We’ve had several conversations in recent days with a number of different parties (all for different reasons*) with respect to the Wambo Wind Farm.

* strange coincidences!

These conversations spiked our interest, so we invested a short amount of time to have a helicopter view, to see what we could see with respect to the WAMBOWF1 unit.

About Wambo Wind Farm

As Stanwell notes here, the Wambo Wind Farm is a 50:50 joint venture between Stanwell Corporation and Cubico Sustainable Investments.  There are two units:

1)  note that WAMBOWF1 is the first of several units to come.

2)  WAMBOWF2 is under construction and will start commissioning soon (if not already)  … but there’s nothing to see there, yet, in terms of the data.

What did we find, in the GSD2025?

Given our GSD2025 was released just under 2 months ago, and contains data to 31st December 2025 over several different pages devoted to the WAMBOWF1 unit (including the two below), I thought I’d start there with the story:

With this image:

1)  It focuses on a two-page span that we term:

(a)  On the left is the ‘A’ Page, which covers a 10-year span of (mostly) monthly data that spans 2016 to 2025 for the unit in focus

(b)  On the right is the ‘B’ Page, which covers a 1-year span of the most recent year (2025) and includes:

i.  Some charts at the top with daily data

ii.  Some tables in the middle with monthly data

iii.  Two specifically focused inserts at the bottom produced with ‘raw’ 5-minute data.

(c)  But remember that there’s also other data for this unit in other sections in the GSD2025.

2)  We’ve snipped out much of the content, in order to focus the reader of some interesting observations

3)  We’ve highlighted:

(a)  To the end of 2025 the unit only reached part-load … presumably part of the commissioning process

(b)  Also highlighted is an outage that lasted much of October 2026 … again presumably part of the commissioning process (I’ve not delved further to check).

(c)  Given that we’ve also been looking at possible instances of high-temperature de-rating of Wind Farm output on Tuesday 27th January 2026 (albeit for some wind farms in the south), there’s a slight hint here that there might have been effects seen at WAMBOWF1 through the hotter periods of calendar 2025?

4)  Our calculations for FCAS costs (and also the apportionment of FPP costs and benefits) are included to assist in other conversations

Delving deeper, in our Software

With the ‘Trends Engine’ within ez2view I also prepared this time-series view of some high-level daily statistics:

There’s a number of things that can be observed here:

1)  but I’ve particularly highlighted some large instances of Dispatch Error through the commissioning process (and perhaps related to the commissioning process).

2)  remember that:

(a) a positive Dispatch Error means under-performance, relative to Target

(b) a negative Dispatch Error means over-performance, relative to Target

(c)  Neither positive or negative is inherently ‘right’ (or ‘wrong’)

… it depends on what the frequency needs are at the time

(d)  Also with respect to the generator’s obligations to follow Targets, there are some nuances here for Semi-Scheduled units

… so please don’t read into any of this any ‘blame’ being attributed.

(e)  And readers should remember that there’s an added complexity, because commissioning was going on for unit 1 through this period.

(f)  But these instances:

i.  might flow through to system frequency implications;

ii.  and they might also flow through to costs (or benefits) the individual generator stakeholder via the recovery mechanisms (including for PFR) in the Frequency Performance Payments methods.

(g)  As Allan described well in his article here, because of the one-sided nature of the Semi-Dispatch Cap arrangements mean that:

i.  under-performance (relative to Target) is expected to occur much more frequently, and at larger magnitude,

ii.  compared to the incidence (and size) of collective over-performance

…  across the Semi-Scheduled fleet.

3)  So it’s the large negative Dispatch Error instances that particularly intrigue me.

4)  Some of the largest instances occurred in March 2026 (i.e. which can be partly understood because output has been growing through progressive hold point constraint levels).

(a)  One such instance is captured in this ‘Unit Dashboard’ widget snapshot from ez2view with the table focused at 02:20 (NEM time) on Tuesday 3rd March 2026:

(b) And there are others…

Time permitting I might explore further, but perhaps offline.