Macag Holdings v Torrens Catchment Water Mgt Board No. Scgrg-99-1635
[2000] SASC 115
•5 May 2000
MACAG HOLDINGS PTY LTD v TORRENS CATCHMENT WATER MANAGEMENT BOARD
[2000] SASC 115
Land and Valuation Division
DEBELLE J. This is an appeal from the Environment Resources and Development Court (“the Environment Court”) on a preliminary question of law. As it was an appeal from an interlocutory order, the appeal may be heard by a single judge of this Court: s 30 of the Environment Resources & Development Court Act 1993. The question is whether a watercourse exists on the appellant’s land.
The appellant owns a substantial parcel of land on the eastern side of Duncan Road, Beaumont. It is in the foothills of the Mount Lofty Ranges not far from Waterfall Gully. On 3 September 1999 the Torrens Catchment Water Management Board issued two notices to the appellant pursuant to s 14 and s 17 of the Water Resources Act 1997. The notice issued pursuant to s 14 required the appellant to take certain action “to maintain the watercourse situated on the land in good condition”. The notice issued pursuant to s 17 required the appellant “to take reasonable steps to prevent damage to the bed and banks of the watercourse situated on the land.” Both notices required the appellant to:
“1.... Take reasonable steps to prevent damage to bed and banks of the watercourse situated on the land; and
2.Refrain from and prevent the placement of any soil, rock or other material in or adjacent to the watercourse.”
The appellant appealed to the Environment Court challenging the validity of the notices on a variety of grounds. One of the grounds was that there is no watercourse on the land. The question of whether a watercourse existed was heard and determined as a preliminary issue. The Environment Court held that a watercourse existed. The appellant now appeals to this Court.
Section 3 of the Water Resources Act defines a watercourse in these terms:
“ ‘watercourse’ means a river, creek or other natural watercourse (whether modified or not) and includes —
(a)... a dam or reservoir that collects water flowing in a watercourse; and
(b) a lake through which water flows; and
(c)... a channel (but not a channel declared by regulation to be excluded from the ambit of this definition) into which the water of a watercourse has been diverted; and
(d) part of a watercourse;”
It is common ground that the channel the subject of this appeal is neither a river nor a creek. The question is whether it comes within the meaning of “other natural watercourse. That question requires regard to be had to what constitutes a natural watercourse at common law.
The principal feature of the appellant’s land is a small valley which runs from south to north. This valley is up to 30 metres deep. The sides are relatively steep and in the main are well covered with trees, bushes and grasses. The valley continues uphill for quite a distance from the southern boundary of the appellant’s land. The land immediately to the south of the appellant’s land has been used by the Burnside Council as a rubbish tip. It has been filled with unconsolidated rubbish over which a topping of soil was placed, thus raising the valley floor. The finished surface of the filling, although relatively flat, is not level. It slopes downwards in a north-westerly direction towards the subject land.
A large concrete pipe, some 300 metres long, has been placed beneath the tip. It runs from the southern end of the filling to its northern end which is close to the southern boundary of the appellant’s land. The pipe was intended to drain water from the upper reaches of the valley. However, part of the pipe is being crushed and it seems that the pipe is no longer functional.
On occasions, water flows along the floor of that part of the valley on the appellant’s land. The Environment Court found that there were signs of past water flows of some significance in a well-defined channel about one metre deep. However, a significant part of that old channel is now used as a path for walking and for trail bike riding so that not a great deal of its original surface remains. The court found that the channel was once a watercourse but that stream has lost its original flow since the filling of the valley to the south. The Environment Court also found that along or adjacent to much of the path and in some parts of the old channel, there was evidence of water erosion which in the greater part consisted of a small but clearly defined channel, varying in depth and width, which ran along the valley floor. Generally speaking, this smaller channel is approximately 200–300 millimetres wide and 100–150 millimetres deep, that is to say, some eight to twelve inches wide and about four to six inches deep. It was common ground that, if a watercourse exists, it is no longer the old channel but the small narrow channel just described.
The water which occasionally flows along this channel has several sources. They are
rainfall which runs off the sides of that part of the valley which forms part of the appellant’s land;
rainfall which runs off the sides of the valley which is south of the appellant’s land; and
water which is collected on Duncan Road and is discharged on to the subject land.
It is necessary to describe more fully the latter two sources.
Rainfall which runs off the sides of the valley south of the appellant’s land and which flows over the filled rubbish tip reaches the appellant’s land in a number of ways. As already mentioned, the pipe beneath the filling is no longer functional. The surface water in the valley flows at different points across the surface of the filling. The surface of the tip is generally graded so that water flows in a north-westerly direction across the tip towards a gate located on Duncan Road near the junction of Duncan Road and Dashwood Road. That flow is assisted by a channel which runs across the northern edge of the tip which prevents water flowing over the edge of the tip and down the graded face and directs it towards the gate. In this way, much of the water from the filled land flows into the kerbing and guttering on Duncan Road. Some of that water is diverted on to the appellant’s land by a small elongated mound called a “berm” and a small channel across the driveway just under the gate. However, the berm has not been well maintained and is not particularly effective. In the result, in times of a light water flow it will, in all probability, direct most of the water from the tip to the subject land. However, in times of a heavier flow most of the water will simply flow over the berm and into Duncan Road.
With the consent of the parties I had a view of the locality, including the rubbish tip, the relevant features of the junction of Duncan Road and Dashwood Road, the kerbing and guttering in Duncan Road and the relevant parts of the appellant’s land. The parties agree that some of the water from the filled land also flows beneath the northern end of a fence near the junction of Duncan Road and Dashwood Road. Three slabs of concrete have been placed on the footpath to direct the flow of water to the kerbing and guttering on the eastern side of Duncan Road, thus adding to the water in Duncan Road.
Water which is collected on Duncan Road comes from several sources. I have already referred to the water which comes from the rubbish tip. There are three other sources of water on Duncan Road. They are water falling on the road itself, water running on to the road from adjacent roads and water running from adjacent residential properties. The water which flows on to Duncan Road finds its way to the subject land at three points. The first two points are cuts in the kerbing and guttering on the eastern side of the road. They have been designed to permit water to flow into the valley. They are located at two points down the slope of Duncan Road. The water which enters the appellant’s land through those two cuttings flows down the slope of the western side of the valley to the floor of the valley within the appellant’s land. At the bottom of the slope of Duncan Road, and close to the north-west corner of the appellant’s land, there is a side entry pit adjacent to the road. It is in the form of a sump into which water flows from the road. The water is then taken by a pipe which empties into the western side of the valley at a point downstream from the appellant’s land. The water from this side entry pit can therefore be disregarded when considering whether a watercourse exists on the appellant’s land. In times of a light flow of water most of the water will be diverted on to the appellant’s land via the cuttings. In times of heavy flow over, the cuttings will take their share but a good part of the water will find its way into the side entry pit.
Mr Hayes QC, who appeared for the appellant, submitted that water flowing from Duncan Road on to the appellant’s land should be disregarded because the water is not surface water but water which has been collected and then discharged on to the land. There is considerable force in the argument but I do not have regard to it since, in the light of amendments to the Local Government Act 1934, there are questions as to the power of a municipal council to continue to be able to discharge storm water on to private land without compensation. These issues were not argued and it is preferable not to agitate them now. I decide the issue on other grounds.
The Environment Court found that any water flowing on to the appellant’s land would not remain in the channel for very long. It said:
“The flow of water along the valley floor depends, of course, upon the extent of the rainfall from which it comes. On many occasions, the amount and timing of the rainfall may be such that there is no flow. On other occasions the flow may be slight. On other occasions, for example, when a rainfall event of a 7–10 year return period occurs, substantial quantities of water are likely to flow along the valley floor. However, the evidence satisfies us that the topographic features of the valley and the nature and extent of the catchment it serves are such that water will not remain in the valley for very long. Leaving events of extreme magnitude aside (the impact of which was not the subject of any evidence before us), we are satisfied that, by large measure, this will apply notwithstanding the quantity of rainfall. Mr Douglas Rudd, a civil engineer who gave evidence on behalf of the appellants, estimated that, on average, there will be approximately twenty ‘events’ in each year when water will flow into the valley and long its floor but that the water is likely to drain away within an hour of rain ceasing to fall on the catchment. This estimate is supported, to some extent, by the rain fall statistics set out in exhibit A7, especially those relating to the monitoring station at Beaumont. Mr Phillip Read, a civil engineer with substantial experience in hydrology who gave evidence for the Board, said that he thought, on some occasions, it may take longer, perhaps 3–4 hours. None of the witnesses suggested that it would take either a day or days.”
The Environment Court found that, although water would flow along the channel only on about twenty occasions in each year, the channel was sufficiently defined to constitute a watercourse. It did so notwithstanding the very small size of this channel and the fact that water would not flow for more than four hours after a rain event. In order to determine whether the court has erred, it is necessary to examine what constitutes a watercourse at common law.
As Windeyer J noted in Gartner v Kidman (1962) 108 CLR 12 at 23, the common law rules applied in England in relation to watercourses are very old so that when applying those principles in Australia regard must be had to the fact that conditions of settlement, of climate and of geography in which the body of customary law developed are very different from those prevailing in many parts of Australia. But, as His Honour also noted, it is beyond doubt that these rules are an important part of the common law that Australia has inherited.
The meaning of the word “watercourse” will depend on the context in which it appears: Doe d Earl of Egremont v Williams (1848) 11 QB 688 at 700, 116 ER 631 at 636. A grant of a watercourse in law, especially when coupled with other words, may mean any one of three things. It may mean the easement or the right to the running of water, it may mean the channel pipe or drain which contains the water, and it may mean the land over which the water flows so the meaning must be shown by the context: Taylor v St Helens Corporation (1877) 6 Ch D 264 at 271, 272 per Jessel MR. Its more natural meaning is a channel in which water flows: Remfry v Surveyor-General of Natal [1896] AC 558 at 560. That is also its usual meaning in common usage. At common law, a watercourse will include a river or a creek as well as a smaller channel and the definition in s 3 of the Water Resources Act picks up each of those meanings. The definition is concerned with a natural watercourse whether modified or not. It is, therefore, unnecessary in this case to consider those instances in which an artificial channel might constitute a watercourse: cf. Gartner v Kidman (supra) at 33 to 34.
In Gartner v Kidman, Windeyer J with whom Dixon CJ agreed delivered the majority judgment. His Honour adopted a passage from the United States text Angell On Watercourses (5th edition, 1854) p 3 stating the common law meaning:
“A watercourse consists of bed, banks and water; yet the water need not flow continuously; and there are many watercourses which are sometimes dry. There is, however, a distinction to be taken between a regular flowing stream of water, which, at certain seasons is dried up, and those occasional bursts of water which in times of freshet or melting of ice and snow descend from the hills and inundate the country”.
His Honour added a quotation from Kerr on Injunctions (2nd edition) at 238 based mainly on Briscoe v Drought (1859) 11 Ir CL 250 at 271. The passage is quoted at greater length and applied in Vinnicombe v McGregor (1902) 28 VLR 144 at 173.
“As distinguished from water of a casual and temporary character, a watercourse is a flow of water usually flowing in a certain direction, and by a regular channel, having a bed, banks and sides, and possessing that unity of character by which the flow on one man’s land can be identified with that on the land of his neighbours. It is not requisite that a stream should flow continuously; it may be dry at times; but it must have a well-defined and substantial existence.”
Earlier in Lyons v Winter (1899) 25 VLR 464 at 465, Hood J had defined a watercourse in these terms:
“To constitute ... a watercourse, as a matter of law, there must be a stream of water flowing in a defined channel or between something in the nature of banks. The stream may be very small, and need not always run, nor need the banks be clearly or sharply defined.”
What is clear from all of these definitions is the requirement that there be a defined channel with something in the nature of banks. These definitions recognise that watercourses are sometimes dry but they do require that there be beds and banks. The definition also distinguishes between flows of water of short or occasional duration and a regular flowing stream. While it is recognised that a flowing watercourse may sometimes be dry, the converse does not apply so that a flow of water along a channel for a few hours is not a watercourse. That is clear from Angell’s reference to a freshet. It is also clear from the distinction drawn in the cases between a stream of water which may be dry at times and water of a casual or temporary character.
The effect of these decisions was summarised by Barwick CJ in Knezovic v Shire of Swan-Guildford (1968) 118 CLR 468 at 475 to 476:
“It seems to me, on an examination of such authorities as exist and of the work of recognized writers on this subject, as well as the decision of the Court, that it is settled that a watercourse consists of a stream with a bed, with banks, and water. That the flow of the water in the stream is intermittent or seasonal will not prevent what would otherwise be a watercourse from being accounted such: but though it is quite true that a watercourse may exist though its bed be dry for some periods, the watercourse, in my opinion, must exhibit features of continuity, permanence and unity, best seen, of course, in the existence of a defined bed and banks with flowing water. It must, in my opinion, essentially be a stream and be sharply distinguished from a mere drain, or a drainage depression in the contours of the land which serves to relieve upper land of excess water in times of major precipitation. It is not enough that the water, when it does flow, does so in what may be seen as a defined course or channel. In the case of a drainage depression, the water being drained off can be expected to flow in the lowest portion of the contours confined by the rising levels of the adjacent land: thus water can be seen when flowing to do so in what could be called a defined channel. If the seasonal rainfall is within an average tolerance in amount and timetable, the flow in the depression may well exhibit some regularity in the depth of water flowing in the contour depression and in the extent to which it spreads as it flows. If there is some such normality in the volume flowing, the impression of a defined channel with limiting margins will be enhanced. If, as I would expect to be the case, there is considerable variation in the rainfall and in the volume and velocity of the water flowing in the depression, the impression of a defined channel may be considerably less. But, in any event, the existence of such a defined channel will not make the drainage depression a watercourse nor the limiting margins of the water’s flow in a rainy season or period ‘banks’ of a stream. Thus, though water when it flows in such a period flows in what can be called a defined channel, such a drainage depression will lack banks and a bed in the proper sense of that term, that is to say, identifiable margins of a continuous and permanent stream which contribute to its unity whether or not water is in fact continuously flowing over the bed. The word ‘occasional’ in the quotation from Angell on Watercourses will here, in my opinion, embrace seasonable events, even if they occur with some regularity from season to season.”
This passage reiterates first, the requirements for a stream with bed and banks as distinguished from depressions and, secondly, that the flow be other than occasional.
The objects of the Water Resources Act include the protection from degradation of watercourses as well as lakes, surface water and underground water: s 6(2)(b)(ii). To that end, the Act empowers a relevant authority to direct the owner or occupier of land to take such action as is specified to maintain a watercourse and imposes duty on owners and occupiers of land to prevent damage to watercourses, a duty which is enforceable by the relevant authority: see s 14 and s 17 of the Act. I do not suggest that these are the only provisions relating to protection of watercourses. Section 7 of the Act also abolishes the common law rules in relation to taking naturally occurring water, which includes water in watercourses, and spells out a statutory regime controlling the taking of water from watercourses. Thus, riparian rights in watercourses are substantially modified, if not abolished and replaced, by the statutory regime: see, for example, s 7(4) and (5). It is unnecessary for present purposes to determine whether riparian rights have been entirely abolished. Ms Murphy, who appeared for the respondent, contended that, since the Act intends to protect watercourses, the definition of watercourse should receive a liberal construction. I do not think it appropriate to accord the definition either a liberal or narrow construction because a distinction has always been drawn between a watercourse and a fold or depression or contour in land to which surface water gravitates and along which it flows. That distinction must be preserved in order that the statutory powers are not used in an unauthorised way.
In Gartner v Kidman (at 27), Windeyer J suggested that the words “watercourse” and “natural watercourse” have a more exact meaning and lesser denotation when the rights of riparian owners are in question than the meaning they may have for other purposes of the law and in common parlance. For example, when rights and obligations relating to occasional flooding by surface water the subject of legal discussion the depressions which provide the natural course or outlet for such waters may be called watercourses. But, as Windeyer J noted in the same passage, the law treats such depressions very differently from watercourses that have the qualities of rivers and streams. As already noted, the Water Resources Act distinguishes between surface water and watercourses. When speaking of watercourses, the Act refers to those watercourses which have the qualities of rivers and streams, that is to say, watercourses which have readily identifiable beds and banks along which water flows, albeit that they may on occasions run dry. That is apparent from the fact that s 7 establishes a new statutory regime to permit water to be taken from a watercourse. This is entirely consistent with the common law definition. The observations of Windeyer J do not mean that mere depressions will necessarily constitute watercourses. The quotation by Windeyer J from Thompson v Andrews 165 NW 9 shows that the more extended meaning is applicable in a narrow context only. Further, as Barwick CJ noted in Knezovic at 476, that looser meaning has no place in a statutory setting where powers over and with respect to watercourses are created.
The Environment Court had regard to the decisions in Gartner v Kidman and Knezovic and to other decisions in which similar views were expressed. It concluded:
“A consideration of these authorities leads us to the conclusion that a channel through which water passes will constitute a watercourse at common law if it exhibits the following characteristics:-
·....... A defined channel having both bed and banks, which channel must exhibit ‘features of continuity, permanence and unity’ sufficient to distinguish it from a mere drain or drainage depression;
·....... Water in it should usually flow in the same direction;
·....... Water must flow through it although such flow may be seasonal or intermittent.”
The court went on to hold that the small channel on the appellant’s land constituted a watercourse. The last of the characteristics identified by the court is incorrect in that the use of the epithet “intermittent” suggests that a watercourse exists where flows of water are occasional or of short duration. The court’s conclusion demonstrates that that is the sense in which the court has used “intermittent”.
“The channel in question is very small. Its catchment is comparatively small, and its nature and extent is such that water only flows along it during and for a short period after rain events. Although not located in a desert or arid area, it is located in a part of Australia which is comparatively dry. The result of this is that the number of flows along it are not numerous and their duration may be short. However, this is consistent with the character, topography and climate of the area in which it lies. In the circumstances and taking note of the comments of Windeyer J, we have come to the conclusion that the flow down the channel, combined with the other characteristics to which we have referred, is sufficient to constitute it a watercourse at common law.”
The Environment Court has misconceived the effect of the authorities. As already mentioned, while a watercourse may sometimes be dry, the converse does not apply so that a flow of water along a channel for a few hours is not a watercourse. Shortly stated, while a watercourse may be dry from time to time, a watercourse is not a flow of water which exists for a few hours only. Given the Environment Court’s finding that water flows only occasionally in the channel on about 20 days in each year for a maximum of four hours, it cannot be a watercourse. The Environment Court has therefore erred in its definition of “watercourse” and in applying the common law definition to the facts.
I do not overlook the fact that one can point to a number of well-known watercourses in this country which are dry for long periods, sometimes for a year or more. Cooper Creek and the Todd River are obvious examples. But each is a watercourse which flows only after days of rain. Each will then flow for at least a period of days, if not longer. Such watercourses are to be contrasted with this channel which flows at most for a few hours only after what the Environment Court called a “rain event”. Furthermore, watercourses such as Cooper Creek have substantial banks. No one characteristic will necessarily determine whether a watercourse exists and it may sometimes be a question of facts and degree whether there is a sufficient flow of water, or a sufficiently regular flow of water, to constitute a watercourse. But a depression in a hills area which flows with water for a very short time after rain is not a watercourse.
There are further errors in the court’s reasoning. First, the dimensions of this channel show that it is very small. The court described it as “small and almost insignificant”, an expression it almost immediately repeated. In the passage just quoted from the court’s reasons, the court described the channel as “very small”. The photographs tendered in evidence and my observations on the view confirm that the channel is very small and almost insignificant. It would, in fact, be more correct to describe the channel as a small, shallow and irregular rut or furrow which runs along this small valley. The sides of this rut or furrow are only 100–150 millimetres high (four to six inches high). It would be a misuse of language to describe them as banks.
The channel (or rut or furrow as I prefer to call it) is, in truth, a drainage depression very similar to many such depressions in the Adelaide Hills – indeed in any hills area – to which surface water, after heavy rain, naturally gravitates and flows down along contours, folds or depressions to a lower point. That flow will often cause a small degree of erosion and thereby create a small shallow and irregular rut or furrow. This is one of many drainage depressions which feed into First Creek which runs along the foot of Waterfall Gully. In truth, the position is that the water which on occasions flows for a few hours through this small valley on the appellant’s land, generally speaking, finds its lowest point and in doing so has made a very slight channel along the valley floor which slopes down relatively steeply from south to north. Water flowing along a series of drainage depressions or folds may ultimately merge and form a watercourse. But this channel is not in any respect a watercourse. If this channel is a watercourse, then so are all such channels in contours or folds in the Adelaide Hills area down which water may run after heavy rain.
In reaching its conclusion, the Environment Court noted that the watercourse can be traced further up the valley upstream of the rubbish tip. In this respect it has also erred. There may be a watercourse south of the rubbish tip but the court has erred in having regard to what is south of the rubbish tip because the rubbish tip interferes to such a substantial extent with the valley floor and the flow of surface water.
For these reasons, the appeal will be allowed. The order of the Environment Court will be set aside and in its place there will be an order in the nature of a declaration that there is no watercourse on the appellant’s land.
96
1
0