Modeling complex spatial dynamics of two-population interaction in urbanization process

Yanguang Chen, Feng XuarXiv 2013 · arXiv

TL;DR

This paper models the complex dynamics of urbanization by developing a two-population (urban-rural) interaction model. It demonstrates that this model can reproduce the period-doubling bifurcations and chaotic behaviors observed in simpler logistic growth models, suggesting that complex dynamics in urbanization may arise from nonlinear interactions.

What Problem Does This Paper Address?

The paper addresses the challenge of understanding the complex, often chaotic, dynamics observed in urbanization processes. While simple ecological models like the logistic equation are known to exhibit complex behaviors such as periodic oscillations and chaos, the underlying mechanisms in real-world systems like urbanization remain ambiguous. Existing models often treat urbanization as a simple logistic growth curve, potentially overlooking the intricate interactions between urban and rural populations that could drive these complex dynamics. The authors aim to bridge this gap by developing and empirically validating a model that explicitly accounts for urban-rural interaction, linking it to the observed complex behaviors.

Methods

The authors propose a nonlinear dynamics model for urban-rural population interaction, inspired by predator-prey models in ecology. They then derive a logistic equation for the level of urbanization from this interaction model. The core methodology involves:

  • Model Development: Creating a system of differential equations to describe the interaction between rural ($r(t)$) and urban ($u(t)$) populations, incorporating nonlinear terms that represent their interaction.
  • Logistic Equation Derivation: Mathematically transforming the urban-rural interaction model into a logistic equation for the urbanization ratio ($L(t)$), showing how complex dynamics can emerge from simpler interactions.
  • Empirical Validation: Using US census data from 1790-1960 to statistically analyze and estimate the parameters of the model. This involves discretizing the differential equations into difference equations for regression analysis.
  • Numerical Simulation: Performing simulations with the discretized model to explore its dynamical behaviors, specifically looking for period-doubling bifurcations and chaotic states as parameter values change.

Datasets

The study utilizes historical US population census data from 1790 to 1960. This data includes rural population, urban population, and total population figures, typically recorded at 10-year intervals. The authors note a change in the definition of "city" around 1950-1970, which might affect data comparability for later periods, but they focus on the pre-1970 data for their primary analysis.

Key Insights and Ideas

  • Interaction as the Source of Complexity: The study strongly suggests that the complex dynamics (period-doubling, chaos) observed in urbanization are not inherent to simple logistic growth but arise from the nonlinear interactions between urban and rural populations.
  • Urban-Rural Dynamics Mirror Ecological Models: The proposed urban-rural interaction model exhibits behaviors analogous to predator-prey models, demonstrating that similar mathematical structures can describe diverse complex systems.
  • Logistic Growth as a Transform: The logistic equation for urbanization level can be derived from the more complex urban-rural interaction model, implying that logistic growth might represent a mathematical transform from complex, nonlinear interactions to simpler, albeit still dynamic, growth patterns.
  • Rural Population as a Key Driver: The model highlights the crucial role of the rural population and the urban-rural interaction in determining the pace and behavior of urbanization, rather than just the urban population itself.

Key Results

  • The developed urban-rural interaction model, when discretized and fitted to US census data, produces results consistent with observed urbanization trends.
  • Numerical simulations of the model demonstrate period-doubling bifurcations and chaotic behavior, mirroring findings from simpler logistic maps, as parameters are varied.
  • The model suggests that the parameter $b$ (related to the interaction term) and the difference $b-a$ are critical in determining the system's behavior, ranging from stable logistic growth to complex oscillations and chaos.
  • The study finds that the Lotka-Volterra and Keyfitz-Rogers models are less suitable for describing US urbanization compared to the United Nations model, which forms the basis of their proposed interaction model.

Limitations and Future Work

  • Closed System Assumption: The model, particularly in its derivation of the logistic equation, often assumes a closed system, which is an idealization. The authors acknowledge that real-world systems like the US are influenced by external factors like migration, which can lead to deviations.
  • Empirical vs. Theoretical Chaos: The paper raises a crucial point: while the models can exhibit chaotic behavior theoretically, the simulation results sometimes lead to unrealistic scenarios (e.g., negative rural population or urbanization ratios > 1). This suggests that true chaos in urbanization might be a theoretical construct rather than a directly observable phenomenon in reality.
  • Future Work: The authors suggest that further research is needed to explore the "essence of chaos" and the potential for a "fire-new logistic transform method" that could bridge simple and complex dynamics. They also indicate that pending questions regarding the connection between chaos, fractals, and urban evolution will be discussed in future reports.

Summary updated

BibTeX
@misc{chen2013modeling,
  author = {Yanguang Chen and Feng Xu},
  title = {{Modeling complex spatial dynamics of two-population interaction in urbanization process}},
  year = {2013},
  eprint = {1309.4860},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1309.4860},
}

Loading related content...

Figure