Following the loss of the Basslink interconnector in late December 2015 there has been a lot of interest in and concern about the state of electricity supply in Tasmania, most of which comes from decades-old hydroelectric systems. This post gives the results of a simple analysis of the recent hydro water storage history, as part of an assessment of the security of electricity supply in Tasmania.
The top level information available to the public does not provide a clear picture of the state of the electricity system in Tasmania. The main media focus is on the figure for total hydro energy storage, which at the time of writing was 3164 GWh, i.e. 21.9% of Full Storage. The significance of this figure, and of the fact that it is rising, is unclear. It is stated that Full Storage is equivalent to 1.5 years of annual average demand, so would the current storage provide full demand for 3-4 months of drought? The hydro energy storage is typically near its minimum at this time of year, so by how much are the reservoirs expected to fill over the next several months?
The weakness of a total figure for energy storage is revealed by the following plot, showing the maximum storage capacity of each reservoir:
The plot shown above reveals that Tasmania has (or rather had) two large reservoirs (Gordon and Great Lake), one basin of intermediate size (Lakes Echo and King William) and many very small ones. Since there is a highly non-uniform connection of generator capacity to storage the single figure for total energy storage is very misleading. For example, if most of the storage was in either Great Lake, or Lake Gordon, then the total hydro MW output could not exceed that of the associated power stations, i.e. no more than around 400 MW, well short of 50% of average demand.
Sadly, Tasmania no longer has two large reservoirs, they have both been severely depleted in recent years. This can be seen from a plot of the recent history of the basin storage levels, which I derived from a spreadsheet available here (click on “Energy Storage – Historical Data”):
The curves in the figure above are simply the numbers in the spreadsheet, summed over the lakes in each of the six hydro basins. The seasonal variation of storage is most apparent in the data for the Derwent system (black curve), on top of which there were severe declines in the two formerly large systems. Recent weeks have been very rainy, which has produced the usual recovery in storage levels at this time of year. I have made very crude eyeball estimates (marked on the figure above) of where the storage levels will be in spring of this year, assuming a rainy autumn/winter.
The fact that total hydro storage is rising does not mean that the reservoir depletion problem has been solved. One of the major reservoirs (Gordon, the red curve) has been essentially emptied, and some recent rain/snow accumulation does not solve that problem. When the next dry spell occurs the smaller reservoirs will soon empty and the larger ones will be required to take over, and the level of the Gordon reservoir would normally go back down again.
What is the overall significance of these current/estimated storage levels? To begin to answer that question I computed how much hydro electricity could be generated solely from the estimated springtime storage, corresponding to the occurrence of a severe drought during spring and summer. I chose to run all hydro generators at 40% of nameplate capacity, in order to (almost) meet typical demand for spring/summer. The results are shown in the following figure:
Hydro could only meet full demand for around 40 days before the first small basin (Anthony Pieman) runs out of water, followed shortly by the loss of any output from Mersey-Forth. A severe drought longer than 6 months would see hydro output fall below 50% of demand. The early step down dates can be extended somewhat by running those generators at lower output, but only at the expense of bringing forward the later step down dates. The total area beneath the black curve will always remain equal to the total energy storage figure.
To properly describe the energy storage situation, figures should be given separately for each basin, together with the associated MW capacity. The black curve shown above can be thought of as the energy storage decay function, with the area beneath the curve composed of a contribution from each basin, a rectangular block whose area is the energy storage figure for that basin. This approach will be further developed here to provide a regular status summary of the Tasmanian electricity system.
Tasmania has already suffered badly from not having enough reserve electricity generating capacity to deal with a credible breakdown in the system, the loss of Basslink. The results shown above suggest to the author that Tasmania no longer has the hydro system produced decades ago by its original designers and engineers, making it very vulnerable to more problems in the credible eventuality of severe droughts.