A new type of dual scale neighborhood based on vectorization for cellular automata models

Although the neighborhood of the cellular
automata (CA) model has been studied in detail, there is a contradiction in the
selection of the neighborhood size, which has not been revealed and addressed.
Small neighborhoods can constrain the shape complexity of simulated landscape,
but they cannot sufficiently characterize the local interactions, while large
neighborhoods do the opposite. This study proposes a new type of dual scale
neighborhood (DSN) based on vectorization to avoid this contradiction, which incorporates
a small neighborhood of pattern constraining (PCN) and a large neighborhood of influence
receiving (IRN). Taking Beijing, Wuhan, and Pearl River Delta as the study areas,
two kinds of CA models, namely, CA model using the original neighborhood
(ORN-CA) and CA model using the DSN (DSN-CA), were constructed based on serial
scalar algorithm and vectorized algorithm, respectively. Comparing their
simulation results and time required, the results show that the DSN enable
users to choose the appropriate neighborhood configuration to obtain the
simulation results with high accuracy and landscape similarity to the actual,
and the vectorization can greatly improve the computational efficiency of
neighborhood effects. Integrating the DSN with the vectorization can significantly
improve the simulation performance and efficiency of CA models.