Featured insights
Analysis
Japan can reach net zero more cheaply with mature renewables than with unproven innovative thermal
New modelling finds a renewable-led transition reaches net zero at 7% lower cost, with stronger economic returns and far less emissions risk than betting on unproven abated thermal.
Analysis
Hydrogen and ammonia are costly for Japan's power sector
Power generation from hydrogen and ammonia plants would come at a significant cost premium and would be best reserved for peak demand and system stability.
Update
China's solar slowdown, seen from space: the Q2 2026 Solar Asset Mapper release
China's detected solar additions halved this quarter, whilst the Philippines posted the fastest growth among Asia's largest solar markets. TZ-SAM's latest data independently corroborates the slowdown reported on the ground.
All articles
Explainer
Building TZ-OSeMOSYS-STEEL: our methodology
How we designed a tool – including data, assumptions, projections, and scenarios – to model the decarbonisation of Japan’s steel sector
Explainer
What is 24/7 Carbon Free Electricity?
24/7 carbon free energy is a way to purchase and account for clean electricity on an hourly basis
Explainer
How can nationally determined contributions help countries reach net zero emissions?
Concepts within NDCs that help policymakers plan for a net-zero energy system
Explainer
How do countries set greenhouse gas emissions limits?
Introducing NDCs – Nationally Determined Contributions – and the underlying assumptions and conditions that shape them
Explainer
What are greenhouse gas emissions limits and carbon budgets?
Why do we need them, who sets them, and how are they used for energy system modelling?
Explainer
From Vision to Voltage with TZ-APG
Tracking Southeast Asia’s grid integration potential with our new, open access model
Explainer
Satellites and steel decarbonisation
How we're using satellite imagery to assess the steel sector's progress on decarbonisation
Explainer
Introducing TZ-SAM: Solar Asset Mapper
A global solar asset dataset, powered by planetary-scale machine learning
