Spatial differential characteristics and driving factors of land urbanization in Anhui Province
ZHANGKe1,2,3,, CHENGJiumiao1,3,, FEILuocheng1,3, HONGDehe1,3 1. College of Geography and Tourism, Anhui Normal University, Wuhu 241000, China2. Shanghai Grand Planning & Design Co.,Ltd., Shanghai 201800, China;3. Center for Land Evaluation and Planning, Anhui Normal University,Wuhu 241000, China 通讯作者:通讯作者:程久苗,E-mail:jmcheng@mail.ahnu.edu.cn 收稿日期:2018-01-19 修回日期:2018-07-15 网络出版日期:2018-10-25 版权声明:2018《资源科学》编辑部《资源科学》编辑部 基金资助:国家自然科学基金青年项目(71603003) 作者简介: -->作者简介:张珂,女,安徽铜陵人,硕士生,主要从事土地利用规划与土地政策研究。E-mail:zkkiki@126.com
关键词:土地城镇化;空间分异特征;驱动因素;安徽省 Abstract Measuring the spatial change of land urbanization scientifically and achieving a new type of urbanization with Chinese characteristics by the trinity of population, land and industry, is the key question to promote the high-quality development of urbanization in China. We explored and analyzed the overall spatial distribution, spatial differentiation pattern and regional spatial disparities of land urbanization by using the spatial autocorrelation indicators adjusted by empirical bayes (EBI) in Anhui Province. And we adopted geographically weighted regression (GWR) supported by R to identify the driving factors, so that it can provide decision-making reference for promoting the quality of new urbanization and balancing the development between regions. We found: ①Regional difference of land urbanization is obvious in Anhui Province,high-value aggregation space form like patch was increasingly emerging with some stadsdistricts of Tongling、Hefei、Ma’anshan、Bengbu、Huangshan、Wuhu、Huainan and Chuzhou as the core. The circular spatial structure extending to the surrounding area was a low-value area with the main agricultural producing area as the core. ②The spatial correlation existed in the level of land urbanization in Anhui Province and local spatial differentiation was significant. ③In addition, land urbanization was affected by the population, industries, investment and output. The influence intensity was characterized by regional differences. Driving types of land urbanization along with Huaihe River could be classified as population-investment type in north of the river and population-industry type in south.
以组间方差最大、组内方差最小为聚类条件,运用Jenks自然断点法对研究区域进行聚类分析[47,48,49]。将安徽土地城镇化水平由低至高划分为5个等级:0.092 70~0.212 10;0.212 11~0.344 40;0.344 41~0.497 00;0.497 01~0.688 10;0.688 11~0.877 60,得到安徽省土地城镇化水平定量空间分布图(图1)。结果显示:安徽省土地城镇化水平总体呈南高北低、东高西低分布态势,且空间结构特征较为显著:以铜陵、合肥、马鞍山、蚌埠、黄山、芜湖、淮南及滁州等地市的部分市辖区为核心,构成土地城镇化高值聚集斑块;以安徽省农产品主产区为核心,向四周渐进式扩展的环绕式空间结构,为土地城镇化水平较低区域。 显示原图|下载原图ZIP|生成PPT 图12015年安徽省土地城镇化水平空间分布示意 -->Figure 1The spatial distribution of land urbanization in Anhui Province in 2015 -->
3.2 基于EBI的全局空间分异格局
采用经验贝叶斯修正的全局自相关指数对研究单元进行空间分析,基于Rook的多边形邻接性生成空间权重矩阵,得到EBI散点分布图及序列经验分布图(图2)。结果显示土地城镇化水平EBI值为0.363 52,Z值为6.254 20,大于正态分布95%置信水平下的验阈值1.96。表明安徽省土地城镇化水平存在空间集聚性且较为显著,即土地城镇化水平较高的县级单元,其邻近单元的属性值亦较高,土地城镇化水平较低的县级单元其周围县级单元的属性值亦较低。 显示原图|下载原图ZIP|生成PPT 图22015年安徽省土地城镇化Moran散点及序列经验分布 -->Figure 2The EBI scatter diagram and sequence experience of land urbanization level in Anhui Province in 2015 -->
3.3 基于EBIi的局部空间差异分析
基于经验贝叶斯修正的局部自相关指数深入探索区域与其周边单元之间的局部空间关联性,生成LISA集聚图(图3)。低低(LL)区主要出现在宿州、阜阳、亳州、六安及淮南,形成“低值聚集面”;由高值县区形成高高(HH)聚集点,包括蚌埠蚌山区、合肥蜀山区、包河区及庐阳区、芜湖弋江区,形成“高值点”;唯一高低(HL)区位于阜阳市颍州区,即颍州区土地城镇化水平远高于邻近县级单元的土地城镇化低值;低高(LH)区主要是马鞍山博望区及当涂县、黄山歙县,三个区域的土地城镇化水平较低,被周围土地城镇化水平高值地区包围,故形成“孤立点”,与土地城镇化水平总体空间分布较为一致。 显示原图|下载原图ZIP|生成PPT 图32015年安徽省土地城镇化EBIi集聚状态示意 注:行政区编码参见图1。 -->Figure 3The EBIi cluster state of land urbanization level in Anhui Province in 2015 -->
4 R语言支持下GWR土地城镇化驱动因素识别
4.1 因子选取
土地城镇化受多重因素影响,机制较为复杂。既有研究成果显示:人口、产业、投资、产出均对城镇土地扩张具有推动作用[50,51,52],但因区域资源禀赋、经济发展水平的不同而存在些许偏差。鉴此,基于现有研究成果及数据可获取性,从人口、产业、投资、产出等四个驱动层次选取驱动因子,如表2所示。城镇化进程中,农村人口向城镇转移,人口城镇化水平可在一定程度上反映人口城乡流动对土地城镇化的影响,故将其纳入人口驱动指标。二三产业比重可反映产业结构调整及科技创新等因素对土地城镇化的影响,故将其纳入产业驱动指标。投资拉动经济发展,地均固定资产投资总额可直接体现以基础设施为主的城市建设对土地城镇化的影响,故选为投资驱动的考察指标之一。投资主体的多元化,推动了区域经济的全面发展,与其紧密相连的地均实际利用外资金额可反映城市投资与土地城镇化的相关关系,故纳入投资驱动指标。地均GDP作为反映区域经济产出的常用指标,可体现经济综合发展水平对土地城镇化的影响,故选取为产出驱动层指标之一。地均工业产值可直接表征区域工业化水平,一定程度上反映工业发展对土地城镇化的影响,故纳入产出驱动层指标。地均公共财政收入的高低是区域公共产品供给能力与产业集聚能力的体现,故将其纳入产出指标层,以表征区域财政收入与土地城镇化的相互关联。地均社会消费品零售总额除反映区域市场发育水平外,同时亦是区域经济活力的重要表现,故选取为产出 指标。 Table 2 表2 表2安徽省土地城镇化驱动因子理论体系 Table 2The theory framework of driving factors of land urbanization in Anhui Province
驱动层
指标层(变量,单位)
预期影响方向
人口
人口城镇化水平(Pop,%)
+
产业
二三产业比重(Bz,%)
+
投资
地均固定资产投资总额(Pgdzc,元/km2)
+
地均实际利用外资金额(Pwz,元/km2)
+
产出
地均GDP(PGDP,元/km2)
+
地均工业产值(Pgycz,元/km2)
+
地均公共财政收入(Pczsr,元/km2)
+
地均社会消费品零售总额(Pshxf,元/km2)
+
新窗口打开 假设上述因子对安徽省土地城镇化发展均会形成有效影响,在统计分析软件SPSS20.0中,将安徽省土地城镇化水平选取为因变量,各驱动因子为自变量,导入回归分析模型,并以逐步回归法(stepwise)进行分析[53,54],回归结果显示地均工业产值、地均公共财政收入、地均GDP等3个变量因共线性较强需剔除(表3),其余5个变量均进入模型。至此,得到安徽省土地城镇化的有效驱动因子指标体系(表4)。 Table 3 表3 表3基于逐步回归结果的安徽省土地城镇化驱动因素排除变量 Table 3The exclusive variables of driving factors of land urbanization based on the stepwise regression results in Anhui Province
模型
Beta In
t
Sig.
偏相关
容差
地均工业产值
-0.017 00
-0.811 00
0.419 00
-0.082 00
0.877 00
地均公共财政收入
0.001 00
0.063 00
0.950 00
0.006 00
0.990 00
地均GDP
0.005 00
0.234 00
0.816 00
0.024 00
0.971 00
新窗口打开 Table 4 表4 表4安徽省土地城镇化驱动因子指标体系 Table 4The index system of driving factors of land urbanization in Anhui Province
空间权重矩阵为GWR运算时的一项重要参数,反映105个研究单元所代表的各要素间的空间关系。空间关系概念化涵盖多种定义,本研究选取符合“地理学第一定律”的反距离权重[55,56,57]。将shp图层导入R语言,调用spdep包进行读取,绘制中心点,设置queen=T,以nbdists函数计算空间邻域距离,nb2listw函数将其转换为空间权重矩阵,最后以plot、colornumeric等命令进行可视化[58,59,60](图4)。由图4可知,各研究单元的空间距离越近(即位于色带的红色端),其权重越大,反之,距离越远(位于色带的紫色端),则权重越小。 显示原图|下载原图ZIP|生成PPT 图4基于R语言的空间权重矩阵可视化注:行政区编码参见图1。 -->Figure 4Visualization of spatial weight matrix based on R -->
4.3 GWR分析结果
以2015年安徽省土地城镇化水平为被解释变量,以5个驱动因子为解释变量,运用ArcGIS10.0软件中的空间关系建模模块,构建GWR模型,其拟合优度为0.813 50,参数估计结果见表5。进一步对标准化残差进行全局自相关分析,其Moran’s I系数为0.076 30,P-Value=0.190 00,Z值检验结果非显著,即标准化残差并未出现空间聚类现象,呈随机分布,GWR模型整体高度适用。 Table 5 表5 表5GWR模型参数估计结果 Table 5Parameter estimation results of GWR model
GWR模型估计结果显示:五个驱动因子的回归系数均为正值,即人口城镇化水平、二三产业比重、地均固定资产投资总额、地均实际利用外资金额、地均社会消费品零售总额的提高对安徽省土地城镇化发展具有正向推动作用,与预期影响方向一致。各因素影响强度为:人口城镇化率>二三产业比重>地均实际利用外资金额>地均固定资产投资总额>地均社会消费品零售总额。值得注意的是,各因素之间区域差异性显著且省内各区域的主导驱动因子也不尽相同。 (1)人口城镇化率。从全省看,县级空间单元的人口城镇化水平与土地城镇化水平呈正相关,相关系数呈“北-中-南”递增态势,且东部高于西部,形成“阶梯式”空间形态,其中低值位于阜亳片区,高值位于铜池、黄山片区(图5a)。深究之,这种区域差异特征与人口的流动密切相关。皖南地区自然资源丰富,乡村旅游休闲产业较发达,实现了旅游带动下的人口就地城镇化,外出流动较少。皖北地区作为省内典型劳动力输出地,跨省市人口外流现象突出,异地城镇化概率提高,当地城镇化的人口驱动力相对较弱。相较以现代工业体系为主的皖中地区及以第三产业为主的皖南地区,其人口城镇化与土地城镇化的正向变化关系则显得较弱。 显示原图|下载原图ZIP|生成PPT 图5基于GWR模型的土地城镇化水平驱动因素回归系数空间分布 注:行政区编码参见图1。 -->Figure 5The driving factors of land urbanization level, regression coefficient spatial distribution map based on GWR model -->
(1)土地城镇化的空间分异特征表现出不同区域间土地资源配置与经济发展水平差异的相互关联。安徽省区域差异特征显著,应制定新型城镇化的区域差异化政策,以期缓解地区差异,实现新型城镇化由高水平向高质量的转变,推动区域协调发展与城乡一体化。在人口-投资驱动型的皖北地区,应推动人口就地城镇化,创建完善的人才支持体系,适当扩大资本投入规模,拉动内需;在人口-产业驱动型的皖中南地区,加快产业结构优化升级,保持土地、人口、产业的紧密联系与协同发展[61,62,63]。 (2)本文从人口、产业、投资、产出等视角对土地城镇化驱动因素进行选取,因指标量化及数据收集难度较大,故并未涉及相关政策因素,指标体系的全面性有所欠缺。因此,如何将政策因素纳入指标体系以进行驱动因素的全面分析,在后续研究中将进行深入探讨。 (3)本文以2015年截面数据为支撑对土地城镇化进行空间分异研究,仅考虑空间特征,未形成系统的时间序列,对土地城镇化空间分异特征在时间轨迹上的变化未做深入分析。在保证数据完整的基础上,尽可能多地选取时间点,全面分析土地城镇化的时空分异特征将为今后主要的研究方向。 The authors have declared that no competing interests exist.
[YuZ N, WuC F.Analysis on spatial characteristics and influence factors of land urbanization from the perspective of ESDA-GWR in Zhejiang province [J]. , 2016, 30(3): 29-36. ] [本文引用: 1]
[WangF, LiC. Analysis on spatial distribution patterns of urban construction land in Guangdong Province based on ESDA [J]. ,2014, 21(2): 167-171+17. ] [本文引用: 1]
[BeiH L, WuC F, FengK, et al. Regional disparity and dynamic evolution of land economic density evidence from the Yangtze River Delta Area [J]. , 2009, 23(11): 1952-1962. ] [本文引用: 1]
[HuoB N.Research on the factors influencing the development of regional urbanization in China-based on the empirical analysis of provincial panel data from 1992 to 2015 [J]. , 2017, 37(4): 76-82. ] [本文引用: 1]
[OuX J, ZhenF, QinY D, et al. Study on compression level and ideal impetus of regional urbanization: the case of Jiangsu Province [J]. , 2008, 27(5): 993-1002. ] [本文引用: 1]
[CaoG Z, WangC J, QiY J.The comparative analysis of urbanization affecting factors among the coastal provinces in eastern China in the transition period [J]. , 2008, 32(6): 1399-1406. ] [本文引用: 1]
[24]
FotheringhamA S, BrunsdonC, CharltonM E.Geographically Weighted Regression: The Analysis of Spatially Varying Relationships [M]. , 2002. [本文引用: 1]
[CaoT B, HuangK L, LiJ B, et al. Research on spatial variation and evolution of residential land price in Nanjing based on GWR Model [J]. , 2013, 32(12): 2324-2333. ] [本文引用: 1]
[ZhangJ, ZhangL F, PuL J, et al. Research on spatio-temporal variation of urban residential land price based on GWR model: a case study of Jiangsu Province [J]. , 2012, 32(7): 828-834. ] [本文引用: 1]
[ShaoY X, LiM C, ChenZ J, et al. Simulation on regional spatial land use patterns using geographically weighted regression: a case study of Menghe Town, Changzhou [J]. , 2010, 30(1): 92-97. ] [本文引用: 1]
[RenG P, LiuL M, FuY H, et al. Spatial differentiation of rural household livelihood capital in metropolitan suburbs based on GWR model: a case study of Qingpu District in Shanghai [J]. , 2016, 38(8): 1594-1608. ] [本文引用: 1]
[ZhangF, KongW.Analysis on spatio-temporal characteristic and influence mechanism of land urbanization in China [J]. , 2014, 32(5): 144-148. ] [本文引用: 1]
[ Gazette of the State Council of the People's Republic of China. Some Opinions on Further Promoting the Construction of New Urbanization [EB/OL]. (2016- 02- 06) [2018- 07- 10]. ]URL [本文引用: 1]
[PanA M, LiuY J.The degree of imbalance between population urbanization and land urbanization of Xiangjiang River Basin [J]. , 2014, 33(5): 63-68. ] [本文引用: 1]
[ Gazette of People's Government of Anhui Province. Notice on the Issuance of the General Plan of New Urbanization in Anhui Province as National Pilot [EB/OL]. (2015- 04- 22) [2018- 07- 10]. ]URL [本文引用: 1]
[LvP, ZhouT, ZhangZ F, et al. Construction and application of land urbanization and corresponding measurement index system [J]. , 2008, 22(8): 24-28. ] [本文引用: 1]
[YanB B.Connotative urbanization: the connotation, restricting factors and the realization way of the complete urbanization [J]. , 2013, 19(6): 88-92. ] [本文引用: 1]
[YuY L, LiuZ Q, LinC H, et al. The impact analysis of the land urbanization to green space rate of built district in China: based on the time series data and the provincial panel data from 1996 to 2013 [J]. , 2017, 23(1): 143-147. ] [本文引用: 1]
[WangL Y, ZhenD, YouB.Research on realization of benign interaction between population urbanization and land urbanization [J]. , 2014, 24(12): 62-69. ] [本文引用: 1]
[ZhuM M, DongY, TianL L, et al. Coupling coordination and spatial pattern among population, economic and land urbanization in Hubei [J]. , 2017, 39(5): 531-538. ] [本文引用: 1]
[GuoF Y, LiC G, ChenC, et al. Spatial-temporal coupling characteristics of population urbanization and land urbanization in Northeast China [J]. , 2015, 34(9): 49-56. ] [本文引用: 1]
[ZhuH Y, ZhangY F.Coupled coordination evolution of population urbanization, industry urbanization and land urbanization of the key cities in Qinba Mountain Area [J]. , 2017, 36(1): 40-44. ] [本文引用: 1]
[42]
MarshallR J.Mapping disease and mortality rates using empirical Bayes estimators [J]. , 1991, 40(2): 283-294. [本文引用: 1]
[LiuY L, LiJ W, HouH P, et al. Study on urbanization rate of urban-rural construction land and its influencing factors: a case study of Hubei Province [J]. , 2014, 32(1): 132-142. ] [本文引用: 1]
[43]
[WangF H.[M]. Beijing: The Commercial Press, 2009. ] [本文引用: 1]
[44]
AnselinL. Exploring Spatial Data with GeoDa: A Workbook [EB/OL]. (2005-03-06)[2018-01-19]. .URL [本文引用: 1]
[ZhangL X, ZhuD L, DuT, et al. Spatio-Temporal pattern evolvement and driving factors of land urbanization in Yangtze river economic belt [J]. , 2017, 26(9): 1295-1303. ] [本文引用: 1]
[47]
JenksG F.The data model concept in statistical mapping [J]. , 1967, 7(1): 186-190. [本文引用: 1]
[WuZ H, LiT.The comprehensive performance evaluation of the high-tech development zone: analysis based on the natural breakpoint method [J]. , 2013, 28(3): 82-88. ] [本文引用: 1]
[ShuB R, LiY L, QuY, et al. Urban land expansion characteristics and its forces under different stages of economic development: a case study of Taicang City [J]. , 2013, 33(7): 155-162. ] [本文引用: 1]
[FengH C, YangQ S.Spatio-temporal characteristics of urban expansion and its driving forces based remote sensing data in Jiansanjiang Reclamation Area [J]. , 2017, 37(8): 1178-1185. ] [本文引用: 1]
[RongH F, FangB.Measurement of the matching degree between urbanization and ecology in Anhui based on barycenter model [J]. , 2017, 31(6): 34-41. ] [本文引用: 1]
[SongL M.Will urbanization promote the upgrading of industrial structure? Based on the empirical analysis of panel data of 30 provinces in 1998-2014 [J]. , 2014, 37(8): 70-78. ] [本文引用: 1]