Fishing News >May 2008

And a river runs through it…but for how long?

Bruno David, Michel Dedual, Murray Neilson, David Kell - ,Department of Conservation

It’s a mild evening in mid summer and there’s a light, warm breeze blowing. We wander across the control structure at the Lake Te Anau outlet and peer down into the massive and mighty Waiau River. It’s such an imposing river, the sheer volume and depth creating dark mystical swirls and vortices that spin off menacingly downstream. From the top of the wall and though the dark, jade coloured water, we watch a large rainbow hanging in mid-water and to one side of the main current. The fish is actually facing downstream away from us. Occasionally a ‘window’ drifts over him and catching a quick glimpse through this portal we can clearly see that he is a male, the pronounced kype easily visible before the window melts away obscuring him again. Incredibly, and despite all the water moving around, the fish is virtually motionless, his large pectoral fins splayed out and body tilted slightly downward as the reverse currents buffet him, his tail barely moving. It’s a sight to behold, such a perfectly evolved design. I turn to my fishing companions and ask ‘why do you think that fish is there?’ We turn back to look at the fish, still wavering slightly. He is surrounded by deep water, complex currents, pressures and upwellings. No response. One thing is for sure though, the fish seems to know what it is doing.

We excitedly head off down the Kepler and are soon engulfed by the beech forest as we wind our way down the track before peeling off to our favourite evening rise spot. There’s little activity at present but we know that just before dusk large numbers of coloburiscus mayflies and hydropsychid caddis will emerge and all hell will break loose. Before long the action kicks off and fish suddenly start slashing away across the whole river channel, many fish smashing emerging insects in impossibly fast water. At one point looking across the channel I see five different fish explode together. The effort just to maintain position in the middle must be huge as the water is fanging past at increadible speed but the rate of prey (caddis) delivery is also very high so it must be worth their while?. I shake the thought off as I send a goddards sedge across the current. As the flyline rapidly arcs away from me and the fly skitters over the top, a rainbow violently smashes it on the way through. This goes on for about an hour before things start to quieten down, but there is still the odd flurry of activity as fish respond to the dying peaks of the hatch. There’s no doubt they are well tuned in to what’s going on. By 11:00 pm some of the fish have switched to a new strategy and in the moonlight countless rings and occasionally heads can be seen (and the ‘clomps’ of jaws heard) in a calm, deep backwater beside the main current. The fish of course are actively seeking and mopping up spent and freshly emerged adult caddis and mayflies in the dead water. It’s midnight and some fish are still at it but we’ve had our fill and we decide to head back.
As we wander back with glow worms illuminating off the earthen walls we begin our discussion about flow velocity and water depth. So I ask ‘what do you think is the optimum flow velocity and water depth for trout in the Waiau river?’Again there is no obvious response as in the space of two hours we saw fish use water flowing in excess of 1m/s and then mop up spent insects in water that was not flowing at all. Water depth didn’t seem to influence fish location as long as the fish was submerged. It’s a difficult question to answer…and more than likely it is probably not the right question to ask. Strangely though the question ‘what is the optimal flow and depth for trout?’ is frequently asked. In a basic ecological sense, fish and other aquatic organisms learn to exploit and track resources depending on the conditions before them. If the conditions are not suitable or resources cannot be found or exploited those organisms will either cease to persist or may be present but at reduced densities. Most fishermen have probably wondered why some rivers seem to support more fish than others or consistently produce larger fish. A large part of this has to do with food production rather than water velocity and depth. In other words the quantity (and quality) of food and how regularly and readily it is accessible to the fish (be it aquatic and or terrestrial in origin) often dictates where fish will be found. Often there are other factors involved such as the threat of predation (e.g humans hunting trout!) which will also dictate how organisms behave. Nevertheless in a very simplistic way, trout and salmon farms are a good example of how you can have high densities of large fish in concrete lined basins that provide little predation threat and no habitat other than clean water and suitable temperatures. On the other hand you will struggle to find fish in a pristine river that has low invertebrate production and has copped a lot of fishing pressure. From a human perspective it is easier to survive alone in a basic hut with the meat safe full than in a mansion with an empty fridge!

The point of these analogies and observations are to spur some thought on the sorts of things that are important not only for fish but aquatic organisms in general. Without a doubt the most essential requirement for all aquatic organisms is of course water. Today the pressure to use New Zealand’s streams and rivers for irrigation and hydro-electric power development is greater than ever. Central and local government are facing an increasingly important question: How much water can be removed from an aquatic ecosystem before significant negative impacts occur? In April 2007, the Ministry for the Environment commissioned a report called ‘Scientific input to a Proposed National Environmental Standard (NES) on Environmental Flows and Levels’. The purpose of the report is to set out a consistent national approach to selecting appropriate methods for setting environmental flows. The Minister for the Environment, Trevor Mallard, released a public discussion document calling for submissions on this proposal on the 29th March this year. The submission period will run for four months, ending on 31 July 2008. The NES will prescribe the methods to be used to set environmental flows and levels in potentially any New Zealand stream river, lake or wetland irrespective of its status (since water has no formal protection). Once the NES is implemented, these methods will apply for the next 10 years. Several methods have been developed internationally in an attempt to address this question. One of the most commonly used is the In-stream Flow Incremental Methodology (IFIM) developed during the 1970s. An important building block of IFIM is the physical habitat simulation modelling (PHABSIM or RHYHABSIM in New Zealand) that estimates mathematically how the physical habitat for target aquatic organisms is affected by flow scenarios. This methodology assumes that the quantity of physical habitat (water velocity and depth) will limit the organisms abundance. This required assumption is rarely if ever tested even though intuitively it seems overly simplistic as highlighted in the analogy at the beginning of this article.

The estimated numbers generated by physical habitat modelling constitute the “currency” that is often used to determine the maximum amount of water that can be taken out of the river and still maintain riverine values or functions. Having such a currency is very useful for decision makers and as a consequence physical habitat modelling, rightly or wrongly, is an appealing method. Unfortunately, assessing whether the new flow regime does infact maintain river values or functions is almost never undertaken. For instance, historically the Waiau downstream of Lake Manapouri naturally had a mean flow of more than 350 cumecs. Physical habitat modelling indicates that to maintain a viable trout fishery 16cumecs (4% of the original flow) is sufficient, but is this sufficient to meet other values and functions further down the catchment that were formerly dependant on the higher mean flows (e.g wetland inundation)?

While this type of ‘designer’ flow approach is intended to address some of the needs of a particular species, often there are multiple values that need to be considered. To illustrate this point take a look at this simple! food web below.

In a Physical Habitat Simulation only the bold elements highlighted above are considered for a target species (e.g brown trout). In contrast, holistic methods (such as the Range of Variability Approach which we will discuss later) consider all elements of this diagram and all species. It’s pretty obvious that there are a whole lot of interactions going on here and each living organism has its own requirements at different stages throughout its life. What’s even harder to comprehend is that the strength of these interactions changes through time and will be influenced by many other environmental variables such as flows, temperature etc. For instance just one high flow event would be sufficient to change many of these interactions. For some organisms a flood event may translate into a positive effect for others it may have a negative effect. In other words these interactions are not static. Nevertheless having a sound understanding of all these interactions can provide very useful insights into why one particular river tends to support more or larger trout than others. It is important to recognise that physical habitat alone (while important) is only part of the answer. For instance, let’s consider the water chemistry box from the diagram above. Irrespective of the suitability of flow or physical habitat available, if a stream is naturally acidic (e.g some streams that drain from catchments high in peat) trout are less likely to do well there. The fact that the water is naturally acidic will also determine the types of invertebrates that will be present and subsequently other organisms that may or may not be tolerant of such conditions. This is not to say that flows are not important here just because trout are not present however, as the organisms that live or persist in these types of streams are still influenced by (or adapted to) the local flow regime.
The main point though is that when setting flows it is important to consider more than just one organism, value or parameter (e.g physical habitat).

In a conservation sense New Zealand’s dynamic landscape and longitudinal position is reflected in the wide diversity of freshwater ecosystem types found throughout the country. River catchments are highly variable both in terms of the rain they receive from north to south and east to west as well their underlying geology. Systems can vary from highly stable spring fed streams with stable flows to highly unstable braided systems that may naturally run dry for some periods of the year. This diversity of ecosystem types (and hydrological conditions) has given rise to a unique assortment of aquatic flora and fauna, many of which have evolved specific adaptations enabling them to occupy particular water-bodies. Good examples of this are the various mudfish species which have evolved to tolerate natural periods of drying in particular wetlands or the ability for some of our non-migratory fish species to burrow down into the gravels of braided river beds when flows recede. Other life history parameters of these endemic organisms (e.g. timing of reproduction, feeding behaviour) are often synchronised or closely tied to events in the natural flow regime. For example floods trigger or facilitate spawning and egg hatching in many native fish species including many of the whitebait species and also trigger their upstream/downstream migratory movements between freshwater and the sea.

While New Zealand boasts a unique assortment of aquatic fauna, our rivers, lakes and wetlands are also used for a diverse array of commercial (e.g hydro, irrigation, domestic water) and recreational activities (e.g whitewater rafting, jet-boating, water skiing, fishing, hunting, and whitebaiting). It is not surprising then that the debate around flow allocation becomes increasingly complex at places where multiple values/activities exist..

For instance let’s consider a realistic albeit hypothetical situation. Let’s say that a particular unmodified stream is valued as a world class rainbow trout fishery but in the lower reaches of the catchment there are naturally inundated wetlands that provide excellent duck shooting and white-baiting opportunities. In the headwaters, permission to build a hydro dam has been given. If the objective of setting flows in this catchment is to provide for all these recreational pastimes how do you go about deciding what the flows should be? There are many things to consider…like…what is the optimum flow for maximising invertebrate production and rainbow trout biomass? How often would the lower wetlands need to be inundated to still provide good duck shooting? When and how frequently should the inundation occur to support vegetation, amongst which whitebait spawning would occur? Do we even know the answers to these questions?. In most cases, particularly when trying to provide for multiple uses and objectives the information may not exist. In complex cases, it is important that scientists acknowledge this. So when you don’t know the answer, what do you do? One commonsense option would be to mimic as close as possible what the natural flow regime was like prior to the dam going in. You then accept that the further you depart from the natural flow regime, the further the risk that the values (rainbow trout, duck shooting, white-baiting) will be effected.

Holistic methods tend to operate on this premise. That is, they recognise how complicated the problem is and as a result tend to be more conservative (in terms of how much water is available for out-of-stream use) than other methods. Holistic methods also recognise that mimicking the natural flow regime (the quantity, timing and variability of a river’s flow), while still allowing the abstraction of some water, is likely to result in better long-term sustainability of these resources.

Holistic methods rely on the use of hydrologic parameters to characterise the essential components of the natural flow regime. That is,. the magnitude, duration, frequency, timing and rate of change of hydrological conditions (data on these parameters are routinely collected by regional councils, as part of their normal monitoring activities). Some holistic methods can calculate how much a proposed flow regime is departing from its natural state. These calculations, like IFIM also provide a “currency” that can be used to negotiate a flow regime. However, because results tend to be more conservative, the use of such methods may not be supported by some commercial sectors. Nevertheless, holistic methods based on a natural flow regime are likely to be a very useful method for applying to rivers which possess highly natural or multiple recreational values as they are likely to leave more water in stream to support natural conditions.

To give an example of their applicability, a recent exercise was run on the hydrological data from the Wairau River in Marlborough. The Wairau River recently underwent a hearing for Trust Power Ltd to divert river flows into a man-made channel run through a series of five turbines for approximately 50 km of river length (both DOC and Fish and Game opposed this application). The project would generate approximately 70 MW of power, approximately 50% of the anticipated national annual increase in power demand in New Zealand. There are multiple high conservation values for the Wairau river, such as threatened braided-river bird species as well as threatened native freshwater fish species and it’s not a bad trout fishing river either. Furthermore native riparian cover in the catchment is also high adding further value to the catchment.

Trust Power’s experts, using the PHABSIM model (an output from the traditional IFIM in-stream method), proposed a minimum flow of 10 cumecs (allowing the diversion of 35 cumecs). This would result in a mid-summer median flow of 17 cumecs (from 30), and a mean annual low flow (7 day MALF) of 9 cumecs (from 16).

In contrast we conducted analyses of historical river flows in the Wairau River over a 25 year record using a holistic method. We identified three main components of the flow regime that are expected to affect the ecology of the river and specifically its relationships with important wildlife species found in this river (e.g. braided river birds). The model is based on predicting a natural range in which the river has historically operated, and identifies the low and high magnitudes of departure of a proposed flow scenario. Holistic analyses indicated that the minimum flow should not fall below 15 cumecs (10cumecs by PHABSIM), average median summer flows should not fall below 22 cumecs (16 cumecs by PHABSIM) and average annual flood frequency should be >14 floods yr-1 (12 by PHABSIM scenario).

Both types of methods allow a significant water harvest of flow from the Wairau River. However, based on the adaptation of communities to the natural flow regimes of the river, and the dependence of both braided river bird and fish populations on the productivity of river invertebrate communities, the holisitc method would recommend both a minimum flow and a median flow that are 50% higher than those from the PHABSIM predictions. In addition, outputs from the Trust Power PHABSIM scenario have suggested that the variability of flows in the river will be unaffected by the harvesting scheme. The holistic model outputs do not agree with this analysis, showing significantly lower flood frequency, and a much greater extension of the period at which the river remains at its low flow. For example, the PHABSIM scenario will extend the period at which the river is at the proposed minimum flow (10 cumecs) from at present, 1% of the time, to 41% of the time. This period would largely occur through the summer period where the potential for elevated stream temperatures to affect fish and invertebrate populations is greater.

While this exercise concentrated on the requirements of the native species involved, we think that most anglers, whitebaiters and duck hunters will recognise the potential benefits of such an approach for the species they harvest and thus for the long-term future of their sport.

However, currently, holistic methods do not appear in the National Environmental Standards (NES) document for setting flows, despite justifiable reasons for their inclusion. It is critical that such methods are included in the NES document particularly if we are to protect waterways with high natural values rather than relying on other less precautionary methods. It is important to recognise that there is no silver bullet answer to setting flows, but one thing is certain, it makes sense to have more tools in the toolbox than not enough. Why dig a hole with a rake when you could use a shovel? The effective use of holistic methods has proven successful in many situations and numerous countries overseas, so why is New Zealand so averse to their inclusion?

Many of New Zealand’s most celebrated pristine wilderness trout fisheries exist under a natural unimpaired flow regime, indeed the natural flow regimes of New Zealand’s rivers and streams are probably amongst the major reasons why trout have done so well in this country. However, one of the problems that we face in this debate, is that some scientists assert that we do know enough about the requirements of our fish species (both native and introduced) and ecosystems to be confident that the use of traditional habitat modelling approaches will be sufficient to ensure their long-term sustainability.

So, what is to be done? Regional Councils currently grant consents up to a maximum term of 35 years, and although the NES will set the in-stream flow methods to be used for a lesser period, 10 years, that is still an appreciable amount of time. Our recommendation is to include the use of holistic methods in the NES document now and to provide for the inclusion of other new methods under development, However, there appears to be opposition to this approach. One weak argument is that these methods have not really been tried here. In our opinion, new holistic methods need to be applied to water development projects, along with appropriate monitoring regimes, to assess their effectiveness. They do, in effect, need to be considered as long-term living experiments. After all, this would be no different from the way traditional in-stream modelling techniques have already been applied – except, of course, that effective monitoring regimes would be in place, right from the start, rather than attempts being made to assess their effectiveness some 20 years later!

So what about it, readers?? In many respects it’s over to you and the submissions you will (hopefully) make to MFE by 31 July this year. Should we take advantage of the new holistic methods (more shovels to add to the tool-shed) which are available, and which will leave more water and more natural flows in our streams and rivers, or should we trust in traditional in-stream methods to look after our future?

Back to Reel Life