Braided River
Update: Court of Appeal affirms High Court’s decision

What is the “bed” of a braided river?

The High Court decision Dewhirst Land Company Ltd v Canterbury Regional Council [2018] NZHC 3338 concerned the correct interpretation of the term “bed” in relation to a river under s2 of the Resource Management Act (RMA). Determining the river bed was “more contentious” in this case because it concerned the Selwyn River, which was a braided river, therefore identifying its banks was more difficult.


In breach of its resource consent, Dewhirst had cleared vegetation up to an existing formed bank, which it had considered to be the edge of the Selwyn River bed, and had created a gravel bund along the line of that bank. It plead guilty to five charges under s 13 and s 14 of the RMA but contested the factual finding that the entire area in question was within the “bed” of the River. A reduction in the area that was considered to be the “bed” would have the effect of reducing the penalty it faced for illegal works in that area.

Correct test for determining the extent of the river “bed”

The High Court considered whether the District Court failed to identify the correct test for determining the extent of the River bed. Following the Council’s approach, the District Court determined the River bed by ascertaining the River’s waters at its “fullest flow” not including land covered by reason of the banks being overtopped. The Council submitted that the banks could not be determined first because of the number of topographic features that could be identified as banks in the area. The Court noted its legal definition of the bed concurred with the Council’s river engineering expert’s determination of the location of the banks.

The High Court held that the District Court applied the wrong legal test to the issue of river bed because it did not give sufficient weight to the observable location of the River’s banks, and appeared to overlook the difference between a river bed and its adjoining flood plain.

The High Court said the correct approach was to determine the river bed through statutory interpretation. Section 2 of the RMA defines the term “bed” in relation to a river as, “the space which the waters of the river cover at its fullest flow without overtopping its banks”. While the term “banks” was not defined in the RMA the Court determined that it was generally defined as “the land alongside or sloping down to a river or lake”. The RMA gave no direction on what “fullest flow” meant in this context, and the Court did not accept its literal meaning as it would extend the “banks” of a river to the flood plains following significant flood events.

Consequently the Court held that one could not apply the s2 definition of “bed” without first identifying where the relevant banks were by way of inspection or otherwise. Once the banks were identified, the correct test for identifying the “bed” was the long accepted “bank to bank” test outlined in Kingdon v Hutt River Board (1905) 25 NZLR 145 (SC) that, “the bed of a river comprises those lands covered by water during the ordinary rainy season but contained within the banks of the river and extending from bank to bank”.

While the Council submitted that a limited definition of a river bed would result in it being disadvantaged jurisdictionally from protecting river environments, the Court considered this concern to be unfounded given other routes available to local authorities to govern general river environments.

Taking into account expert evidence of the location of the river “bed”

The High Court also considered whether the District Court took into account an irrelevant matter when considering the appropriate flow for determining the extent of the river bed. The Council submitted this was not a question of law but the High Court accepted the question because it concerned the relevant considerations that could be taken into account when determining the proper legal test. The District Court had taken into account the advice of Council officers that “a 50-year flow period was an appropriate starting point for indicating what the boundary of the bed should be”. The High Court found that taking into account Council’s evidence was an irrelevant consideration, therefore the legal test undertaken was awry.

Environment Canterbury (ECan) has said that it will appeal the High Court decision to bring clarity to the legal definition of a “riverbed”, which will have broad effects for river users and in particular ECan’s efforts to protect braided rivers.