Fractional reaction-diffusion modeling and machine learning for vegetation pattern analysis in Junggar Basin under...

Area of Science:

• Ecological modeling
• Climate science
• Geophysics

Background:

• Climate change significantly impacts vegetation patterns globally.
• Understanding vegetation-climate dynamics is crucial for predicting ecosystem responses.
• The Junggar Basin faces unique environmental challenges influencing its vegetation.

Purpose of the Study:

• To develop a refined vegetation-climate dynamic model incorporating vegetation physiological processes.
• To investigate the impact of climate change on vegetation patterns in the Junggar Basin.
• To predict future vegetation growth under various climate scenarios.

Main Methods:

• Developed a vegetation-climate dynamic model using a fractional-in-space diffusion model.
• Integrated key climatic factors: precipitation, temperature, and CO2.
• Performed numerical simulations with Junggar Basin data and machine learning algorithms for future predictions.

Main Results:

• An inverse relationship was found between the fractional-order coefficient and Turing instability domain size.
• Heat stress, water, and CO2 fertilization significantly influence vegetation growth.
• SSP1-2.6 is favorable for vegetation growth; SSP2-4.5 and SSP5-8.5 suppress it, with SSP5-8.5 showing the fastest desertification.

Conclusions:

• The developed model provides insights into vegetation-climate interactions.
• Climate change, particularly under SSP5-8.5, poses a significant threat of desertification in the Junggar Basin.
• Targeted mitigation and adaptation strategies are necessary to preserve vegetation cover.