Connect with us

Published

on

India’s carbon dioxide (CO2) emissions from its power sector fell by 1% year-on-year in the first half of 2025 and by 0.2% over the past 12 months, only the second drop in almost half a century.

As a result, India’s CO2 emissions from fossil fuels and cement grew at their slowest rate in the first half of the year since 2001 – excluding Covid – according to new analysis for Carbon Brief.

The analysis is the first of a regular new series covering India’s CO2 emissions, based on monthly data for fuel use, industrial production and power output, compiled from numerous official sources.

(See the regular series on China’s CO2 emissions, which began in 2019.)

Other key findings on India for the first six months of 2025 include:

  • The growth in clean-energy capacity reached a record 25.1 gigawatts (GW), up 69% year-on-year from what had, itself, been a record figure.
  • This new clean-energy capacity is expected to generate nearly 50 terawatt hours (TWh) of electricity per year, nearly sufficient to meet the average increase in demand overall.
  • Slower economic expansion meant there was zero growth in demand for oil products, a marked fall from annual rates of 6% in 2023 and 4% in 2024.
  • Government infrastructure spending helped accelerate CO2 emissions growth from steel and cement production, by 7% and 10%, respectively.

The analysis also shows that emissions from India’s power sector could peak before 2030, if clean-energy capacity and electricity demand grow as expected.

The future of CO2 emissions in India is a key indicator for the world, with the country – the world’s most populous – having contributed nearly two-fifths of the rise in global energy-sector emissions growth since 2019.

India’s surging emissions slow down

In 2024, India was responsible for 8% of global energy-sector CO2 emissions, despite being home to 18% of the world’s population, as its per-capita output is far below the world average.

However, emissions have been growing rapidly, as shown in the figure below.

The country contributed 31% of global energy-sector emissions growth in the decade to 2024, rising to 37% in the past five years, due to a surge in the three-year period from 2021-23.

India’s CO2 emissions from fossil fuels and cement, million tonnes of CO2, rolling 12-month totals. Source: Analysis for Carbon Brief by CREA. (See: About the data.)

More than half of India’s CO2 output comes from coal used for electricity and heat generation, making this sector the most important by far for the country’s emissions.

The second-largest sector is fossil fuel use in industry, which accounts for another quarter of the total, while oil use for transport makes up a further eighth of India’s emissions.

India’s CO2 emissions from fossil fuels and cement grew by 8% per year from 2019 to 2023, quickly rebounding from a 7% drop in 2020 due to Covid.

Before the Covid pandemic, emissions growth had averaged 4% per year from 2010 to 2019, but emissions in 2023 and 2024 rose above the pre-pandemic trendline.

This was despite a slower average GDP growth rate from 2019 to 2024 than in the preceding decade, indicating that the economy became more energy- and carbon-intensive. (For example, growth in steel and cement outpaced the overall rate of economic growth.)

A turnaround came in the second half of 2024, when emissions only increased by 2% year-on-year, slowing down to 1% in the first half of 2025, as seen in the figure below.

Bar chart showing that India's CO2 emissions growth has slowed sharply since 2024
Year-on-year change in India’s half-yearly CO2 emissions from fossil fuels and cement, %. Source: Analysis for Carbon Brief by CREA. (See: About the data.)

The largest contributor to the slowdown was the power sector, which was responsible for 60% of the drop in emissions growth rates, when comparing the first half of 2025 with the years 2021-23.

Oil demand growth slowed sharply as well, contributing 20% of the slowdown. The only sectors to keep growing their emissions in the first half of 2025 were steel and cement production.

Another 20% of the slowdown was due to a reduction in coal and gas use outside the power, steel and cement sectors. This comprises construction, industries such as paper, fertilisers, chemicals, brick kilns and textiles, as well as residential and commercial cooking, heating and hot water.

This is all shown in the figure below, which compares year-on-year changes in emissions during the second half of 2024 and the first half of 2025, with the average for 2021-23.

Bar chart showing that India's power sector drives marked slowdown in CO2 growth
Year-on-year change in India’s half-yearly CO2 emissions from fossil fuels and cement, million tonnes of CO2. Bars show the half-yearly average for 2021-23 along with the periods July-December 2024 and January-June 2025. Source: Analysis for Carbon Brief by CREA. (See: About the data.)

Power sector emissions fell by 1% in the first half of 2025, after growing 10% per year during 2021-23 and adding more than 50m tonnes of CO2 (MtCO2) to India’s total every six months.

Oil product use saw zero growth in the first half of 2025, after rising 6% per year in 2021-23.

In contrast, emissions from coal burning for cement and steel production rose by 10% and 7%, respectively, while coal use outside of these sectors fell 2%.

Gas consumption fell 7% year-on-year, with reductions across the power and industrial sectors as well as other users. This was a sharp reversal of the 5% average annual growth in 2021-23.

Power-sector emissions pause

The most striking shift in India’s sectoral emissions trends has come in the power sector, where coal consumption and CO2 emissions fell 0.2% in the 12 months to June and 1% in the first half of 2025, marking just the second drop in half a century, as shown in the figure below.

The reduction in coal use comes after more than a decade of break-neck growth, starting in the early 2010s and only interrupted by Covid in 2020. It also comes even as the country plans large amounts of new coal-fired generating capacity.

Chart showing that India's power sector CO2 just fell for only second time in half a century
Electricity generation from coal, terawatt hours per year. Source: NITI data portal.

In the first half of 2025, total power generation increased by 9 terawatt hours (TWh) year-on-year, but fossil power generation fell by 29TWh, as output from solar grew 17TWh, from wind 9TWh, from hydropower by 9TWh and from nuclear by 3TWh.

Analysis of government data shows that 65% of the fall in fossil-fuel generation can be attributed to lower electricity demand growth, 20% to faster growth in non-hydro clean power and the remaining 15% to higher output at existing hydropower plants.

Slower growth in electricity usage was largely due to relatively mild temperatures and high rainfall, in contrast to the heatwaves of 2024. A slowdown in industrial sectors in the second quarter of the year also contributed.

In addition, increased rainfall drove the jump in hydropower generation. India received 42% above-normal rainfall from March to May 2025. (In early 2024, India’s hydro output had fallen steeply as a result of “erratic rainfall”.)

Lower temperatures and this abundant rainfall reduced the need for air conditioning, which is responsible for around 10% of the country’s total power demand. In the same period in 2024, demand surged due to record heatwaves and higher temperatures across the country.

The growth in clean-power generation was buoyed by the addition of a record 25.1GW of non-fossil capacity in the first half of 2025. This was a 69% increase compared with the previous period in 2024, which had also set a record.

Solar continues to dominate new installations, with 14.3GW of capacity added in the first half of the year coming from large scale solar projects and 3.2GW from solar rooftops.

Solar is also adding the majority of new clean-power output. Taking into account the average capacity factor of each technology, solar power delivered 62% of the additional annual generation, hydropower 16%, wind 13% and nuclear power 8%.

The new clean-energy capacity added in the first half of 2025 will generate record amounts of clean power. As shown in the figure below, the 50TWh per year from this new clean capacity is approaching the average growth of total power generation.

(When clean-energy growth exceeds total demand growth, generation from fossil fuels declines.)

Bar chart showing that clean-energy expansion is close to matching demand growth overall
Columns: Six-monthly growth in clean-energy generation, by source, TWh. Dashed line: Average growth in electricity demand, 2021-2024, TWh. Source: CREA analysis of figures from the NITI data portal, with added capacity converted to expected annual generation based on average capacity factors calculated from monthly capacity and generation data.

India is expected to add another 16-17GW of solar and wind in the second half of 2025. Beyond this year, strong continued clean-energy growth is expected, towards India’s target for 500GW of non-fossil fuel capacity by 2030 (see below).

Slowing oil demand growth

The first half of 2025 also saw a significant slowdown in India’s oil demand growth. After rising by 6% a year in the three years to 2023, it slowed to 4% in 2024 and zero in the first half of 2025.

The slowdown in oil consumption overall was predominantly due to slower growth in demand for diesel and “other oil products”, which includes bitumen.

In the first quarter of 2025, diesel demand actually fell, due to a decline in industrial activity, limited weather-related mobility and – reportedly – higher uptake of vehicles that run on compressed natural gas (CNG), as well as electricity (EVs).

Diesel demand growth increased in March to May, but again declined in June because of early and unusually severe monsoon rains in India, leading to a slowdown in industrial and mining activities, disrupted supply-chains and transport of raw material, goods and services.

The severe rains also slowed down road construction activity, which in turn curtailed demand for transportation, construction equipment and bitumen.

Weaker diesel demand growth in 2024 had reflected slower growth in economic activity, as growth rates in the industrial and agricultural sectors contracted compared to previous years.

Another important trend is that EVs are also cutting into diesel demand in the commercial vehicles segment, although this is not yet a significant factor in the overall picture.

EV adoption is particularly notable in major metropolitan cities and other rapidly emerging urban centres and in the logistics sector, where they are being preferred for short haul rides over diesel vans or light commercial vehicles.

EVs accounted for only 7.6% of total vehicle sales in the financial year 2024-25, up 22.5% year-on-year, but still far from the target of 30% by 2030.

However, any significant drop in diesel demand will be a function of adoption of EV for long-haul trucks, which account for 32% of the total CO2 emissions from the transport sector. Only 280 electric trucks were sold in 2024, reported NITI Aayog.

Trucks remain the largest diesel consumers. Moreover, truck sales grew 9.2% year-on-year in the second quarter of 2025, driven in part by India’s target of 75% farm mechanisation by 2047. This sales growth may outweigh the reduction in diesel demand due to EVs. Subsidies for electric tractors have seen some pilots, but demand is yet to take off.

Apart from diesel, petrol demand growth continued in the first half of 2025 at the same rate as in earlier years. Modest year-on-year growth of 1.3% in passenger vehicle sales could temper future increases in petrol demand, however. This is a sharp decline from 7.5% and 10% growth rates in sales in the same period in 2024 and 2023.

Furthermore, EVs are proving to be cheaper to run than petrol for two- and three-wheelers, which may reduce the sale of petrol vehicles in cities that show policy support for EV adoption.

Steel and cement emissions continue to grow

As already noted, steel and cement were the only major sectors of India’s economy to see an increase in emissions growth in the first half of 2025.

While they were only responsible for around 12% of India’s total CO2 emissions from fossil fuels and cement in 2024, they have been growing quickly, averaging 6% a year for the past five years.

The growth in emissions accelerated in the first half of 2025, as cement output rose 10% and steel output 7%, far in excess of the growth in economic output overall.

Steel and cement growth accelerated further in July. A key demand driver is government infrastructure spending, which tripled from 2019 to 2024.

In the second quarter of 2025, the government’s capital expenditure increased 52% year-on-year. albeit from a low base during last year’s elections. This signals strong growth in infrastructure.

The government is targeting domestic steel manufacturing capacity of 300m tonnes (Mt) per year by 2030, from 200Mt currently, under the National Steel Policy 2017, supported by financial incentives for firms that meet production targets for high quality steel.

The government also imposed tariffs on steel imports in April and stricter quality standards for imports in June, in order to boost domestic production.

Government policies such as Pradhan Mantri Awas Yojna – a “housing for all” initiative under which 30m houses are to be built by FY30 – is further expected to lift demand for steel and cement.

The automotive sector in India is expected to grow at a fast pace, with sales expected to reach 7.5m units for passenger vehicle and commercial vehicle segments from 5.1m units in 2023, in addition to rapid growth in electric vehicles. This can be expected to be another key driver for growth of the steel sector, as 900 kg of steel is used per vehicle.

Without stringent energy efficiency measures and the adoption of cleaner fuel, the expected growth in steel and cement production could drive significant emissions growth from the sector.

Power-sector emissions could peak before 2030

Looking beyond this year, the analysis shows that CO2 from India’s power sector could peak before 2030, having previously been the main driver of emissions growth.

To date, India’s clean-energy additions have been lagging behind the growth in total electricity demand, meaning fossil-fuel demand and emissions from the sector have continued to rise.

However, this dynamic looks likely to change. In 2021, India set a target of having 500GW of non-fossil power generation capacity in place by 2030. Progress was slow at first, so meeting the target implies a substantial acceleration in clean-energy additions.

The country has been laying the groundwork for such an acceleration.

There was 234GW of renewable capacity in the pipeline as of April 2025, according to the Ministry of New and Renewable Energy. This includes 169GW already awarded contracts, of which 145GW is under construction, and an additional 65GW put out to tender. There is also 5.2GW of new nuclear capacity under construction.

If all of this is commissioned by 2030, then total non-fossil capacity would increase to 482GW, from 243GW at the end of June 2025, leaving a gap of just 18GW to be filled with new projects.

When the non-fossil capacity target was set in 2021, CREA assessed that the target would suffice to peak demand for coal in power generation before 2030. This assessment remains valid and is reinforced by the latest Central Electricity Authority (CEA) projection for the country’s “optimal power mix” in 2030, shown in the figure below.

Chart showing that India's power sector CO2 could peak before 2030
Electricity generation by fuel, TWh per year. Source: Historical generation from NITI, projection for the fiscal year 2029-30 from CEA. The trajectories from the latest data to 2029-30 are based on assuming steady annual growth rates for generation from each technology. The CEA projection is aligned with the target of reaching 500GW non-fossil capacity by the end of 2030.

In the CEA’s projection, the share of non-fossil power generation rises to 44% in the 2029-30 fiscal year, up from 25% in 2024-25. From 2025 to 2030, power demand growth, averaging 6% per year, is entirely covered from clean sources.

To accomplish this, the growth in non-fossil power generation would need to accelerate over time, meaning that towards the end of the decade, the growth in clean power supply would clearly outstrip demand growth overall – and so power generation from fossil fuels would fall.

While coal-power generation is expected to flatline, large amounts of new coal-power capacity is still being planned, because of the expected growth in peak electricity demand.

The post-Covid increase in electricity demand has given rise to a wave of new coal power plant proposals. Recent plans from the government target an increase in coal-power capacity by another 80-100GW by 2030-32, with 35GW already under construction as of July 2025.

The rationale for this is the increase in peak electricity loads, associated in particular with worsening heatwaves and growing use of air conditioning. The increase might yet prove unneeded.

Analysis by CREA shows that solar and wind are making an increasing contribution to meeting peak loads. This contribution will increase with the roll-out of solar power with integrated battery storage, the cost of which fell by 50-60% from 2023 to 2025.

The latest auction held in India saw solar power with battery storage bidding at prices, per unit of electricity generation, that were lower than the cost of new coal power.

This creates the opportunity to accelerate the decarbonisation of India’s power sector, by reducing the need for thermal power capacity.

The clean-energy buildout has made it possible for India to peak its power-sector emissions within the next few years, if contracted projects are built, clean-energy growth is maintained or accelerated beyond 2030 and demand growth remains within the government’s projections.

This would be a major turning point, as the power sector has been responsible for half of India’s recent emissions growth. In order to peak its emissions overall, however, India would still need to take further action to address CO2 from industry and transport.

With the end-of-September 2025 deadline nearing, India has yet to publish its international climate pledge (nationally determined contribution, NDC) for 2035 under the Paris Agreement, meaning its future emissions path, in the decades up to its 2070 net-zero goal, remains particularly uncertain.

The country is expected to easily surpass the headline climate target from its previous NDC, of cutting the emissions intensity of its economy to 45% below 2005 levels by 2030. As such, this goal is “unlikely to drive real world emission reductions”, according to Climate Action Tracker.

In July of this year, it met a 2030 target for 50% of installed power generating capacity to be from non-fossil sources, five years early.

About the data

This analysis is based on official monthly data for fuel consumption, industrial production and power generation from different ministries and government institutes.

Coal consumption in thermal power plants is taken from the monthly reports downloaded from the National Power Portal of the Ministry of Power. The data is compiled for the period January 2019 until June 2025. Power generation and capacity by technology and fuel on a monthly basis are sourced from the NITI data portal.

Coal use at steel and cement plants, as well as process emissions from cement production, are estimated using production indices from the Index of Eight Core Industries released monthly by the Office of Economic Adviser, assuming that changes in emissions follow production volumes.

These production indices were used to scale coal use by the sectors in 2022. To form a basis for using the indices, monthly coal consumption data for 2022 was constructed for the sectors using the annual total coal consumption reported in IEA World Energy Balances and monthly production data in a paper by Robbie Andrew, on monthly CO2 emission accounting for India.

Annual cement process emissions up to 2024 were also taken from Robbie Andrew’s work and scaled using the production indices. This approach better approximated changes in energy use and emissions reported in the IEA World Energy Balances, than did the amounts of coal reported to have been dispatched to the sectors, showing that production volumes are the dominant driver of short-term changes in emissions.

For other sectors, including aluminium, auto, chemical and petrochemical, paper and plywood, pharmaceutical, graphite electrode, sugar, textile, mining, traders and others, coal consumption is estimated based on data on despatch of domestic and imported coal to end users from statistical reports and monthly reports by the Ministry of Coal, as consumption data is not available.

The difference between consumption and dispatch is stock changes, which are estimated by assuming that the changes in coal inventories at end user facilities mirror those at coal mines, with end user inventories excluding power, steel and cement assumed to be 70% of those at coal mines, based on comparisons between our data and the IEA World Energy Balances.

Stock changes at mines are estimated as the difference between production at and despatch from coal mines, as reported by the Ministry of Coal.

In the case of the second quarter of the year 2025, data on domestic coal has been taken from the monthly reports by the Ministry of Coal. The regular data releases on coal imports have not taken place for the second quarter of 2025, for unknown reasons, so data was taken from commercial data providers Coal Hub and mjunction services ltd.

Product-wise petroleum product consumption data, as well as gas use by sector, was downloaded from the Petroleum Planning and Analysis Cell of the Ministry of Petroleum & Natural Gas.

As the fuel dispatch and consumption data is reported as physical volumes, calorific values are taken from IEA’s World Energy Balance and CO2 emission factors from 2006 IPCC Guidelines for National Greenhouse Gas Inventories.

Calorific values are assigned separately to different fuel types, including domestic and imported coal, anthracite and coke, as well as petrol, diesel and several other oil products.

The post Analysis: India’s power-sector CO2 falls for only second time in half a century appeared first on Carbon Brief.

Analysis: India’s power-sector CO2 falls for only second time in half a century

Continue Reading

Climate Change

Efforts to green lithium extraction face scrutiny over water use 

Published

on

Mining companies are showcasing new technologies which they say could extract more lithium – a key ingredient for electric vehicle (EV) batteries – from South America’s vast, dry salt flats with lower environmental impacts.

But environmentalists question whether the expensive technology is ready to be rolled out at scale, while scientists warn it could worsen the depletion of scarce freshwater resources in the region and say more research is needed.

The “lithium triangle” – an area spanning Argentina, Bolivia and Chile – holds more than half of the world’s known lithium reserves. Here, lithium is found in salty brine beneath the region’s salt flats, which are among some of the driest places on Earth.

Lithium mining in the region has soared, driven by booming demand to manufacture batteries for EVs and large-scale energy storage.

Mining companies drill into the flats and pump the mineral-rich brine to the surface, where it is left under the sun in giant evaporation pools for 18 months until the lithium is concentrated enough to be extracted.

The technique is relatively cheap but requires vast amounts of land and water. More than 90% of the brine’s original water content is lost to evaporation and freshwater is needed at different stages of the process.

One study suggested that the Atacama Salt Flat in Chile is sinking by up to 2 centimetres a year because lithium-rich brine is being pumped at a faster rate than aquifers are being recharged.

    Lithium extraction in the region has led to repeated conflicts with local communities, who fear the impact of the industry on local water supplies and the region’s fragile ecosystem.

    The lithium industry’s answer is direct lithium extraction (DLE), a group of technologies that selectively extracts the silvery metal from brine without the need for vast open-air evaporation ponds. DLE, it argues, can reduce both land and water use.

    Direct lithium extraction investment is growing

    The technology is gaining considerable attention from mining companies, investors and governments as a way to reduce the industry’s environmental impacts while recovering more lithium from brine.

    DLE investment is expected to grow at twice the pace of the lithium market at large, according to research firm IDTechX.

    There are around a dozen DLE projects at different stages of development across South America. The Chilean government has made it a central pillar of its latest National Lithium Strategy, mandating its use in new mining projects.

    Last year, French company Eramet opened Centenario Ratones in northern Argentina, the first plant in the world to attempt to extract lithium solely using DLE.

    Eramet’s lithium extraction plant is widely seen as a major test of the technology. “Everyone is on the edge of their seats to see how this progresses,” said Federico Gay, a lithium analyst at Benchmark Mineral Intelligence. “If they prove to be successful, I’m sure more capital will venture into the DLE space,” he said.

    More than 70 different technologies are classified as DLE. Brine is still extracted from the salt flats but is separated from the lithium using chemical compounds or sieve-like membranes before being reinjected underground.

    DLE techniques have been used commercially since 1996, but only as part of a hybrid model still involving evaporation pools. Of the four plants in production making partial use of DLE, one is in Argentina and three are in China.

    Reduced environmental footprint

    New-generation DLE technologies have been hailed as “potentially game-changing” for addressing some of the issues of traditional brine extraction.

    “DLE could potentially have a transformative impact on lithium production,” the International Lithium Association found in a recent report on the technology.

    Firstly, there is no need for evaporation pools – some of which cover an area equivalent to the size of 3,000 football pitches.

    “The land impact is minimal, compared to evaporation where it’s huge,” said Gay.

    A drone view shows Eramet’s lithium production plant at Salar Centenario in Salta, Argentina, July 4, 2024. (Photo: REUTERS/Matias Baglietto)

    A drone view shows Eramet’s lithium production plant at Salar Centenario in Salta, Argentina, July 4, 2024. (Photo: REUTERS/Matias Baglietto)

    The process is also significantly quicker and increases lithium recovery. Roughly half of the lithium is lost during evaporation, whereas DLE can recover more than 90% of the metal in the brine.

    In addition, the brine can be reinjected into the salt flats, although this is a complicated process that needs to be carefully handled to avoid damaging their hydrological balance.

    However, Gay said the commissioning of a DLE plant is currently several times more expensive than a traditional lithium brine extraction plant.

    “In theory it works, but in practice we only have a few examples,” Gay said. “Most of these companies are promising to break the cost curve and ramp up indefinitely. I think in the next two years it’s time to actually fulfill some of those promises.”

    Freshwater concerns

    However, concerns over the use of freshwater persist.

    Although DLE doesn’t require the evaporation of brine water, it often needs more freshwater to clean or cool equipment.

    A 2023 study published in the journal Nature reviewed 57 articles on DLE that analysed freshwater consumption. A quarter of the articles reported significantly higher use of freshwater than conventional lithium brine mining – more than 10 times higher in some cases.

    “These volumes of freshwater are not available in the vicinity of [salt flats] and would even pose problems around less-arid geothermal resources,” the study found.

    The company tracking energy transition minerals back to the mines

    Dan Corkran, a hydrologist at the University of Massachusetts, recently published research showing that the pumping of freshwater from the salt flats had a much higher impact on local wetland ecosystems than the pumping of salty brine. “The two cannot be considered equivalent in a water footprint calculation,” he said, explaining that doing so would “obscure the true impact” of lithium extraction.

    Newer DLE processes are “claiming to require little-to-no freshwater”, he added, but the impact of these technologies is yet to be thoroughly analysed.

    Dried-up rivers

    Last week, Indigenous communities from across South America held a summit to discuss their concerns over ongoing lithium extraction.

    The meeting, organised by the Andean Wetlands Alliance, coincided with the 14th International Lithium Seminar, which brought together industry players and politicians from Argentina and beyond.

    Indigenous representatives visited the nearby Hombre Muerto Salt Flat, which has borne the brunt of nearly three decades of lithium extraction. Today, a lithium plant there uses a hybrid approach including DLE and evaporation pools.

    Local people say the river “dried up” in the years after the mine opened. Corkran’s study linked a 90% reduction in wetland vegetation to the lithium’s plant freshwater extraction.

    Pia Marchegiani, of Argentine environmental NGO FARN, said that while DLE is being promoted by companies as a “better” technique for extraction, freshwater use remained unclear. “There are many open questions,” she said.

    AI and satellite data help researchers map world’s transition minerals rush

    Stronger regulations

    Analysts speaking to Climate Home News have also questioned the commercial readiness of the technology.

    Eramet was forced to downgrade its production projections at its DLE plant earlier this year, blaming the late commissioning of a crucial component.

    Climate Home News asked Eramet for the water footprint of its DLE plant and whether its calculations excluded brine, but it did not respond.

    For Eduardo Gigante, an Argentina-based lithium consultant, DLE is a “very promising technology”. But beyond the hype, it is not yet ready for large-scale deployment, he said.

    Strong regulations are needed to ensure that the environmental impact of the lithium rush is taken seriously, Gigante added.

    In Argentina alone, there are currently 38 proposals for new lithium mines. At least two-thirds are expected to use DLE. “If you extract a lot of water without control, this is a problem,” said Gigante. “You need strong regulations, a strong government in order to control this.”

    The post Efforts to green lithium extraction face scrutiny over water use  appeared first on Climate Home News.

    Efforts to green lithium extraction face scrutiny over water use 

    Continue Reading

    Climate Change

    Maryland’s Conowingo Dam Settlement Reasserts State’s Clean Water Act Authority but Revives Dredging Debate

    Published

    on

    The new agreement commits $340 million in environmental investments tied to the Conowingo Dam’s long-term operation, setting an example of successful citizen advocacy.

    Maryland this month finalized a $340 million deal with Constellation Energy to relicense the Conowingo Dam in Cecil County, ending years of litigation and regulatory uncertainty. The agreement restores the state’s authority to enforce water quality standards under the Clean Water Act and sets a possible precedent for dozens of hydroelectric relicensing cases nationwide expected in coming years.

    Maryland’s Conowingo Dam Settlement Reasserts State’s Clean Water Act Authority but Revives Dredging Debate

    Continue Reading

    Climate Change

    A Michigan Town Hopes to Stop a Data Center With a 2026 Ballot Initiative

    Published

    on

    Local officials see millions of dollars in tax revenue, but more than 950 residents who signed ballot petitions fear endless noise, pollution and higher electric rates.

    This is the second of three articles about Michigan communities organizing to stop the construction of energy-intensive computing facilities.

    A Michigan Town Hopes to Stop a Data Center With a 2026 Ballot Initiative

    Continue Reading

    Trending

    Copyright © 2022 BreakingClimateChange.com