A public-facing analysis of Pakistan's exposure to re-gasified LNG supply shocks — and the curtailed domestic gas lever that can recover 20–28 GWh/day during disruption. Data: NEPRA monthly generation reports and NTDC SCED curtailment sheets, May–July 2025.
The system is stressed — but the curtailed gas lever changes the math.
RLNG supplied 16–18% of total electricity generation across June–July 2025. An hourly supply-stack simulation shows a complete RLNG blockade forces peak shedding of 2,851 MW at midnight under the full mitigation package. Reallocating 200 mmcfd of domestic gas from CNG, industrial captive, and non-critical sectors into the dual-fuel CCGT fleet cuts that peak by 25–51% across scenarios and the daily energy shortfall by a similar margin.
16–18%
RLNG share of generation
Jun–Jul 2025, verified against NEPRA
3,823 MW
Peak hourly RLNG load
00:00 — when solar is gone and demand stays high
20 GWh
Daily curtailed gas recoverable
Central estimate, 200 mmcfd reallocation
2,851 MW
Peak shed — 100% blockade, w/ gas lever
Down from 3,823 MW without gas (−25%)
Why the gas lever is the first-order policy action. Unlike coal substitution, reallocated gas flows into the same RLNG-fired CCGTs that provide grid stability. The plants keep running, frequency regulation is preserved, and only the upstream fuel tap switches. Coal plants cannot do this — wrong ramp rate, wrong location, wrong role. A 3-hour MoE directive to SNGPL/SSGC recovers roughly 20 GWh/day. Coal acceleration over weeks recovers 1 GWh/day. The gas lever is 20× faster and 20× larger.
Data integrity note. This dashboard uses verified NEPRA figures for March, June, and July 2025. Peak MW shedding is calculated via an hourly supply-stack dispatch simulation (not a daily energy average × peak factor). The May 2025 figure in the underlying memo was interpolated (no NEPRA file was available in source data) and has been corrected from 8,500 GWh to a seasonally-adjusted ~11,200 GWh. Cargo estimates use electrical-output conversion (~380 GWh/cargo at 42% avg efficiency), not the thermal 700 GWh/cargo figure used in the earlier memo.
Part 1 — How Dependent Are We?
RLNG share of generation, month by month.
In absolute terms, RLNG generation rose from 1,528 GWh in March to 2,438 GWh in July — tracking the summer demand ramp. As a share of generation, RLNG is remarkably stable at 16–18% despite total generation nearly doubling. This means the system cannot grow out of its RLNG dependency without new baseload capacity.
Monthly generation mix (GWh)
RLNG as share of total (%)
Source-wise generation, July 2025 (NEPRA verified)
Source
GWh
Share
YoY
MoM
Hydel
5,668
40.1%
+6%
+5%
RLNG
2,438
17.3%
-18%
+10%
Coal (Local)
1,503
10.6%
0%
0%
Nuclear
1,405
9.9%
-29%
+2%
Coal (Imported)
1,140
8.1%
0%
-18%
Gas
1,093
7.7%
-7%
+13%
Wind
592
4.2%
+33%
+13%
RFO
108
0.8%
+6%
-28%
Solar
105
0.7%
-5%
-1%
Others
71
0.5%
-30%
-13%
Total
14,123
100%
-5%
+3%
Part 2 — The Hidden RLNG Layer
Not all RLNG is replaceable. This is why coal isn't a clean swap.
RLNG generation has two distinct components: demand-driven (fills consumer load at prevailing tariffs) and OMO-driven — Out-of-Merit-Order generation that exists solely to maintain grid voltage and frequency stability. OMO RLNG cannot be replaced by coal because coal plants are being curtailed precisely to hold the grid stable. This distinction is the reason simple "switch to coal" solutions fail in practice.
Monthly OMO curtailments (GWh, from NTDC SCED)
Why plants are curtailed (stability vs. congestion)
What OMO actually looks like on the ground. In May 2025, China Hubco (coal) was curtailed on 18 separate days to honor HVDC+HVAC safe evacuation under the SCS strategy table. In July, Liberty Power was curtailed on 4 consecutive days to avoid overloading the 220 kV GKR–NGKR circuit. Each curtailment is replaced by RLNG because it's the only dispatchable source with the ramp rate to hold frequency. Grid reinforcement — not fuel switching — is the structural fix.
Grid code clauses invoked (for technical readers)
OC 7.5.1, OC 7.5.9: Stability considerations and must-run generation
SDC 1.7.4, SDC 1.7.3: System dispatch code provisions for economic dispatch deviation
OC 6.7.2, OC 3.1.1: Transmission line loading limits and congestion management
Part 3 — The Cargo Math
At 6–7 cargoes per month, there is no buffer.
Port Qasim's regasification terminal runs at 5 MTPA nameplate, which translates to roughly 7 standard LNG cargoes per month. Pakistan's power sector alone needs 6–7 in peak summer — leaving essentially no buffer for shipping delays, weather, or geopolitical friction. One missed cargo (~20 days of dispatch) triggers Scenario 2 shedding.
Conversion: 1 standard LNG cargo ≈ 65,000 tonnes × 14 MWh/tonne thermal × 42% avg electrical efficiency ≈ 382 GWh electricity. Port Qasim 5 MTPA capacity ≈ 7 cargoes/month at standard size.
Part 3.5 — The Curtailed Gas Lever
Pakistan has 200 mmcfd of hidden swing capacity.
Domestic natural gas is currently allocated across seven priority tiers by SNGPL and SSGC. Power sits third, below household and commercial. During an RLNG disruption, 150–250 mmcfd can be emergency-reallocated from lower-priority tiers — CNG stations, industrial captive power, cement/general industry, and non-critical fertilizer units — into the power sector's dual-fuel CCGT fleet. Precedent exists: the 2022–23 winter gas load management exercise recovered similar volumes for priority use.
Current gas allocation by sector (%)
Emergency reallocation potential (mmcfd)
Reallocation ladder — what each source yields
Source sector
mmcfd (range)
Electrical output (GWh/day)
Political feasibility
Time to activate
CNG stations
40–60
5.2–7.8
High — precedent exists
24–48 hours
Industrial captive power
60–100
7.8–13.0
Medium — compensation likely needed
3–7 days
Cement / general industry
20–40
2.6–5.2
High — seasonal flexibility
3–5 days
Non-critical fertilizer
20–40
2.6–5.2
Lower — food-security concerns
7–14 days
Commercial / misc.
10–20
1.3–2.6
Medium — ad-hoc measures
2–5 days
Total reallocation potential
150–260
19.5–33.8
Conservative central case: ~200 mmcfd → 20 GWh/day
Why this lever beats coal ramp-up. Unlike coal substitution, reallocated gas flows into the same RLNG-fired CCGT plants that currently provide grid stability (OMO) services. The plants keep running, the frequency/voltage regulation capability is preserved, and the only operational change is switching the upstream fuel tap from the RLNG terminal to the domestic pipeline network. Coal plants cannot do this — their ramp rates and location make them ill-suited for OMO. This is why the gas lever carries 4× the mitigation weight of the coal lever (20 GWh/day vs. 0.9 GWh/day).
Caveats on the gas lever. (1) Reallocation is a policy action, not an operational toggle — it requires explicit government direction to SNGPL/SSGC and compensation frameworks for affected sectors. (2) Fertilizer diversion during urea production season carries food-security implications and is the hardest tier to access. (3) Winter reallocation potential is roughly 40% lower due to domestic heating priority. (4) Industrial captive diversion may cause knock-on manufacturing output losses that partially offset grid savings — included in the scenario cost figures below.
Part 4 — Scenario Analysis (Hourly Supply-Stack Model)
What happens when cargoes don't arrive?
Three disruption scenarios are modeled using an hourly supply-stack dispatch simulation. For each hour, we calculate RLNG lost (from July's verified hourly profile) against a mitigation stack of gas reallocation, coal ramp, and HFO. Residual shortfall at each hour becomes load shedding at that hour. Click a scenario to see the impact.
939 MW
Peak hourly shed (no gas: 1,912)
16.1 GWh
Daily energy shed (no gas: 36.1)
~805k
Households at peak (60% resi share)
PKR 4–6B
Monthly economic cost
Hourly mitigation stack — RLNG lost vs. each offsetting lever
Stacked bars show how much RLNG loss at each hour is absorbed by gas reallocation (green), coal ramp (amber), and HFO (blue). The red line on top is residual load shedding. Middle-of-day hours have no coal/HFO mitigation (those plants are already running at baseline), which is why gas reallocation is the dominant lever during 08:00–16:00.
Shedding profile — with vs. without gas lever
Monthly cost breakdown (PKR billion)
Scenario comparison — with vs. without gas reallocation
Scenario
Daily loss
Peak MW (no gas)
Peak MW (w/ gas)
Peak reduction
Daily GWh shed
Monthly cost
Mild · 50% loss
39.3 GWh
1,912
939
−51%
16.1 (was 36.1)
PKR 4–6B
Moderate · 70% loss
55.1 GWh
2,676
1,704
−36%
31.9 (was 51.9)
PKR 8–12B
Severe · 100% blockade
78.6 GWh
3,823
2,851
−25%
55.4 (was 75.4)
PKR 15–22B
Why gas efficacy falls as disruption deepens. Gas reallocation delivers a fixed 20 GWh/day regardless of scenario severity. At 50% loss (39 GWh/day gap), that's 51% of the problem solved. At 100% blockade (79 GWh/day gap), it's 25%. The implication: gas reallocation is most effective as a first-response lever — it can keep a mild-to-moderate disruption from escalating into crisis-grade shedding. For severe blockades, gas must be paired with structural measures (industrial load-shifting, premium dispatch, rotational schedules). Peak shedding concentrates at midnight because that's when RLNG runs hardest — solar is absent, demand stays elevated, and gas reallocation hits its plant-capacity ceiling.
Part 5 — Your Scenario
Build a custom disruption scenario.
Adjust the sliders to model any disruption size and duration. The calculator uses the same mitigation assumptions as the scenario analysis (coal ramp of 0.9 GWh/day, 500 MW HFO emergency deployment, industrial load-shifting).
50%
15
150
10%
PKR 2.3B
Total economic cost across the disruption window.
Load shedding peak: ~500 MW · Affected households: ~140,000 · Total energy shed: ~475 GWh · Gas contribution: ~20 GWh/day
Part 6 — Why Peak Hour Is The Binding Constraint
3× the volatility, 2× the cost, one shared dependency.
Marginal cost at peak (17:00–23:00) averages PKR 18.66/kWh — roughly double the morning rate and with 3× the volatility. This is the signature of a system running without reserve margin: every incremental MW pulls from the most expensive plant still available, which is almost always RLNG-fired.
July 2025 — Hourly demand vs. marginal cost
Source: NTDC Security Constrained Economic Dispatch, July 2025. Demand peaks at 21,238 MW at hour 23; marginal cost peaks at PKR 27.83/kWh at the same hour.
PKR 18.66
Avg peak MC (17–23h)
per kWh
PKR 9.26
Avg morning MC (06–12h)
per kWh
36.6%
Peak hour volatility (CoV)
3× morning volatility
Part 7 — Mitigation Roadmap
PKR 1.5–2B of prep, PKR 5–8B/month of avoided losses.
A well-coordinated mitigation package — coal supply acceleration, industrial load-shifting contracts, SCED optimization, and grid reinforcement — pays back in 2–3 months under any sustained disruption scenario. The specific actions and timelines:
Days 0–3 · Immediate (highest impact)
Emergency gas reallocation via SNGPL / SSGC directive
Issue MoE directive to SNGPL and SSGC for emergency reallocation of 150–200 mmcfd from CNG, industrial captive, and cement sectors to power sector CCGTs. Expected yield: +20 GWh/day within 48–72 hours. Activate compensation framework for diverted sectors (PKR 3–5B/month standby fund). This single action reduces peak load shedding by 26–52% across scenarios.
Days 0–7 · Immediate (secondary)
Coal procurement & load-shedding protocols
Activate spot market purchases for 0.5M tonnes from South Africa, Indonesia, or Australia (PKR 500M–1B). Issue NTDC circular with Scenario 1/2/3 shedding templates to all DISCOs with 48-hour implementation authority. Brief Minister, SAPM Energy, and PM's office. Brief HFO-capable plants (Kotri, Guddu, Sukkur) for 500 MW standby.
Reduce single-source Qatar dependency to 80–85% via spot and medium-term contracts (Australia, USA, Oman). Negotiate 3-month demand curtailment agreements with 5–8 major industrials (Hubco, Fauji Fertilizer, DG Khan Cement, Lucky Cement) for 300–500 MW flexible load. Coal plant readiness surge: +1.2–1.5 GWh/day online by week 6.
Months 3–12 · Structural
Gas pipeline capacity, LNG terminal expansion, grid reinforcement
Pre-position gas-reallocation protocols (standing MoE notification to SNGPL/SSGC) so the 48-hour trigger becomes 12-hour. Port Qasim regasification capacity expansion from 5 to 7–8 MTPA ($150–200M, 18–24 months). High-impact 220/500 kV grid reinforcement (GKR–NGKR, Gatti Autos, Multan Autos) to reduce OMO requirement below 1% ($400–600M). NTDC-controlled demand response platform: 50–100 MW fast-response capacity at PKR 5–10/kWh.
Part 8 — Escalation Protocol
When to pull which lever.
Three escalation triggers tied to observable cargo-arrival signals. Each level maps to a pre-defined response with pre-authorized spending and shedding authority.
Level 1 · Advisory
Cargo delay > 10 days
Activate coal supply acceleration task force. Brief Energy Minister and NTDC MD. Monitor buffer stock — trigger alert if < 10 days supply remaining. No consumer impact at this stage.
Level 2 · Precautionary
1 cargo lost (~20 days supply)
Issue Scenario 2 (70% loss) shedding schedules to DISCOs within 72 hours. Engage major industrial consumers with load-shifting contracts. Activate emergency coal procurement. Brief cabinet.
Level 3 · Emergency
2+ cargoes lost (6+ weeks)
Implement full Scenario 1 (100% loss) load shedding nationwide. Activate all coal, HFO, and hydel resources. Public energy emergency declaration. Coordinate with State Bank on forex reserves for alternate LNG procurement.
Methodology & Data Sources
How we arrived at these numbers.
Step 1 — RLNG generation (NEPRA verified)
Monthly RLNG dispatch is sourced directly from NEPRA Monthly Power Generation Reports. Verified values: March 2025 = 1,528 GWh, June 2025 = 2,216 GWh, July 2025 = 2,438 GWh. The May 2025 figure is interpolated (no NEPRA file was published/available in the source dataset for that month) at ~2,000 GWh based on the seasonal Mar→Jun ramp.
Step 2 — OMO curtailments (NTDC SCED)
Out-of-merit-order generation is proxied by summing all plant-level curtailments reported in NTDC's monthly Security Constrained Economic Dispatch (SCED) sheets. Curtailments are driven primarily by grid stability (OC 7.5.1, OC 7.5.9) and transmission congestion (OC 6.7.2, OC 3.1.1). When a coal or hydel plant is curtailed, the replacement generation is typically RLNG — the only source with sufficient ramp rate to hold frequency stable.
Step 3 — Cargo conversion
Standard LNG cargo size: 65,000–75,000 tonnes. Thermal energy content: ~14 MWh/tonne (LHV basis). At Pakistan's fleet-average CCGT efficiency of 42%, electrical output per cargo is ~380 GWh. Port Qasim 5 MTPA nameplate ≈ 7 standard cargoes per month. Earlier analyses using 700 GWh/cargo conflated thermal energy with electrical output and should be treated as upper-bound only.
Step 4 — Curtailed gas reallocation (central addition)
Pakistan's domestic gas (SNGPL + SSGC aggregate) is allocated across seven priority tiers. Under the Economic Coordination Committee (ECC) gas allocation framework, reallocation during emergencies is permissible with MoE direction. Reallocation potential by sector (summer conditions): CNG 40–60 mmcfd (high feasibility), industrial captive 60–100 mmcfd (medium), cement/general 20–40 mmcfd (high), non-critical fertilizer 20–40 mmcfd (lower feasibility), commercial/misc. 10–20 mmcfd. Total central-case: 200 mmcfd. Conversion to electrical output: 200 × 1.03 mmbtu/mmcfd ÷ 8,000 btu/kWh ≈ 26 GWh/day gross; after plant efficiency and pipeline losses, we use 20 GWh/day as the conservative central case applied to all scenarios.
Step 5 — Hourly supply-stack model (revised)
Peak MW shedding is now calculated using an hourly supply-stack dispatch simulation, not a daily energy average. For each of the 24 hours:
Step A: Derive the July 2025 hourly RLNG profile by scaling NTDC's June thermal hourly shape to match July's verified 2,438 GWh monthly RLNG total. Peak RLNG = 3,823 MW at 00:00; minimum = 2,726 MW at 12:00.
Step C: Apply mitigation stack in merit order — (1) gas reallocation: 20 GWh/day budget dispatched proportionally to hourly loss, capped at 4,000 MW plant capacity ceiling; (2) coal ramp: +200 MW during 17:00–22:00 (peak hours only, reflects slow ramp physics); (3) HFO emergency: +500 MW during 19:00–22:00 (peakers).
Step D: Residual shed at hour h = RLNG lost at h − gas at h − coal at h − HFO at h.
Peak MW shed = maximum over 24 hours. Daily GWh shed = sum over 24 hours.
This replaces the earlier approach that approximated peak MW from daily energy deficit × a peak factor, which understated true peak by ~35%. The memo's original figure of 2,100 MW for 100% blockade is now corrected to 2,851 MW under the same mitigation assumptions.
Households experiencing blackout at peak hour = peak MW shed × 1,000 × 0.60 (residential share of shedding) ÷ 0.70 (kW average peak residential load). This gives a simultaneous-snapshot figure. Rotational shedding means different households experience blackout over the course of the day; total unique households over 24 hours can be 3–5× this snapshot. The dashboard reports the peak-hour snapshot.
Step 7 — Economic cost (revised)
Monthly cost components: (1) industrial output loss — dominant component at higher scenarios, calibrated at ~PKR 0.20B per GWh shed (cement, steel, textile, commercial); (2) gas-reallocation compensation to diverted sectors — ~PKR 1.2B/month for 200 mmcfd (CNG, captive, cement, commercial); (3) emergency fuel procurement premium — spot coal and HFO price differential; (4) coal spot premium — for accelerated 0.5M tonne procurement; (5) grid ancillary services — frequency regulation, voltage support, blackstart preparation.
Known limitations & caveats
May 2025 total generation is interpolated, not observed. Refer to June/July figures for verified analysis.
OMO replacement by RLNG is conservatively assumed 1:1. In practice, some curtailments may be offset by reduced demand or imports.
Gas reallocation potential is based on summer conditions. Winter potential is ~40% lower due to domestic heating priority. Fertilizer diversion during urea-production seasons carries food-security costs not fully priced in.
Industrial output loss assumes no demand-side response beyond contracted load-shifting. Informal load curtailment could materially reduce this loss.
Cargo conversion uses fleet-average efficiency. Specific plant-level calculations may vary ±15%.
Economic cost excludes second-order effects (tariff adjustment, circular debt escalation, social impact from diverted-sector output losses).
Gas reallocation assumes SNGPL/SSGC pipeline capacity to redirect volumes — broadly true in summer when domestic demand is lower, but localized pipeline bottlenecks could constrain specific plant access.