Where is land most valuable to the power system, and how do land constraints reshape the path to carbon neutrality?

The study treats land as a production factor whose value varies by location. It quantifies how making more land available for wind and solar affects system costs. The most valuable land lies near major demand centres in eastern and central China, where local development can reduce reliance on distant resources and long-distance transmission.

A carbon-neutral power system is technically feasible under all three land-supply scenarios: conservative, baseline and open. However, where renewables are built, which technologies are used and how much the transition costs differ substantially. China can have abundant resources overall while still facing a shortage of suitable land near demand centres.

How can global net-zero power systems deliver both deep decarbonization and energy access?

The study brings deep decarbonization and energy access into a single optimization framework. Its results show that a global net-zero power system can technically meet the electricity requirements for decent living in every region. Wind and solar resources in low-income regions can help expand electricity services.

Where should AI data centres and desert clean-energy systems be built together?

The study proposes building off-grid wind, solar and storage systems alongside data centres in deserts. Sending computing results through fibre-optic networks reduces the need to transmit electricity over long distances. Remote sites can supply 24/7 clean power for foundation-model training, while sites closer to demand centres can serve latency-sensitive inference.

How should capacity planning, hourly operation, land policy and market design adapt to several terawatts of wind and solar?

China's power system could technically achieve substantial negative emissions in 2060. However, high shares of wind and solar make system conditions more variable. Alongside investment in storage, transmission and negative-emission technologies, firm generation needs new compensation mechanisms to ensure adequate capacity.

Where should wind and solar be built during the 15th Five-Year Plan, and how much storage and interprovincial transmission will they need?

The study models renewable expansion from 2026 to 2030 on a 0.1° grid with hourly time series. In the baseline scenario, wind and solar generate approximately 4.9 PWh in 2030, approaching the approximately 5.3 PWh generated by coal.

How do land policy, resource quality and grid connections shape the location of several terawatts of wind and solar?

The study integrates subprovincial renewable siting, demand centres, transmission and full-year system operation in a single planning model. Land availability affects developable potential, system costs and technology mixes.

With 5,000–8,000 GW of wind and solar in China, how much pumped-storage and hydropower flexibility do conventional models miss?

The national model represents reservoir levels, environmental flows, linked hydropower cascades, pumped-storage operation and detailed renewable resources. Capturing this flexibility changes estimates of capacity needs, system costs and investment returns substantially.