BEIJING: Chinese researchers have created a hybrid framework capable of tracing on-road carbon dioxide emissions in real time with 30-metre resolution.
This technology is currently being implemented in Shenzhen within Guangdong Province and may expand to additional cities for carbon reduction assessment.
Urban expansion and population mobility have driven continuous increases in road carbon dioxide emissions, creating challenges for climate regulation and public health.
Previous carbon emission inventories suffered from coarse spatial resolution according to Wang Li, corresponding author and researcher at the Chinese Academy of Sciences.
This limitation made capturing fine-scale emission variations from different road segments or over time particularly difficult.
Accurately tracing emission sources or explaining their causes became even more challenging with previous methods.
Developing precise monitoring methods for multi-factor analysis of on-road carbon dioxide levels is crucial for effective reduction efforts.
Wang and his team combined Panoptic-Artificial Intelligence with a mobile observation framework to predict hourly 30-metre resolution carbon dioxide concentrations.
Their system provides daytime dynamic carbon dioxide increment predictions across urban traffic networks.
This innovation integrates AI with panoramic cameras, high-precision greenhouse-gas analysers and meteorological sensors for synchronous multi-source data acquisition.
The framework collects data on road carbon dioxide concentrations, traffic volumes, building layouts, vegetation cover and meteorological conditions during mobile surveys.
Researchers achieved an average identification accuracy exceeding 93% for emission sources through this technology.
The framework quantifies individual factor influences including traffic conditions, surrounding land cover and meteorological variables.
This capability clearly reveals the spatio-temporal dynamics and driving mechanisms behind carbon emissions.
Wang described the technique as an innovative AI deployment in environmental monitoring enabling a multi-dimensional carbon monitoring system.
The system combines conventional emission inventories with satellite-based greenhouse-gas monitoring technologies for comprehensive coverage. – Bernama-Xinhua