Journal of Geo-information Science >
Urban Expansion Cellular Automata Simulation by Coupling Hierarchical Driving Mechanism of Cell and Patch Scales
Received date: 2023-05-12
Revised date: 2023-07-08
Online published: 2023-09-05
Supported by
National Key Research and Development Plan(2022YFC3800804)
National Natural Science Foundation of China(42171411)
Open Fund Project of Key Laboratory of the Ministry of Natural Resources for Research on Rule of Law(CUGFZ-2207)
Guangdong Science and Technology Strategic Innovation Fund (the Guangdong-Hong Kong-Macau Joint Laboratory Program(2020B1212030009)
Requirements for the territorial spatial planning of the new era pose new challenges to the relationship between urban land expansion and agricultural and ecological spaces. Urban expansion is closely related to social and economic development as well as ecological environment protection. It is of great significance to study the urban expansion for promoting the urbanization process. The simulation of urban expansion can provide scientific support for formulating territorial spatial planning policies. The spatial characteristics of urban construction land patches are important factors affecting the expansion of urban land and are at different levels against the cell level. Therefore, hierarchically coupling the spatial characteristics of cells and patches can improve the simulation accuracy. Taking the Shanghai metropolitan area as an example, this paper constructs a cellular automata model (Spatial Network - Hierarchical Generalized Linear - Cellular Automata, SN-HGLM-CA) that takes into account both the spatial network structure of urban construction land and the hierarchical relationship of variables. Firstly, the spatial characteristics of patches are extracted through the spatial network model of urban construction land. The hierarchical relationship between variables is considered by using the HGLM to obtain the suitability probability of urban land expansion, and then embedded into the cellular automata model for urban land expansion simulation. This study draws the following research conclusions: ① The spatial network model based on urban construction land patches is of great significance for identifying the spatial importance characteristics of urban construction land and mining its expansion patterns; ② The HGLM model can reveal the hierarchical relationship between cells and patches during urban construction land expansion, which helps to improve the simulation accuracy of the cellular automata model; ③ The simulation of urban land expansion based on the SN-HGLM-CA model achieves good results in terms of simulation accuracy and landscape morphology, with the landscape similarity index exceeding 95%. In addition, compared to the SN-Logistic-CA model, the figure of merit value of our simulation increases by about 6.61%, which indicates that our simulation not only accurately reproduces the actual layout of urban construction land, but also improves the compactness of patch distribution. This paper explores the law of urban expansion from the spatial network structure characteristics of urban expansion, which can help urban planners to determine the reasonability of current trend considering the development status of urban expansion, and provide reference for the delineation of urban development boundaries.
WANG Yuying , WANG Haijun , ZHOU Xingang , ZHANG Bin , ZHOU Xiaoyan . Urban Expansion Cellular Automata Simulation by Coupling Hierarchical Driving Mechanism of Cell and Patch Scales[J]. Journal of Geo-information Science, 2023 , 25(9) : 1784 -1797 . DOI: 10.12082/dqxxkx.2023.230261
表1 驱动因子数据信息Tab. 1 Spatial variable factor data information |
分类 | 驱动因子含义 | 数据来源 | 时间/年 | 原始数据值范围 | 数据预处理 |
---|---|---|---|---|---|
地形地貌 | 海拔高度/m | 中国科学院资源环境科学数据中心 | 2015 | -76.96~1 548.99 | 归一化 |
坡度/° | 2015 | 0~65.34 | 重采样、归一化 | ||
道路交通 | 距铁路距离/m | Openstreetmap | 2015 | 0~89 511.10 | 欧氏距离、归一化 |
距高速公路距离/m | 2015 | 0~91 105.50 | 欧氏距离、归一化 | ||
距城市主干道距离/m | 2015 | 0~92426.80 | 欧氏距离、归一化 | ||
距城市次干道距离/m | 2015 | 0~70 250.10 | 欧氏距离、归一化 | ||
距城市支路距离/m | 2015 | 0~41 121.00 | 欧氏距离、归一化 | ||
距高架快速路距离/m | 2015 | 0~79 939.20 | 欧氏距离、归一化 | ||
距市中心距离/m | 2015 | 0~125 522.00 | 欧氏距离、归一化 | ||
距区县中心距离/m | 2015 | 0~77 802.20 | 欧氏距离、归一化 | ||
社会经济 | 公里格网GDP/万元/km2 | 中国科学院资源环境科学 数据中心 | 2015 | 780.81~2 069 174.62 | 重采样、归一化 |
公里格网POP/人/km2 | 2015 | 151.80 ~ 41 863.10 | 重采样、归一化 |
表2 驱动因子多重共线性检验Tab. 2 Multicollinearity test list of driving factors |
变量名称 | VIF | 变量名称 | VIF |
---|---|---|---|
海拔高度 | 1.651 | 距城市支路距离 | 1.374 |
坡度 | 1.630 | 距高架快速路距离 | 2.311 |
距铁路距离 | 2.226 | GDP | 3.876 |
距高速公路距离 | 1.955 | POP | 3.809 |
距城市主干道距离 | 2.120 | 距市中心距离 | 1.908 |
距城市次干道距离 | 1.619 | 距区县中心距离 | 1.445 |
表3 2000—2020年上海大都市圈城镇建设用地空间网络模型参数Tab. 3 Parameters of urban land spatial network model in Shanghai metropolitan area from 2000 to 2020 |
年份 | 面积/km2 | 节点数/个 | 搜索半径/m | 边数量/条 |
---|---|---|---|---|
2000 | 2 404.56 | 861 | 22 720 | 31 690 |
2005 | 3 365.91 | 815 | 23 472 | 27 784 |
2010 | 5 343.97 | 746 | 25 693 | 26 382 |
2015 | 5 819.49 | 796 | 24 438 | 29 258 |
2020 | 6 577.09 | 700 | 25 531 | 23 036 |
表4 SN-HGLM变量参数识别结果Tab.4 Parameter identification of SN-HGLM |
驱动因子 | 系数 | P值 | 空间交互系数 | 空间交互P值 |
---|---|---|---|---|
空间重要性 | 0.814 | 0.495 | - | - |
坡度 | -1.616 | 0.101 | -7.942 | 0.032 |
距铁路距离 | 3.409 | 0.312 | 14.766 | 0.234 |
距高速距离 | -0.734 | 0.835 | 8.048 | 0.515 |
距主干道距离 | 21.606 | 0.055 | 99.125 | 0.027 |
距支路距离 | -0.501 | 0.903 | 22.805 | 0.108 |
距高架距离 | 3.910 | 0.312 | -1.868 | 0.893 |
海拔高度 | -1.472 | 0.000 | - | - |
距次干道距离 | -10.831 | 0.000 | - | - |
距市中心距离 | -1.352 | 0.000 | - | - |
距区县中心距离 | -5.620 | 0.000 | - | - |
GDP | -23.906 | 0.000 | - | - |
POP | 11.080 | 0.000 | - | - |
表5 2020年上海大都市圈城镇建设用地扩展模型精度对比Tab. 5 Comparison of accuracy of urban land expansion models in Shanghai metropolitan area in 2020 |
Logistic-CA | SN-Logistic-CA | ANN-CA | SN-ANN-CA | SN-HGLM-CA | |
---|---|---|---|---|---|
OA | 0.983 4 | 0.983 4 | 0.984 1 | 0.984 3 | 0.984 0 |
Kappa | 0.919 1 | 0.919 2 | 0.922 5 | 0.922 9 | 0.921 8 |
FoM | 0.234 3 | 0.234 5 | 0.245 3 | 0.247 3 | 0.250 0 |
表6 2020年上海大都市圈城镇建设用地模拟景观指数对比Tab. 6 Comparison of landscape index of urban land expansion models in Shanghai metropolitan area in 2020 |
NP | PARA | ENN | LPI | /% | |
---|---|---|---|---|---|
真实值 | 661 | 127.662 3 | 1 514.147 8 | 1.593 1 | - |
SN-Logistic-CA | 686 | 130.249 7 | 1 384.005 2 | 1.652 7 | 95.46 |
SN-ANN-CA | 675 | 122.478 7 | 1 420.680 5 | 1.621 6 | 96.46 |
SN-HGLM-CA | 675 | 115.975 3 | 1 463.112 8 | 1.626 7 | 95.81 |
[1] |
刘新智, 周韩梅. 土地资源配置如何影响城市经济效率——基于中国286个地级以上城市工业数据的实证检验[J]. 中国土地科学, 2022, 36(9):49-58.
[
|
[2] |
陈理庭, 蔡海生, 张婷, 等. 基于Markov-FLUS模型的饶河流域土地利用多情景模拟分析[J]. 生态学报, 2022, 42(10):3947-3958.
[
|
[3] |
郭远智, 李许红. 基于随机森林模型的黄河流域城市建设用地结构时空演化及其驱动机制研究[J]. 地理科学进展, 2023, 42(1):12-26.
[
|
[4] |
刘小平. “土地利用模拟”专栏导言[J]. 地球信息科学学报, 2020, 22(3):516.
[
|
[5] |
冯一凡, 李翅, 李宇, 等. 黄河下游滩区周边城镇空间扩展特征与驱动机制[J]. 地理研究, 2023, 42(4):955-976.
[
|
[6] |
李丹, 胡国华, 黎夏, 等. 耦合地理模拟与优化的城镇开发边界划定[J]. 中国土地科学, 2020, 34(5):104-114.
[
|
[7] |
卞子浩, 马小雪, 龚来存, 等. 不同非空间模拟方法下CLUE-S模型土地利用预测——以秦淮河流域为例[J]. 地理科学, 2017, 37(2):252-258.
[
|
[8] |
田贺, 梁迅, 黎夏. 基于SD模型的中国2010―2050年土地利用变化情景模拟[J]. 热带地理, 2017, 37(4):547-561.
[
|
[9] |
王海军, 夏畅, 张安琪, 等. 基于约束性CA的大都市郊区城镇增长的情景模拟与管控——以武汉市江夏区为例[J]. 地理科学进展, 2016, 35(7):793-805.
[
|
[10] |
|
[11] |
吴佩君, 刘小平, 黎夏, 等. 基于InVEST模型和元胞自动机的城市扩张对陆地生态系统碳储量影响评估——以广东省为例[J]. 地理与地理信息科学, 2016, 32(5):22-28,36,2.
[
|
[12] |
罗紫薇, 胡希军, 韦宝婧. 基于多准则CA-Markov模型的城市景观格局演变与预测——以上杭县城区为例[J]. 经济地理, 2020, 40(10):58-66.
[
|
[13] |
|
[14] |
刘荣增, 陆文涛, 杜力卿. 基于ANN-CA模型的郑州城市空间拓展研究[J]. 城市发展研究, 2019, 26(12):77-85,49.
[
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
王海军, 武悦, 邓羽, 等. 基于城市流和层级性的城市群扩展模型构建[J]. 地理学报, 2021, 76(12):3012-3024.
[
|
[20] |
|
[21] |
郭旭东, 谢俊奇, 何挺. 基于斑块层次的土地利用变化对土地质量影响的初步分析——以河北省张北县为例[J]. 地理科学进展, 2006, 25(3):116-127,140.
[
|
[22] |
|
[23] |
王海军, 张彬, 刘耀林, 等. 基于重心-GTWR模型的京津冀城市群城镇扩展格局与驱动力多维解析[J]. 地理学报, 2018, 73(6):1076-1092.
[
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
|
[31] |
周俊, 凌昌隆. 长江中游城市群网络联系度与度中心性时序变化研究(2001—2016)[J]. 长江流域资源与环境,2021, 30(3):544-553.
[
|
[32] |
刘涛, 陈忠, 陈晓荣. 复杂网络理论及其应用研究概述[J]. 系统工程, 2005, 23(6):1-7.
[
|
[33] |
|
[34] |
|
[35] |
|
[36] |
|
[37] |
|
[38] |
李从欣, 张再生, 李国柱. 解决多重共线性的新思路:路径分析[J]. 统计与决策, 2013(1):28-30.
[
|
[39] |
乔文怡, 黄贤金. 长三角城市群城镇用地扩展时空格局及驱动力解析[J]. 经济地理, 2021, 41(9):162-173.
[
|
[40] |
熊健, 范宇, 张振广, 等. 区域协调与空间治理背景下的上海大都市圈空间协同规划编制创新探索[J]. 城市规划学刊, 2022(2):76-82.
[
|
[41] |
|
[42] |
|
[43] |
|
[44] |
|
[45] |
|
/
〈 | 〉 |