New release of EMI dataset – Full update of the hydrological modelling dataset (data to 31 Dec 2020)

  • Last post 27 September 2021
Matthew Keir posted this 28 July 2021

The Authority has released an update of the Hydrological Modelling Dataset (HMD) that includes data to the end of 2020. This release is a full update of the dataset revisiting the historical records and correlations. As 2021 has reminded us, an appreciation of the significance of hydro generation and the inherent variability of hydro inflows is fundamental to understanding the electricity market, managing security of supply, and planning the New Zealand Electricity system. The HMD provides detailed information that supports the efficient operation of the market, monitoring and policy development, and planning as NZ transitions to a 100% renewable system that delivers for consumers.

The HMD includes the following datasets:

  • Infrastructure and hydro constraint attributes (including access to contingent storage)
  • Flows (including inflows, outflows, tributary flows, and flows at a junction or point)
  • Storage and spill information

This release includes the following enhancements:

  • Historical spill data is now available at daily resolution aligning it with information in other datasets
  • A flow series for the natural Lake Onslow inflows has been added to support any modelling of potential schemes

More information is available in the reports included with the HMD. Users of the dataset are encouraged to post feedback - especially if they believe there is an error or something has been missed in the infrastructure and hydro constraints file.

The Authority appreciates the ongoing cooperation of the hydro generators and other data providers in compiling and publishing this dataset.


Order by: Standard | Newest | Votes
luckiestmanalive posted this 02 August 2021

This is great news, thanks, Matt! I especially appreciate the addition of daily spill data so everyone has access to the information required to analyse the pricing decisions of the generators who are the kaitiaki of our hydro lakes.

  • Liked by
  • Callum M
davidw posted this 27 September 2021

Hi Matthew,

I'm having a look at this updated data and I'm getting an apparent increase in spill across most series of about 1.6x. I assume it's a units/conversion error on my part as it's reasonably consistent throughout. The 2015 and 2018 updates used "m3", which I take to mean cubic metres of water (i.e. a quantity of water, rather than a flow rate as used for flow data). The 2021 update uses Mm3, which I assume to simply be million cubic metres of water (i.e. the previous unit x 1,000,000). Is there some cumecs/volume conversion process I'm missing here?

I note the comment in the documentation that the 2021 series also includes weekly data, so perhaps this is leading to some double counting? However, that doesn't match the size of the increase, and the "weekly data" isn't obvious to me when looking at the data series.

Any idea what I'm missing here?

Cheers, Dave

Matthew Keir posted this 27 September 2021

Hi David,

Thanks for pointing this out - you've found an error in the old datasets with the units/conversion (the conversion treated Mm3 and m3/s (ie like flows) giving rise to the 1.6 factor). The documentation appears to be incorrect also as we do not include a weekly series given this was superseded with the daily series available now (it should just be the sum of seven days). The comparison report (3a) should address the differences between the old weekly series and the newer daily series but may have undertaken the comparison prior to conversion. The current data and units are correct.

Apologies for the confusion.