the bridge long beams run with the road, supporting the bridge deck.  They project down from the bridge deck.  The stopbank builders built the crest as far up beneath the bridge as they could.  Then they filled in the spaces above the stopbank and between the beams with bags filled with cement and gravel.  These harden to concrete.  It is as if the spaces are filled with large flat rocks.

[17]     Applying  and  enlarging  the  advertising  adage  that  a  picture  is  worth  a thousand words, I reproduce three images adduced in evidence, which depict the situation:

Fig 1: bridge substructure, downstream of bridge, looking north towards Foxton.

[18]     Figure 1 shows that the side of the bridge comprises (1) a fascia (with safety parapet above) and (2) a series of longitudinal beams beneath (and inset from) the fascia.  The beams project 480 millimetres below the fascia.  This is the appropriate point to repeat some statistics that emerged in evidence:

(a)      The designed stopbank level at its conjunction with the trestle bridge was 5.2 metres above mean sea level.  That is also approximately the road level of the bridge.   Although settlement had occurred on the crest of the stopbanks adjacent to the trestle bridge, those deficiencies had been identified in May 2002.  Repair work was conducted in June

2003.   I find at the time of the February 2004 flood, the stopbank heights were likely to be at the 5.2 metres design level on either side of the trestle bridge.

(b)The underside of the bridge (i.e. lowest point, the bottom of the downward projecting beams) is 4.52 metres above mean sea level. Thus, obviously, the bridge cuts through the stopbank.

(c)      A 1 in 100 year flood (i.e. flood levels for which the floodway was designed) would reach 4.63 metres above mean sea level at the bridge. At that level the flood waters would reach and lap the beams below the bridge fascia.  But the floodwaters would still be contained within the stopbanks.

(d)The peak level of the February 2004 flood at the bridge was slightly higher than the design level.  It reached 4.65m above mean sea level. This  level  was  reached  at  or  shortly  after  8.00  pm  Monday

16 February 2004.  That still left 550 millimetres freeboard between the flood level and the 5.2 metre stopbank crest - more than the design

level of 450 millimetres freeboard.

Fig 2: cement bag infill under bridge at northern stopbank abutment.

[19]     Figure 2 is a recent photograph of the northern end of the trestle bridge, at the point it abuts the stopbank.     That is the bank opposite the one that failed.  It shows how cement bags have been used to fill the gaps between the stopbank and the arches  formed  by  the  longitudinal  beams.    While  representational,  it  is  not  an accurate picture of the position at the southern end of the trestle bridge in 2004.  The northern end has a crossbeam4 at or about the crest of the stopbank at that point.  At the southern end, however, the bridge straddled the stopbank. The piers were set part way up the slopes either side of the bank.   No piers and crossbeam met on the

stopbank crest there.  The cement bags had to be built right up to the very underside

4      That is the structure joining the piers to the bridge (forming the three low archways shown in Fig

2).

of the deck of the bridge.  This is made clearer in the next figure, which is a cross- section drawing provided by one of the expert witnesses called by the Council.

Fig 3: diagram produced by Mr N S Luxford, geotechnical engineer.

 

Fig 3: cross-section diagram prepared by Mr Noel Luxford depicting probable southern stopbank/trestle bridge conjunction in 2002

[20]     The upper part of figure 3 is shown on a north/south plane (i.e. as if one was standing below the bridge, looking towards the stopbank).   The lower part of the diagram is viewed on an east/west plane (as if one was standing on the stopbank, looking towards the bridge).  Although the upper part of the diagram shows some settlement of the stopbank adjacent to the bridge, Mr Luxford drew his diagram as if set in 2002. As I have already said, I am satisfied that the stopbank was rebuilt to its design level either side of the bridge in 2003.

LMS funding

[21]    The floodway stopbanks lie on land owned by the Manawatu-Wanganui Regional Council.  LMS operating costs and capital improvements are funded by a special rating scheme.   Eighty per cent is funded through targeted rating systems, paid for by those who benefit directly from flood protection.  The remaining 20 per cent is funded from the Council’s general rate.  Beneficiaries are represented on a Scheme Liaison Committee.   The Council is required to consult with the Liaison Committee before including the Scheme budget in the Council’s draft annual plan.

[22]     The LMS budget for 2003/04 made provision for maintenance ($984,238), capital works ($70,773), flood damage reinstatement ($2,299,353) and “management/supervision/inspection” ($845,107).  That cost was then spread, on a differential basis, across 77,000 individual ratepayer beneficiaries.

LMS review

[23]     In 1992 the Council commenced a comprehensive review of the LMS.  Stage

1, which began that year, concerned the Palmerston North region where the potential flood damage was greatest.  Stage 2 related to the rural reaches of the river, from Ashhurst (above the city) down to the river mouth near the Moutoa basin. A key part of the stage 2 review was a geotechnical investigation by Riley Consultants Ltd.

The Riley report

[24]     The  Council’s  brief  required  Riley  to  assess  stopbank  conditions  and  to identify high and medium priority sections requiring more detailed investigation. Higher priority stopbanks were to have permeability checks by digging test pits or drilling.   Medium priority stopbanks would also be tested where preliminary investigation indicated necessity.   The brief also required the consultant “to check that services under, through or adjacent to the stopbanks do not compromise their security”.  In particular, it was to:

...check for any signs of deterioration or damage to the adjacent stopbank or structure itself which may impair the security of the stopbanks.

[25]     Riley’s report on the risk of failure of stopbanks, including those with which we are concerned, was presented in July 1994.   The Riley report is essentially a geotechnical one.   A preliminary examination of the stopbanks was followed by more intensive investigation.  Twenty-eight test pits, 40 hand auger bore holes, 20 penetrometer measurements and a further 10 machine bore holes, were drilled.  From these investigations Riley was able to express a view on constituent materials in the stopbanks and their engineering properties.   Its assessment of the probability of failure, in the event of a “bankfull” flood was “low to moderate” for the relevant southern floodway stopbank, immediately above the bridge.   Below the bridge the

risk was “low”.  By contrast, Riley expressed concern about the state of the stopbank on the northern side of the floodway.  This incorporated sandhill material.  The risk above the bridge was denoted as “very high.”   That below, “moderate”.   Riley concluded:

The assessed risk of failure for a flood stopbank crest (i.e. at the point of over topping) is generally low.  Some areas of high risk occur.  These are generally  applications  under  threat  from  undermining  in  the Manawatu/Oroua Rivers, and sections of the Moutoa floodway right bank which are at risk from piping failure in the foundations.  These sections of the Moutoa floodway coincide with sandhills.

[26]     “Piping” happens when permeable materials in a stopbank permit water to flow through, and then course through, the stopbank.  The consequence is that the stopbank will erode at the exit point.

[27]     Riley did not identify a particular risk associated with the trestle bridge, the focal point of this case.

Stage 2 of the LMS review

[28]     Following the Riley report the Council’s senior design engineer, Mr Graham Doull, undertook an investigation of the LMS stopbank heights.  He used a Danish computer modelling technique.  By these means he located 5.5 kilometres of LMS banks lower than the 1 in 100 year design flood height.  A further 20.7 kilometres had less than the required 450 millimetres of freeboard.

[29]     Mr Doull’s work established that the floodway stopbanks were below the required freeboard level at two points.   First, near the Stratton property, some 4.6 kilometres from the river mouth.  Secondly, at the trestle bridge, some 1,200 metres from the river mouth.  The Council’s Group Manager of Operations, Mr Allan Cook, gave evidence that minor earthworks were undertaken in June 2003 to reinstate the design crest level of the floodway.  The Council’s records indicate a repair cost of

$1,411.   Mr Cook could not say precisely where that money had been spent.   It appears likely however that the deficiencies at the bridge were fixed.  He also noted that a more substantial freeboard reinstatement had also been undertaken approximately 3 kilometres upstream of the bridge, at a cost of $12,700.   It is

reasonably clear that this work reinstated the stopbanks near the Stratton property.  It was there that the greatest deficiency in freeboard existed.

[30]     Ultimately  the  plaintiffs  make  no  complaint  about  the  height  of  the stopbanks, save immediately under the trestle bridge.

[31]     In  1998  Mr  Doull  prepared  a stage 2  LMS  review for  consideration  by Council.  Based on the Riley report, Mr Doull prepared a list of works that could be done to further improve flood protection.  Broadly the categories of work were the raising of stopbank heights to design height, prevention of undermining by rock armouring, prevention of piping failures where porous layers in the stopbank soils had been identified, and some stabilisation of structures.   Mr Doull undertook a cost/benefit analysis and presented five upgrade options.   Option 1 involved all works on the list being undertaken, but at a cost of $28 million.  The least costly option was Option 5, in which only a few of the very high priority works were to be undertaken.

[32]     There was no evidence before me that any of the options, including the most expensive Option 1, involved works specifically at the trestle bridge.

Public consultation

[33]     In 1999 the Council consulted with ratepayers about the options.  The Easton family  interests5   made  a  submission  in  favour  of  Option  4.     That  involved expenditure of $6.1 million.   It does not appear that the second plaintiff made a submission.

[34]    A hearing committee convened to consider submissions.   Eventually it recommended Option 4 be adopted.  But with two additional items.  The total capital cost was to be $6.6 million. Annual maintenance was also to be increased, from $1.1 million  to  $1.5  million.   The option  would  result  in  a 50  per  cent  increase in beneficiaries’ rates.  The works were to be carried out over a six year period, starting

in 2000.

5      The first plaintiff company was not yet conducting the family farming business.

[35]     The major thrust of Option 4 was to bring the stopbank crest height in all cases to 450 millimetres above the design flood level.   That accounted for 65 per cent of the expenditure.  Option 4 included the reinstatement work near the trestle bridge and the Stratton property that I have already described.6

The February 2004 flood

The flood generally

[36]     As  noted  earlier7   the  February  2004  flood  was  the  third  largest  on  the Manawatu River since recording began in the 19th  century.  It was the largest flood since 1902.  The tributary Oroua River experienced its largest ever recorded flood. The floodwater was greatly in excess of the design levels of its stopbanks.   The

Oroua stopbanks failed in five places.   The resulting flooding was said to be “extensive  and  destructive”.     Failures  also  occurred  on  the  Manawatu  River stopbanks at Koputaroa, on Kauri Creek, and on the Moutoa floodway at the trestle bridge.  It is that failure that directly concerns us in this case.

[37]     Heavy  rain  had  occurred  in  the  catchment  in  late  January,  and  in  early February 2004.   The catchment was already very wet prior to the 15-16 February storm.  Heavy rain fell on 14, 15 and 16 February.  At Moutoa, the Council’s rain gauge recorded 62.6 millimetres across those three days.   The total for the month was 250 millimetres - compared to a monthly mean of 67 millimetres.  By way of comparison, 45 millimetres fell in January and 22 millimetres in March.  Throughout the rest of the year the highest rainfall month was June, with 131 millimetres.  The February rainfall event was, therefore, exceptional.

[38]     Measured  at  Palmerston  North,  the  Manawatu  River  was  running  at  a maximum flow rate of 3,500 cumecs.  As at 2004, the river system provided for a 1 in 100 year flood at 3450 cumecs flow.  On that basis, therefore, the February 2004

flood was, by a small margin, an over-design flood.  The consequence was that the

6 At [29].

7 At [6].

February 2004 flood rose into the 450 millimetre design freeboard level in some places.

[39]     Council  officers,  including  Mr  Doull,  used  a  sophisticated  computer modelling system to identify likely river flows during the period of heavy weather. By midnight on Sunday 15 February, it was appreciated that the Moutoa sluice gates would need to be opened, to divert river flow down the Moutoa floodway. Landowners were advised, so they could move stock.  The gates were opened at 7.00 am on Monday 16 February.   Soon after, it became apparent that the Oroua River was in a state of exceptional flood. As Mr Doull put it:

During  the  morning  of  16  February  it  was  abundantly  clear  that  the Manawatu  River  would  be  carrying a  flood that  at least  approached its design flood, and would possibly exceed it.

[40]     The peak flow past Palmerston North occurred at about 6.30 pm on Monday

16 February, although the river remained within 100 millimetres of its peak between midday and 9.00 pm that day.   The peak flood downstream at the Moutoa trestle bridge occurred somewhat later of course, at or after 8.00 pm.

[41]     As noted at [18], the peak floodwaters reached 4.65m at the bridge.  But the bridge itself offered only 4.52m clearance.  So the bridge structure intruded into, and constricted, the flood flow.

Moutoa basin flooding

[42]     Before  anyone  appreciated  that  there  had  been  a  failure  in  the  Moutoa floodway stopbank, local farmers and others were reporting to Council that ponding was occurring near the trestle bridge.  That is, in the Moutoa basin, to the south of the floodway.

[43]     I now turn to the eye-witness accounts of the stopbank failure.  They are of considerable importance when eventually the issue of causation is considered.8

8      At [195-223].

(1) Mr Jamieson

[44]     Mr Lee Jamieson lives in Levin. At about 5.00 pm on the evening of Monday

16  February 2004,  he  went  with  his  family to  watch the  flood  on  the Moutoa floodway.  They stood on the top of the southern stopbank, downstream of the trestle bridge, near the cycleway.  About 10 metres downstream of the bridge.  This part of the stopbank, where Mr Jamieson and his family were watching the flood, collapsed spectacularly the following morning.9   Mr Jamieson said:

I could see and hear water flowing over the top of the stopbank under the bridge.  I am positive it was flowing under the bridge and not on either side. It was approximately half a metre to one metre wide and 20 centimetres deep.

Under  cross-examination  by  Mr  David  Heaney  SC  for  the  Council,  however, Mr Jamieson confirmed that at that stage the flow was really more in the nature of a “slow trickle” beneath the bridge and above the cement bags.  At that stage he could see no ponding of water in the Moutoa basin, to the south of the stopbank. That was, of course, three hours before the peak of the flood.

(2)   Mr Stratton

[45]     Mr Darryl Stratton farms in the Moutoa basin.   His farm is south of the southern stopbank.   It lies between State Highway 1 and the plaintiffs’ farms.  At

1.45 am on the morning of Tuesday 17 February, Mr Stratton decided to check his paddocks.  He drove west along the Whirokino Road towards the highway.  The road skirts the edge of the Manawatu River.  The point where the road passes beneath the trestle bridge is about 500 metres south of the stopbank.

[46]     As Mr Stratton attempted to drive under the trestle bridge, he came upon ponded water.  He assessed it to be about 1.5 metres deep.  He decided not to venture further.  He told me that he had lived in the area all his life.  He had never seen water ponding to that extent.  He believed for the road to have filled up like that, water had

to have come down from the floodway.  Certainly no one suggested to me that the

9 See [61].

water could have come from the river.   Its stopbanks to the south of the Moutoa basin held fast throughout the flood.

[47]     Mr Stratton went home and rang the Council.  The Council officer asked him to check again.  So at about 2.00 am Mr Stratton went back to the scene.  This time he did not try to drive under the trestle bridge.  Instead he drove up an access road onto the bridge, over the highway and back down onto Whirokino Road west of the bridge.   He then went for a walk along the southern stopbank back towards the trestle bridge.  He began to feel extremely uneasy.  He could see water on both sides of the stopbank.  The floodway water was “roaring”.  He rang the Council on his cell phone and reported the situation.  He said he could not see any stopbank breach, but he was well west of the bridge when he made the phone call.  He then went home again.  But not for long.

[48]     At about 3.00 am Mr Stratton and his partner looked out their bedroom window.  They could see sheets of water across their farm.  The water was being held back by a metre high cow race between the stopbank and the house.  As they watched, the water crossed the race “like a wave”.   Mr Stratton and his family evacuated the property.  In about two minutes the water was knee deep inside their house.

(3)  Mr Doug Easton

[49]     Mr Doug Easton is a dairy farmer.  He is related to Mr Ian Easton, the driving force behind the first plaintiff.  Mr Doug Easton’s main farm is near Waitarere, north of Levin.  But he also has two smaller blocks in the Moutoa basin area, and he leases

500  acres  within  the  floodway  itself.    As  the  weather  worsened  on  Monday

16 February he moved his stock out of the floodway, to higher ground on one of his Moutoa basin blocks.  At 3.20 am on the Tuesday morning he received a call from a relative reporting that water was lapping around Mr Stratton’s house.  As he still had

700 cows in the basin, he headed there straight away.

[50]     Standing on the trestle bridge at the point it meets the southern stopbank, he looked upstream.  He saw what appeared to be a breach in the bank upstream of the

bridge, with “water pouring through”.  Concerned for his safety, he headed off the bridge.  Returning to his block west of the highway he found water in his paddocks, perhaps a metre deep.  The cattle in the paddocks were in distress.  Mr Easton was able to open some gates.   The stock headed off to higher ground, the Manawatu River stopbank to the south.

[51]     Then, between 5.30 and 6.00 am the same morning, Mr Easton was standing on the river stopbank, along with his stock.   He looked back at the floodway stopbank:

I could see quite clearly water pouring through the bank on the eastern [i.e. upstream] side of the bridge.  I could not see under the bridge because of the water that was flowing through the bank on the eastern side.  I could see bits of the stopbank upstream of the bridge breaking off as the water scoured it away.

(4)  Mr Richard Barber

[52]     Mr Richard Barber farms to the north of the Moutoa floodway.   He drove down to the bridge, arriving at the north end at 5.30 am.  This was about the same time as Mr Doug Easton’s observations, but of course from the opposite end of the bridge.

[53]     Mr Barber  got  out  of his  vehicle to  talk  to  a roading worker.   As  they watched, he could see water flowing out of the floodway upstream of the bridge, into the basin.    He also  saw it  flowing over the  stopbank  on  the downstream  side. Unfortunately Mr Barber was a very early witness, and neither counsel nor the Court asked him to locate precisely where he saw this water flowing, relative to the bridge. That issue was to prove important later in the trial.

[54]     Mr Barber left the flood scene at that point.  He went to fetch his father.

(5)  Mr Stuart Barber

[55]     Mr Stuart Barber10 and his son Richard arrived at the bridge at about 6.15 am. They were in the left lane, coming from the north, so Mr Barber Snr’s attention was first drawn upstream of the bridge.  He said:

I could see clearly brown water flowing across through the bank on the eastern [i.e. upstream] side of the trestle bridge.

He said he thought the bank immediately upstream of the bridge had “disappeared". They then parked and crossed the road.   Just as his son had, Mr Barber also saw water  flowing  through  the  stopbank  on  the  downstream  side  of  the  bridge. Mr Barber’s evidence ultimately was unclear whether this initial downstream breach was immediately adjacent to the bridge or not.

(6)  Mr Owen Bonis

[56]     Mr Bonis is the Levin branch manager for Higgins Contractors Limited.  He was responsible for closing the trestle bridge during the night.   There had been concern about the safety of the bridge.  Mr Bonis noted:

The ... water was right up close to the underside of the bridge and debris which normally occurs with floods was starting to be caught in the structure of the bridge.

[57]      At about 5.30 am on the Tuesday morning Mr Bonis stationed himself at the south end of the bridge, near the Manawatu River stopbank.  From that location he could see the floodway stopbanks about 500 metres north.  His perspective would have been similar to that of Mr Doug Easton, although a little more to the west.  A colleague from Higgins was posted at the north end of the Moutoa floodway.  It was that colleague presumably who spoke to Mr Richard Barber.

[58]     From his position on the highway, Mr Bonis could not see under the bridge. But he could see a breach in the stopbank downstream of the bridge.  In his evidence

10 Mr Barber’s background is discussed at [162].

he said he thought he could still see intact stopbank between the bridge and the downstream breach.   Mr Bonis seemed to have no recollection of the  upstream breach seen by Messrs Doug Easton, Richard Barber and Stuart Barber.

(7)  Mr Graeme Anderson

[59]     Mr Anderson was a works supervisor for the Council at the time of the flood. He was in charge of the Shannon work gang.  He was responsible for inspecting the Moutoa floodway stopbanks.  I will comment later on his evidence in relation to that responsibility.11    But on the Tuesday morning he detoured to the bridge, to look at the flood, while driving his wife to work.  Mrs Anderson started work in Foxton at

6.00 am, so they were at the north end of the bridge a little after 5.30.  At that stage Mr Anderson saw a shallow breach in the stopbank very near the cycleway, about 10 metres downstream of the bridge.   He is sure that the 7-8 metres of the stopbank immediately downstream of the bridge was still intact.   He had his binoculars with him.  He used them to survey the stopbank, in which of course he had a particular interest.   He does not seem to have paid particular attention to the upstream side, however. That was all he saw at that stage.

[60]     He dispatched Mrs Anderson to work and returned at about 6.00 am.   He remained at the site for the next 40 minutes or so.  He saw the breach downstream of the bridge enlarge until a catastrophic failure occurred.

Catastrophic stopbank failure

[61]     It  is  common  ground  that  at  approximately 6.30  am  on  the  morning  of Tuesday 17 February 2004 the southern stopbank, between the Moutoa floodway and the Moutoa basin, gave way downstream of the bridge.  The collapse was described by several of the foregoing eye-witnesses as “spectacular”.   The stopbank “blew out”.  Water rushed from the floodway into the Moutoa basin, and proceeded to back

up the basin towards the plaintiffs’ farms.

11 At [165].

[62]     The breach, ultimately, reached 23 metres upstream from the centreline of the bridge, and 17 metres downstream. The bridge survived, intact.

Sutherland report

[63]     Following the February 2004 flood and the various failures in the Council’s flood protection systems, the Council commissioned a review.   The convenor was Dr A J Sutherland, Dean of Engineering at the University of Canterbury.  There were three other engineering members of the team.  One of them was Mr P B Riley.  His firm had undertaken the 1994 geotechnical review.12   The review report is known as the Sutherland report.   It was delivered in April 2004.   It is a mixture of expert analysis  and  reportage.    It  refers  liberally  to  information  obtained  from  eye- witnesses.  Some of those eye-witnesses gave evidence in this case, but not all.  The Sutherland Committee was asked to consider the possible causes of failures, and to

recommend appropriate forms of repair to avoid recurrence.

[64]     On the reportage front, the Committee report contains some information that was not in evidence before me, and which  I must  put aside.   For instance, an unnamed witness who reported a 300 millimetre water level drop at 6.00 am, before the major downstream breach of the stopbank.   Similarly, two eye-witnesses who heard a “large bang” at 1.30 am the same morning.  I shall disregard all that.

[65]     As  to  the  failure  of  the  stopbank  at  the  trestle  bridge,  the  Committee concluded that the likely cause of failure was “the interaction between the bridge structure and the stopbank”.   The Committee noted that the upstream water level striking the soffit caused a “pulsing” in the flow and increased velocities under the bridge and along the stopbank face.  Any debris caught on the bridge would have exacerbated the situation.  The Committee felt the presence of the bridge piers may have contributed to local scouring.  But the experts who gave evidence before me all agreed that if local scouring had been a significant concern, earlier floods should

have identified that.

12 See at [25].

[66]     The Sutherland Committee considered a possible explanation of the reported flooding in the Moutoa Basin early Tuesday morning was a failure “at the vertical interface between the bridge structure and the stopbank”.  The pulsing nature of the flow, in conjunction with “observed strong wave action” could have exploited a weakness at the interface, and could have caused sufficient scouring to lower the bank and cause the considerable overflow observed at 3.00 am.   The Committee discounted seepage as a cause of failure, in the absence of direct evidence to support it.  The Committee recommended improved armouring of the bank/bridge interface, and more effective sealing of the area under the bridge via a flexible waterproof butynol liner.

Parties

Plaintiffs

[67]     The first plaintiff, Easton Agriculture Limited, is a family-owned company. One of its directors, Mr Ian Easton, gave evidence.  He has farmed in the area all his working life.  The farm relevant to this case is located in the Moutoa basin on the Foxton Shannon Road.  It is about 250 hectares in area.  It was formerly farmed by Mr Easton’s parents.   The family has been there since 1961.   The Eastons grow mainly potatoes, onions, squash, grass seed and cauliflower.   They also run dry stock:  cattle and some sheep.

[68]     Mr Easton is a highly regarded farmer.  In 2005 he won the Hill Laboratories Harvest Award.   He is committed to sustainable farming, and operates a policy of strict crop rotation.  He plants crops for two years, and then grasses the paddock for the next five.

[69]     The   second   plaintiff’s   farm,   known   as   Bussitons,   is   tucked   against

Mr Easton’s at its north eastern boundary.  It is a somewhat smaller operation, about

67 hectares, and I heard less about it in evidence.  Mr Eveleigh has been farming in the Opiki area for approximately 50 years.  He devoted himself to two crops: onions and potatoes.

[70]    Both the Easton and Eveleigh farms are in an excellent area for potato production.   The area is largely frost-free.   The summer is moist because of the Ruahine ranges just to the east.  The two farms are situated on Kairanga silt loam. That is some of the best soil there is for potato production, because potatoes can be stored in the ground until needed.  The crop can mature, but need not be harvested immediately.

[71]     The breached stopbank flooded the Moutoa basin.  A vast amount of water flowed back up the basin from west to east.   The area covered by water has been estimated at 2000 hectares.  The water volume is not known.  Floodwaters ran up open drains cut into the basin by drainage boards and farmers in earlier years.  That exacerbated the flooding. Almost all Mr Eveleigh’s farm was flooded.   So was a large part of Mr Easton’s.  Mr Easton had lifted a crop of onions for harvest prior to the flood.  They floated on the floodwaters and ended up piled a metre high against a fence line. The floodwaters covered the plaintiffs’ farms for up to 12 days.

[72]     Although there was argument over the extent and duration of flooding (which would need to be resolved as part of any quantum hearing), it is undenied that the flood caused substantial damage to the plaintiffs’ properties and crops.

Defendant

[73]     The defendant is the Manawatu-Wanganui Regional Council.  For reasons not given  in  evidence,  it  prefers  to  call  itself  the  “Horizons  Regional  Council”.    I imagine that reflects the fact that the region is rather flat.

[74]     The Council is responsible for 33 river and drainage schemes.  Each scheme is a separate entity, funded largely by targeted differential rating systems.13     The LMS is the largest of the 33 schemes.  It accounts for 45 per cent of total scheme expenditure.

[75]      These  schemes  were  previously  managed  by  the  Manawatu  Catchment

Board and the Rangitikei/Wanganui Catchment Board.   But in 1989 there was a

13 At [21].

reorganisation of local government in the area.  The Local Government (Manawatu- Wanganui  Region)  Reorganisation  Order  1989  dissolved  the  catchment  boards. Their functions, duties and powers under the Soil Conservation and Rivers Control Act 1941 were transferred to the Council.

Pleadings

Claim

[76]     The plaintiffs claim that following the opening of the Moutoa sluice gates, and the filling of the floodway on 16 February 2004, “the stopbank on the true left hand side14  of the floodway progressively failed and breached in the vicinity of the trestle bridge, [and] the flood waters escaped the floodway and flowed back upstream”. The Council admits that allegation.

[77]   The plaintiffs then allege the stopbank failure occurred because of the inadequate condition of the stopbank under the trestle bridge (“in particular the interface between the stopbank and the bridge”), the floodway landform geometry, the alleged poor condition of the cement bag infill under and in the vicinity of the bridge, the flood waters lapping onto the bridge beams, and the consequent hydraulic effects resulting from those various conditions.   The Council denies all those allegations.

[78]     The plaintiffs advance their case on four bases:  negligence, private nuisance, Rylands v Fletcher liability, and breach of statutory duty.   I will traverse each, briefly.

[79]     Negligence:  the plaintiffs allege that the defendant was negligent in failing to “note or in any way evaluate the conditions of the stopbank under the bridge” when it reviewed the LMS in 1997 and 1998; failing to model the bridge when assessing the hydraulic capacity of the floodway; and “failing to generally maintain to design

standard and/or monitor the condition of the stopbank under the bridge”.

14     The “true left” and “true right” banks of a river or other waterway are defined by looking

downstream.

[80]   The plaintiffs’ opening confirmed the focus of their case on the area immediately beneath the bridge.   In particular, on the existence of an alleged gap above the cement bags, and beneath the underside of the bridge deck.   Mr John Upton QC for the plaintiffs submitted:

In summary, the stopbank had narrowed under the bridge and the cement bags on top of the bank under the bridge had apparently settled leaving a significant gap.  The stopbank at the bridge was in an inadequate condition, so that when the flood level came to the bottom of the bridge beams, water would simply have washed out and overflowed under the bridge.

But the narrowing of the stopbank was not ultimately considered by any of the

experts (including the plaintiffs’ expert) as causative.15   I discount it accordingly.

[81]     Furthermore, in closing Mr Upton made it abundantly clear that the plaintiffs were staking their claim firmly on deficiencies in the Council’s “maintenance and monitoring” of the stopbank.  Mr Upton expressly disavowed any assertion that the stopbank construction was actionable.

[82]     The plaintiffs’ claim in negligence was not confined to breach of the bank.  It also extended to the Council’s response to the emergency, both in its immediate response to the risk of flooding while the progressive failure was occurring, and in preventing certain remedial work being undertaken by the plaintiffs in an attempt to block the flow of water up the Moutoa basin.

[83]     Nuisance:  the plaintiffs plead that the facts give rise to an action of private nuisance, causing loss. That is about the substance of that pleading.

[84]     Rylands v Fletcher liability:   the plaintiffs also plead that escape of water from land under control of the Council, resulting in damage, creates liability on a Rylands v Fletcher basis.

[85]     Breach of statutory duty:  finally, the plaintiffs plead breach of statutory duty. The relevant duty is that under s 126 of the Soil Conservation and Rivers Control Act

1941:

15 See [209].

It shall be a function of the River Catchment Board to minimise and prevent damage within its district by floods and erosion.

[86]     The  first  plaintiff  claims  damages  of  $1,412,467.    The  second  plaintiff,

$1,127,920.  The damages claims are confined to loss of income from sale of flood- damaged crops.  No damage for harm to other property, or for reduced production in later years, are claimed.  Interest and costs are sought.

Defence

[87]     The Council denies each of the claims.   It advances a number of defences. One, based on the Land Drainage Act 1908, was abandoned in closing.   Those remaining were four in number.

[88]     Statutory defence to all claims other than negligence:  the Council contends that rights of action other than negligence are not available to the plaintiffs, because of s 148 of the Soil Conservation and Rivers Control Act 1941.  I will return to this topic shortly.16

[89]     Rylands v Fletcher:   natural user:   the Council pleads that the use of the Moutoa floodgates and floodway for the purpose of draining the district was an ordinary, natural use of the land.  On that basis it would fall within an exception to the rule in Rylands v Fletcher.

[90]     Rylands v Fletcher:  act of God:  if the foregoing defences do not prevail, the Council pleads that the storm event was of such magnitude as to constitute an act of God, for which it is not liable.

[91]     Limitation  Act  1950:     finally,  the  Council  pleads  that  the  plaintiffs’ allegations are time-barred by s 4 of the Limitation Act 1950, to the extent they relate to any act or omission by the defendant occurring prior to 24 January 2002. That is six years before the original statement of claim was filed.

[92]     I turn now to the issues arising in this case.

16 At [94].

Issues

[93]     The issues arising in this case are four in number:

(a)       Can the Council be liable other than in negligence?

(b)Did the Council owe the plaintiffs a duty of care in its monitoring and maintenance of the stopbank?

(c)       Was the Council negligent?

(d)      Did the Council’s negligence cause the plaintiffs’ loss?

Issue 1:  Can the Council be liable other than in negligence?

[94]     It is best to address this issue first.

[95]     Section  148(1)  of  the  Soil  Conservation  and  Rivers  Control  Act  1941 provides:

No Board shall be liable for injury to any land or other property caused   without   negligence   of   the   Board   by   the   accidental overflowing of any watercourse, or by the sudden breaking of any bank, dam, sluice, or reservoir made or maintained by the Board.

Submissions

[96]     I start with the Council.   It submits, in short, that claims against it arising from breach of the stopbank can be maintained (if at all) only in negligence.  It says that the other three claims – private nuisance, Rylands v Fletcher and breach of statutory duty – are subsumed by statute into a single potentially permissible claim of negligence.  But, as we shall see17 the Council also denies that a duty of care can

arise at all in this case.

17 At [132].

[97]     Ms Sarah Macky argued this part of the case for the Council.  She submitted that s 148 reflects a policy judgment made by Parliament to preclude liability of a strict nature for catchment boards (and therefore for regional councils assuming their functions under the Act) in major rainfall events.   That is because those bodies simply have no control over such events.   She submitted that to hold catchment boards strictly liable for losses arising out of massive natural events would be unduly harsh, and would fly in the face of the express words of the statute.  The provision thus prevents a finding of liability under common law actions for nuisance, Rylands v Fletcher and breach of statutory duty in this case.

[98]     Ms Macky was unable to refer me to New Zealand authority directly on point.  In Burke v Waikato Regional Council18 a District Court affirmed the right of a property owner to bring an action in negligence for inundation resulting from the alleged failure of a floodgate structure maintained by the regional council in that case.  That case involved a striking-out application, and does not take things much further.  The same could be said for a High Court decision Atlas Properties Ltd v Kapiti Coast District Council.19    That was a claim arising from a storm which had caused under-road culverts to overflow. The council sought to rely on s 148.  Durie J held that the section was not available to it as a defence.   It was not a catchment board for the purposes of s 148.  Moreover, the work undertaken by the council in relation to the culvert was not work contemplated as part of the functions of a catchment board.  Those conclusions clearly were correct.  The decision was upheld on appeal, albeit on different grounds.20

[99]     Turning now to the plaintiffs, Mr Upton submitted s 148 of the Act applies only to conduct carried out without neglect.  In this case the plaintiffs claim that the Council’s neglect or negligence was causative  of their loss.   Accordingly s 148 would not apply.

[100] Secondly, Mr Upton submitted that the Local Government (Manawatu- Wanganui Region) Reorganisation Order 1989 provided only for the Council to take

18     Burke v Waikato Regional Council [1996] DCR 897.

19     Atlas Properties Ltd v Kapiti Coast District Council HC Wellington CP 172/00 19 December

2001, per Durie J.

20     CA30/02, 20 June 2002.

over the “functions, duties and powers” of catchment boards under the Soil Conservation and Rivers Control Act 1941.  Said Mr Upton, “That order does not include the immunity or statutory protection available in s 148”.   The approach contended for by the plaintiffs, he submitted, provides a “better fit” with s 247H of the Local Government Act 1974 and s 191 of the Local Government Act 2002. These provide that a local authority is not entitled to create a nuisance, or deprive any person of any right or remedy that person would otherwise have in respect of a nuisance.

Analysis

[101]   I reach three conclusions.

[102]   First, the Council is the “Board” for the purposes of s 148 of the Act.  This was admitted by the plaintiffs.   They of course contend that the responsibilities imposed on such a Board in s 126 gives rise to an actionable statutory duty.  So they cannot have it both ways, and do not try to.

[103]   It  is  worth  setting  out  why  the  Council  is  the  “Board”.     The  Local Government (Manawatu-Wanganui Region) Reorganisation Order 1989 gave effect to a reorganisation scheme advanced by the Local Government Commission.  This dissolved a number of territorial authorities in the Manawatu and Wanganui regions. Three united councils, a city council, numerous borough district and county councils, two catchment boards and five drainage boards.   All dissolved.   The property of certain of these bodies (and in particular of the Manawatu Catchment Board and the Moutua Drainage Board) was vested in the new regional council.  More importantly,

the Order provides:21

The  functions,  duties,  and  powers  of  the  Manawatu-Wanganui  Regional

Council shall be:

...

(d)      The  functions,  duties,  and  powers  of  a  Catchment  Board  and  a

Regional  Water  board  under  the  Soil  Conservation  and  Rivers

21     Cl 16(1)(d).

Control Act 1941 and the Water and Soil Conservation Act 1967 or any other Act.

[104]   This was a commonplace reform in 1989.  As far as I am aware, there are no catchment boards left in New Zealand.  All have been subsumed into regional councils.    Nonetheless,  the Act  remains  in  place and  imposes certain  duties  on “catchment boards”.  Section 126 is one such provision. Another, s 145, provides for claims against catchment boards for land damaged or injuriously affected.22   Section

148 follows shortly after that provision.

[105]   In my view the intention of Parliament was simply that s 148 apply to any body undertaking the lawful functions, powers, and duties of a catchment board, under the Act, in maintaining watercourses or banks.

[106]   Secondly, s 148 means what it says.  The Council’s liability for damage to property resulting from “the sudden breaking of any bank” (which is what happened here) is dependent on proof of negligence.   In my view this has the statutory consequence of effectively displacing other forms of liability, apart from negligence. The provision operates not so much as a defence to the common law forms of action other than negligence, as an ouster of them.  It follows that my conclusion in relation to s 148 disposes, also, of any claims in nuisance and breach of statutory duty.

[107]   Thirdly, even if I had not reached the conclusion that s 148 excluded liability under the rule in Rylands v Fletcher, I would have held that the rule does not apply in this case.

[108]   It is worth reminding ourselves that the rule in Rylands v Fletcher represented not so much an advance in the law in 1866, as a reversion to a more medieval approach to tort liability.   One in which liability was absolute, regardless of negligence.   By 1866 negligence had become the effective organising principle of

actions on the case.   Neglect – negligence – characterised actions on the case, in

22     Soil Conversation & Rivers Control Act 1941, s 145.

contrast to trespass which tended to focus upon intentional wrongdoings.  Rylands v Fletcher, as the late Professor A W B Simpson23  points out, was an action brought squarely  on  the  basis  that  Mr  John  Rylands’  reservoir,  built  above  old  coal workings,24  was “carelessly and negligently constructed”.25     At first instance an arbitrator had found Mr Rylands’ contractors negligent.   But not Mr Rylands personally.   The former were not worth pursuing.   The latter was not vicariously liable for their actions.  Mr Fletcher therefore pursued Mr Rylands in his personal capacity to the Court of Exchequer in banc (where he lost) and thence (but now on the basis that liability should be strict) to the Court of Exchequer Chamber26 (where he won). There, Blackburn J said, famously:27

We think that the true rule of law is, that the person, who for his own purposes brings on his land and collects and keeps there anything likely to do mischief if it escapes, must keep it at his peril, and that if he does not do so, he is prima facie answerable for all damage which is the natural consequence of its escape.

[109]   Thus a negligence case was transformed into a strict liability one.  Professor Simpson’s study offers a possible explanation for this jurisprudential diversion.  In the decade before Rylands v Fletcher was decided, two other large reservoirs had failed, and far more catastrophically.  First, in 1852, the Bilberry Dam:28

Construction of this decrepit, ill-designed, and ill-maintained structure had begun in 1839, and its collapse at 1 a.m. on Thursday, February 5, 1852, came as no surprise; some fifty or more people were standing at or near the dam at the time, confidently expecting it to go.  Others, more optimistic or fatalistic, had retired to bed below it.

Seventy-eight died and there was widespread property loss.   Then the Dale Dyke embankment, near Sheffield (and overseen by the nephew of the Bilberry Dam

engineer) failed in 1863.   The death toll this time was 238.   The property owner

23     Professor Brian Simpson was a pre-eminent historian of the common law. Lamentably he died earlier this year, on 11 January 2011, precisely 150 years (and one month) after the actionable event in Rylands v Fletcher.

24     Through which, on 11 December 1860, the water burst downwards, emptying Mr Rylands’

reservoir but inundating Mr Thomas Fletcher’s coalmine.

25     See A W B Simpson Legal Liability for Bursting Reservoirs: The Historical Context of Rylands v Fletcher (1984) 13 J Leg Stud 209 at 212.

26     Occupying the same essential function as that performed by the Court of Appeal, when constituted in 1875.

27 (1866) LR 1 Exch 265, 279-280.

28     Op cit n 25, at 219.

claimants briefed Sir Hugh Cairns QC.   It was he who, as Lord Cairns, affirmed Blackburn J’s decision when Rylands v Fletcher reached the House of Lords.  In the ensuing compensation proceedings,29  the dam owner was represented by counsel who acted for Mr Rylands.   And the Dale Dyke claimants were represented by counsel who thereafter was for Mr Fletcher.

[110]   Since Rylands v Fletcher we have seen the rule applied to a number of community or privately conveyed utilities:   water,30  sewerage,31  electricity,32  and gas.33    As Professor Smilie has pointed out,34  the cases in which liability for these activities has arisen have tended to be ones involving bulk conveyance, as opposed to domestic installations. But the dividing line is by no means distinct.

[111]   Due regard to the context and content of Rylands v Fletcher might suggest that  this  rule  concerning  the  escape  of  dangerous  things  might  itself  be  more confined.  Its proper function is to impose, exceptionally, strict liability for escaping hazards that are of their nature (1) very dangerous and (2) (perhaps) unexpected, so that the adjacent land owner cannot make satisfactory alternative provision – e.g. by insurance.

[112]   In Australia the High Court of Australia abandoned the rule in Rylands v Fletcher in Burnie Port Authority v General Jones Pty Ltd.35   It held that it had been absorbed within general principles of negligence, public nuisance and trespass.  In doing so it returned some way to the orthodox position ex ante Rylands v Fletcher. Although  Burnie  was  a  majority decision,  the  High  Court  of Australia  has  not

revisited the issue.

29     The dam owner was subject to a wide-ranging and strict compensation provision in the private

Act of Parliament authorising construction of the dam: s 68 of the Sheffield Waterworks Act

1853. The provision had been inserted in the bill following the Bilberry Dam disaster: Simpson, op cit n 25, at 234-237.

30     Charing Cross Electricity Supply Co v Hydraulic Power Co [1914] 3 KB 772 (CA); Irvine & Co Ltd v Dunedin City Corporation [1939] NZLR 741 (CA).

31     Smeaton v Ilford Corporation [1954] Ch 450 (Ch D).

32     Hillier v Air Ministry [1962] CLY 2084 (a case where cows were electrocuted by high voltage

cables laid beneath the plaintiff ’s fields).

33     North Western Utilities v London Guarantee & Accident Co [1936] AC 108 (HL); Hanson v

Wearmouth Coal Co [1939] 3 All ER 47 (CA).

34     Smillie, “Non-natural Use” [2011] NZLJ 88.

35     Burnie Port Authority v General Jones Pty Ltd (1994) 179 CLR 520 (HCA).

[113]   In England, the House of Lords was invited to take a similar approach in Transco plc v Stockport Metropolitan Borough Council.36    But it declined the invitation.   In an earlier appeal, in Cambridge Water Co Ltd v Eastern Counties Leather plc, Lord Goff had said that, as a general rule, the imposition of strict liability is more appropriately a matter for Parliament than the Courts.37    The same might also be said of abolition of the rule.   But confinement in accordance with original purpose is another thing.  The House of Lords did exactly that in Transco when Lord Bingham said:38

Bearing in mind the historical origin of the rule, and also that its effect is to impose liability in the absence of negligence for an isolated occurrence, I do not think the mischief or danger test should be at all easily satisfied.  It must be shown that the defendant has done something which he recognised, or judged by the standards appropriate at the relevant place in time, he ought reasonably to have recognised, as giving rise to an exceptionally high risk of danger or mischief if there should be an escape, however unlikely an escape may have been thought to be.

[114]   In  New  Zealand,  the  Court  of  Appeal  too  has  declined  to  follow  the Australian abrogation option.   In Autex Industries v Auckland City Council39  the defendant council was sued in respect of a burst water main.  It submitted that New Zealand should follow Burnie and treat Rylands v Fletcher as having been subsumed into the law of negligence.  A majority of the Court contemplated the possibility that the concept of non-natural user might be different now from when the similar fact pattern in Irvine & Co Ltd v Dunedin City Corporation40 was decided.  The minority appeared to consider Irvine still good law, and remained critical of the approach taken  in  Burnie.    They  cited  with  approval  this  observation  of  Professor  John Fleming, supporting Rylands v Fletcher as a vital component of tort theory:41

Negligence  deals  with  the  wrong  way  of  carrying  out  an  activity,  the residuary risk of which it is not unfair for victims to shoulder themselves. Strict liability deals with activities which even when carried out with due care, retain an abnormal risk and could be deemed negligent as such but for their countervailing utility.

36     Transco plc v Stockport Metropolitan Borough Council [2004] 2 AC 1 (HL).

37     Cambridge Water Co Ltd v Eastern Counties Leather plc [1994] 2 AC 264 (HL) at 305.

38     Transco plc v Stockport Metropolitan Borough Council [2004] 2 AC 1 at 11.

39     Autex Industries v Auckland City Council [2000] NZAR 324 (CA).

40     See note 30 above.

41     (1995) Tort Law Review 56 at 60.

[115]   Certainly, as the House of Lords held in Transco, there will be a residue of cases where strict liability properly applies.  As noted earlier, such cases involve the imposition of exceptional hazard and (from the adjacent owner’s perspective) unexpected outcome.42     Take, for instance, the lawful manufacture or storage of explosives.   If through no fault of anyone at all they explode, and devastate the neighbourhood, are the neighbours to be denied compensation for want of proof of

fault?

[116]   It may well be time, however, to reconsider whether the rule should apply to commonplace  utilities,  such  as  water  and  sewerage.    Particularly  where  their presence is obvious to adjacent landowners.  It may be that Courts will eventually conclude that liability for the failure of such systems is dependent in all cases on proof of negligence.

[117]  Regardless however of these philosophical considerations, two particular responses to the Rylands v Fletcher claim, additional to s 148, can be made in this case.

[118]   First, a floodway built through a flood-prone district is not a non-natural use of the land.   The very necessity for its construction suggests that it is not.   The consequence of not having constructed the floodway would, in the context of this storm  event,  likely have  been  inundation  of  the  Moutoa  basin  and  surrounding district.  The works do not of their nature engage hazard; rather they seek to defray hazard that natural causes have already created.   In a context where works are constructed  to  remove  (or  at  least  diminish)  a  risk  otherwise  present,  it  seems counter-intuitive to regard them as creating an exceptional hazard.  And, moreover, their presence is immediately apparent to adjacent landowners.  Those landowners knew they were a community work, and that they were not completely flood-proof in nature.   The community set the extent to which capital should be spent on the stopbanks.  It can be taken to have had an informed appreciation of the general risks

associated with them.

42     If the hazard is simply stored, for instance, the adjacent owner may not be aware of the activity’s

existence at all.

[119]   Such a view is supported by the decision of La Forest J (with whom Dickson CJ agreed) in Tock v St John’s Metropolitan Water Board.43    In that case there had been a ferocious rainstorm.   A nearby stormwater drain failed to work, and the plaintiff’s basement was flooded.   La Forest J concluded that the installation of public drainage systems was “an indispensable part of the infrastructure necessary to support urban life” and could not be characterised a non-natural use.44

[120]   The  second  response  is  that  the  very  statutory  duty  relied  upon  by  the plaintiffs  to  found  their  breach  of  statutory  duty  claim  creates  a  fundamental objection to Rylands v Fletcher liability.  Section 126 of the Soil Conservation and Rivers Control Act 1941 provides that “it shall be a function of every catchment board to minimise and prevent damage within its district by floods and erosion”.  As

Lord Scott put it in Transco:45

... it is, I think, worth reflecting on why it is that an activity authorised, or required, by statute to be carried on will not, in the absence of negligence, expose the actor to strict liability in nuisance or under the rule in Rylands v Fletcher. The reason, in my opinion, is that members of the public are expected to put up with any adverse side-effects of such an activity provided always that it is carried on with due care. The use of the land for carrying on the activity cannot be characterised as unreasonable if it has been authorised or required by statute. Viewed against the fact of the statutory authority, the user is a natural and ordinary use of the land. This approach applies in my opinion,  to  the  present  case.  The  council  had  no  alternative,  given  its statutory obligations to the occupiers of the flats, but to lay on a water supply. Strict liability cannot be attached to it for having done so.

Conclusion

[121]   For  the  reasons  given  above,  which  centre  upon  s  148  of  the  Soil

Conservation and Rivers Control Act 1941, the Council’s liability to the plaintiffs in

this case is dependent on proof of negligence.

43     Tock v St John’s Metropolitan Water Board [1989] 2 SCR 1181 (SCC).

44 At [74]. The remaining members of the Supreme Court determined the case on other grounds.

45     Transco plc v Stockport Metropolitan Borough Council [2004] 2 AC 1 at 33.

Issue 2: Did the Council owe the plaintiffs a duty of care in its monitoring and maintenance of the stopbank?

Duty of care

[122]   The essential case for the plaintiffs is that the Council owed them a duty of care to monitor and maintain (to design standards) the condition of the stopbank.46

[123]   I dismiss from consideration allegations concerning duties of care in the conduct of the 1997/1998 LMS review.  Likewise modelling of hydraulic capacity of the floodway.   All  of these would  be time-barred,  and  were not  the subject  of extensive evidence or submission.  As I have said earlier, the plaintiffs’ focus is not on construction,47  but on subsequent monitoring and maintenance of the stopbank adjacent to the trestle bridge.

[124]  I also dismiss allegations relating to the response of the Council to the emergency.  Although pleaded, this aspect was but faintly pursued.  There was no evidence of any cogency to suggest that the Council’s response to these unexpected and catastrophic floods was other than impeccable.   I need not waste time now analysing the nature of the duties they owed in respect of their flood response.

[125]   So  the  pertinent  question  here  is  simply  whether  the  Council  owed  the plaintiffs a duty of care in monitoring and maintaining the stopbanks adjacent to the trestle bridge.

Submissions

[126]   The plaintiffs submitted that the Council owed them a duty of care in the monitoring and maintenance of stopbanks, for the following reasons:   the Council had  a statutory duty to  protect  and  maintain  its  waterways;48   it was  funded  by

ratepayers, who were a particular identifiable class; ratepayers being levied for flood

46 See [79]-[82] above.

47     Including reconstruction, there being community consensus to adopt Option 4: see [31]-[35]

above.

48     Soil Conservation and Rivers Control Act 1941, ss 10, 126, 133.

protection purposes were in “an almost contractual relationship with the Council”; ratepayers   were   entitled   to   assume   that   the   Council   would   carry   out   its responsibilities appropriately; and if monitoring   and maintenance was not carried out properly at critical areas such as the bridge/stopbank interface, then serious damage was likely (along with significant risk to the community).

[127]   Turning to issues of policy rather than proximity, the plaintiffs submitted that there were no political or policy issues involved, and that the proposed liability related to purely operational matters.  Mr Upton submitted:

The Council has assumed responsibility for maintenance and monitoring in circumstances where ratepayers were entitled to assume that the Council has the skills and resources to do so properly.  Further, the ratepayers themselves had no capacity to assess the risks nor to directly protect themselves from risks of the present kind.

[128]   The sole case cited by the plaintiffs was Atlas Properties v Kapiti Coast

District Council.49 I have discussed that decision already, at [98].

[129]   The Council, on the other hand, submitted that what is contended for by the plaintiffs is a novel duty of care.  Counsel had not found any decided case in New Zealand where a regional council or catchment board had been held to owe a duty of care to farmers in respect of flooding caused by breach of a stopbank.   Citing, in particular, the Court of Appeal decisions in South Pacific Manufacturing Co Ltd v

NZ Security Consultants & Investigations Ltd50 and Rolls Royce New Zealand Ltd v

Carter Holt Harvey Ltd,51 the Council submitted that the overall consideration must be one of policy.  It was submitted that if a duty was imposed in this case, then “it will open the floodgates to so many claims after every flooding event that the end result  would  be  a  massive  cost  to  the  community  as  a  whole  and  obviously significant resources being dedicated to litigation”.

[130]   The Council submitted that it was not correct to view this case as an attack on the operational management of the LMS.   The management of that scheme was

49     Atlas Properties v Kapiti Coast District Council HC Wellington CP 172/00 19 December 2001, per Durie J.

50     South Pacific Manufacturing Co Ltd v NZ Security Consultants and Investigations Ltd [1992] 2

NZLR 282 (CA).

51     Rolls Royce New Zealand Ltd v Carter Holt Harvey Ltd [2005] 1 NZLR 324 (CA).

dictated by the funding available, something voted upon by the community.  Thus the case dealt with significant policy and resourcing issues.  Ms Macky submitted:

It  is  not  for  the  Courts  to  override  the  wishes  of  the  community  in determining what level of protection the community is prepared to pay for. The imposition of a duty by the Courts, if it requires greater expenditure than approved by the community, will be a burden to the community by way of increased rates.   ... [I]t is not for the Courts to exercise this function but rather the community through implementing policy as envisaged by the Act.

Analysis

[131]  I conclude that a duty of care does exist in respect of monitoring and maintenance of the floodway stopbanks.  I reach this conclusion for three reasons:

[132]   First, there is s 148 of the Soil Conservation and Rivers Control Act 1941.52

Two points can be made about s 148:

(a)      it would be a remarkable consequence if s 148 were to have the effect of  excluding  causes  of  action  other  than  negligence,  but  that negligence  itself  could   not  then  arise.     The  Council’s  stance effectively would render it immune for all practical purposes from civil liability; and

(b)the purpose of s 148 (and its adjacent provisions) is in part to confirm that bodies such as the Council should be liable if they are negligent in the construction or maintenance of stopbanks.

[133]   Secondly, there is the fact that the Council explicitly undertakes monitoring and  maintenance  of  the  stopbanks,  and  budgets  for  that  activity.   A significant percentage of the annual rating of landowners benefiting from the LMS scheme (and therefore paying the lion’s share of its cost) is spent on maintenance.53   Council work gangs inspect the banks on a regular basis.  No major policy or resource allocation issue arises.   The allocation has been made already.   The Council monitors and

maintains.

52 See at [95].

53     See at [21] and [34].

[134]   And it has little choice but to do so.  It was common ground among the expert witnesses called for the parties that stopbank security is dependent on good maintenance practices. As Mr David Hamilton, a consulting engineer with particular experience of catchment boards and regional councils, put it:

Good practice for a regional council that owns and operates flood control and drainage schemes is to ... provide for maintenance of the scheme to meet defined levels of service.

And:

A well-run  scheme  which  involves  extensive  stopbanks  would  typically involve regular inspections with each part of the stopbanks in the scheme being inspected at least once in every 24 month period or as required as a result of specifically noted issues.

[135]   It is proper to infer in such circumstances a duty of care in the performance of that operational function.   At the very least those who pay rates for maintenance services are proximate.

[136]   Harm to adjacent landowners between the stopbanks and the Manawatu River to the south, following failure to meet that duty, is wholly foreseeable.   I do not understand the Council to have suggested otherwise.  In my view the Council had a duty of care to monitor and maintain the stopbanks so that the community could reasonably be assured that they would conduct flood waters in ordinary circumstances.  The consequences of failure to do so plainly include significant risks to people as well as to property.   Witness the invidious experience of the Stratton

family.54     The Council was well placed to undertake the duty.   Indeed stopbank

monitoring was already a regular responsibility of the local work gang overseer. Broader policy reasons do not exist to negate this duty.  Rather, to negate the duty would conflict with the underlying legislation.

[137]   The proposition advanced by Ms Macky for the Council that this would “open the floodgates to so many claims after a flooding event”, causing a “massive cost to the community” is more a reflection of the extent of danger that would arise

from disregard of the duty of care, than a reason not to impose it.  The point is more

54 See at [48].

relevant to determination of the relevant standard, rather than the existence of a duty, of care.

[138]   The fact that the community itself determines  the extent of maintenance services to be performed will significantly influence, although not determine, the extent of the duty imposed.  I will consider that under Issue 3.

[139]   Thirdly, although the nature of the breach of duty alleged in this case is an omission to repair, landowners in New Zealand may be held to owe a duty to maintain their land and to take positive steps to prevent harm to adjacent landowners resulting from the operation or use of their land.  The principle finds expression in

cases like Goldman v Hargrave,55  Landon v Rutherford56  and Leakey v National

Trust.57

[140]   The former cases were ones of negligence by omission in failing to respond adequately to a fire (itself not caused by negligence).58   The latter concerned the fall of unstable soils from one hillside property to another lying below. Although Leakey was brought in nuisance, the discussion of the relevant duty of care is equally transferable to negligence.   The Privy Council decision in Goldman demonstrates that proposition, drawing as it does upon authorities in both forms of action.  What Goldman holds is that a landowner has a general duty of care to a neighbouring occupier, in relation to hazards occurring on that land (and regardless of cause).  The

duty is to take such steps as are reasonable for a person in the shoes of the landowner to prevent or minimise the risk once the landowner is (or should have been) aware of it.

Conclusion

[141]   To impose a duty of care on the Council in this case, in respect of the monitoring and maintenance of the stopbanks, and in favour of adjacent landowners,

is supported by the underlying legislation. And it is orthodox as a matter of common

55     Goldman v Hargrave [1967] 1 AC 645 (PC).

56     Landon v Rutherford [1951] NZLR 975 (HC).

57     Leakey v National Trust [1980] QB 485 (CA).

58     See also Wilson & Horton Ltd v Attorney General [1997] 2 NZLR 513 (CA).

law.  In short, there is sufficient proximity between the Council (as the owner of the stopbanks), and the adjoining landowners and occupiers, that it is just and reasonable that a duty of care to monitor and maintain the stopbanks should exist.

[142]  I find, therefore, that the Council owed the plaintiffs a duty of care in monitoring and maintaining the floodway stopbank.

Issue 3: Was the Council negligent?

Standard of care

[143]   The applicable test in determining whether the Council has breached its duty of care is that set out in a judgment of Venning J in Body Corporate 189855 v North Shore City Council:59

The standard is that stated in Askin v Knox [1989] 1 NZLR 248, namely the standard to be regarded as reasonable at the relevant time for the officers carrying out the task in issue. I remind myself that the parties and the Court are approaching the matter with the benefit of hindsight and the knowledge that now exists of the various issues that have contributed to the leaky building crisis.

[144]   The question is what could reasonably have been expected of the Council and its staff in the period prior to the flood in February 2004.  As Heath J said in Body Corporate 188529 v North Shore City Council,60  care must be taken to ensure that, when assessing the standard of care expected from a council and its officers, the bar is not set too high.

[145]   In particular, the standard should generally reflect constraints the community itself has imposed on the Council in terms of funding.   I have already held this affects  the  scope  of  the  duty  of  care.    Likewise  it  influences  the  standard  of

performance the Council owes those to whom it owes a duty.

59     Body Corporate 189855 v North Shore City Council CIV-2005-404-5561 HC Auckland, 25 July

2008 at [75].

60     Body Corporate 188529 v North Shore City Council [2008] 3 NZLR 479 at [55].

Submissions

[146]   The plaintiffs’ argument was focused very substantially upon a contention that there was a gap between the cement bags on top of the stopbank, and the underside of the trestle bridge.  I have noted this point already.61

[147]   The  existence  of  the  gap  was  the  linchpin  of  the  plaintiffs’ case.    As

Mr Upton put it in cross-examination of a Council witness:

Messrs Easton and Eveleigh’s claim to the Court is on the basis that the

problem started under the bridge and then developed and spread from there.

[148]   In particular, the stopbank was intended to have a watertight seal between its top and the underside of the bridge.  Such a seal was depicted in the original 1958 design drawings.  A watertight seal was, said Mr Upton, “central to maintaining the bank’s integrity in a design flood when the water in the floodway was intended to lap the bridge beams”.

[149]   In summary:

The plaintiffs say that the stopbank collapse was initiated under the bridge with some sort of failure at that point, the precise mechanism of which is not important

[150]   The plaintiffs also contended that there were other contributing causes: a narrowing of the stopbank adjacent to the bridge;62  the presence of a raised berm (caused by drain clearings) in the floodway floor upstream of the bridge; and a cycleway ramp intruding into the floodway some 20 metres downstream of the bridge.

[151]   The  plaintiffs’  theory  of  their  case  was  therefore  dependent  to  a  very substantial degree on sustaining three propositions:  (1) that the gap existed; (2) that the presence of the gap was negligent; and (3) that the gap was causative of the ultimate failure of the bank (recognising that it would do them no good if it was

simply one of a series of independent causes of the stopbank failure).  Propositions

61 See at [80].

62 Which I have already dismissed, as none of the expert witnesses supported it: see at [80].

(1) and (2) must be dealt with under the current Issue 3; proposition (3) is at the heart of Issue 4.

[152] Turning to the Council, it devoted considerable effort and evidence to responding to proposition (1).   That is, it sought to establish that no gap in fact existed beneath the bridge.  That contention turned on evidence given by Mr Graeme Anderson.  I discuss Mr Anderson’s evidence on this topic later in my judgment.63

[153]   In the alternative, if a gap existed then the Council’s response to proposition (2) was that it had nonetheless taken reasonable care in its management and monitoring of the stopbanks.   In particular, by obtaining the Riley report in 1994. The Council submitted that Riley had been instructed to check services through or adjacent  to  stopbanks,  to  check  the  permeability  of  high  priority stopbanks  for potential piping failure zones, and to recommend works and management practices to improve the security of the stopbank system.  Nothing in the Riley report would have indicated a problem in the vicinity of the trestle bridge.   Moreover, the monitoring and maintenance regime implemented by the Council was the subject of evidence.   It received the approval of the Council’s expert witness, Mr David Hamilton.  Mr Hamilton concluded the Council had acted diligently.

[189]   On the evidence before  me, therefore,  I could  not  have found the drain causative of the stopbank failure.   Nor would I have found that a foreseeable consequence of the drain was erosion of the stopbank.  Accordingly, even assuming the deposition of the drain clearings amounted to “monitoring and maintenance”, I would not have found that to constitute a breach of duty.

[190]   Cycleway:  Downstream of the bridge, some 10   metres   or   so   distant,   a cycleway runs parallel to the highway.  Of course, cyclists must mount the crest of the stopbank to go upon their journey.  So ramps were built on either side.

[191]   Mr Williams suggested, and again with caution, that the ramp in the floodway might have contributed to eddying.  But, he said, he did not know how much erosion could be said to have resulted.  Mr Hamilton’s evidence was that the cycleway ramp would have resulted in a slight generation of vortices at the point of intrusion into the flood waters.  But it would not have affected things as far back as the bridge.

[192]   The construction of the ramp, which lies beyond the scope of maintenance and monitoring, was at most a potential contributor to the downstream failure.  But there was simply insufficient evidence for me to conclude that, but for the construction of the cycleway, the floodway stopbank would not have failed at that point.   Viewed with foresight, let alone hindsight, I would not have found it foreseeable that the cycleway ramp would cause erosion of the stopbank.

[193]   Accordingly, I discount the breach allegations relating to the cycleway and drain.

Conclusion

[194]   I conclude that the failure of the Council to identify and remedy the gap between the cement bags topping the stopbank, and the underside of the bridge, in the course of its routine monitoring and maintenance of stopbanks, was in breach of its duty of care to the plaintiffs.  In that respect alone I find the Council to have acted negligently.

Issue 4:  Did the Council’s negligence cause the plaintiffs’ loss?

Causation

[195]   A negligent defendant will only be liable if its conduct causes damage.  The

defendant’s action must be a substantial and material cause:   Johnson v Watson.74

That  is,  it  “must  have  had  a  real  influence  on  the  occurrence  of  the  loss  or damage”.75  The Courts apply a “but for” test.76     Would the loss or damage have occurred but for the negligent act or omission?   A robust approach is taken.   As Glazebrook J said in Accident Compensation Corporation v Ambros:77

It must, however, always be borne in mind that there must be sufficient material pointing to proof of causation on the balance of probabilities for a court to draw even a robust inference on causation. Risk of causation does not suffice.

[196]   It is on this issue that the plaintiffs’ claim runs into greatest difficulty.

Submissions

[197]   The plaintiffs submitted that there were not two independent causes of the stopbank failure, but that “the whole of the events at, around and under the bridge on the night of 16/17 February 2004 were intimately connected and all part of the same causative incident”.  The initial flow from the floodway “would have been through the gap between the bags and the underside of the bridge and through any gaps that opened up between the bags as a result of settlement”.  The stopbank would have eroded along the floodway face and (as a result of overflows) along the southern face of the stopbank outside the floodway. This caused the bank to weaken and breach.

[198]   The most likely explanation for the breach at the bridge was said to be that scouring at the bridge had created a gap in the stopbank on both sides of the bridge and “was starting to erode the base of the stopbank to the side of and under the

concrete  bags”.    The  plaintiffs  referred  here  in  particular  to  Mr Doug  Easton’s

74     Johnson v Watson [2003] 1 NZLR 626 (CA).

75 Ibid at [18].

76     Accident Compensation Corporation v Ambros [2008] 1 NZLR 340 (CA).

77 Ibid, at [70].

evidence of water pouring through a breach immediately upstream of the bridge.  It was also submitted that I should not find that there was a separate downstream breach.   Instead it was probable that there had been a progressive breach, both upstream and downstream, emanating from the conditions under the bridge.

[199]   The Council, on the other hand, contended that the cause of the failure, initially, was probably water working its way through loose (perhaps hand compacted)  materials  immediately  upstream  of  the  bridge.    It  was  this  erosion initially that allowed substantial volumes of water into the Moutoa Basin.  On the other  hand,  the  Council  submitted,  the  sandbags  under  the  trestle  bridge  were “almost certainly still intact” at that stage.

[200]   The  Council  also  submitted  that  over-topping  downstream  of  the  bridge resulted in a sequence of failures near the cycleway, and again between the cycleway and the bridge.  But, submitted the Council, there was no evidence to suggest that any of these failures were connected to the gap under the bridge, on which the plaintiffs had built their case.

Analysis

[201]   To  set  the  scene,  I  should  observe  that  I  was  faced  with  somewhat diametrically opposed evidence. At least to begin with.

(1)  Mr Gary Williams

[202]   Mr Williams maintained that the most logical cause of the failure of the stopbank in the vicinity of the trestle bridge was water flowing through the gap, eroding the stopbank, and causing progressive failure both upstream and downstream of the initial breach:

When the water started lapping onto the bridge, and up against the beams, there would no longer be a free air surface under the bridge and pressure flow would have developed.  This, though, requires some heading up or rise in water level to drive the pressure flow, and there is a loss of flow efficiency at  this  point.    A surging  in  the  surface  levels  takes  place  under  these conditions, which generates pulses, waves and eddying in the water.  This surface fluctuation, eddying and wash effect would have interacted with the

complex beam arrangement of the bridge, giving rise to wave impact and slosh effects, and forcing water in different directions.   Overflows would, therefore, have occurred through the gap under the deck slab and beams at the stopbank.   Once pressure flow had developed at the bridge, overflow under the bridge would have become inevitable.

...

Once floodwaters in the floodway had reached the underside of the bridge, stopbank failure of the bridge may have been inevitable – without any other aggravating influence – because of the overtopping of the sandbag areas under the bridge.

[203]   Although  expressing  his  views  in  evidence  with  a  greater  degree  of dogmatism than the other expert witnesses, Mr Williams did qualify his views in three important respects.

[204]   First, Mr Williams accepted that he could not be certain about the source of the flow of water in the basin reported by Messrs Stratton and Doug Easton in the early hours of Tuesday morning.  Mr Williams accepted that these had to have come from around the bridge.  He noted, correctly, that the eye-witnesses’ evidence was not consistent.  His “best assessment” was that it came from under the bridge.

[205]   In this respect I consider Mr Williams had been influenced in his opinion, and thus understandably misled, by the initial evidence of Mr Jamieson.  Mr Jamieson’s initial observation of water “flowing over the top of the stopbank under the bridge” – and  not  on  either  side  –  “approximately half  a  metre  to  1  metre  wide  and  20 centimetres deep” was later significantly qualified in cross-examination.   Then he accepted that what he had seen at that stage was only a “slow trickle”.  This was, of course,  at  5.00  pm  on  Monday  evening  –  three  hours  or  so  before  the  peak floodwater level was reached.

[206]   Secondly, the experts caucused before giving evidence.  The caucus produced a useful joint report, agreeing on a number of technical issues.  Then the evidence that followed was given in a panel format – otherwise known as a “hot tub”.  Both the caucusing and panel processes resulted in a greater degree of consensus than would have been the case had evidence been given serially.   It also reined in any predisposition to rigid adherence to briefs.  That was important because in some key areas (e.g. Mr Jamieson’s evidence, and the issue over whether there was indeed a

gap), things had been clarified by factual witnesses before the experts’ turn arose.  In my view,  it  is  always  preferable  for  expert  witnesses  to  follow  fact  witnesses, whether or not a panel (or “hot tub”) approach is adopted.

[207]   In  this  case  a  critically  important  concession,  and  point  of  consensus, emerged during the caucusing.  It was that all experts agreed:

...  that  the  sandbags  may  still  have  been  in  place  at  the  time  of  the catastrophic failure.  [i.e. at 6.30 am on Tuesday morning].

[208]   Thus it followed that, with the cement bags potentially still in place at 6.30 am, the large amounts of water in the basin at 2.00 – 3.00 am had to have come from an over-topping or breach either upstream or downstream of the bridge.  It could not possibly have come just from water forcing itself through the narrow gap above the bags (even allowing for additional volumes flowing around the bags).  That could not have been sufficient to cause the huge flow which “raced like a wave” across Mr Stratton’s farm at 3.00 am.

[209]   The third respect in which Mr Williams qualified his views was to withdraw his earlier suggestion that the narrowing of the stopbank under the bridge may have made the area more prone to scour and erosion.  The experts also agreed this point in their joint caucus report.   They said, “we do not consider that the smaller cross- section of the stopbank under the bridge was a contributing factor to the failure”.

(2)  Mr Noel Luxford

[210]   The first point Mr Luxford made was that even allowing for the gap, reducing the effective freeboard beneath the bridge, the cement bag level was still slightly higher than  the peak  floodwater level  at  the bridge.   The  gap  lay between  the underside of the bridge (approximately 4.9 metres) and the top of the cement bags (approximately 4.75 metres).  The water level at the peak of the flood was agreed as being 4.65 metres at the point it met the bridge.

[211]   It was not suggested by Mr Luxford that water would not escape over the top as a result of slop – given the disturbed water patterns beneath the bridge caused by the intruding beams and debris.   Plainly it would, and in reasonably significant

quantities.  But in his view there would not be any continuous pouring of water over the gap.  It was more a case of water bypassing between the bags – as demonstrated by  Mr Jamieson’s  observations  of  a  “slow  trickle”  at  5.00  pm  on  the  Monday evening (when the water was not of course yet at its peak).

[212]   Mr Luxford also concluded that even if there were gaps between the cement bags, “any flow through those gaps would have been very unlikely to have displaced what would by then have effectively been weak rocks moulded into each other”.  As I have already noted, the experts all agreed that the bags may still have been in place at 6.30 am, just before the catastrophic downstream failure.

[213]   Mr Luxford also considered that the cement bags should “easily withstand water flowing over the top of them even if there was a gap between them and the underside of the bridge”.   In good part this was due to the slow velocity of the floodwaters, due to the shallow gradient of the floodway. Thus he opined:

I would be very surprised if the failure sequence was due to erosion and displacement of those concrete “rocks” causing failure.  It would be much more likely in my opinion that failure would emanate at some other point on the stopbanks.

[214]   Mr Luxford said:

I cannot tell why the stopbank failed downstream of the bridge and from the evidence I have seen I doubt if anyone else can.  I understand there has been some suggestion that the surging under the bridge or interface caused by the bridge was the trigger point for the failure of the sandbag portion of the stopbank under the bridge.   That seems to me to be unlikely because the stopbank on the right hand side downstream of the bridge did not fail and I would have expected the effects of the waterway under the flood event both beneath the bridge and downstream to be the same as the left hand side and yet no failure was experienced on the right hand side.  I suspect that it must have been an unseen weak point within the stopbank downstream of the bridge which eventually initiated failure.

In my opinion it is unlikely that failure of the southern stopbank on 17

February 2004 would have commenced by displacing the sandbags under the

Whirokino trestle bridge.   It is much more likely that failure would have been initiated at the point within the silt and sand fill forming the stopbank material either side of the trestle bridge.

[215]   In panel evidence Mr Luxford expanded on what the “weak points” might have been.  In particular he now focused on the fact witnesses’ evidence of:

(a)      upstream breach – i.e. the evidence of Mr Doug Easton, Mr Richard Barber and Mr Stuart Barber.  This confirmed the presence of water flowing over the stopbank upstream of the bridge from 5.30 am or thereabouts on the Tuesday morning.  Indeed, probably much earlier given Mr Stratton’s evidence.   Mr Doug Easton’s evidence was particularly notable:  he was standing to the south of the stopbank and

could see “bits of the stopbank upstream of the bridge breaking off”;78

and

(b)separate downstream breach – i.e. the evidence of Mr Richard Barber, Mr Stuart Barber, Mr Owen Bonis and Mr Graeme Anderson.   This confirmed a distinct breach before 6.00 am on the downstream side of the bridge.   That is, ahead in time of the catastrophic failure in the same vicinity at 6.30 am.  Understandably after the passage of seven years, their evidence as to the location of that breach varied.  But the majority of these witnesses put the breach at a point remote to the bridge.  Mr Anderson put it 7-8 metres downstream; Mr Bonis and Mr Stuart Barber rather closer (although I found Mr Barber’s evidence on

location equivocal in the end).  As noted before,79 Mr Richard Barber

was not questioned about location.

[216] Considering these matters, Mr Luxford opined that it was possible that the bridge (which of course pre-dated the construction of the stopbanks) had caused difficulties for those constructing the banks in the early 1960s. In particular in compacting the fill immediately adjacent to the bridge. As a result, it was possible that the stopbank in that location would have contained voids due to the fill being relatively loosely compacted. Here the stopbank may have been prone to “piping” in the way described at [26]. Mr Luxford observed:

At  the  same  time  we  would  have  seen  water  saturating  the  stopbank materials and coming through any voids.   This is when I think internal erosion would have been occurring in the stopbank materials completely separate from any influence of the sandbags.   Seepage connected with the existence of the trestle bridge/sandbag interface material could also have

78 See at [51].

79 At [53].

existed because it would have been difficult to achieve ideal compaction of the stopbank materials particularly at the interface between the bridge beams and the stopbank and in places under the soffit of the deck of the bridge. Even with the topping of the stopbank undertaken by Horizons [i.e. in 2003] there  would  still  have  been  some  voids  left  in  those  natural  stopbank materials which would have enabled the water to traverse through the natural stopbank materials.  Weak points will have been created and in places fine materials would be washed out giving rise to the risk of failure at the point of those weaknesses.

This failure probably occurred because the original stopbank material settled over time and contained voids which allowed water to penetrate and traverse through so as to weaken the stopbank in the vicinity of the catastrophic failure zone.  We do not know and nor will we ever know how that water entered the stopbank.   It could have had its origin in a weak compaction point or some extraneous material within the original stopbank.  There could have been a number of entry points for water to eventually find its way to the centre of the catastrophic failure zone.  The bridge being where it was may have had an influence because it would have created some difficulties in compaction.   Such a mechanism is unconnected with any voids and gaps between the sandbags and the bridge.

[217]   Mr Luxford pointed out (and I agree) that the Council would have had no reason to appreciate these potential weaknesses.   None had been identified by the

1994 Riley report.80    That had not identified any voids in the southern stopbank in

the vicinity of the bridge, despite bore holes having been drilled in the stopbank within 200m of the bridge, both upstream and downstream.  Riley in fact identified potential issues in the northern stopbank, near the bridge.   That was because that stopbank incorporated sandhill material.81    But the northern bank held fast in the flood.

(3)  Mr David Hamilton

[218]   Mr Hamilton also rejected Mr Williams’ theory of a breach “emanating” from the gap under the bridge.  He concluded that the critical breach was downstream of the bridge (in accordance with the fact witnesses’ evidence). And that (as is normal) it would have spread laterally both up and downstream. As he put it:

When a breach occurs there will normally be some lateral spread in both upstream and downstream directions.   As the breach develops the breach widening will occur to a greater extent in the upstream direction due to the downcutting and headwards erosion of the fill (given similar bank materials

80 See [24]-[25].

81 See at [25].

in  both  directions).   The  actual  breach  is  therefore  most  likely  to  have occurred more than 6 metres downstream of the bridge centreline.

[219]   Mr Hamilton considered Mr Luxford’s suggested mode of failure “logical and sensible”, and that it provided a “reasonable explanation of the sequence of events”.  Both Mr Hamilton and Mr Luxford considered that the catastrophic failure at 6.30 am was caused by an unseen weak point downstream of the bridge.  What that was they could not say.  The flood had washed that evidence away.  It was not likely to have been the cycleway ramp.  For a brief period Mr Hamilton thought it might have been an old telecommunications duct running through the stopbank at the extreme downstream end of the breach.   But he abandoned that view during the panel discussion.   It would have been inconsistent with his overriding theory of a breach working both ways from the weak point.

(4)      Assessment of the evidence

[220] The burden of proof as to causation, in accordance with the principles at [195] above, lies throughout on the plaintiffs. They must show on the balance of probabilities that, but for the negligently overlooked gap between the underside of the bridge and the cement bags, the Moutoa basin would not have been inundated by the February 2004 flood.

[221]   I  do  not  consider  that  the  plaintiffs  are  able  to  discharge  that  burden. Following the foregoing review of the evidence, my reasons can be stated succinctly:

(a)      It is clear that by about 5.30 am on Tuesday 17 February 2004 the stopbank had eroded significantly upstream of the bridge.  A number of witnesses gave evidence of this, but the clearest evidence was from Mr  Doug  Easton,  a  witness  called  by  the  plaintiffs  themselves.82

Mr Easton saw water pouring through the bank on the upstream side

of  the  bridge,  and  “bits  of  the  stopbank  upstream  of  the  bridge breaking off as the water scoured it away”.

82     Although Messrs Richard Barber and Stuart Barber likewise gave evidence to this effect.

(b)The expert evidence did not establish any convincing link between this initial upstream breach and the existence of the 150 millimetre gap between the cement bags and the underside of the bridge, further downstream.   In this respect the plaintiff’s theory, restated at [197]- [198]  above, was  profoundly unsatisfactory.    On the one hand,  it recognised the critical importance of the evidence of the upstream failure,  attributing it  to  “scouring action” at  the bridge.      But  its evidence could not connect that initial failure to the actionable gap beneath  the  bridge.    In  other  words,  it  could  not  explain  it  as something that had “emanated” from the gap.  But nothing apart from the gap had been shown to be actionable.

(c)      Likewise,  by  early  Tuesday  morning,  a  significant  breach  was occurring downstream of the bridge.  The exact location of that, after so many years, is in dispute.  But I am satisfied that the evidence of witnesses called for both sides identified the existence of such a breach.  It also appears likely that the initial downstream breach did not reach continuously back to the bridge.   That is, there was still some intact stopbank between the bridge and the initial downstream breach.  Again, that suggests there was not a continuous failure of the stopbank originating at the gap.

(d)The 150 millimetre gap would still have been above the maximum water level during the flood, putting to one side the disturbed water patterns beneath the beams.  Of course I do not put those effects aside at all.  But I do consider that what was likely to have occurred under the bridge was splash and slop through the gap, rather than a continuous erosive overflow.

(e)      The preceding conclusions are then reinforced very substantially by the  joint  position  taken  by the  experts  that  as  at  the  catastrophic failure downstream, at 6.30 am on Tuesday 17 February 2004, the cement bags may well still have been intact beneath the bridge.  That

joint assessment severely undermines the plaintiffs’ causation theory

of a failure emanating from the gap.

(f)      Finally, while the exact cause of the failure is not now known, we do know that this was a flood event of a magnitude that exceeded the design standard of the LMS.   A standard that the community had bought into in approving Option 4.  As I said earlier, this flood event was always going to challenge the capacity of the floodway and river system  to  cope.    The  fact  that  ultimately they  did  not  cannot  be attributed with any clarity to the limited deficiency identified in the Council’s maintenance processes.

[222]   I therefore agree with the tenor of the evidence given by the experts called by the  Council  that  it  was  unlikely  that  the  gap  beneath  the  bridge  caused  the catastrophic failures that occurred to the stopbank both upstream and downstream of the bridge.  The gap was insufficiently material to cause those failures.  It is more probable than not that those failures were caused by factors independent of the gap, exposed in the course of an exceptional 1 in 110 year, over-standard weather event. And, that being so, it is more probable than not that the Moutoa basin would have flooded to exactly the same extent on Tuesday 17 February 2004, had the gap not been there.

[223]   At this juncture, therefore, the plaintiffs’ claim must fail.

Conclusion

[224]   For the foregoing reasons, I find:

(a)       The effect of s 148(1) of the Soil Conservation and Rivers Control Act

1941 is that claims against the Council arising from breach of the

Moutoa floodway stopbank can be maintained only in negligence.

(b)Reinforced by that same provision, the Council owed the plaintiffs a duty of care in monitoring and maintaining the floodway stopbanks.

(c)      The central and overwhelming focus of the plaintiffs’ case was that the Council had breached that duty of care by failing to identify and repair a gap above the cement bags topping the stopbank crest, under the trestle bridge on the southern side of the floodway.

(d)As a matter of fact there was indeed a gap of approximately 150 millimetres in that location.

(e)      The failure of the Council to identify and remedy that gap, in the course of its routine monitoring and maintenance of the stopbanks, was in breach of its duty of care to the plaintiffs.

(f)       In that respect, alone, I find the Council to have acted negligently.

(g)The plaintiffs however failed to prove on the balance of probabilities that the presence of the gap caused the catastrophic failures that occurred to the stopbank.   Those failures, both upstream and downstream  of the bridge,  were more  probably caused by factors independent of the gap.

(h)Accordingly, it is more probable than not that the floods that damaged the plaintiffs’ crops would have occurred to exactly the same extent had the gap not been there.

Disposition

[225]   Judgment is entered for the defendant.

[226]   The defendant is entitled to costs on a category 2 basis.  If the parties cannot agree costs within 15 working days, the defendant is to file a memorandum within a further 5 working days.   The plaintiffs are to respond within a further 5 working days.

[227]   I thank counsel for their assistance.

Stephen Kós J

Solicitors:

Wadham Goodman, Palmerston North for Plaintiffs
Heaney & Co, Auckland for Defendant