Forecasting carbon dioxide emissions in Chongming: a novel hybrid forecasting model coupling gray correlation...

Area of Science:

• Environmental Science
• Data Science
• Climate Change Research

Background:

• Accurate prediction of regional carbon dioxide (CO2) emissions is crucial for achieving global carbon neutrality.
• Shanghai Chongming faces unique environmental, economic, and energy consumption factors influencing its CO2 output.
• Existing methods for CO2 accounting may suffer from data scarcity and inaccuracies.

Purpose of the Study:

• To develop and validate a novel hybrid model for predicting CO2 emissions in Shanghai Chongming.
• To analyze the influencing factors of CO2 emissions in the region.
• To provide technical support for local carbon reduction policies and sustainable development.

Main Methods:

• Grey relational analysis was used to determine the influence of economic activity, natural conditions, and energy consumption on CO2 emissions.
• A dual-channel pooled convolutional neural network (DCNN) with feature stacking was employed to capture spatial data features.
• A Gated Recurrent Unit (GRU) network was utilized to analyze the temporal dynamics of the identified features.

Main Results:

• The hybrid model demonstrated high precision, low error, and good stability in predicting CO2 emissions using accounting data.
• Carbon emissions in Chongming showed an increasing trend from 2000 to 2022, consistent with existing research.
• The proposed method effectively addresses challenges related to limited accounting data and traditional calculation inaccuracies.

Conclusions:

• The developed hybrid model offers a precise and stable approach for regional CO2 emission prediction.
• The findings provide valuable insights into Chongming's carbon emission trends and contributing factors.
• This research offers effective technical support for implementing targeted carbon reduction strategies and promoting sustainable development in the region.