Known Limitations and Assumptions
Every number here is only as fresh as its source dashboard's last build
The Desk doesn't fetch anything live — at build time it pulls each source
dashboard's own latest published data (via a private mirror of each
dashboard's data store, data_kit.ensure_lake()), so a KPI here is exactly
as current as that source dashboard's own most recent successful rebuild,
not "as of right now." If a source dashboard's own last refresh failed or
is running behind schedule, the Desk's figure for it will lag until that
dashboard catches up — the Desk itself doesn't know it's showing a stale
figure and won't visibly flag it as such.
SIN gas generation is a sum, not a published figure
Same as ONS Balances: there is no published national "SIN" gas-generation series. This KPI sums ONS's four published subsystems, exactly as the ONS dashboard's own SIN figures do.
PLD ≠ CMO
Shown side by side here for comparison, but they're different figures from different sources — see PLD Prices for the distinction.
The spark calculator's assumptions
- 9,400 kcal/m³ for natural gas heat content is Brazil's standard calorific value, not a plant- or region-specific figure — same constant ONS Balances uses for its own estimated-gas-consumption figures.
- 1,800 / 2,500 kcal/kWh heat-rate presets are typical CCGT/OCGT industry figures, not specific to any real plant.
- The default gas price is whichever figure is available first, in this order: the latest GUS trade price, then POC's 7-day average price, then an illustrative 1.2 R$/m³ so the calculator never loads with an empty field. Whichever one populated the field is worth checking before reading too much into the default — it's a starting point for your own number, not a recommendation.
- The calculator computes an implied CVU from whatever gas price and heat rate you enter — it isn't a published CVU for any specific plant, and doesn't claim to be one. The CVU override field is there specifically so you can substitute a real plant's own published CVU instead.