Dear Ministry of Business, Innovation & Employment,
Via your organisation's web site https://www.mbie.govt.nz/building-and-en... a copy of file renewables.xlsx can be obtained, and my attention for the moment is directed at the entries under the heading Energy Transformation, specifically those for Wind. You offer calendar year totals in PetaJoules, and starting with the year 2000 (before which there was just the Brooklyn hill turbine and Hau Nui and Tararua had just started), the table below show your figures followed by mine, from a national database of electricity data. Apologies for the layout, but fixed-spacing typefaces are not available here.
Year: MBIE other
2000: 0.43 0.50
2001: 0.50 0.48
2002: 0.56 0.55
2003: 0.53 0.52
2004: 1.30 1.29
2005: 2.21 2.20
2006: 2.24 2.22
2007: 3.35 3.31
2008: 3.81 3.76
2009: 5.31 5.26
2010: 5.89 5.84
2011: 7.05 6.66
2012: 7.48 6.87
2013: 7.28 6.71
2014: 7.96 7.01
2015: 8.51 7.22
2016: 8.43 7.74
2017: 7.53 6.88
2018: 7.44 6.79
2019: 8.12 7.49
2020: 8.30 7.70
2021: ?.?? 8.56

As you can see, apart from the year 2000, your figures are consistently higher and often by more than ten percent. My values are based on the summations of half-hourly generation data, as metered, easily obtained via a system called Gnash. (Gnash.exe for windows, Gnash for GNU/Linux)
So my question is 1) where have your values come from, and 2) what is included and what not. For instance, have any values been based upon the nominal installed capacity (with some estimate of an annual load factor, say), and have you included some sort of data/estimate for miscellaneous wind turbines scattered about, perhaps supplying a factory? For instance, there are some wind turbines in the Seaview industrial area that might supply internal load and thus not the public supply even though offsetting the draw on the public supply.
By contrast, the half-hourly data available to Gnash are clearly deficient: wind turbine generation at Mahinerangi and Horseshoe Bend is lumped in with hydro generation (if it is counted at all) and so not identified as "wind", and whole wind farms are omitted entirely, as at Flat Hill (near Bluff), Weld Cone, and Lulworth, as well as for various individual locations such as atop Brooklyn. And the absence of data for the new wind farm at Turitea will shortly be the subject of an OIA request to the Electricity authority.

Thus, I expect Gnash's data to be "low", but, not by so much as appears to be the case. So, what makes up your data?

Yours faithfully,
Nicky McLean.

Dear Anome,
Thank you for your acknowledgement. It would be nice to have the name of a human somewhere in it, or perhaps it was generated by computer. Still, it seems the process has begun.

Yours sincerely,
Nicky McLean (a human: Mr.)

Dear Ms. King,
Thank you for your reply. I'm guessing that your organisation's figures derive in whole or part from those provided by the Electricity authority, and you will have difficulty in obtaining straight or helpful responses from them describing just what their data series contain and do not contain. Certainly, I do and you can view the lack of progress on the FYI front should this be of interest.
My suspicion is that something has been double-counted in the forming of your totals, as the data series offered by the Electricity authority frequently involve changes of name. Making sense of this mess takes a lot of patience.

Incidentally, the new public holiday would allow you one extra calendar day for your response "within twenty working days", if my finger-counting is correct, but no matter. I continue to look forward to learning how your figures were derived. Next week or the following.

Yours sincerely,
Nicky McLean

Dear Ms. King,
Rather than drift in the fog, here is a schedule of the wind generation data available, collated from data published by the Electricity authority and its predecessors. It can be elicited by entering "stats GEN.Wnd.~" to Gnash
01/10/2003-31/3/2008 GEN.Wind.GebbiesPass
01/02/2005-31/3/2008 GEN.Wind.Southbridge
29/08/2006-31/3/2008 GEN.Wind.Tararua.TeRereHau.Linton
01/03/2005-31/3/2008 GEN.Wind.Wellington.Brooklyn
These data series ended with March 2008 not because they ceased operating, but because the Electricity authority re-contracted some activities, and forgot to include the requirement that data be supplied. Thus, when ems ceased handling the task, data were no longer supplied. The Te Rere Hau windfarm was odd, because it was connected via a limited capacity transmission line, apparently towards Linton. When the wind farm was greatly enlarged (starting later in 2008), a new connection was used, called TeRereHu.Grid as below. There is no indication whether or not the limited connection towards Linton continued to be used for a limited portion of its capacity, and the Electricity authority is declining to provide information.
01/01/1999-31/8/2012 GEN.Wind.HauNui
Data for this wind farm was originally supplied with a code name of HAU (for Hau Nui) but then this was later abandoned for a code name of GYT (for Greytown), a petty annoyance - except that some of the GYT data was 7% higher than the HAU data (where they overlapped in time), this apparently being some mad notion of deeming power generated at Hau Nui being as if supplied from Greytown, with a 7% loss (an incorrect estimate process at that) being offset. One imaginees that most of the power from Hau Nui gets no further north than Martinborough, which is a long way from Greytown. Whatever those games, data have no longer appeared since 2012, yet the wind farm also continues to operate.
09/04/2009-31/5/2022 GEN.Wind.Makara.Line2
07/03/2009-31/5/2022 GEN.Wind.Makara.Line3
01/09/2015-31/5/2022 GEN.Wind.Ohariu
15/08/2001-31/5/2022 GEN.Wind.Tararua.Bunnythorpe
24/02/2007-31/5/2022 GEN.Wind.Tararua.Grid
01/12/1999-31/5/2022 GEN.Wind.Tararua.Linton
21/07/2004-31/5/2022 GEN.Wind.Tararua.TeApiti
88/09/2008-31/5/2022 GEN.Wind.Tararua.TeRereHau.Grid
19/07/2021-31/5/2022 GEN.Wind.Tararua.Turitea
08/11/2010-31/5/2022 GEN.Wind.TeUku
16/11/2020-31/5/2022 GEN.Wind.Waipipi
01/06/2007-31/5/2022 GEN.Wind.WhiteHill
In these data series, "Makara" is also known as "West Wind" currently, and "Ohariu" as "Mill Creek". I prefer to use names related to location, because the location doesn't change - while the name of a company or an owner does. Data supplied via the Electricity authority endlessly churns through name changes, with obvious risks: that some data are lost, and, that some stations are counted more than once via different names. For instance, provoked by this, Gnash records GEN.Wind.HauNui.N and GEN.Wind.HauNui.Notional as parallel series for Hau Nui - these series should *not* be added together in an attempt to ascertain the total generation from Hau Nui.

There are multiple locations for which the Electricity authority does not supply data, and it doesn't want to make any effort to do so. Nor does it care to seek out the missing data on the early operation of a number of wind farms, even when other data indicate that there must be such generation. Thus, load supplied from the Linton substation drops even though there are no generation data for the first few months. Similarly at Wilton due to early generation from Ohariu. Discrepancies in the accounting are of no concern to the Electricity authority.
Notably, generation at the Mahinerangi wind farm joins hydro generation from the Deep Stream project in the local network connected to Waipori, itself linked to multiple hydro generators. This generation may or may not augment the generation data related to Waipori. Some power enters the national grid at Berwick, and some continues by a tripled local transmission line to Dunedin, with offshoots to Mosgiel, etc. along the way. A tangle. But whichever, the wind generation at Mahinerangi is not identified as such.
Similarly, the wind farm associated with the Horseshoe Bend hydro power station is not separately identified. The Teviot scheme's accountancy is unclear as well, though not as confusing as at Waipori: its data also ends with March 2008...

Then there is wind generation entirely ignored, and not just on the Chatham Islands. Wind farms such as Flat Hill (by Bluff), Mount Stewart (SW of Waipori), Lulworth and Weld Cone (S of lake Grassmere) plus lone turbines such as at Gebbie's Pass and Southbridge and Seaview. These last won't have much effect, but once one starts on completeness,...
This was why I asked whether your data includes estimates of annual production, as for locations without explicit metered data.

In other words, my calculation's details are explicitly knowable, and the summation is clearly deficient. What of yours?

Yours sincerely,
Nicky McLean

Dear Ministerial Services,
Ms. Katty Lau of the Ombudsperson's office has been pursuing this matter, with the latest reference number of 0014609 (leading zero digits seem important to them), and has passed on your response, which was to name the wind turbine installations whose generation data forms a part of the annual energy totals from wind generation. The list is
Brooklyn Wind Turbine
Chatham Islands Wind
Flat Hill
Gebbies Pass
Harapaki
Hau Nui
Horseshoe Bend
Kaiwera Downs (Stage 1)
Lulworth Wind
Mahinerangi Wind Farm
Mill Creek
Mt Stuart Wind Farm
Southbridge Wind
Tararua Wind Farm (three stages)
Te Apiti Windfarm
Te Rere Hau
Te Uku
Turitea Wind Farm
Waipipi Wind Farm
Weld Cone Wind
West Wind
White Hill

It can be seen that the list includes locations for which no data have been published by the Electricity authority, or for which data have ceased being published even though the locations are still in operation. These locations are as named in my earlier detailed description of the problem (I use slightly different names in some cases), and there is one surprise in your list: the wind farm on the Chatham Islands, a place apparently of no interest to the Electricity authority but cherished by MBIE.

Your list explicitly does not include information on the annual generation at each site, and presumably, this is intentional. As a result there can be no comparison between my figures for annual generation and yours with a view to identifying matches and concentrating on discrepancies, the object of my initial request.

After an earlier meeting with Mr. Gareth Wilson and Gerard van Rensberg I supplied a copy of a data file PowerStations.kmz which locates and names power stations, and it fingers a few additional wind turbines not in your list - but they are small, single-turbine sites and so won't make much difference. (My lists do include indications of individual capacity)

My full objective in this matter if to obtain the standard half-hourly generation data that are missing, and I shall be attempting to prod the Electricity authority into doing a proper job. There is a lot of interest in the actual capacity of various sorts of electric power generation, and I am surprised that your organisation has not insisted on being supplied with comprehensive data.

Yours sincerely,
Nicky McLean

Dear Ministerial Services,
Belatedly, the skull sludge finally remembered a wind farm I hadn't named, nor does it appear in your list. It is called Awhitu, near the coast south-west of Genbrook around 37°15'51"S 174°39'37"E and its construction appears to have faltered again. Two wind turbines are visible in a satellite image of August 2023, and there is what looks like a foundation pad for a third nearby. The wind farm may now be called Waiuku, which town is not much closer than Glenbrook. With wind turbines in place, one would imagine that power generation would follow, but, no half-hourly generation data have been published for it - that I can identify, anyway.

The file PowerStations.kmz shows it as "Awhitu?"

Yours faithfully,
Nicky McLean