THE PREDICTED "CLIMATE-HYDRO-ELECTRICITY INDEX"
for the following selected months# in 2011/2013 issued on 24 November 2011, are as follows".
Commentary: November 2011
With the economy remaining weak and restricting demand, prices should be restrained for most of the forecasting period (to May 2013). There is little difference between predicted spot and contract prices except over the winter 2012 period.
Generally low spot and contract prices, however low inflows into South Island lakes in January to March could have a delayed effect on spot prices, which could rise in winter in both the South Island and the north of the North Island (although not to historically high levels). Restricted transmission suggests a different picture for the south of the North Island with spot prices increasing earlier, but dropping over winter.
2013Generally low spot and contract prices through to May, the end of our forecast period. Contract prices could increase in April and May, as these do not take into account the anticipated higher than average inflows predicted at this time
Long range rainfall forecasts for several seasons ahead for Tauranga, Turangi, New Plymouth, Hokitika , Queenstown, and Milford Sound prepared by John Maunder ( last updated on November 24, 2011) are available at:
The PowerFutures Group is a Group of three consultants who have developed a system for forecasting the weekly spot and contract price of electricity in New Zealand 1-24 months ahead.
The PowerFutures Group was originally proposed by the late David Cook a well known consultant in the electricity industry who died in November 2001. The three remaining consultants in the group have further developed the concepts which David Cook had proposed when setting up the group.
The group comprises Dr John Maunder, in Tauranga who has extensive experience in the meteorological and climatological business in New Zealand, Canada, Australia, Ireland, the United States, and the World Meteorological Organization( email@example.com); Dr Jonathan Lermit, a consultant of Wellington, who has considerable experience in the electricity industry and in particular in the supply and demand aspect of the industry( firstname.lastname@example.org, ; and Dr Blair Fitzharris, Emeritus Professor of the University of Otago, who developed a "snow model" for calculating the amount of water stored as snow in the mountains of the hydro-catchment areas, and has very wide international experience in various aspects of climate science (email@example.com).
The Group makes predictions of the water available for hydro-electricity production in NZ, based mainly on the forecast state of the atmosphere over NZ and the surrounding oceans using a variety of techniques - including using forecasts of the Southern Oscillation Index (El Nino/La Nina), sea surface temperatures, sunspot numbers, and linking such forecasts to rainfall forecasts, and the forecast state of the snow volume in the mountains which will be available for melting into the hydro lakes. These hydro water predictions, together with other supply and demand data including forecasts of the temperature, are in turn used to forecast both the contract and spot price of electricity on a weekly basis 1-24 months ahead.
The main intention of our predictions is to provide an estimate of the difference in price that can be attributed to the forecast of inflows and temperatures actually eventuating (the spot price), compared with the generation industry not having any detailed knowledge of the future inflow sequences and temperatures (other than the historical data and some short-term climate and hydrology forecasts), and having to make contractual arrangements in the absence of any medium to long-term predictions of hydro inflows (the contract price).
Our model provides two sets of prices, an anticipated spot price and a contract price which uses no hydro prediction (other than the historical data). Because the difference between the two prices is in the assumed knowledge about hydro conditions and temperatures, differences between the prices can be significant when hydro conditions and temperatures are expected to be significantly different from the average historical data.
The contract prices thus derived are from the average price arising from (ideally) all possible inflow sequences, assuming that the generators do not know the final outcome. In contrast, the spot prices assume that the predicted inflows and temperatures actually occur.
Because generators have to be risk averse, they have to charge a risk premium in contracting to supply (and will endeavour to not over contract so as to avoid having to buy on the spot market). The magnitude of this premium is not covered in our model, but users will presumably have an idea of its magnitude, and can compare the contract price which our model provides; with the contract price offered by generators.
The model thus concentrates on estimating the hydrological effects, and to a lesser extent temperature factors; it does however, also handle other significant parameters of the New Zealand electricity system: notably,the hydro/thermal mix; time-of-use demand and prices; major transmission constraints; hydro uncertainty; and the market behaviour of generators and consumers.
The prices which our model forecasts are average weekly values, and are averaged out over New Zealand. However, internally the model calculates the time-of-use prices, and it is possible to provide a time-of-use breakdown according to specific requirements.
Similarly, geographical differences can be modeled, although, because of the nature of the model, this is likely to be less accurate. These arise both from transmission losses, and from transmission constraints. Currently, we split the load into the South Island, and the North Island into two halves (to approximate the transmission constraint across the centre of the island).
Any model of this nature has limitations, and users need to be aware of these. For example, there are significant factors, such as plant breakdown which cannot be modeled. Additionally, some of the information, such as fuel costs, have to be estimated in the absence of publicly available data. Also generators are modeled as pure profit maximizers (subject to their contractual obligations, which again have to be estimated), whereas real market behaviour may well be different from thi'UA-26008507-1's.