India's grid must match its solar ambitions | The coal mines that leak gas

ELI5/TLDR

India built solar capacity faster than it built the grid to absorb it. The result this April: power was nearly free at noon and rationed at 10 p.m. Storage, transmission lines, and grid-stabilising hardware are all years behind the panels. Separately, India’s coal expansion is leaking enormous amounts of methane — a gas 80 times worse than CO2 in the short run — and nobody in government is clearly in charge of stopping it, even though the capture technology has been proven in India since the late 1990s.

The Full Story

A grid that succeeds and fails on the same day

On 25 April, India set an all-time peak demand record of 256 GW at 3:38 p.m. About a fifth of that came from solar. Seven hours later, the same grid fell 4.2 GW short. The night before, 5.4 GW short. In the second half of April, 13 of 15 nights ran shortages — but not when demand peaked. The shortages came later, after the sun had gone.

You can see it in the wholesale market. On some April afternoons, real-time prices on the Indian Energy Exchange collapsed to one or two paise per unit. The same week, evening prices kept hitting the regulated ceiling of Rs 20.

Why the day is now the problem

Solar peaks at noon. Demand peaks in the evening. That mismatch is structural, not a bug. The old grid was designed for coal, hydro and nuclear — plants that could be ramped on demand and whose heavy spinning turbines acted as physical shock absorbers. Renewables do neither. They generate when nature lets them, and they connect to the grid through power electronics that pass every flicker straight through.

Roughly 40% of the energy on India’s grid now comes from these “inertia-free” sources. Between January 2022 and July 2025, there were 68 separate occasions where more than 1 GW of renewable power was suddenly dumped onto the grid at once. Too much electricity is its own problem — voltage and frequency spike, and the grid trips.

Three problems, three buildouts behind schedule

A grid built for renewables has to solve three things simultaneously, and India is behind on all three.

Moving power across the country. Solar lives in Rajasthan, Gujarat, Karnataka and Andhra. Wind lives on the coasts of Gujarat and Tamil Nadu. Demand lives elsewhere. The Ministry of Power planned 51,000 circuit-km of new interstate transmission back in 2022, including 8,000 km of high-voltage DC corridors at Rs 2.44 lakh crore. Inter-regional transfer capacity is supposed to climb from 120 GW today to 168 GW by 2032. It still won’t be enough. Rajasthan alone is struggling to evacuate 60 GW of renewable energy. Every day between 11 a.m. and 2 p.m., about 4 GW of commissioned renewable capacity is simply switched off because the wires can’t carry it.

For context: China’s state grid corporation plans to spend $115 billion a year on its grid over the next five years. India plans to spend roughly $18 billion across the entire decade.

Storing power for after sunset. India has just over 7 GW of pumped storage and a bit over 1 GW of grid-scale batteries. The target is ten times that by 2032 to serve roughly 411 GW of demand. Pumped storage takes five to seven years to build. Battery rentals fell 35% in one year (from Rs 2.26 lakh to Rs 1.48 lakh per MW per month), which helps, but probably not enough to close the gap on time.

A softer lever is time-of-day tariffs: charge more in the evening, less at noon, and let consumers shift their usage. India introduced these in 2023, mandatory for commercial and industrial consumers above 10 kW from April 2024. Enforcement remains patchy.

Keeping the grid stable. This is the least visible piece. Renewables don’t provide inertia, so the grid needs synthetic inertia — synchronous condensers and grid-forming inverters. India commissioned its first synchronous condenser only recently. The CEA opened its first public consultation on grid-forming inverter standards in January. One stopgap proposal is to run thermal plants at a 40% floor instead of 55%, freeing up 34 GW of headroom to absorb solar surges. The trade-off is hardware damage — NTPC has logged 692 boiler tube leakages between 2022 and 2025, partly from this kind of cycling.

Coal that won’t quit, and the gas it leaks

India produced over 1 billion tonnes of coal in FY25 and wants to hit 1.5 billion tonnes by 2030. Coal still powers more than 70% of electricity generation. The most aggressive expansion is underground — output is meant to nearly triple to 100 million tonnes a year by FY29.

Underground seams sit on the richest deposits, including coking coal for steel. They also leak. Coal seams contain methane; mining cracks them open and lets it out. Over a 20-year window, methane warms the planet more than 80 times as much as the same mass of CO2. By 2029, Ember estimates India’s coal mine methane will have a short-term warming impact of about 138 million tonnes of CO2-equivalent — roughly the annual emissions of every truck and bus on Indian roads in 2021. Underground mines, despite producing only 3.5% of the coal, account for 15% of the methane.

The problem isn’t technology, it’s the org chart

India ran a successful methane capture pilot at the Moonidih underground mine in Jharia between 1996 and 2010. Three vertical wells, captured gas, ran generators, lit up a mine colony. Several follow-up studies confirmed the approach works in Indian conditions. In the fifteen years since, zero commercial-scale projects have been built.

The reason, according to Ember, is governance. Coal mine methane in India falls into an institutional gap. The Ministry of Coal sets policy and controls Coal India, CMPDI (the technical body) and the CCO (production monitoring). Separately, DGMS handles mine safety (and treats methane as an explosion hazard, not an emission), MoEFCC handles environmental clearance, DGH oversees coal bed methane (legally a different thing, even though it’s the same gas from the same seams), and NITI Aayog handles general planning. The CMM Clearinghouse, set up with the US EPA, can convene meetings but has no enforcement power.

The downstream effects are predictable. Emissions are estimated from sample factors across 99 mines, then averaged — there’s no mine-by-mine public data, no integrated reporting platform, no independent verification. Abandoned mines aren’t counted at all. The Ministry of Coal plans to shut 147 mines by 2028. None of their methane will appear in any official tally.

The economics of inaction are notable. India spent $13.3 billion importing 31 billion cubic metres of natural gas in FY24. Ember’s conservative scenario says capturing 1.6 million tonnes of CMM cumulatively by 2030 could save up to $980 million in gas imports. The IEA’s higher numbers would put annual savings above $1 billion by decade-end. Australia rewards facilities that beat a 100 kt CO2 baseline with tradable credits. The EU has banned methane venting from drainage stations outright. India has neither.

Tidbits

Adani Ports plans Rs 13,000 crore on marine services (dredging, harbour tugs, offshore vessels) and is entering European subsea cable-laying. India has set up a $1.5 billion maritime insurance pool to back Indian vessels in Red Sea waters and reduce dependence on London-based war-risk underwriters. China has filed a WTO complaint against India’s domestic content rules for solar cells and modules, following an earlier January complaint about EV and battery component incentives.

Key Takeaways

April 2025 marked the moment India’s grid started failing structurally rather than from undercapacity — gluts at noon, shortages at night.
The renewables target is largely on track; everything around renewables (transmission, storage, inertia hardware) is years behind.
India’s planned grid spend over a decade is less than what China spends in two months.
Coal isn’t slowing down — output target is 1.5 billion tonnes by 2030, with underground tripling.
Methane capture from coal mines has been proven in India since the 1990s. Nothing commercial has been built because no single agency owns the problem.
Captured methane could offset roughly $1 billion of LNG imports a year by 2030, before any carbon credit upside.

Claude’s Take

This is a good piece of explanatory journalism, well-sourced from Ember and CERC, with the right instinct: the bottleneck has moved. The grid story is the more interesting half — it captures a transition most people miss, where success on the generation side creates a new and harder set of problems on the absorption side. The framing of “three jobs of a grid” is a clean way to organise it.

The CMM story is more polemical and slightly less convincing in places. Calling governance reform “the lowest hanging fruit” understates how hard inter-ministerial reorganisation actually is in India — it’s the lowest hanging fruit in the sense that the technology is solved, but the political fruit is well out of reach. The $1 billion savings number is also a soft anchor: it assumes the captured gas displaces imported LNG at premium prices and ignores the capital cost of building drainage infrastructure across hundreds of dispersed mines with thin, dipping seams. Real, but not free money.

The unspoken throughline is the same in both halves: India keeps scaling the easy thing (panels, mines) and underinvesting in the harder coordination layer (grid, governance). That pattern is worth watching across other parts of the economy.

Score: 7/10. Clean reporting, useful frame, slightly thin on the economics of the methane side.

India's grid must match its solar ambitions | The coal mines that leak gas | The Daily Brief #465