Analysis: Ramokgopa plays the gas procurement trump card from his energy plan deck

What a difference a year makes. At Windaba 2025, Minister Ramokgopa spoke of electrons flowing from endless windmills. Fast-forward to the same event a year later, and that electricity will now come from gas molecules and battery chemicals first, and solar-powered sprites (wind is also solar-powered, btw) at a later date.

“We are announcing a determination of 4,600 megawatts of battery energy storage system, and we further are announcing 5,000 megawatts of gas to power,” Ramokgopa explained.

“This represents a combined allocation of 9,600 megawatts. It will make no provision for new wind or solar capacity. The purpose is to address the immediate need for storage flexibility in agile dispatchable supply whilst reducing the risk that further variable generation will increase electricity curtailment.”

He had an answer to the logical next question already loaded in his cranium. “A generating plant without dependable fuel cannot provide dependable electricity … the proposed gas allocation must therefore be implemented through a coordinated fuel infrastructure and generation programme,” he said, filling in some of the blanks about the looming gas cliff and South Africa’s lack of domestic production.

“What we'll be announcing is that we’ll be co-locating load. So instead of relying on transmission infrastructure, we are able to match the load to the power packs,” which settles the dispute about not building out the grid enough, while nullifying the benefits of chasing gas pipe dreams.

Going through changes

Oddly, Ramokgopa is deviating from the alignment between his Integrated Resource Plan (IRP) 2025 and Gwede Mantashe’s 2024 Gas Master Plan. Both frameworks specified 6,000MW of gas-fired power as part of the mix by around 2030 – with the IRP 2025 splitting the allocation evenly between Eskom and independent power producers.

The new procurement determination accelerates the plans for large-scale gas-to-power (G2P) projects to serve as an anchor load for midstream liquefied natural gas (LNG) import infrastructure.

It then advances this idea to include market-led site selection that matches G2P and storage locations directly with transmission capacity and industrial demand nodes.

Wait, but what about the gas cliff?

Glad you asked. For the non-gas nerds: Sasol’s natural gas supply from the Pande and Temane fields in Mozambique, alongside methane-rich gas supplies in KZN and Mpumalanga, will decline sharply or cease for third parties by 2026-2028.

The 5,000MW G2P determination was launched in October 2026, with major new G2P plants and LNG terminals expected to reach commercial operation in 2029–2030. Bridging this shortfall will require interim emergency LNG arrangements via Mozambique/Matola.

The alignment goes deeper still. The Industrial Gas Users Association South Africa (IGUA-SA) Gas Roadmap and its Gas Master Plan 2024 repeatedly emphasise that LNG import infrastructure cannot achieve financial close without massive, bankable anchor demand (read: offtake agreements are the currency of energy projects).

The 5,000MW G2P allocation directly matches this requirement, providing the baseline offtake (around 20-130 petajoules per year) necessary to de-risk terminal investments.

Both national planning frameworks point out that G2P is the fastest, most cost-effective dispatchable solution to back up variable renewables and balance peak grid loads – but take that with a giant cup of salt because these are gas planning frameworks.

The magic lies in the determination’s directive to align generation with gas corridors, which matches the Gas Roadmap’s designated priority corridors: Richards Bay (KZN), Rompco (Gauteng/Mpumalanga), and the West/South Coast.

Imperfect agreement

The Gas Master Plan and industry studies specify that South Africa needs approximately 12,000MW to 13,000MW of gas-fired power by the mid-2030s to fully replace retiring coal and safeguard against deep multi-day renewable droughts (days when the wind doesn’t blow or the sun doesn’t shine – Mantashe gets very dramatic about baseload and dispatchable power).

This current 5,000MW procurement covers only the initial phase of national demand. Additionally, the industrial users’ Gas RoadMap envisages imported LNG as a short-term bridge (2027-2035), transitioning to domestic West Coast gas (285-469 petajoules a year) and Block 11B/12B (Luiperd-Brulpadda, about 65 petajoules a year) after 2035 to achieve low-cost energy (they’re calling $6-$8/GJ low-cost).

Ramokgopa’s determination prioritises near-term coastal LNG import nodes without formally linking the procurement framework to mid-term indigenous gas field development or offshore pipeline aggregation.

A quick and dirty cost-benefit analysis:

During severe load shedding periods (FY2023–FY2024), Eskom’s open-cycle gas turbines (OCGTs) burned up to 3.3 billion litres of diesel annually.

In the last financial year, coal EAF improved, and Kusile Unit 6/Medupi Unit 4 returned to service, while OCGT spend dropped to R7.1-billion (saving R10.6-billion). So far this financial year (April-October 2026), diesel spend dropped by an additional 80.49% (by R4.79-billion to R1.16-billion), with the OCGT load factor falling to 1.02%.

Eskom’s diesel OCGTs cost R5,935/MWh in FY2026 (IPP OCGTs: R6,048/MWh). By contrast, combined-cycle natural gas generation operates at a fraction of this variable cost.

Diesel costs R200-R300/GJ, while imported LNG delivered inland/coastal costs R105-R140/GJ ($10-$11/GJ). West Coast Gas offers R90-R120/GJ (that magical $6-8/GJ).

Replacing diesel with natural gas cuts fuel costs by approximately 50%.

And we’re almost part of the way there. Zululand Energy Terminal at Richards Bay is a joint venture between Vopak Terminal Durban, Reatile Group and Transnet Pipelines, which signed a 25-year terminal operator agreement with Transnet National Ports Authority in February 2025.

They also signed a heads of agreement with Eskom as the foundation customer for its planned 3,000MW G2P project in the Richards Bay Industrial Development Zone.

South Africa’s existing 3.8GW fleet of liquid-fuel peakers can also be rapidly adapted for natural gas – the OCGTs are actually designed to run on both fuels.

So yes, frontloading gas and battery energy storage was quite an about-turn for those who followed Ramokgopa’s earlier plans closely, but there is more than enough underlying scaffolding to make at least some of Mantashe’s wildest gas dreams a reality. DM