By Daniel Collins
Last week, the Land and Water Forum released its third and final report on water management in New Zealand. It is a substantial piece of collaborative work with 67 recommendations. Number 29 is that allocation limits be set by taking into account “any flow and water level fluctuations caused by seasonal or other climate variations”. While this primarily refers to natural variability, such as the Interdecadal Pacific Oscillation, it’s also important to consider climate change. And along the same lines, last year’s National Policy Statement for Freshwater Management stated the need to account for the “foreseeable impacts” of climate change.
This is an important issue, as climate change is expected to bring about a raft of changes to New Zealand’s freshwaters (more details on that soon). Among these changes are reductions or increases in the amount of water available for use. Also importantly, climate change makes assessments of future water resources less certain.
So how should resource managers set allocation limits for long-term consents in the context of climate change, accounting for both a change in supply and an increase in uncertainty?
To explore this issue, I propose the following method. It is still in its formative stages, so feedback is welcome.
Let’s start by considering a hypothetical New Zealand river. Its allocation limit is currently set at 40 m3/s. And let’s put aside any complications like priority rights.
Now suppose that results from a climate change impact assessment indicate that allocable flow will reduce by 7% by 2050. This is a middle-of-the road projection, associated with a moderate greenhouse gas emissions scenario and using the median result from 12 global climate models (GCMs). But if you account for the uncertainty of the scenarios, the GCMs and the hydrological models that convert climate changes into runoff changes, then the impact could be anywhere between a 4% and a 12% reduction. That is, it is almost certain that the allocable flow will drop by 4%, it will likely drop by a further 3%, and it might drop by another 5% again.
To set a conservative new allocation limit, first reduce the existing limit by 12% to 35.2 m3/s You can be pretty sure that this water will still be available in 2050 and so you should have no qualms about allocating it for the longest possible duration under the RMA of 35 years (2012 + 35 = 2047). This gives water users the confidence to invest in long-term infrastructure, and it will mean that over-allocation is unlikely to occur.
Second, take an additional 5% of the water (2 m3/s) and allocate this for a shorter period of time, say 10-15 years. It is likely that this water will also be available in the future, but we can’t be as sure. For those water users who are willing to accept the higher risk, they should be allowed to, thus making better use of the available resource.
(If the climate change projections were for an increase in water availability, the same method applies, but the numbers are shifted in the opposite direction.)
Every few decades or so, the long-term allocation limit is re-assessed and changed as needed. Every 10-15 years or so, the short-term allocation limit is also re-assessed and changed as needed.
This allocation scheme meets users’ needs for long-term consents for most of the water (the “certain” water), giving them the confidence to invest in long-term infrastructure, while also allowing them to seek additional water if they are not too risk averse. The scheme also allows the limits to be managed adaptively as new information comes to light – new data on water availability or better climate change projections. And finally, it means that the social, cultural and environmental limits are met whatever happens with climate change, and that the detrimental effects of over-allocation are avoided.
In terms of climate change adaptation, the scheme ticks the boxes of adaptive management and balanced risk-based assessment, and is robust to uncertainties in climate change. As far as I can tell, it also meets the different stakeholders’ needs while accounting for the realities of climate change (that is, change plus uncertainty).
But what do you think? Your feedback would be appreciated in refining this time-dependent allocation scheme.