Data for region Newfoundland (Canada)

[q--]

Current generation

Current Island System Generation

so for Newfoundland only

It seems there is an interconnector, yet, curiously the page linked at the beginning says "Current system generation = current demand on the system". This may or may not mean imports are included in the figure

---

Generation + forecast data of only a few days old is available in PDF format here.

Edit: the graph in this PDF does include imports/exports via the interconnector

Edit 2: the graphs in the PDF are in a vector format; I could extract them with Inkscape - here's the green line from the Jan 1, 2024 report as an SVG image:

---

Installed capacity, from the TSO website (Via here - "The Newfoundland and Labrador System Operator (NLSO) is responsible for the safe and reliable operation of the bulk electric system in Newfoundland and Labrador, including the administration and provision of transmission on the NL Transmission System pursuant to the Multi-Party Pooling Agreement (MPPA).")

This one seems to be for Labrador as well, then.

Page 14 (16 in the PDF) of this has a map with installed generating facilities, including their type. Coupled with the installed capacity document linked above, it should be possible to determine what type of generation the facilities are and where they are located.

This page also mentions installed capacity. The grid map it links to is dated 2011 instead of the one from 2017 in the previous link.

---

"Energy Supply Monthly Reports" which have the generation by unit in GWh, imports, and data on electricity bought from other parties, both for the specified month and YTD, which should be helpful when trying to calculate the emission factor. (Appendix B, last page.)

---

Possibly interesting: quarterly reports, haven't looked through these yet

[jarek]

Still no fuel data, eh?

The Maritime Link to Nova Scotia is fairly new. Kind of sad if they didn't manage to export much on it.

[pierresegonne]

Hey @q-- Thanks for the investigation and the breakdown,

From what I see, getting the live data is only possible through the first link which shows the current generation = demand (let's ignore the interconnector at first). With the sources you have found, it would be possible as a first approximation to estimate the breakdown of capacity by production type and apply it on the live demand as unknown production type with a custom co2eq factor.

Then if the data sources from Newfoundland become better, we could improve on the parser/production breakdown.

The task could therefore be broken down into:

• [x] Create zone for Newfoundland -> #3092
• [ ] Determine average production breakdown by production type [EDIT] Could use https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=2510001501&pickMembers%5B0%5D=1.2&pickMembers%5B1%5D=2.1 to make it easier
• [x] Create parser for live generation from https://nlhydro.com/system-information/supply-and-demand/ probably needs beautiful soup to parse the html -> #3077
• [ ] Compute average co2eq factor for unknown

Any comments/objections/improvements?

[q--]

Looks good. It's unfortunate that the generation data from statcan is Newfoundland+Labrador instead of just Newfoundland, but since it's nearly all hydro and only a little thermal the impact won't be too much.

Also, I noticed you used data for Jan 2017-March 2020 in #2565, I was thinking it might be best to take just one year to account for any changes in installed capacity, what do you think?
In any case I'd use 12-month groups of data instead of using January as the cutoff point to account for seasonal differences in electricity demand. E.g. May 2019-April 2020

By the way, is there a guide to calculating the static carbon intensity somewhere?

---

Edit: just discovered that you could get the biomass/non-renewable breakdown by enabling these options at statcan. This data is only available as of January 2020, though

---

Here's a link to my customized selection at statcan

[q--]

I also found the gas/oil breakdown for the thermal bit. Data on that site seems up to 2019 so sadly doesn't account for the biomass part.

I tried tampering with the dataset parameter to set it to a month in 2020, but that did not change the output.

[jarek]

Please remember that the Statscan generation data is for Newfoundland and Labrador. Bulk of the generation is a giant hydro plant in Labrador which is only connected to Quebec's grid, the 5.5 GW https://en.wikipedia.org/wiki/Churchill_Falls_Generating_Station

Per Statscan's Table 25-10-0021-01 (previously CANSIM Table 127-0008), about 70% of all generation in the province is exported - most of it to Quebec and never reaches Newfoundland: https://www150.statcan.gc.ca/t1/tbl1/en/cv!recreate.action?pid=2510002101&selectedNodeIds=1D2,3D1&checkedLevels=1D1&refPeriods=20140101,20180101&dimensionLayouts=layout2,layout3,layout2,layout2&vectorDisplay=false (note that the undersea connection to Nova Scotia, the other possible export destination, wasn't yet live in 2018 when this data was taken)

My post from 2017 may be a little out of date but might still be helpful in parts: https://piorkowski.ca/rev/2017/06/canadian-electricity-co2-intensities/

[pierresegonne]

Hey guys :)

We have split up the NF and LB zones and with the help of the parser created by a contributor we could add LB on the live map :)

We still need to get a better static estimate of the unknown breakdown before that though

[Kongkille]

I've tried to understand the current status of this issue. We did split the zones in #3092

So i think next steps are:

1. Create a live parser for Newfoundland demand/generation
2. Compute static breakdown and CO2 factor

[Hurc2000]

My background knowledge of Newfoundland & Labrador comes from having worked on operations documentation for NLH for over 2 years preparing for the Muskrat Falls project. In that time, I had access to their systems. I can't provide any direct information I worked with, but have used that background knowledge to compile a rather complete profile of their systems, the Island Interconnected System (IIS), and the Labrador Interconnected System (LIS), from public info, with one important very random sample.

I will break the 2 systems up in separate posts, which will take a little while to organize.

[Hurc2000]

I see you know where to find the Newfoundland IIS live demand value. It does provide an accurate value based on my direct NLH observations, and I believe it is the value without the Maritime Link export added on to it as seen in the daily status reports. It's been a couple of years since I finished that work. The way to verify this is to save some spot readings and then compare them to the reports for the same time stamp. Unfortunately, NLH has not posted any reports for February 2023. Strange. If you have questions about how to read the trends, just ask.

NLH doesn't serve the complete island. There is a second entity, Newfoundland Power, which also serves some customers and does own smaller amounts of generation and transmission. However, all of that info is reported to NLH for demand and source reporting. Unfortunately, none of this resource mix is reported publicly.

The Island's generating capacity sources are as follows:

[Hurc2000]

Now here is the useful part. Over the course of my work, I captured three random screen caps of the resource mix on the IIS. This was pre-LIL deliveries but ML was in service, very lightly used. However, there were a few instances where NF sold some hydro during the low demand days, and purchased the odd time when oil was running.

So note, little to no oil in the low demand months, say under 800 MW of IIS demand.

I have 3 samples of overview. I copied the data as it was displayed. These point to about 20% oil use during the higher demand periods. With LIL now in service, NF does receive some benefit to reduce oil once the NS allocation is met. Hence I have added the last column displacing 100 MW of oil with excess LIL delivery.

Edit: Should have provided some date info for the samples. In sequence, Feb 2018, Nov 2019, Feb 2020.

Based on this, I'd say using the 12-15% oil for a general mix CO2 number is a good estimate with LIL limited to single pole operation up to 430 MW. This leaves 100 MW or so to displace NF oil. Sure it will be higher with lower LIL deliveries or higher demand, but will also be zeroed many low demand days.

[Hurc2000]

The Labrador system is a completely different animal. Call it 100% hydro. Diesel sources are only used for transmission loss backup or in isolated communities.

The LIS, including Muskrat Falls, is synchronously tied to the Quebec grid and any excess generation flows out the three 735 kV lines from Churchill Falls. Demand peaks at just over 400 MW in the winter, with a low demand of about 175 MW. There is no public data that I am aware of. The Labrador Island Link HVDC is set at whatever flow is planned or permitted to the Island.

Something to be aware of if making a static CO2 and generation profile here, Quebec counts the delivery on the lines as its internal generation. In fact, Churchill Falls is currently included on Electricity Maps as part of Quebec's total hydro. https://www.hydroquebec.com/generation/generating-stations.html

That's all for now.

[Hurc2000]

Just thought of something else to use for Island energy mix, but unfortunately may miss some non-NLH sources. However, it can produce an oil % estimate too.

From the 2021 NLH annual report. https://nlhydro.com/wp-content/uploads/2022/06/Newfoundland-and-Labrador-Hydro-2021-Annual-Report.pdf#page=37

Remove Churchill Falls from the equation and Menihek. Some Muskrat Falls energy was likely stranded in Labrador but most would head to the Island. Using the full amount for the Island can be a rough proxy for energy perhaps not captured in the report. Also, most of the diesel burn would be remote communities in Labrador but we'll count for the Island as well.

This equation results in an Island oil sourced energy of 10%. It will only go down from there in the future.

I won't dwell on this further. Now searching for some tidbits for other areas. Cheers!

[jarek]

For energy generated in 2021, I calculate 34253 + 4495 + 727 + 45 + 1988 + 44 to be 41552 but the total is given as 39564, do you know why?

[VIKTORVAV99]

For energy generated in 2021, I calculate 34253 + 4495 + 727 + 45 + 1988 + 44 to be 41552 but the total is given as 39564, do you know why?

If you exclude Muskat Falls the figures line up so maybe they are double counting this somehow in the individual categories but not in the total? That's just a guess on my part though.

[Hurc2000]

It certainly looks like Muskrat Falls is not included, and there may be a good reason why. With the delays in the commissioning of the plant itself, and the ongoing limitations of the Labrador-Island Link, they are at contract commitment limits of the Nova Scotia block of energy delivery commitments. Nova Scotia paid for the Martime Link, without which Newfoundland could not handle the LIL delivery losses to the island. The ML will automatically provide frequency support as it is the grid forming HVdc Voltage Sourced Converter technology.

You will see that contractual relationship in the daily reports, LIL to ML, as I have already pointed out. Nova Scotia has paid for this energy up front.

From a 2021 news item: "Nova Scotia Power ratepayers have been paying for the $1.57-billion Maritime Link since 2018, even without receiving a single electron from Muskrat Falls until a few weeks ago.

They paid $109 million in 2018, $115 million in 2019, and $144 million in 2020. Customers will shell out $172 million this year for the Link, which is owned by Nova Scotia Power parent company, Emera. The cost has already been factored into rates.

The 35-year "20 for 20" deal guarantees Nova Scotia at least 20 per cent of the electricity from Muskrat Falls in return for paying 20 per cent of the cost in the form of the Maritime Link.

Emera's share of the overall cost is now considerably less than 20 per cent. It completed the Maritime Link on budget while Muskrat Falls costs have exploded by billions of dollars. An original budget of $6.2 billion is now $10.2 billion."

As far as the annual reduction in the oil (thermal) use the last few years, up to 2021 it has been related to reductions of heavy industrial loads such as the Come By Chance refinery rather than Muskrat Falls energy.

[Hurc2000]

I did the Island demand comparison and that data point is the total demand without the Maritime Link export added to it.

[q--]

Interestingly, the daily reports previously used to call the purple line "Island System Demand Plus LIL/ML Exports", but now it's called "Island System Demand Plus ML Exports". (It was still called the former in January 2023, but in June 2023 (and possibly earlier) it was called the latter.) If that means we can be sure that exports are only through the Maritime Link interconnector, this data (if a parser is ever built) could be used to fix the problem that we don't know whether Nova Scotia is importing from New Brunswick or Newfoundland anymore since the ML link to Nova Scotia went live.¹ Well, looks like we get the data on the NS-NB interconnection from New Brunswick Power, not from NSPower. So we don't actually have to worry about that.

---

Also interesting: the graphs in the PDF are in a vector format; I could extract them with Inkscape - here's the green line from the Jan 1, 2024 report as an SVG image: . This should make it easier to build a parser based on the PDF reports.

---

¹ Update: A different source providing real-time, but not historical, data is also available, see https://github.com/electricitymaps/electricitymaps-contrib/issues/6317

[q--]

More good news: there are "Energy Supply Monthly Reports" which have the generation by unit in GWh, imports, and data on electricity bought from other parties, both for the specified month and YTD, which should be helpful when trying to calculate the emission factor.

[jarek]

As pointed out on https://github.com/electricitymaps/electricitymaps-contrib/pull/5100, the real-time Newfoundland reporting page has been updated and now has:

• Newfoundland ("Island System") demand, with note "This number does not include what we are generating to supply customers in Labrador or our export commitments for the Maritime Link and Hydro Quebec." so I take it's the actual Newfoundland consumption. Currently it's 762 MW
• Muskrat Falls generation: "This power travels across LIL [Labrador-Island Link] to serve customers on the Island Interconnected System and to fulfill our commitments to Emera (as power flows over the Maritime Link to our neighbours in Nova Scotia). Generation that is surplus to these commitments can be used to serve customers in Labrador and/or exported to external markets in Canada and the US." Right now it's reported as 464 MW
• Churchill Falls generation, this is mostly going to Quebec (per comments about it might even be accounted for in Quebec reporting): "Approximately 90 percent of this power flows to Quebec and neighboring markets in Canada and the US, while the remaining power serves customers in Labrador." - doesn't say anything about going over LIL to Newfoundland

Some of these are accounting distinctions rather than physical realities.

We could assume that Labrador generation is equal to Muskrat Falls generation if we assume Churchill Falls is reported by Quebec - there aren't any other major generating stations in Labrador: no hydro larger than 20 MW, there's one 25 MW diesel turbine in Happy Valley, and a 8 MW diesel gen in L'Anse au Loop - pretty much dwarfed by Muskrat's Falls 824 MW.

But there's still missing data for our purposes:

• Labrador demand is not known, thus we don't know the exact amount of Labrador-Island interchange. It is certainly relatively small, but I don't know how much we want to assume
• This page doesn't tell us how much is going from Newfoundland to Nova Scotia. Potentially the Nova Scotia parser could tell us but we'd have to double-check
• We still don't know actual Island generation, nor what it's made up of. We could assume that vast majority of Muskrat Falls is going to Newfoundland and if Nova Scotia reports its transfer, then Island generation would be Island demand + Nova Scotia exchange - Muskrat Falls generation, but it's still not broken down into sources so the carbon intensity would be only estimated from medium-term statistics as well.

So there still isn't much real-time data here other than Newfoundland (CA-NL-NF) consumption.