Invest1 distinct publisher3 min readUpdated
Peak demand hit 75.99 GW on August 15 against a 98.8 GW official projection. The number that matters is solar's 14.3 GW swing between a sunny peak and a rainy holiday.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
South Korea's daily peak power demand in the electricity market came in at 75.99 GW on August 15, according to the Korea Power Exchange [1], against a pre-summer official projection that this year's peak would climb to 98.8 GW [2]. That is 76.9 percent of the forecast [20], and the gap says less about forecasting accuracy than about which variable actually decides whether the grid has a bad week.
The heat itself behaved as expected. Daily peak first crossed into the 90 GW range on August 3 at 91.1 GW, then set this year's high of 95.3 GW on August 7 [4][5]. Typhoons forming around the peninsula then broke up the heat-dome high and temporarily pulled temperatures down [7], and reduced industrial load during the summer vacation season contributed as well [8]. Since August 7, daily peaks have moved sideways in a 76 to 83.7 GW band [6].
On paper the headroom is comfortable. Summer supply capacity secured by authorities stands at 107 GW, and experts quoted by Seoul Economic Daily say even a sharp demand jump is unlikely to produce a large-scale blackout [9]. Measured against this year's high, that is 11.7 GW of slack, about 12 percent [21]. The all-time record is 97.12 GW, set in the summer of 2024 [3], which sits 1.82 GW, or roughly 2 percent, above the August 7 reading [22][10].
The number that deserves the attention is on the supply side. At 12:30 p.m. on August 7, solar output reached 26.4 GW, its largest share of the mix at 26.6 percent [11]. At noon on August 15, with rain clouds over the country, solar peaked at 15.6 percent of the mix and just 12.1 GW [12]. That is a 14.3 GW difference between two days nine days apart [23], consistent with the report's estimate that weather can move solar output by 10 to 15 GW, a gap that must be filled entirely by LNG or coal [13].
Both days so far this month have been correlated in the convenient direction: hot and sunny, or wet and mild. The combination that has not occurred yet is humid and heavily overcast without the rain that suppresses cooling demand. In that case, load near the August 7 level would arrive with solar at 12 GW rather than 26, leaving dispatchable plant to cover roughly 14 GW more than it did on the highest-demand day of the summer [23]. Analysts cited in the report make the same point in weaker form: cloudy, humid days are harder to manage because solar output falls while air-conditioning demand rises [14].
Near term, the Korea Meteorological Administration expects rain clouds left by Typhoon Dolphin, the season's 13th, and Typhoon Chanhom, the 15th, to drop up to 200 millimetres, mainly in southern regions, over the Liberation Day holiday [16]. Highs around 33 degrees Celsius nationwide, including Seoul, are forecast for the following week [17], and a renewed expansion of the retreating North Pacific high could restore the double heat-dome pattern [18]. Some forecasts put the summer's peak in late August, once the holiday rain stops, the heat returns and vacations end [15]. Authorities are running an emergency response system on the assumption that the year's high could be matched if the heat comes back [19].
Watch the noon numbers rather than the daily headline peak: load against solar share, and how much thermal ramp the operator has to schedule to cover a low-irradiance afternoon. A 98.8 GW forecast against 107 GW of capacity leaves 8.2 GW of margin [24], and a 14 GW solar swing is larger than that margin.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
Experts say a sharp jump in demand is unlikely to cause a large-scale blackout because the summer power supply capacity secured by authorities stands at 107 GW.
Peak power demand in South Korea's electricity market reached 75.99 GW on August 15, according to the Korea Power Exchange.
Authorities had initially projected that this year's peak demand would climb to 98.8 GW.
The all-time high for Korean peak power demand is 97.12 GW, recorded in the summer of 2024.
Daily peak demand first entered the 90 GW range on August 3, at 91.1 GW.
Daily peak demand reached this year's high of 95.3 GW on August 7, four days after first entering the 90 GW range.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Specific official figures, single outlet, unnamed expert layer
Core load and generation numbers are precise and attributed to named institutions (Korea Power Exchange for demand and solar mix, Korea Meteorological Administration for the weather outlook), and the derived arithmetic follows directly from those figures. Evidence is capped by the cluster containing exactly one source item with no independent corroboration, and by the interpretive claims resting on unnamed 'experts', 'analysts' and 'some forecasts'.
Solar already a material, measured share of a live national grid
Adoption is observable in operating data rather than announcements: solar supplied 26.4 GW and 26.6 percent of the Korean mix at midday on the highest-demand day and 12.1 GW / 15.6 percent on a rainy day, on a system whose secured summer capacity is 107 GW and whose peaks ran 76-95.3 GW. That is real deployment at scale within one national market; it is not evidence of adoption beyond Korea, and the cluster gives no installed-capacity, storage or fleet-composition detail.
Mildly overstated: risk framing runs ahead of the observed slack
The cluster's forward-looking frame - record territory, deadly heat, a possible double heat dome and a late-August peak - is heavier than what the measured data shows, since the reported peak came in roughly 23 GW under the official projection and secured capacity of 107 GW sits about 11.7 GW above this year's high. The overstatement is modest rather than large because the same source concedes a blackout is unlikely and quantifies the genuinely tight variable: a 10-15 GW solar swing that must be backfilled by LNG or coal.
No disclosed stakes in the supplied material
The single source item discloses no funding, ownership, commercial relationship or vendor position for any actor, and the interpretive voices ('experts', 'analysts', 'some forecasts') are unnamed, so their interests cannot be characterised. Attributing motives to the authorities who both set the 98.8 GW projection and secured the 107 GW capacity would require inference the supplied text does not support.
Moderate: hard numbers, thin sourcing, unresolved forecast
Confidence is supported by the precision and institutional attribution of the load, mix and weather figures, and constrained by one-publisher sourcing, anonymous interpretive attributions, an unresolved late-August peak, and no disclosed incentive information to weigh.
invest
Seoul's debt write-off fund rests on a Western precedent analysts say is not there1 distinct publisher
invest
Korea's COFIX climbs a fourth month, pushing variable mortgage ceilings to 5.89%1 distinct publisher
invest
The Bank of Korea's North Korea desk is now an economic security desk1 distinct publisher
invest
Korea's household credit clears 2,000 trillion won just as the BOK starts hiking again1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
en.sedaily.com
1 article · August 15, 2026