Water management options for urban and rural Australia
Leveraging Australia’s water information
Tuesday 2 March 2010
Rob Vertessy has a PhD in fluvial geomorphology from the Australian National University. He spent the first 20 years of his career as a researcher in CSIRO specialising in catchment hydrology, and is widely published in that field.
Rob served as Chief Executive of the CRC for Catchment Hydrology (2002–04) and Chief of CSIRO Land and Water (2004–07).
In late 2006, Rob was seconded to the Department of Prime Minister and Cabinet to advise on the establishment of a national water information strategy. In mid 2007, Rob joined the Australian Bureau of Meteorology as Deputy Director (Water), to establish a new water information division, arising out of the Australian Government’s Water for the Future program.
Rob is currently a Board member of eWater CRC Pty Ltd.
Leveraging Australia’s water information
Rob Vertessy: I thank you all for coming this evening, and I thank the Academy for its very kind invitation.
[SLIDE: Our water resources are changing]
Our water resources are changing. This indeed is the imperative to attain much better optics on water resources than we have had in the past.
How are our water resources changing?
[SLIDE: Australian Temperatures]
First, let's look at some of the temperature or climatological signals. These aren't going to be news to many of you. I am sure you are all keen watchers of the weather and climate. We have gone through some pretty systematic increases in temperature over the last five decades. In fact, each decade has been warmer than the last for the last five in a row. On average there is close to about a degree overall temperature rise in Australia. We have had quite a run of records, as you know, over the last decade or so. So things are hottin' up.
[SLIDE: Australian Temperatures (map)]
The average temperature for Australia belies significant local temperature extremes – sometimes as much as 1.5 to 2 degrees over the last 50 years. There are pockets of Australia that have really warmed up, unfortunately in many of the areas where some of our valuable water resources are.
[SLIDE: Australian Rainfall]
With rainfall, when we look at an average pattern across the nation there isn't such an obvious trend. If anything, there is a slight increase in rainfall across the country, when averaged over the whole continent.
However, that belies a very serious rainfall deficiency that is occurring in virtually all of the populated parts of Australia, including the Murray Darling Basin, our food bowl.
[SLIDE: Australian Rainfall (map)]
The north‑west of Australia has got quite a bit wetter over the last few decades, which is why we don't see an overall average decrease in rainfall.
[SLIDE: Rainfall elasticity of streamflow]
This is particularly bad news for water resources because Australian catchments, particularly those in the south‑east and south‑west, are notorious for their degree of sensitivity to subtle changes in rainfall.
In fact, a 10 per cent decrease in rainfall tends to manifest itself as up to a 20 to 35 per cent decrease in run‑off. This is why our water resources are feeling it.
[SLIDE: Yearly inflow to Perth Dams including Southern sources]
This is best exemplified by looking at the inflows into some of the storages for our major cities. This slide shows Perth’s water supply system. Plotted on the slide is a time series of annual inflows dating from 1911 to more or less the present. On the Y axis you can see the total annual inflow into the system. There has been a steady decrease since about the mid 1970s, when a climate change set in in that part of the world.
Those catchments receive 70 per cent less run‑off today than they did in the period prior to 1974. That is what has put Perth into overdrive in terms of finding alternate water supplies such as desalination and enhanced groundwater use.
[SLIDE: Total annual water flowing into Melbourne's main water supply storage reservoirs]
Unfortunately, that change seems to have swept east and now Melbourne is suffering a similar experience. The magnitude of the response is not so grave as yet. Nonetheless, there is a very significant drop in the inflows into the Melbourne water supply system – probably around 40 per cent less than Melbourne was accustomed to prior to the mid 1990s. These are very challenging effects for water resource managers, and also for the general public who must bear the brunt of water restrictions, et cetera.
[SLIDE: How low did we go?]
This slide gives an indication of how low we have actually gone. It plots the minimum storage levels in the major supply systems in the Murray Darling Basin (MDB), in Brisbane, Perth, Melbourne and Sydney. All of these systems have gone perilously close to bottoming out. Sydney the least so, at about 35 per cent; Melbourne has got down to about 26 or 27 per cent; Perth down to 20 per cent; Brisbane 18 per cent; and the MDB about 19 per cent at its worst case. The little red dots indicate where they have rebounded up to now. So some of those systems are well and truly out of the woods for the time being, but other systems, like the MDB and Melbourne, are still on a bit of a knife edge.
This is one of the reasons why there is such a radical water reform program underway in Australia today to augment water supplies, and also to rationalise the way in which that scarce water resource is divvied up against competing users.
[SLIDE: The big eight water scarcity factors]
The worst of it though is that it isn't just a bad run of climate. There are many fundamental systemic changes that have been going on, some of which will only manifest themselves as decreased streamflow decades from now.
Growing urban demand is probably the most significant one. The capital cities of Australia grew at something like 2 per cent last year. An extra 250,000 people shifted to cities and required water. You would be well aware of the Treasury projections for very significant population growth, going out to 35 million by 2050. That will exert very significant demands on urban water supplies.
We have been through a period of significant over‑allocation of water in rural systems for irrigation. You may have seen the graphs that Don Blackmore showed earlier in this lecture series, showing the proliferation of dam development and the proliferation of allocations of water to irrigators. That has taken a great dip in the last few years as allocations in some systems have declined to zero. But, nonetheless, there are many expectant users out there with significant investments in infrastructure who do expect to have water allocated to them.
Plantations have expanded and have also exerted downward pressure on water resources, in that they transpire more water and, therefore, intercept water from reaching streams.
Bushfire recovery impacts are an effect that have probably been positive in the last few years, but will shortly become negative as the regrowth eucalyptus vegetation establishes itself and the evapotranspiration increases. Unfortunately the areas that have been burnt tend to be in highly productive water supply catchments, where the effects of fire are likely to be maximal.
So in the decades ahead we can expect reduced streamflows in a lot of the productive water producing parts of the Murray Darling Basin.
The environmental flows imperative. We have really only come to our senses in the last decade about the criticality of reserving water in our rivers for aquatic ecologic assets. Governments have finally woken up to this and are now allocating significant energy and resources and, indeed, water to those environmental assets for the future. But that means less for the consumptive pool.
Small farm dams are seemingly innocuous but there are so many of them. Each debits streaming flows, to some extent, by anything between 1.5 and 3 megalitres per year for every megalitre of dam storage.
There are over a million small farm dams in the Murray Darling Basin alone. Add these up and it actually makes a profound contribution to a changed water balance.
Finally, uncapped groundwater extraction. One of the perverse consequences of imposing a cap on surface diversions in the Murray Darling Basin has been a proliferation of groundwater bores. In fact, it has not been confined only to rural areas. It is also actually becoming increasingly common in urban areas, like Perth and Melbourne, where people who are constrained by water restrictions have been increasingly sinking bores.
Groundwater accounts for about 25 to 30 per cent of total water use in Australia now. It has increased markedly over the last decade or so.
[SLIDE: The need for water information]
Let me turn now to the need for water information. I would hope it is obvious, after that introduction, but I do want to go through this because I am sure the question on everyone's mind is: don't we have enough of the stuff already? Aren't water managers on top of it?
[SLIDE: Diverse data are needed to understand our water resources]
First of all, let me define what I mean by water information.
The top row of this slide illustrates that there are quite a number of different physical hydrology data parameters. These can be climate, streamflow, groundwater levels and pressure, water held in storages and water quality parameters. Things like sediment load, salt load, nutrient content, phosphorus, nitrogen and so on are physico‑chemical parameters, which are important to the understanding of the physical hydrology of the continent.
The second row shows what you might class administrative type data, such as water entitlements. There are tens of thousands of different water entitlements and hundreds of different water entitlement products out there.
There are probably well over 100 water restriction-setting entities. Water use, some of it metered, much of it estimated. And finally, a burgeoning water trading market. Last year the value of the water trading market was $2.8 billion. Some 4000 megalitres of water were traded on the market last year. It's virtually been doubling in value year on year for the last four years. So water trading has taken off in a very big way.
On the bottom row is what you might call a whole lot of contextual or inferential information which is vital to actually understanding water balances in the country. Much of this is derived from remote sensing and some mapping programs, designed to understand the distribution of vegetation or landcover, the distribution of farm dams, and soil properties and their hydrologic properties, et cetera. All critical pieces of the puzzle for understanding the water balance of Australia.
[SLIDE: Good water information is the key]
Why is good water information important? Well, it is absolutely vital to virtually all of the pillars of the national water reform agenda as set out in the National Water Initiative and as programmed into the Water for the Future Program.
Firstly, properly functioning water markets. We know from the world of financial markets that equity is generated in markets when good information exists and particularly when the information is symmetric; buyers and sellers can see the same kind of information.
Secondly, infrastructure. There is a $30 billion urban water infrastructure program in play at the moment. This year alone $14 billion worth of urban water projects are in play. It's staggering. It's probably about 40 per cent of the value of what's going on in the mining sector at the moment, just to give you a frame of reference. So it's really big business.
There are many other infrastructure programs in planning at the moment. You will have heard of some of the more popular ones, like the north‑south pipeline in Victoria. There is a $6 billion program of rural infrastructure about to get underway, funded under the Water for the Future Program.
Back to the slide. I'll skip over flood warnings for the moment but I will come back to it a bit later. Prudent environmental flow allocations. The federal government is now investing $3 billion in water purchases for the environment and is already three years into that 10‑year program; 40 per cent of the way through the program. They have acquired very significant amounts of environmental water and they need good intelligence on when and how to actually allocate it to environmental assets to leverage value out of that substantial public investment.
Fair pricing and equitable sharing of the resource. We are in a very interesting phase at the moment with the new arrangements for the management of the Murray Darling Basin. You would be aware there is a new federal agency, the Murray Darling Basin Authority. Its first task is to come up with a Basin plan. In that plan it will set sustainable diversion limits that will guarantee water resource allocations to key environmental assets.
Without question, there will have to be some reductions in the diversions of water for the consumptive pool. Working out how to do that is really technically challenging. It is also socially challenging as well. But there is a lot of careful technical work that has to be done to get that right and to refine the plan as things go on.
Finally, greater efficiency in water use. Another pillar of the water reform agenda is to assist cities and rural enterprises to use every drop as effectively as possible. There is a science and technology dimension to working out how best to do that and there are a lot of economic and social aspects as well.
[SLIDE: Water information. Who wants it or needs it?]
You might think that all of the major government departments that are acquitted with the responsibility to roll out that reform agenda have all of the information they want. To some extent they do have that information. I don't believe any of the governments have got a good holistic picture, but there is also a lot of other people who actually need water information and would benefit from it. They include public agencies with very limited technical skills or resource basis, catchment management authorities and water boards, and local councils that have very limited technical ranks and not enough resources to invest in major ICT infrastructure or models or modellers, et cetera.
The community: a very important group of people who have been very much overlooked in the water information scene. We do need to inform what is a vital public debate. A lot of rancor that we see in the media, I believe, is largely attributable to the fact that those that are engaged in the debate don't have good data on what is going on.
I believe we can make a great public contribution by providing better information to the public to make informed, or develop informed, opinions about what is going on.
There is very real utility value in recreation – aquatic recreational uses such as sailing. And do you really want to be moving to an area that is going to be devoid of water storage in the next few years?
Water dependent businesses, agricultural enterprises, industrial enterprises that require water can actually benefit a lot and can minimise their risk by getting involved in trading and hedging, even in futures markets and derivatives markets. Good optics on water resources can help these water-dependent industries make good solid choices and minimise their risk.
The point I'm trying to make here is that water information collected for one purpose by a particular government entity for its particular business needs has immense value when it's repurposed for other applications. That, indeed, is the underlying thesis for this whole program.
[SLIDE: Navigating the water information jungle]
But again you may ask, isn't there enough water information? Can't we get access to it? I can see information online. I can see dam levels. I can go to some government websites and I can actually see real‑time streamflow levels and so on. But I liken it a little bit to walking into a jungle, staggering between various villages where each of the local inhabitants have a great understanding of what's going on immediately around them, all speaking a different language, however, and virtually none of them having a good perspective on what's going on beyond their immediate area. There very well may be bulldozers on the edge of that forest mowing it down, yet many are oblivious to those changes.
So navigating the water information jungle for an organisation such as my own that now has this task is really challenging. There are a lot of players. In fact, there are 240 that are required to give us water information at the moment. But there are yet more collecting water information.
It is even more fragmented than that because many of the bigger agencies are internally fragmented into groups that deal with surface water, groups that deal with groundwater, groups that manage the storages, groups that do water quality studies and ecologic studies and so on. So it's not 240 conversations, it is actually more like 1000 conversations that have to be held across the nation.
As we go to lasso all that data and bring it into a national repository we are struck by the bewildering array of data formats that are out there. Some types of data, like streamflow, tend to be in good standard enterprise data management systems. But there are other forms of information, like water licences, that are in very queer idiosyncratic bespoke systems, and getting data out of them is challenging.
All data systems that we have confronted are plagued by a general lack of metadata. They rely very heavily on the internal business knowledge of an individual engineer or hydrologist or technician in that organisation. It is very difficult to give the data to anyone and expect them to be able to understand it.
In particular, we have been struck by how poor the geo‑coding is for monitoring stations. There are probably some 7000 stream gauging stations opened and closed in the country. There are probably 100,000 groundwater levels. There are probably about 20,000 climate stations and there would be hundreds of thousands of water extraction points around the country. About 5 per cent of those things are properly geo‑coded. So actually identifying where the hell all these things are is tough.
Data supply systems are generally very immature. These organisations haven't set up their business processes to export data to anyone. The data quality is unclear and it hasn't been systematically recorded. About 30 per cent of the big agencies that we have looked at have got very high quality ISO accredited types of quality assurance programs, but most don't. For many data types in fact there is no QA process whatsoever.
[SLIDE: The Water Information Value Ladder]
What we are trying to do strategically is pretty much summarised in this graph. In essence we are trying to climb this mythical object known as the Water Information Value Ladder. Down the bottom you can see things that are really quite commonplace across the water sector: monitoring; some quality assurance; archiving of data and some local reporting.
Above that, though, there are some higher value functions that are less common: sharing the information; integrating the information, such as putting surface water and groundwater data together; augmenting the information with modelling. Although we have a huge array of monitoring points in the country it is still woefully inadequate to compute water balances for the continent. You have to rely on models to do that in areas that are ungauged, unmonitored, et cetera.
High level analysis, such as trend analysis in data, is far more uncommon and has much higher value than the lower order functions.
Finally, national reporting and the high art of hydrology, which is forecasting into the future. These are rare things but particularly precious. And this is very much the orientation of this program, to assist Australia to climb this Value Ladder and leverage much more information out of what is collected today.
[SLIDE: Australian Government Bureau of Meteorology]
Enter the Bureau of Meteorology as part of the Water for the Future Program. A bit over two‑and‑a‑half years ago we were given our new mandate. Just to remind you, however, the Bureau is a lot more than just a weather service. We are also the gatherers and the owners of the national climate record. We have been active in the water domain already, running the national flood warning and forecasting service. We were the people that issued the recent national tsunami alerts.
We also run a lesser known but really exciting national ocean current prediction program. We run a very sophisticated numerical model of all the ocean currents around the country to help with navigation for shipping and to aid the understanding of environmental processes in the ocean. We run the national tidal centre, which has the benchmark tide gauges around the country. We also compute the computations for the tidal predictions. Now we have this national water information service.
[SLIDE: Components of the Bureau's observation system]
Why was this gig given to a weather outfit, you might ask? Well, we did have some grounding in hydrology – but not a lot. I think it is because – I am sure it is, because I was one of the people that recommended it – of the great technical competence of the Bureau of Meteorology. It runs an absolutely awesome observing system out there, which is an amalgam of ground‑based, space‑based, airborne, ship‑based and ocean‑based observations. Virtually all of which are collected in real‑time, gathered in the Bureau's information system, and shared freely with the rest of the world, with the other meteorological services that also rely on these observations.
These observations are also used in sophisticated numerical weather prediction models and climate change models, so the Bureau has a long pedigree in assimilating environmental data into physical models and producing forecasts of what may be ahead minutes, hours, days, months and even years.
[SLIDE: Bureau website hits]
We are also very good at pushing information out to the public. This graph is just totally staggering. It shows the number of hits on the Bureau's website, peaking at about 2.5 billion per month.
If you go to any of the website hit league tables you will commonly see the Bureau in the top 10. They will tell you that the Bureau accounts for 40 per cent of all government website traffic. The next in line is Centrelink at about 6 per cent. So there is a lot of daylight between us and Centrelink.
Of course the hits come at peak load times, when there are extreme weather events taking place. Much of the popularity of the website is attributable to the radar, which is insanely popular. But, nonetheless, the public sees the Bureau's website as a key portal to find out information about the environment. We hope we can raise its popularity a notch more with some good water information as well.
[SLIDE: Our mandate]
Our mandate comes through the Water for the Future Program. We have been given $450 million over 10 years. This has been used to recruit 130 new staff, mainly hydrologists, to build a whole lot of new information systems and to support important research and development – noting that the innovation required to do this is yet to be developed. We can't just take good water information technology off the shelf, it must be developed by our research partners CSIRO.
An $80 million chunk of that funding is to be allocated back to the states to assist them in improving their monitoring networks. The program is underpinned by good solid legislation in the Water Act. Part 7 of the Water Act ascribes some very significant powers to the Bureau. It can request water information to be provided to it from virtually anyone. The definition of 'water information' in the Act is pretty broad. It also sets out some obligations. It mandates that we must provide that information free to the public. It also mandates that we must publish a National Water Account.
The water regulations are pointed to from the Act. They include the list of organisations that must supply water information to the Bureau, by when, what type of information, on what repeat or refreshed schedule and in what format.
Although it is a big stick, it is also a fragile thing insofar as it very much depends on the goodwill of the data gatherers. We put a lot of energy into working with the 240 agencies, to help them with the burden that this is placing on them and to incentivise them to give us the information in a timely manner and to help us understand it.
[SLIDE: Key questions we will answer] – 1st slide
The kinds of questions we are asking are as follows: how much water is going to be available in different parts of the country today, and how does that compare with the past? How is the quality of the water in the rivers and aquifers changing? And how is the pattern of run‑off and flood risk changing as a result of what we are doing to the land and what is happening to it with climate change?
They are what you might call physical hydrology type questions.
[SLIDE: Key questions we will answer] – 2nd slide
Then there are questions to do with the utilisation of the resource. First of all providing clarity on who is entitled to use water under what conditions, and to actually document how much they are using. How much water is being traded, and to where, to inform those water markets. And now, importantly, to actually understand how much is being purposefully allocated to the environment as a part of environmental flow releases. So these are indicative of the practical day‑to‑day questions that our program is meant to answer.
[SLIDE: Our new water functions]
Our functions are as shown on the slide. The top three are the core enduring functions. The first is to set standards for water data. This includes things as simple as just a water dictionary that sets a consistent national terminology for hydrology and related terms. But it can also go into setting standards for data collection, measurement techniques, data curation methods and so on, and possibly even hydrologic modelling.
The second function is to become the primary information custodian, to collect that data from the many people that measure it in the field, build a national repository and make it freely accessible to people.
Note that our role does not include any monitoring whatsoever. We rely on all of the current data collectors to continue with that task.
The third function is to then do something useful with that information, to create value‑added products that have some economic, social and environmental benefits.
Two other functions which are temporary for the first half of the program are to assist the water data collecting agencies to improve their monitoring networks, and also to invest in some very vital R&D that is needed.
[SLIDE: Our ongoing legacy water functions]
I’d like to remind you that we do have some ongoing legacy water functions that pre‑date this program. One of them is to provide hydro‑meteorological advice to the water engineering fraternity. We are the people that create these so‑called intensity frequency duration curves, which are the mainstay of water engineering design. Any water engineering thing that you have to design, be it a gutter on a house roof or a culvert or a major bridge all require reference to these IFD curves and an understanding of what the rainfall spectra are for the area in which you are doing the design.
We are in the grip now of the revision of the Australian rainfall and run‑off, the bible for water engineering. It has not been revised since 1987. Under this new program the Bureau is able to take on the task of doing a massive revision to the IFD component of that. We will probably be trebling the number of data points that are in that IFD analysis as a result of the data that we are collecting under the regulations.
The other important function is the flood warning and forecasting service. It is a 24/7 operation and it operates for about 300 locations around the country that are flood prone. It is a very sophisticated system of about 4500 telemetered installations. Some are managed by the Bureau, many are managed by councils and state governments. In a typical year we will issue something like 1200 or so formal flood warnings and a lesser number of flood watches.
[SLIDE: Leveraging water information]
This slide is to re‑emphasis the point that this program is about leveraging water information, and that we are trying to focus on value‑added or niche activities that are not so common elsewhere across the water sector.
[SLIDE: Modernisation and Extension of Hydrologic Monitoring Systems Program]
We are now into the third year of the five-year M&E program. We have already allocated $50 million of the program. The funding has gone to a very large number of organisations and projects, largely to improve the currency, comprehensiveness and accuracy of the hydrologic monitoring that goes on.
In the last year a large chunk of money has also been dedicated to the agencies supplying data, to help them simplify the task of sending the data. They have adopted a new national standard that we have developed, a water data transfer format. We put energy and funds into spatially enabling that water information, getting geo‑coding of the sites worked out, so we know where the data actually is in the continent.
We have also started to resource some small teams within major lead water agencies, to help us with the development of the National Water Account.
For the first time in Australia we now have Strategic Water Information Monitoring Plans for every jurisdiction, so we now have an overall view of the monitoring infrastructure that exists in each state and, when put together, the continent.
[SLIDE: M&E Program funding profile]
This is the first time that we have been able to transcend the individual inventory of a small number of large government departments that have systematically recorded what is measured, where, and for what purpose, and sought to document where the observing systems need to be augmented. That has been a fantastic outcome of the program.
[SLIDE: WIRADA]
WIRADA, the Water Information Research and Development Alliance, is a joint venture with CSIRO. The Bureau is investing $30 million over five years in WIRADA and that is being matched by $20 million from the CSIRO.
[SLIDE: WIRADA portfolio]
WIRADA commenced in July 2008. It involves four CSIRO divisions via the Water for a Healthy Country Flagship. It brings together 40 full‑time equivalent scientists from CSIRO, giving us a huge powerhouse of innovation to drive this program. There is some really exciting research coming out of WIRADA.
[SLIDE: Provider data]
In a nutshell, what we are doing is as follows: we are taking a bunch of diverse water information from providers. We are putting it into a large water information system, which we are calling Australian Water Resources Information System (AWRIS), and we are pushing it out to a series of information products. I'm going to talk about each of these pillars.
For a moment, though, I want you to focus on the middle of the diagram, under the heading ‘AWRIS’. You can think about this as a technology stack, with a national hydrologic database at the bottom, based on the stuff we are sweeping up on the left, and fusing that with the National Climate Database, which the Bureau is the custodian of. AWRIS will be spatially enabled with the Geospatial Fabric and augmented with hydrologic models, so that we can compute seamless water balances across the nation.
The information products that we are going to produce are really reports that look back from today into the past, or forecasts looking forward in time.
[SLIDE: Administrative boundaries]
The spatial enablement through the Geofabric is absolutely critical. Unless we can put all of that point series data into a rich geospatial environment that can be queried and sliced and diced as we see fit, it is going to be very hard to utilise that data.
You can think of the Geofabric as a specialised geographic information system for the nation, which has all of Australia's hydrofeatures stacked into it in a consistent way, with no boundary matching problems across jurisdictions or water management businesses.
It will include administrative data, the location of rivers, lakes, wetlands and so on. We will come up with a consistent national reporting unit system based on a high resolution digital elevation model (DEM), which has been built by CSIRO as part of the WIRADA alliance. It will include a lot of the contextual spatial information, like soils information, landcover and so on. It will also include the locations of those many hundreds of thousands of monitoring points and extraction points, as well as the water infrastructure like supply channels, drainage channels and so on.
[SLIDE: 1050 sites]
At some point in time, maybe in a year or so, all of the information in AWRIS will be able to be reflected back out to the public free of charge, through a comprehensive water data warehouse. This slide shows a wireframe mock‑up of what that might look like. It will have a spatial navigation pane, it will have a shopping cart that you can put the data into in a sophisticated manner to filter that data, so that you can get whatever you want.
This alone, we think, will save the consulting industry and the R&D sector significant heartache by pulling water information sets together for their day‑to‑day work.
[SLIDE: Liberating water data]
One thing that I am really excited about is that we have been able to largely cut through the data licensing hassles that everyone has to put up with.
We recently wrote to all 240 water data suppliers and exhorted them to sign up to the creative commons attribution licence. It is an open data sharing arrangement, which is now fast becoming an international standard. There are different flavours of creative comments. The attribution version is the most liberal. The only thing you have to do is to acknowledge the owner of the information. You are otherwise free to utilise it for whatever purpose you like.
Of the 240 people we have written to, over 150 have written back to us and said, 'We're in'. So it's a breakthrough in liberalising the free sharing of water information. We are confident that we will be able to get most of the others in the fullness of time.
[SLIDE: Water reports from AWRIS]
The type of water reports that we are going to be putting out are shown on this slide on a time axis, going from today on the right to decades past on the left. And some examples of reports that we are going to be pushing out are dashboards, which are current summaries on the web; situation reports, which are integrated information summaries for particular basins, like the Murray Darling Basin, that we will probably punch out on a monthly basis; the National Water Account and the national water resource assessment will be annual products. I will speak about the Water Account in a moment. That will talk about the last water year. The water resource assessments will take a more long‑term view, looking at the trends in water resources condition and availability and use.
[SLIDE: All Australian public storages]
The first dashboard that we are going to punch out in about three months from now is a water storages dashboard. We are in the shake‑down phase of this at the moment. This slide is a graphic of all of Australia's public storages that exceed 1 gigalitre in capacity. There are about 500 of them. Until now there hasn't even been a list of these things. We have had to create that list.
As you can see, they are mainly distributed in the areas of high population and agricultural production. They vary greatly in size, everything from a gigalitre up to a teralitre. The data for the ones shown in green will be released in the first version of this water storages product. The data for the ones shown in red will be released some months later.
In the first release, however, we will capture about 90 per cent of the total water storage by volume in the country. It is about 300 storages, but it is about 90 per cent of the total volume.
[SLIDES: Surface water storages] (1st, 2nd and 3rd slides)
These three slides are pretty accurate screen grabs of what the system looks like. This will be a web page on the Bureau's website. The first slide shows water storages. You will be presented with a national map and you will be able to get aggregate summaries at a state level, or at a city level (second slide).
The third slide shows how you will be able to zoom all the way down to individual storages and be able to look at time series and all of the rich metadata. There will be one website showing all of Australia's water storage information, using a consistent terminology.
[SLIDE: Planned (2010) AWRIS releases]
In the months ahead there will be something similar put together for water restrictions and river flows, and probably later in the year also something on water quality and groundwater levels. We will gradually enhance the water storages product with greater functionality. It is hard getting the first product out but once the first version is stable we will enhance its functionality quickly.
[SLIDE: The national water account]
I want to talk about the National Water Account. This is probably the most important thing we have to do as it is a statutory obligation of the director of the Bureau to publish this each year. This is going to be a seminal document for water reform in Australia.
It will disclose the availability, the rights to take and the actual take of water on a water‑year basis, equivalent to a financial year basis.
We have developed, through the Water Accounting Standards Board, an independent expert body reporting to us a water accounting conceptual framework and a preliminary water accounting standard which later this year will become an established formalised national standard for water accounting.
[SLIDE: Methods Pilot for the National Water Account 2009 Test Regions]
We have already piloted the methods on seven major regions, as shown on this slide, going from the insanely small, like the Onkarparinga, to the hairily large, like the Murray Darling Basin; encompassing rural areas and local areas, such as south‑east Queensland and the Melbourne region. And also looking, in one instance, at a heavily groundwater-oriented system in the Gnangara Mound in Perth. We have had some very rich learning through this exercise.
[SLIDE: Water Balance framework]
In essence, what we have done in the Water Account is to construct a conceptual water balance for the nation. This slide shows that we have partitioned it between a landscape store, a surface water store, a groundwater store and a water transport system.
[SLIDE: Chart of accounts line items]
We have assiduously established a chart of accounts that consists of a series of line items that enumerate pretty much every water balance term that you could ever imagine. The objective of the water accounting exercise is to populate that chart of accounts, using observational data and using hydrologic models, and to do that systematically across Australia year on year, so that we can intimately understand how the water balance is changing.
[SLIDE: Pilot National Water Account]
Our experiences have been documented in the Pilot National Water Account. The first edition was published in December 2009 and the second edition in February 2010. It encompasses four of the seven reporting regions. Three of the regions that we did the work on aren’t reported on in this pilot report. It is a thick and fairly inaccessible document. It is downloadable in PDF format. It is evident to us, through this pilot exercise, which is very much constructed as a learning exercise, that web delivery of this product is going to be absolutely necessary.
[SLIDE: Water Assets and Liabilities]
We have already started mocking up wire frames for what this will look like. This will allow you to collapse and expand different water balance terms or to quickly shift from one reporting area to another to make comparisons. And work is now underway to develop this system in time for the publication of the first formal account at the end of this year.
[SLIDE: Water forecasts from AWRIS]
I want to conclude by talking about forecasting. Again, there is a range of different products: the flood forecasting we do already; a 10‑day flow forecasting service, which is a new thing that will be constructed as an augmentation of the flood service, adding some probabilistic seasonal streamflow forecasting services; and, finally, long‑term water balance scenario analyses.
[SLIDE: Our rainfall forecast for today]
I want to begin with floods. This slide shows our rainfall forecast for today. As you can see, the numerical weather prediction models were predicting intense rainfall in south‑east Queensland – between 50 and 200 millimetres or so.
[SLIDE: Flood warning service. Recent rainfalls - QLD]
This is what our flood site shows what happened today. These are the recent rainfalls in the Queensland area. The red dots show rainfall exceeding 100 millimetres in the last 24 hours.
[SLIDE: Flood warning service. Current river levels – QLD]
This is the manifestation in real‑time flood levels, again recorded on the Bureau's website. Those areas shown in red are undergoing major flooding as of 1.15pm this afternoon. The areas shown in orange are experiencing moderate flooding. There is nothing happening in the blue areas.
What you can't see on the public access website is that the Bureau punches out flood level predictions for all the people in the state emergency services and agencies and councils who are involved in the flood management process.
[Flood forecast for Charleville]
This graph is really constructed of two parts. On the left of the purple vertical bar is the trace of the flow levels. The model in blue, and the model in red, the little bars at the bottom are the rainfall. And the red line to the right of the vertical bar is the flood prediction, or the flood level prediction. This provides intelligence to those involved in the emergency planning process of what's actually coming.
So the inputs to this system are the numerical weather prediction models, the rainfall radar and of course the network of all of the gauging stations, rainfall and runoff.
[SLIDE: ... and what gets done with the forecast]
And this is what happens with it. State emergency services take up this information and use it to manage public safety. (At the moment I understand there are people being evacuated in the Charleville area.) This service has been operating quite well for quite a few years now. We are about to turbocharge that service to develop a new set of models with enhanced accuracy and coverage. We will probably look at the extent of inundation as well as just flood levels in rivers, and also to increase the lead times in the forecasts.
[SLIDE: Predicting seasonal streamflows]
Moving on to seasonal streamflows. You can't be deterministic about what is going to happen three months out. It is, by definition, a probability or an odds game. But if you can get that more or less right it's hugely valuable. If you can give guidance to people who need to know what is going to happen in three‑months' time it can be hugely valuable.
Fortunately we have some skill in climate forecasting. There is a whole lot of synoptic climatology that we know after many years of looking at the weather. Things like El Niño, La Niña and the Indian Ocean dipole. The more exotic things like the blocking highs and the subtropical ridge and east coast lows, these are all phenomena that we know about and we can inject our understanding of these systems into hydrologic models to try and foresee what might be ahead for the next three months.
[SLIDE: Bureau's seasonal climate outlook service]
For many years we have been issuing these types of climate outlooks. We issue them each month for three months ahead. In this example we are showing the chance of exceeding the median rainfall. People can be critical of this because it is often a 50/50 kind of thing. With hydrologic forecasting we can in fact be quite a bit more emphatic than in the climate predictions alone, because the catchments have a memory, and you can capture and utilise that memory in a flow forecast.
[SLIDE: Experimental seasonal streamflow forecast ...]
We now have an experimental seasonal flow forecasting system. This is only available to registered users at the moment. The intent is to make it publicly available at the end of the year, once we can prove that it really adds some value.
This slide is indicative of the kind of thing that you can do. Here we are showing the high water production areas of the Murray system – the Murrumbidgee, the upper Murray and the Goulburn. These tercile plots are showing the probability that the next three months of streamflow will be either below average, near average or above average.
This particular set of forecasts, which we have just published online for the period February to April, shows that there is a good chance of streamflow being well below average to the north, with the odds of better streamflow being more likely as you move south, down into the Goulburn system.
[SLIDE: Murray River at Biggara]
This is the type of intelligence that we provide a water manager. On the left of the slide is the information that water managers historically have at their disposal. There would be a record of the most flow that was ever attained in that three‑month window – in this case it was in 1974. The least flow was the blue line for 1983. And there might be a few different types of averages, a long‑term average, a 10‑year average and last year's record, shown in green.
Normally, a water resource manager would just go on that historical information. But on the right of the slide is the output of our seasonal streamflow forecasting model, which plots the likely flows of probability distribution function.
The historical distribution is shown in pink and the forecast distribution, which is much tighter and more emphatic, is shown in blue. Water resource managers are pretty excited about this, as are farmers, water traders, hedge fund operators, and all kinds of people who can really draw some value out of these types of forecasts.
[SLIDE: Climate impact on water availability in the MDB – Historical streamflow]
This slide shows CSIRO's work on long‑term streamflow prediction. Many of you will have heard of the sustainable yields assessment work – the most ambitious hydrologic analysis study ever undertaken in Australia. It was a Herculean exercise, where they systematically documented the historical streamflows for all major basins of the Murray Darling Basin System.
[SLIDE: Climate impact on water availability in the MDB Decrease in future streamflow]
With the aid of various climate change scenarios and demand management scenarios, CSIRO also had a look into the future of what the likely prognosis was for streamflows. This is the high art of water resources management, and somewhere where we will go in the long‑term future, once we have nailed the short‑term forecasting and seasonal forecasting.
[SLIDE: Close to the 3‑year mark]
In summary, we are close to the three‑year mark. We have had a hell of a job finding 130 new water specialists in a pretty tough labour market. But we have got some great people and they are coming up to speed. We have over 5 million files that have come into the Bureau and we are busily sorting them out. There is a very vigorous program of modernising observing systems going on. It is a bit of an administrative challenge, I must say, to deliver on this. But it is having huge benefits in terms of improving the quality and coverage of the data that we are receiving.
[SLIDE: Good water information is the key]
This R&D program is one of the most exciting things I have seen during my career in the water game. We have a really sophisticated water information system, AWRIS. The early stages of development are proving to be a lot tougher than we ever imagined, but it is going to deliver immense value to the nation when it is completed. Our water accounting foundations are set and many of our new forecasting services are taking form.
The pillars of water reform that depend so heavily on good water information are going to be strongly sustained into the future by this program.
[SLIDE: Thanks for listening]
Discussion
Chair (Graham Farquhar): Wow, free data access! Transparent data! I think there has been a lot of disservice done to science recently from the lack of transparent data and the inability of scientists to see what other scientists have done and to be able to check their calculations. So this is an incredibly important move in the right direction. I'm delighted that you kept the notion of keeping it free in times when so many things are commercialised. It's important that basic data like this be available to so many people other than scientists.
Question: You described the enormous challenge of getting the data and definitions across a very large community of organisations. That's often difficult to do even with a small group of organisations. Can you tell us a little bit about the kind of governance mechanisms you are using to try to achieve those sorts of agreements on those sorts of issues?
Rob Vertessy: You are right; it is difficult to do culturally. However, I have been struck by the goodwill of the agencies out there that are participating. They look at this program and they can see very clearly that it is a game changer. They understand intimately the hassle that has existed in the past with sewing this stuff up. They have always been frustrated, of course, when the Commonwealth has come in and tried to make sense of it, only to kind of jumble it up.
I have to say there is a lot of motivation out there to do it. Nonetheless, it does require pretty formal governance about how to do it. You are right to point out that we do have to have governance structures.
We have various types of committees. We have an overarching committee called the Jurisdictional Reference Group on Water Information, which you might think of as a kind of council of war, where all of the states and the Commonwealth come together three times a year for a couple of days on each occasion and work through the high level strategy. Then that is supported by specific expert panels that we are putting in place to work on particular technical facets.
At the moment we have three panels running. One is concerned with the water data transfer format, which is an XML schemer, basically a big data model for all the water information. Another panel is concerned with the Geofabric, to get the spatial enabling framework right. And the third panel is concerned with a water information dictionary, which in fact is the base lexicon for all of the terms that will be used in all of the products and services that we produce. And we hope one day these terms may gradually weave their way back into state water legislation – which is a pretty hard thing to unwind, but we hope that one day people will see the value of using a nationally consistent terminology.
Question: It’s very interesting to hear what's happening. Even though they are only a very small portion of the continent’s surface, cities are one of the areas where water issues are becoming really focused, partly because of where they are sitting relative to the changes in water availability.
As part of your future scenario testing are you seeing that this material that you are providing will be able to be used at a scale where people looking at different development patterns for cities can start to understand more quickly with stronger evidence as to the sorts of impacts that they are going to put on catchments, the sorts of restrictions that you might need to demand in terms of people's use within cities by expanding the particular allocations? Do you see it being part of a water pricing system that will be singly used to limit the people?
Rob Vertessy: Yes. Absolutely. I would be bitterly disappointed if it didn't add significant value to that process. Every urban water utility and state government that is in the game at the moment would be pulling their hair out worrying about what water supply augmentation option to take and when to take it; whether to defer the investment by a year or not can mean millions of dollars on the balance sheet. Likewise they are always vexed with the water restrictions conundrum. Do we go now at the well‑known trigger point or do we wait a little bit, or when might we relax it? You can see the calculus going on in the papers.
I'm very confident that this early warning intelligence and these better optics can assist those people to be a lot more confident in the hard choices that they are having to make. I'm super confident of that.
Question: Great presentation, thanks Rob. Presumably there will also be a role for programming systematic identification of knowledge gaps. And I'm thinking in particular of the dearth of information on aquifer hydro‑geology mapping and also the urban issue, the new emerging issue of micro‑climate management.
Rob Vertessy: Good point. Built into the ethos of the National Water Account is the concept of disclosure notes. The disclosure notes are to be prepared for each line item in the chart of accounts and are meant to explain the methods that were used to derive the estimates, but also to ‘fess up’ to the inaccuracies and the unknowns in the process, which are considerable.
I am very confident that for several years to come you will be able to drive a car through the Water Account. It's a hugely challenging task to complete water balances for areas which are so sporadically and unevenly measured. The reliance on models and a lot of base assumptions is great. So a big part of this is actually revealing where the knowledge is weak.
I think not only is that good honest practice, but it is also potentially a great stimulus for people to invest more in the monitoring infrastructure.
That's my big concern, in fact, that although we seem on first appearances to have a great monitoring infrastructure in Australia, it is well short of what is actually needed to pin down water balances. And if we want them, as we say we want them in the national water initiative, we have got to do a hell of a lot more monitoring.
Question: Thank you for giving us a briefing on the resources of water in Australia. You said much less about the use of the water, and I particularly noticed in one of your pie charts that you talked about water scarcity factors but didn't include industry and car generation, which I believe use very large proportions of valuable water. Are all these major users of water going to provide you with accurate data on their use, and how are they to be controlled in the future?
Rob Vertessy: Yes, really good point. To take the final question, probably yes in the future. At the moment pretty well all of the people listed in the regulations are public agencies. We are getting some information from power generators but it is restricted to water held in storage as opposed to the water use.
Likewise, at the moment we are not procuring anything from the mining sector in terms of their use. These are known gaps. We are being strategic about it. We are not rushing into trying to scoop up absolutely everything until we can master what we are getting from the public sectors. But it will be increasingly important that we do cover those users.
Some of you may be aware that the Australian Bureau of Statistics also publish 'The Water Account'. But that is very much about the utilisation of water in the economy. It is based on survey methodology by profiling different sectors of the economy and estimating what their water use is.
The ABS put together a very good national picture of the relative use by different parts of the economy. We won't be duplicating that. In fact, we will be supplying them with a lot of the base physical hydrology data. We will also be building up a large database of the actual metered water use of those users – at least the ones that are metered. We will have to put in place hydrologic modelling methods to estimate what isn't being metered, which is a pretty substantial amount of water, unfortunately.
Question: I am interested in the scenario of testing part of your ambit, in particular the urban water security problem. Some would say the concept of urban water security is rapidly changing – the paradigm is changing. Is that something foremost in your mind, essentially the concept of cities as catchments as we don't use a lot of the water that runs off our cities now. I expect that will change. Is that something the Bureau is looking at?
Rob Vertessy: You are making the point that the concept of water security in cities is changing. I think you are implying that there is a relative shift away from catchment supply sources to a more diversified source of water through, say, stormwater, desalination, recycling, et cetera. So, yes, it is a very different game emerging in cities.
In fact it is rather striking how quickly desalination has come on. By 2012 we will have something like 4800 megalitres per annum on stream, scalable to about 6700 megalitres on stream. That is about 30 per cent of the 4800 megalitres and about 30 per cent of the 2007-08 water consumption. So in a way cities are buttressing themselves against climate-induced fluctuations.
It is a bit less of an issue to be able to have good foresight about what is going to be running off catchments and more of an issue now to work out how one can integrate diverse sources of supply. And also, I suppose, there are downstream issues like amenity value of water courses and the like.
They are probably things that we are not going to deal with much. We will certainly account for those new sources and supply. We collect that information under the regulations, in so far as it has to be injected into a water balance to do all of the water accounting. But I don't think we are going to be in any way an advisory service to water utilities on how to do water sensitive urban design or to make choices about which new supply augmentation option is best.
Question: Evaporation rates seem to be a poor cousin amongst the data. Is it worth any comment?
Rob Vertessy: It is a really important term, particularly in the water accounts where we are trying to compute regional water balances in areas where we are not measuring streamflow very well. It is vital that we do get a good accurate coverage of evapotranspiration. In fact, our colleagues at CSIRO are injecting quite a lot of effort into that problem at the moment, trying to come up with a gridded daily evapotranspiration mapping model for the country.
The objective is to build basically a data cube on a 5 km grid of the actual evapotranspiration for the entire continent. To have that go back as far as the satellite coverage inputs will allow, but to update that on a daily basis, as an input into our water balance calculations. We are giving it a red hot go, but it is tough stuff.
Question: Thank you for that talk. It was excellent. I am sure your work has long been needed.
I'm a farmer, so you are going to get a bit of a farm perspective on this. You mentioned farm dams in the context of run‑off. I'm sure you are aware that not all water in farm dams comes from run‑off. A lot of the farm dams which I've built, for instance, come from spring water as well as run‑off. So it's a combination of the aquifer and the run‑off.
We are up on top of the tablelands. The springs flow slowly if they are not opened up, but once they are opened up and go into dams then a whole different flow transpires.
I suppose, as farmers, we get a bit concerned that people look at our dams and they think, oh, well, that water is being hoarded out there. But it does overflow eventually and it does keep running downstream. Of course, if you introduce things such as water trading markets you are going to encourage the hoarding of water. This is perhaps one of those areas of policy where things are working in diametric opposition to each other. Perhaps that needs to be thought through a bit.
Are you taking account of the fact of those different sources of water within farm dams – no doubt you are – but could you just clarify that? Also, in regard to the monitoring, I don't know whether you have given thought to the use of farms in terms of the monitoring, because your closest station to us is 300 metres lower than we are. We get a completely different rainfall pattern where the water actually forms up on the tablelands to run down into the streams, but your monitoring stations down in the valley are located where not as much of that is happening.
Rob Vertessy: Let me take the last question first. I'm really enamored with the idea of the extent to which community observers can contribute to this story. In my mind I can see a second wave downstream, once we sort out this jumble of public information from agencies. I can really see the opportunity to get the community involved in this kind of 'crowdsourcing' – I think that is the trendy new Web 2.0 term for it.
In effect we have been doing this for a hundred years with rainfall. We have got 5000 community observers working for the Bureau, making daily rainfall measurements. It is absolutely astonishing. Our rainfall record for the continent would be nowhere near as good without those measurements.
I have every confidence that if there was a good framework, a single institution and some standards in place we could leverage huge value out of the community in making hydrologic measurements.
We have seen some values already, of course, with programs like Water Watch. It has been a little variable but I think there have been some programs that have been absolutely topnotch and some good data has come out of them.
We are dealing with water quality parameters which are typically much harder to do well with limited kit. But I have every confidence the day will come when citizens of Australia will contribute to this rich water information base.
As for farm dams, I really don't want to single them out or demonise them at all. I fully understand the criticality of them for agriculture. They are just absolutely essential. So I hope no‑one thinks that I'm on a crusade against farm dams. But it is important to understand what their role in the water balance is. Whether they are intercept and surface flows or groundwater flows is really immaterial. The fact is that they are part of the cycle now and they've got to be represented in water balance terms if we want to have an accurate water balance for the nation.
Plantations and flood diversions are other things in the rural landscape that also have to be factored in.
