[00:00:00] Bridget Scanlon: Welcome to the Water Resources Podcast. I am Bridget Scanlon. In this podcast, we discuss water challenges with leading experts, including topics on extreme climate events, over exploitation, and potential solutions towards more sustainable management. I would like to welcome Cath Moore. To the podcast, Cath is a Senior Groundwater Modeler at GNS Science in New Zealand, and she previously worked at CSIRO in Australia for many years and also worked in consulting.
Her research focuses on modeling of groundwater resources, including both quantity and quality aspects, with the detailed emphasis on uncertainty quantification. The models are generally geared toward decision makers and policy makers addressing many critical issues. Thank you so much, Cath, for joining me today.
[00:00:59] Cath Moore: Thank you for having me and letting us talk.
[00:01:02] Bridget Scanlon: So Cath, many of our listeners may be U. S. if there are any listeners. So I think it would be nice if you could maybe give a little bit of background about New Zealand, help them understand what the climate, the land use, and some basic things like that to help provide context for the groundwater systems.
[00:01:23] Cath Moore: Sure. So New Zealand's a very narrow country in the South Pacific. It's at its widest, I think it's 450 kilometers at its longest, about 1600 from North to South. It's a temperate climate, but there's quite a diverse range from in the south we have fjords like Sounds and then up in the north we have sandy beaches and this sort of subtropical climate.
So there's quite a lot of variation and and then they're a long little country. It's About 270, 000 square kilometers, and it's separated into two main islands, the North Island and the South Island. Most of New Zealand's population of about 5.5 million is in the North Island, and then about 25 percent is in the South Island.
The country lies on the edge of the Pacific plate and the Australian plate. So we have quite a lot of tectonic influence in our landscape and topography. We have, In the South Island, the Southern Alps, which dominate the island, and that they're being pushed up every year as the Pacific plate is subducted under the Australian plate.
And in the north island, the subduction zone is pretty much under the north island, and we have this big central volcanic plateau. And so our rivers flow from those topographic, volcanic and our features sort of dominate our river systems. So in the North Island, our major rivers tend to flow from that central plateau out to the coast, and the biggest ones to the west coast, the Waikato and the Whanganui and in the South Island we have a whole range of thundering, huge rivers coming off the Alps, many of them going to the east coast, but some, going to the west coast, the Clutha River, the Taieri River, there's many huge rivers in the South Island (Fig 1).
The elevation contrast is quite big. It is particularly in the South Island, up to almost four kilometers in elevation in the South Island and the Alps, and then towards the east of the South Island where the Alps come down to reasonably large extensive plains, and those plains in the South Island are often filled with alluvial gravels which make up our aquifer systems.
And those alluvial gravels are glacial deposits, so they're quite permeable and highly productive aquifer systems often.
Our rainfall is, varies from about 250 millimetres on the east side of the South Island up to about 4 metres on the outside. And in one particular town, in Milford Sound, we get six metres of rain a year. And in a narrow band on the Southern Alps, we get up to more than ten metres a year. So it's, it's quite variable across the region.
[00:04:43] Bridget Scanlon: Then there's a lot of agriculture, right, which would impact the water resources, right? I mean, everybody thinks about sheep in Australia, wool and mutton and things like that.
And, and it was interesting, Cath, I'd heard in the past that the impact of exporting mutton from New Zealand was better for global warming than raising sheep and stuff in the UK or something like that. I was looking at some other papers yesterday and sort of similar results. It seemed like the large transport doesn't really figure into it as much as other aspects of emissions and other things.
So, but you guys do have a lot of agriculture.
[00:05:24] Cath Moore: We do a lot of agriculture. It's up to, agriculture makes up to 70 percent of our exports, so it's a huge part of our export economy. And much of that is, is dairy more recently, but traditionally sheep based, and still is quite a lot of sheep-based agriculture.
The, the increase in dairy has led to a huge increase in irrigation, which has led to quite a big impact on our water supplies and our river flows and the health of our rivers. But in terms of water and agriculture, water is said to make up about 11 percent of our GDP, water related industry. So it's quite a big, in terms of agricultural industry.
So it's quite a big contribution, direct, direct contribution to our GDP from agriculture alone.
[00:06:20] Bridget Scanlon: Yeah. And you mentioned the climate up to four meters of rain in some areas, but then as low as 250 millimeters in the East side of the South Island. So that's a huge range. So when you hear, you see the Alps and the South Island, you think, oh man, they have buckets of water.
They have snow melt and these huge rivers that you can see on Google Earth, which are just crazy. But still, you can have water shortages. You can have drought and water shortages, and then maybe shifting from surface water use to groundwater use. And, and so even with such a large range in climate.
[00:07:01] Cath Moore: That's correct. So within the same country, we have floods and droughts at regular intervals. Most of the floods, but not all of the floods, would tend to be on the west coast, and most of the droughts tend to be on the east coast. However, there's plenty of places on the east coast that fluctuate between floods, droughts, and a recent example is Cyclone Gabriel that occurred a year, just over a year ago, and that affected quite a lot of the East Coast and, and this year many of those regions are in drought, so, and we've looked back at historical records and you can read in 1870 there was a big drought and then there was a big flood, so it's a long term problem.
[00:07:49] Bridget Scanlon: Right, so managing those extremes, a lot of times we hear people saying scarcity all the time, but really I think it's trying to manage the ups and downs and stabilize it for water resources. Another thing like the US, I mean, California and the West Coast, we see a lot of atmospheric rivers and, and it seems like they also figure prominently into rainfall in New Zealand, South Island, maybe more coming up from the South and looking at the arrows and some of these huge atmospheric rivers contribute a lot to rainfall.
[00:08:25] Cath Moore: That's correct. The atmospheric rivers contribute more than any other weather pattern to our total rainfall and pretty much dominate all of our extreme rainfall events. Not all of them, but, but most of them. And of course, climate change, that's meant to increase the frequency and duration and the strength of those atmospheric river events.
So yeah, I can't say too much about atmospheric rivers because they're not my area, but they directly impact our area of work. Right. They're a big deal. Yeah. 94 percent of extreme precipitation is atmospheric.
[00:09:12] Bridget Scanlon: Right. And so that makes it more challenging to manage water resources when you get so much intense rainfall over such a short time period.
And then you can have the summer periods maybe when it's dry or whatever. But your main area, and I know I pushed you to provide this context, so that's probably outside of your comfort zone a little bit, but thank you for setting the stage for understanding the water resources. And you mentioned that the river systems emanating from the plateau areas or the Alps and stuff, but groundwater is very important in New Zealand, and maybe you can describe a little bit about
the aquifer systems and I've been very impressed with the mapping and everything and really interesting and some of the more recent advanced analysis that you guys have been doing with airborne geophysics and stuff, which is fascinating.
[00:10:01] Cath Moore: Yeah. So the general aquifer systems in New Zealand. So we have about 200 aquifer systems, but only a few of those are enormous.
So we have a few really enormous aquifer systems, and the Canterbury Plains would be one of them. The gravels, it's an alluvial gravel system, which are our most productive aquifer systems. generally as gravel and sand. The Canterbury system is up to two kilometers deep and water flows are often the ground water velocities are often 100 meters a day. So very, very rapid and very productive aquifer systems. And we also see that in the North Island and, and in little pockets of the country. The other aquifer systems that we have are limestone karst systems, and they are very localized, but, but can be quite productive. And then volcanic systems, which in around the Auckland area and the North Island, they're, they're traditionally basalt. And then we also have big ignimbrite. So the planes is another one, which is quite productive as well. So we have, they're our main, our main systems, but with the alluvial gravels aquifers being the big, most productive ones. And that issue that you're talking about, the atmospheric rivers and then the lake of atmospheric rivers and the kind of flood drought issues, that, that has become a real issue for those alluvial gravel systems so that the flood management schemes that have sort of straightened out rivers to try and get the floodwaters out of the plains have, it seems that they're slightly reducing the aquifer recharge because we're not slowing down that water and letting it seep into the aquifer, so there's a real tension between flood protection and drought prevention work in those other ecosystems that we're grappling with at the moment.
[00:12:08] Bridget Scanlon: And so you mentioned a large aquifer system in the South Island and the the Canterbury plains and Christchurch, the capital is sort of in the middle of that. So two kilometers deep, that's amazing. But then how rapidly the water flows up to a hundred meters a day.
I mean, that's like a river, that's not like what we traditionally think of as an aquifer. So, I mean, our, our systems are most like rivers or karst systems, like the Edwards Aquifer, things like that, where the water just zips through. But so that, and we traditionally think of porous media aquifers having longer storage and providing a buffer against extremes and stuff.
But if the flow rates are so fast, then you're not getting those advantages maybe from your aquifer systems, from some of them anyway. That's correct.
[00:12:59] Cath Moore: So if they're deep, like the Canterbury Plains, they do still provide that buffer because there's such great storage. But in terms of, say, pathogen transport, it's a very poor buffer for pathogen transport when it's moving around.
[00:13:16] Bridget Scanlon: Right. So the ability to reduce contaminant transport to then, and particularly pathogens is low. And so then you're more vulnerable to contamination and you have to be very careful. And I guess you had an example of that in the North Island, right? Havelock? Absolutely.
[00:13:36] Cath Moore: that's correct. Yeah, we did. We had a big outbreak. It was like our version of the Walkerton outbreak where half the town became ill and, and one person died. And it was across the town and, and the country lot money that everything had to close down. You couldn't have any processing of food or anything at that time.
So yeah, that was in, in Hawke's Bay, but even, even in Canterbury, it's been well known for a long, a long time that in the areas where water isn't treated, not the city supplies, but the small domestic supplies, we just have a well in your property, the rate of campylobacter or other pathogen infections was disappointingly high.
Yeah. For a first world country, it was higher than it should have been. Right. Because of those very rapid changes, transport times and having a septic tank for your own and well, or other sources of contamination nearby to meet that. Right,
[00:14:50] Bridget Scanlon: I mean, you turned it into a learning experience then, and now you have a treatment plant and the transparency.
And so I think that goes a long way for the public to have confidence in the water quality because you've got a lot of treatment now, maybe UV and chlorination, maybe more than before. And so we, we come out stronger sometimes from these situations and we learn from them when we act on them. So I think that's a good thing, right?
[00:15:20] Cath Moore: Yeah, that's true. That's so true. And actually, it's in the town near Havelock North. The district council have set up this amazing water center and the have sort of this garden that's like the alluvial plains and the sort of landscape that with all of these educational sort of resources along the treatment plant. And they have videos of the system and, and people talking about it. And then instead of having the treatment system. And within a building that you can't see and all of the treatment system is glassed so you can look in and look at the people chlorinating your water and look at the people putting the UV treatment on the water.
So it's the, yeah, you're right. They've a great educational resource.
[00:16:14] Bridget Scanlon: Right. And to help the public understand what the issues are and fecal bacteria, E. coli or Campylobacter. I guess that case was from sheep and heavy rain and all of that sort of thing. So, so we can learn from these experiences.
It's unfortunate, but long term, then maybe we'll come out stronger. And you mentioned the Walkerton case in Canada, which was very famous. And I think they also improved their regulations and everything to manage. those types of systems. So that's on the water quality side of things, and you've done a nice job of describing some of these different types of aquifer systems.
And then more recently, you guys have been doing a lot of work on mapping these aquifer systems using airborne geophysics, the SkyTEM system, and maybe you can describe that a little bit, Cath.
[00:17:07] Cath Moore: Sure, so it's only in the last three or four years that we've started to use the SkyTerm technique. And so that's an airborne electromagnetic system inducing a current through the ground and the resistance that we get to that current tells us something about the type of rock.
So that has been flown in the Tonga planes and where we've done the most work so far, but also the rural mahana planes in Northland and in smaller sort of pockets of aquifer systems across the country. So we've got about six regions at the moment where we have data and the insights that we're getting are really cool, so they tell you a lot about the big structure of those aquifer systems. So in the, the Hirotonga plains, one of the things that I found exciting was that there was a major fault and everybody always knew that there was a major fault there, but the SkyTEM data showed just this massive 300 meter drop off in the basement at that area. And it also showed that, sort of provided another line of evidence for things that were being hypothesized, like the offshore flow boundary, like in some areas, the, the water age dating was indicating we may have more offshore flow here than say over here. And the SkyTEM data show in a really clear high permeability channel out towards the coast compared to other parts of the coast. So it tells us quite a lot about the basic structure of the system, but also about the relative hydraulic conductivity across that system. And it's exciting to say, but also important for water allocations.
[00:19:05] Bridget Scanlon: Right. Yeah. Looking at some of the reports, I was able to find a couple of them online. And so it seemed like, was the maximum depth that you could see maybe to 300 meters or deeper in, and they're nice cross sections and everything.
[00:19:23] Cath Moore: Yeah, it's about 300 meters as we go deeper, we can go deeper, but it becomes less and less accurate. So, the more shallow, the more accurate those profiles. The big thing now is how they use that information to influence decisions. So it's quite costly to fly. And so, people, especially now that we have a bit of a recession in our country, people want to know. How much money are we saving by spending this money on SkyTEM? So that's a study that we're doing now looking at the kind of economic benefits of having greater certainty in our water resources and how much we can allocate. What does that mean economically in terms of lost opportunity costs and sunk costs that you can't realize? So it's an exploration that we're doing at the moment.
And that requires making a direct connection between that extra information that you have about the conceptualization of the system and how that relates to water shortages or overallocation.
[00:20:39] Bridget Scanlon: Right. So, I mean, they do airborne geophysics. They have done airborne geophysics a lot in California, Rosemary Knight, and then Jeff Payne in our group has done a lot of airborne geophysics, contaminant studies, or seepages of saltwater, all sorts of things.
And I mean, before they do them, they figure out, well, if you did some ground-based studies, how much time it would take to do those ground-based surveys versus the time. So, when you're doing a fairly large survey, then it makes sense to do airborne and yours were helicopter, right? SkyTEM with a helicopter. Sometimes it's a fixed wing airplane. So, but you're being pushed then to show the return on investment. And so that can be challenging because sometimes we don't think about those things. We're not economists. But it provides a view into the subsurface that is incredible. I mean, when you go down a cave and you can see what the rocks are like around you and stuff like that's probably the only way we can see oftentimes what's going on in the subsurface or a downhole camera down a borehole or something like that.
But this 3D visualization and helping you understand and develop conceptual models of where the water is flowing, how much water is being lost from the river and recharging the groundwater are constraining that conceptual model. I think having that information is extremely valuable. Have you found that in some of your modeling exercises?
[00:22:09] Cath Moore: Definitely, yeah, it's fantastic. So another example was, there'd been a kind of traditional conceptualization of a system where there was a huge amount of recharge coming from the river into the aquifer that everyone acknowledged, and then there was this minor recharge zone, which was not that minor, it was still kind of significant, that was hypothesized.
And then the chemistry data was sort of showing, well, it looks more like limestone rather than coming from the river. So how can that be so? And then the SkyTEM data just showed it really well. So it showed that, there was, the aquifers was not really existing at that point. There was just a sliver of alluvial gravels on the top. And then, and then the true recharge zone, there was that great depth of growth. And we never saw that before the SkyTEM data. So yeah, just sometimes, sometimes it doesn't reveal something new, but removes all doubts. And sometimes it reveals something new, so yeah.
[00:23:11] Bridget Scanlon: Well, it's nice to, yeah, it's nice to confirm what you thought.
And it's a good check on your conceptual understanding of how the flow system is working, where the water is coming in and going out and all of that. And of course, you mentioned another way that you do that this forensic type of analysis is with water chemistry and isotopes and you guys are globally like number one with tritium age dating and stuff like that.
And I was wondering the name of your organization, GNS is part of that means nuclear, geological and nuclear. So was it from that background that you have such a good tritium lab?
[00:23:51] Cath Moore: So the Tritium lab is one of our nuclear sciences, I guess the sort of carbon and noble gases. And, but the, the Tritium lab in particular, so the, the International Atomic Energy Agency sends out these blind samples around the world. So people don't know that this is their blind sample, and they do that every so often. And because the New Zealand lab was developed for very low concentrations, because we didn't have that big background legacy of the bomb peak that's in the Northern Hemisphere. Here we have those fast moving systems, so it well and truly was never as big and it well and truly worked its way out of the system.
We needed to be able to analyze at lower and lower concentrations. So the methods that we have were developed and the tools that we have were developed for those very, very low concentrations, which were only possible because we're in the Southern Hemisphere. And then when they tested that, when the International Atomic Energy Agency tested it, we always come out the best.
[00:25:05] Bridget Scanlon: Are you guys using tritium-helium dating?
[00:25:07] Cath Moore: y Yeah. But, especially the Quantulus counter. That's the low, low level liquid scintillation spectrum. Again, I can't tell you that much about it. All I know is that it's very, very accurate to 0.02 tritium units is what I was reading.
[00:25:24] Bridget Scanlon: Yeah. And you would need it to use it as a tracer down there because when we were looking at bomb tracer, tritium data, bomb tracer. Tritium and chlorine 36 in soil zones, we could see that big peak from the bomb pulse in the 60s from the testing that occurred in the northern hemisphere in the 60s. But with your aquifer systems anyway, that would be gone long ago. And so it's good that you guys have this detailed tritium sampling. But you also use nitrogen and phosphorus tracers, right, from agricultural diffuse sources of agricultural input, bringing in nitrogen and phosphorus.
[00:26:06] Cath Moore: Yeah, we do. That's been really interesting.
So this is something that we've only just started to do, where we work with some people who were looking at this chemistry assisted base flow assessment method, they called it. And they developed this method for looking at the sources of nitrate and phosphorus entering surface water. That was the purpose of the NP development.
And they sort of defined three main end members of the NEPA mathematically and conceptually of nitrate and phosphorus sources going into surface waters. And so to analyze this, they would get time series of flow, time series of nitrate, time series of phosphorus in the river, usually at the bottom part of the catchment. And it was interesting to see conceptually, not spatially, but conceptually where those sources of nitrate and phosphorus were coming from. But we decided to turn that around and let's just look at this as a base flow separation tool and so use the nitrate and phosphorus as a tracer and see how much does groundwater actually contribute to surface water? Like we have these other base flow separation methods that indicate about 50 percent or 40 percent or round those numbers. And we found that across the whole flow duration curve from low flows to very high flows, groundwater is medium depth, shallow to medium depth and deep groundwater
made up to 90 percent of the flow within that river at about the 75th and 80th percentile going from low flow to high flow. And it was only once you go to those really high flows that surface water runoff started to dominate the contribution of water in the river. So that was surprising for us.
We're used to thinking of groundwater as something that definitely is happening at low flows, but not that it was happening at such high flows and not to that extent.
[00:28:09] Bridget Scanlon: So, yeah, that's amazing. Because I mean, you see in the South Island, you have water coming off the Alps and you think it'd be a surface water dominated system and you think it would just be all surface water flow and then during the dry summer time or when the low flow period when it's dry, maybe the groundwater contribution would be maybe 40 or 50 percent. So it must have been really crazy when you got these results then. So all the way up to the 90th percentile of flow, all the way up to these very high flows, groundwater is contributing like 80 percent or more to the surface water.
It's incredible.
[00:28:45] Cath Moore: It is. It is incredible. It's not in every catchment. So far we've looked at 56 catchments across the country, but it is incredible. We were shocked by it, even though we've looked at that kind of stuff all of our lives. I know. And we still don't completely understand where and how that's occurring.
So we're putting quite a lot of future effort into nailing down spatially, what does that mean, and verifying it with other methods. But yeah, the analysis so far tells us that. So it certainly has raised questions, even if the exact numbers aren't right. We no longer trust that 40, 45 percent estimate that was the traditional, this is what New Zealand rivers are about.
That means that the net and block recharge is a much bigger component of our water resources, I thought. At some point, most water goes through the ground in those catchments.
[00:29:45] Bridget Scanlon: Right. And it discharges to the surface water then in the plains.
[00:29:51] Cath Moore: And then discharges to the surface water in the plains, yeah.
And so you've written quite a lot about the importance of conjunctive management, and it really plays to that.
[00:30:03] Bridget Scanlon: Right, so are the regulations for surface water and groundwater separate? Do you, in New Zealand, do you combine, is there combined regulations? Do they acknowledge the connectivity between surface water and ground water?
[00:30:19] Cath Moore: They do, in part, yeah. So there's always stream depletion conditions on ground water consents. But, in all parts of the country. But in terms of allocation volumes, bulk allocation volumes. Occasional volumes, often that's treated separately. Not always, but often. So you get a situation where there are groundwater wells that are quite far from a river, so then there's no direct stream depletion impact.
But they are drawing down groundwater, and next season The recharge will be compromised by that reduction in storage, and so low flows will occur more frequently. So there's a kind of indirect effect that's not really taken account of.
[00:31:10] Bridget Scanlon: Right, and so just go back on some of what you were saying there, so you're talking about hydrograph separation.
So you have stream flow, the total stream flow, and then you want to figure out how much of it is groundwater discharging to the stream and how much of it is surface water. So the groundwater discharge to the stream is called base flow. And then with the chemistry data, then you were able to figure out that actually 80 percent or more was coming from the groundwater and that only 10 to 20 percent was from surface water, except at very high surface water flows.
And then, so one of the things that I often talk about, you're correct, is that we need to be managing these two things together, both surface water and groundwater, to optimize. And so seeing that connectivity, and I know in a study in a recent study that the US Geological survey did in Mississippi where they were pumping more groundwater than they were in the California Central Valley, but the, the GRACE satellite data and their more recent groundwater model indicated that that groundwater pump, which was just mostly capturing surface water and the groundwater storage, aquifer storage, was not really declining at a regionals.
So that connectivity is very important for managing water resources. So that's amazing. You think we know everything that we're done and dusted and then you come along and do this analysis and it just turns everything upside down. So I think people must have been very surprised, including yourselves.
Yeah, I don't think anyone actually believes it.
Right. Yeah. So, so your real forte, Cath, is groundwater modeling. And, and now you are bringing, you're really trying to work directly with decision makers and policy makers and working with stakeholders on applying these models. And that can be very challenging, but you are adapting your modeling approach then to be able to do this and developing faster modeling techniques and then talking to stakeholders and indigenous people.
Maybe you can describe some examples of that. That would be great. Thanks.
[00:33:27] Cath Moore: The first thing that everybody already knows is that all of our models are wrong. So starting from that point, we're trying to ensure that they're at least more useful in terms of supporting the decisions that need to be made. And that requires, It's partly designing the model for the decision that needs to be made, so kind of putting that decision as the compass point that we're designing that model for, instead of just pulling a model off the shelf and hoping that it will do, even if it's scale and construction structure might be quite inappropriate.
So that's one, one thing that we're trying to do, but that's obviously. People think, oh my, my goodness, it's cost a million to build this model. How can we build a different model for every decision? So we've put quite a lot of effort into making the build of models quite cheap by automating the basic build up models, where you just have a shape file in terms of what you're interested, what area you're interested in, and then you have all of these scripts that interrogate geological databases, water level databases, water pumping databases, stream network databases, so that you can spin up a basic model build without a lot of cost. And there's still skill there in terms of what that structure should be given the decision that you need to make and that's when the skill comes in. But an actual build can be done within a short time rather than months.
So that's one thing, and all of those are linked to the PEST history matching tools and uncertainty quantification tools that have been developed over time, mostly by John Doherty and more recently by Jeremy White in conjunction with John and separately from each other. So we've built these tools that are able to harness the other cutting edge tools in our industry to better support decision making.
We've also looked at how to bring in unusual or novel data, so the typical data that we would use in a groundwater model is we went history match to what groundwater levels, maybe flow losses along the stream, maybe stream flows. But we're trying to think of how, how would we use that SkyTEM data directly in the model?
How do we use those novel isotopes? Data directly in the model and the three ways that we've been exploring is
- either just using it as a conceptual information
- using that to inform what our boundary conditions and aspects of the system may be like,
- but we're also looking at directly pulling it into our models where we think it will improve the accuracy of our decisions.
That is significant so that for the SkyTEM for example, we've been doing that in two ways.
- One, we try and infer hydraulic properties from this. The resistivity distributions and then import those hydraulic properties into the model instead of the more geological model and with a lookup table of hydraulic properties.
- And the other way is to directly bring in the resistivity data itself and do a kind of joint inversion with the, so if you're finding, Model parameters that allow you to match that resistivity signal as well as ground levels and flows and so on. And similarly with the, say, the tritium data, we have some models that are just know it, know that that wood is younger, that wood is older and conceptually represent that.
But also we are directly history matching to that tritium which requires that the input signal, the tritium is bought into our models, including the bomb peak and, and then directly history matching to those tritium concentrations, which takes a lot more time. So that kind of looking at the trade offs in terms of different model design, different history matching regimes is something that we're looking at to look at what best informs the decisions that we need to make and where possible reduces the uncertainty of those predictions so that those models are most useful for those decisions that need to be made. So that's the modeling aspects. But what one part of models is the trust of those models and having communities actually trust what kind of information those models provide, spit out and also still having faith in the models despite the big uncertainty that they'll have even with all of this novel new information that we're using to try and constrain those stochastic distributions?
And so one thing is that we've done is look at co development of our models. And in particular, in a couple of catchments in the project that we're working on. And we've actually learned more than just you know how they gain trust we've we've also looked at the well, it's highlighted the importance of looking at different data even different from our SkyTEM and tracer data and in this case, the local community, in this case it was a local indigenous Maori community who have a lot of information in both the court system, court records, but also just in the oral histories, of how that system used to be and that's indicated that the surface water and groundwater system used to be highly connected and now they're highly disconnected and that caused us to go back and look at early soil maps to show where there were swamps and where there are no longer swamps and look at deforestation impacts and drainage impacts and so we've looked at quite different data than what we would have looked at in a water allocation model in that area.
So just because of the questions that they're asking, they sort of wanted us to see what would be required to restore the stream in their little area to what it was.
[00:40:02] Bridget Scanlon: yeah, this was the system in Hawkes Bay. Is that correct?
[00:40:08] Cath Moore: Yeah. Yeah. So it was traditionally a stream that used to flow all of the time, and then suddenly it was not flowing at all for eight months of the year.
And that was causing quite a lot of stress to the community that lived there, and also to the ecology of the eels and whatever lived in the stream. And, yeah, so they wanted to know what should they to to restore that. How could they restore that stream and also to try and figure out who was to blame, I suppose, which is something that we're looking at.
But more than who is to blame, maybe more informing what did that stream actually look like? realistically, given all we know, could we constrain those ideas? Because there'd be one community, one indigenous community, one area that would say stream never used to flow in the 50s, and another area that said it always used to flow.
So people's memories are fallible, and so we're trying to, But, but they're also rich sources of information, they said, so that we're trying to use that extra information, allowing for it's precision and allowing development models that inform decisions.
[00:41:27] Bridget Scanlon: So, so it seemed like you're doing a lot more than you would have been doing in a traditional modeling exercise or, and so you went from like pre European times and then you're doing snapshot models of different periods.
Yes. So pre-European, or maybe, I don’t know, back to pre Maori, I don't know. And then the deforestation episodes. And then you had dike irrigation and, and then you had maybe groundwater fed irrigation and groundwater depletion disconnecting the surface water from the groundwater. And so the surface water then is recharging the groundwater.
And so then the low flows and the streams are much lower than they were before. But so it's great to have a tool that you can test some of these hypotheses and these scenarios with, and then that the community could see the development of this tool and develop trust in it. And so some things maybe aren't physically possible, or you're finding with the data that you have. And then people say, well, I want to go back to baseline.
But what is the baseline that you want? Because I mean, in the case of the Central Valley, it was a wetland. They drained it to develop agriculture. Do you want to go back to it being a wetland? Not really. So it's nice that you can examine these different periods with such a rapid model that you're not making careers out of each one, and then give a synthesis of how the land has changed, how the water management has changed, and the land use has changed, and see the impacts on the water, and then try to explain to them how the current system with the low flows in the stream has evolved.
[00:43:15] Cath Moore: Correct. That's it, it is exciting. They also, I think just trying to design a model to look at the things that they care about. That’s been quite a strange but validating thing for them to provide a numerical tool like the councils have, like the board managers have, and now there's a tool that same tool that has been designed to look at what they find important.
So there's a really nice river restoration paper that says who, whoever's asking the questions of, of what, what it should be has, has sort of the most power and it's allowed them to be asking those questions for themselves, so.
[00:43:58] Bridget Scanlon: Yeah, yeah. And do you think that the expansion of dairy has anything to do with the more exploitation of groundwater in this actual system?
And so declining groundwater levels are not real. This is not a dairy area that you're looking at.
[00:44:14] Cath Moore: Yeah, this, this is not a dairy area. It's a horticultural and viticultural area. So that directly across this the road from this stream that's drying up. There's a whole lot of vineyards and they're quite a prestigious wine growing area, the Bridge Pa Triangle.
And so they're the nearest sort of culprits for drying up that stream. But it's actually not simple, as we were talking before.
[00:44:44] Bridget Scanlon: I think I was looking up the name of some of those wines. Is it like Paritua or something like that? Yeah,
[00:44:50] Cath Moore: Paritua. Yeah, and that's the name of the stream that goes down to Paritua.
[00:44:54] Bridget Scanlon: Okay, okay. I was thinking maybe you could send me a bottle sometime.
[00:45:01] Cath Moore: At our last field trip, we had a bottle. We wanted to try it.
[00:45:09] Bridget Scanlon: Well, it's really exciting how you're being able to bring in, incorporate all of these different types of data. And I get another aspect of the modeling is the uncertainty estimation.
And so with anything that you say, you're trying to put boundaries on. How certain you are about that or not. And so you're running, because you have a rapid model, then you can run hundreds or thousands of simulations then, and, and get a handle on the uncertainty. So I think that's also very important when you're talking to stakeholders and helping them with decisions and policies and stuff like that.
[00:45:49] Cath Moore: That's correct. That does really help. And also. So having more rapid sort of targeted models that run fast is, is good. And also there's new methods for analyzing uncertainty that are, more rapid as well. But because traditionally uncertainty analysis has been a very costly, time consuming process, so the ability to to analyze it more quickly has made it more widespread, which is a good thing and most people want to know what's the risk of this going wrong. When we first started doing uncertainty analysis, lots of people said, people don't want to know that our models are wrong. But then it turned out that everybody knew our models were wrong. And it actually helped maintain the trust by expressing that. And the new ways of analyzing uncertainty so that can be done in a more cost effective way have helped just put that information out there on a more regular basis and people are used to using it now in the same way they use uncertainty with weather forecasts, they can very much cope and I like to know that this is more uncertain than that.
[00:47:09] Bridget Scanlon: So much, you have given us a nice broad picture of what's going on in New Zealand with the climate and the topography and all of these things, and then the river systems and, and the groundwater aquifers. How do you think going forward, do you think you would move towards the, are you optimistic that they'll be able to manage the water resources better and and minimize contamination.
I mean, there are a lot of agriculture, nitrogen, phosphorus, algal blooms. You have almost, is it almost 4,000 lakes in New Zealand and nitrogen and phosphorus discharging to some of those and algal blooms and, and things like that. So it's a lot to manage and it's such a dynamic system. It seems like you don't get the benefit of a slow moving long residence time system. So it must be challenging. How do you think going forward? You've got a lot of new tools in your toolbox.
[00:48:07] Cath Moore: Yeah, I think they will help a lot. They are really helping a lot. So I do feel optimistic. I think we are going to a time where things may need to be more engineered.
Our problems are bigger and occurring more quickly. We need to be able to adapt in a more dynamic way. So we can't sort of look at something over three years and say, okay, it's going to be like this. Nothing is stationary anymore. So we need to be able to assimilate information and data more rapidly and say, okay, we were going there, but now we need to go here.
If we can't. continually have a viable drinking water supply or, or economy in this region. So I think we're setting ourselves up well with the tools that we're developing. Yeah, I have hope, but I think it's not an easy challenge, but I have hope.
[00:48:59] Bridget Scanlon: Right, right. So it's, it's an interesting place. You're on the ring of fire with the plate boundaries and the earthquakes and volcanoes and stuff like that. But you've got quite a lot of rain and beautiful countryside and a lot of agriculture and you have a population that's sort of similar to Ireland, but you have about four or five times the area of Ireland. So. A bit more real estate to work with, but it's fascinating to hear about how these systems work.
And now you're working more with the indigenous Maori people and, and trying to help them understand the system and how it works and, and hopefully develop a stronger relationship with them going forward. So thank you so much, Cath, for joining me on this podcast. And our guest today is Cath Moore. She's a senior modeler at GNS Sciences in New Zealand. And as you've heard, doing great work. So thank you so much.
[00:49:56] Cath Moore: Thank you.
