Anthony Gar-On Yeh

Neural-network-based cellular automata for simulating multiple land use changes using GIS

This paper presents a new method to simulate the evolution of multiple land uses based on the integration of neural networks and cellular automata using GIS. Simulation of multiple land use changes using cellular automata (CA) is difficult because numerous spatial variables and parameters have to be utilized. Conventional CA models have problems in defining simulation parameter values, transition rules and model structures. In this paper, a three-layer neural network with multiple output neurons is designed to calculate conversion probabilities for competing multiple land uses. The model involves iterative looping of the neural network to simulate gradual land use conversion processes. Spatial variables are not deterministic because they are dynamically updated at the end of each loop. A GIS is used to obtain site attributes and training data, and to provide spatial functions for constructing the neural network. The parameter values for modelling are automatically generated by the training procedure of neural networks. The model has been successfully applied to the simulation of multiple land use changes in a fast growing area in southern China.

Economic Development and Agricultural Land Loss in the Pearl River Delta, China

Abstract The Pearl River Delta is developing very rapidly in the last two decades since the adoption of economic reform and open-door policy of China in 1978. Concomitant to this development is the rapid change of landscape in both urban and rural areas. The loss of valuable agricultural land by the encroachment of urban development, especially massive construction sites from land speculation, is very severe recently. This paper examines the relationship between economic development and agricultural land loss in the Pearl River Delta, using Dongguan as a case study. It is found that agricultural land loss has been much aggravated by land speculation as a result of the property bloom in the Pearl River Delta that was induced by the property boom in Hong Kong in the early 1990s. The urban sprawl in the Pearl River Delta is also related to other economic factors, such as rural industrialization, rise of localism, influence from Hong Kong, transport improvement, and lack of land management and monitoring system. There is an urgent need to develop a sustainable land development strategy to protect the fertile agricultural land from further unnecessary losses, especially from land speculation.

A constrained CA model for the simulation and planning of sustainable urban forms by using GIS

A constrained cellular automata (CA) model based on ‘grey cells’ for the simulation of different types of urban forms and developments is developed within a raster GIS. Remote sensing and GIS are used to supply information on environmental constraints and locations of growth centers for the simulation. The model is able to consider different criteria, such as urban forms, environmental suitability, and land consumption for the planning of sustainable cities. Seven different types of urban forms and developments that range from compact–monocentric, compact–polycentric, compact–monocentric–environmental, compact–polycentric–environmental, dispersed, highly dispersed, to very highly dispersed developments were simulated by using the model. The model can generate urban forms and developments with fractal structures that are close to real cities. The simulated patterns were evaluated by using different cost indicators related to sustainable development principles.

Sustainable land development model for rapid growth areas using GIS

There has been worldwide concern for sustainable development especially after the 1992s Rios UN Conference on Environment and Development. Rapid encroachment of urban development on valuable agricultural land will have great influence on whether sustainable development can be achieved. A sustainable land development model is developed using GIS in an attempt to control urban sprawl under rapid rural urbanization. The objective of the model is to ensure that equity between generations and efficiency in land use can be achieved in land development so that development can be sustained in the future. The model is used to study the impacts of agricultural land loss in Dongguan in 1988-93, a rapidly growing area in the Pearl River Delta of China. The impact of unplanned urban growth is evaluated by comparing the actual development with optimal development derived from the sustainable land development model. Land use problems are identified in both spatial and time dimensions as some land conversions are found...

Changing Spatial Distribution and Determinants of Land Development in Chinese Cities in the Transition from a Centrally Planned Economy to a Socialist Market Economy: A Case Study of Guangzhou

The spatial distribution and locational characteristics of land development have changed dramatically in Chinese cities since the land reform of 1987 which allowed the paid transfer of land-use rights—i.e. land leasing. This has led to the rapid transformation of the urban spatial structure of Chinese cities. There is an urgent need to study the general trend of such changes and their policy implications. However, due to the lack of data, such investigations lag far behind the rapid land development in Chinese cities. This paper attempts to examine the new spatial pattern of land development in Chinese cities and its determinants by studying land development in Guangzhou before (1979-87) and after (1987-92) the land reform by analysing data obtained from aerial photographs with the aid of GIS techniques. The determinants of land development are analysed using a logistic regression model. It is found that there has been significant acceleration of urban redevelopment and urban sprawl in Guangzhou since the adoption of the new land-leasing system in 1987. The changing spatial distributions and determinants of land development suggest the emergence of new locational behaviours of land development in Chinese cities in the transition from a centrally planned economy to a socialist market economy.

An integrated remote sensing and GIS approach in the monitoring and evaluation of rapid urban growth for sustainable development in the Pearl River Delta, China

Abstract The publication of the Brundtland Report in 1987 has led to worldwide concern over sustainable development. Sustainable development is particularly important to regions that are undergoing rapid economic development, such as the Pearl River Delta and other rapidly developing regions in China, Malaysia, Thailand and Asia. Geographical information systems (GIS) and remote sensing are useful tools in the formulation, implementation and monitoring of urban development in the move towards a sustainable development strategy. Since the adoption of economic reform in China in 1978, cities in the Pearl River Delta are developing very rapidly. Economic growth is around 20% a year, making it one of the highest growth areas in the world. Accompanying economic development is rapid urban growth. This paper attempts to demonstrate the application of the integration of GIS and remote sensing in urban growth management in the Pearl River Delta by using them to monitor and evaluate land development. Remote sensing...

Data mining of cellular automata's transition rules

This paper presents a new method to discover knowledge for geographical cellular automata (CA) by using a data-mining technique. CA have the ability to simulate complex geographical phenomena. Very few studies have been carried out on how to determine and validate the transition rules of CA from observed data. The transition rules of traditional CA are usually expressed by mathematical equations. This paper demonstrates that the explicit transition rules of CA can be automatically reconstructed through the rule induction procedure of data mining. The explicit transition rules are more intuitive to decision-makers. The transition rules are obtained by applying data-mining techniques to spatial data. The proposed method can reduce the uncertainties in defining transition rules and help to generate more reliable simulation results.

Calibration of cellular automata by using neural networks for the simulation of complex urban systems

This paper presents a new cellular automata (CA) model which uses artificial neural networks for both calibration and simulation. A critical issue for urban CA simulation is how to determine parameter values and define model structures. The simulation of real cities involves the use of many variables and parameters. The calibration of CA models is very difficult when there is a large set of parameters. In the proposed model, most of the parameter values for CA simulation are automatically determined by the training of artificial neural networks. The parameter values from the training are then imported into the CA model which is also based on the algorithm of neural networks. With the use of neural networks, users do not need to provide detailed transition rules which are difficult to define. The study shows that the model has better accuracy than traditional CA models in the simulation of nonlinear complex urban systems.

Integration of genetic algorithms and GIS for optimal location search

Optimal location search is frequently required in many urban applications for siting one or more facilities. However, the search may become very complex when it involves multiple sites, various constraints and multiple‐objectives. The exhaustive blind (brute‐force) search with high‐dimensional spatial data is infeasible in solving optimization problems because of a huge combinatorial solution space. Intelligent search algorithms can help to improve the performance of spatial search. This study will demonstrate that genetic algorithms can be used with Geographical Information systems (GIS) to effectively solve the spatial decision problems for optimally sitting n sites of a facility. Detailed population and transportation data from GIS are used to facilitate the calculation of fitness functions. Multiple planning objectives are also incorporated in the GA program. Experiments indicate that the proposed method has much better performance than simulated annealing and GIS neighborhood search methods. The GA method is very convenient in finding the solution with the highest utility value.

Errors and uncertainties in urban cellular automata

Errors and uncertainties are important issues in most geographical analyses and modelling processes. Cellular automata (CA) have been increasingly used for modelling geographical phenomena, such as the evolution of urban systems. Urban simulation frequently involves the inputs of a large set of spatial variables from GIS. The errors of data source in GIS can propagate through CA modelling processes. Moreover, CA models themselves also have modeling uncertainties because they are just an approximation to reality. These uncertainties have impacts on the outcome of urban simulation. Identification and evaluation of these errors and uncertainties are crucial for understanding and implementing the simulation results of urban CA modelling. It is found that some of the characteristics of errors and uncertainties in urban CA are quite unique which are not present in traditional GIS models. The study can help urban modelers and planners to understand more clearly the characteristics and implications of CA modelling.