Mechanistic spatio-temporal point process models for marked point processes, with a view to forest stand data.

Area of Science:

• Spatial statistics
• Ecological modeling
• Point process theory

Background:

• Spatial point processes model object locations.
• Marks associated with points add quantitative information (e.g., tree size).
• Existing models may lack interpretability for complex spatial relationships.

Purpose of the Study:

• To establish an equivalence between marked spatial point processes and spatio-temporal point processes.
• To develop a framework for constructing interpretable parametric statistical models.
• To enable tractable inference for ecological spatial data.

Main Methods:

• Identification of marked spatial point processes via conditional intensity functions.
• Development of parametric statistical models based on this identification.
• Application of maximum-likelihood-based inference techniques.

Main Results:

• Demonstration of the identification method for marked spatial point processes.
• Construction of easily interpretable parametric models.
• Tractable inference for complex ecological spatial data.

Conclusions:

• The proposed framework effectively links spatial and spatio-temporal point process models.
• This approach enhances the interpretability and analytical tractability of ecological spatial data models.
• Facilitates robust statistical inference for spatial ecology studies.