ISSN 1794-6190 e-ISSN 2339-3459 https://doi.org/10.15446/esrj.v27n3.1056532 EARTH SCIENCES RESEARCH JOURNAL Earth Sci. Res. J. Vol. 27, No. 3 (Septiembre, 2023): 289 - 297 EN V IR O N M EN TA L G EO LO G Y Analysis of Green Infrastructure Network Pattern Change in Zhengzhou Central City Based on Morphological Spatial Pattern Analysis Zhetao Xiao1, Donghui Zhao2, Lijun Hao2* 1.School of Architecture, North China University of Water Resources and Electric Power, Zhengzhou 450046, China 2.School of Art and Design, North China University of Water Resources and Electric Power, Zhengzhou 450046, China *Corresponding author: haolj35@ncwu.edu.cn.com ABSTRACT: Manuscript received: 26/12/2022 Accepted for publication: 31/08/2023 How to cite this item: Xiao, Z., Zhao, D., & Hao, L. (2023). Analysis of Green Infrastructure Network Pattern Change in Zhengzhou Central City Based on Morphological Spatial Pattern Analysis. Earth Sciences Research Journal, 27(3), 289-297. https://doi.org/10.15446/ esrj.v27n3.1056532 Keywords: Landscape architecture; Morphological Spatial Pattern Analysis; Green Infrastructure; Zhengzhou central city Green infrastructure is generally regarded as an effective way to maintain regional ecological security, and its construc- tion method plays an essential role in its function realization. In this study, the central urban area of Zhengzhou was taken as the research area. The land use type maps of the study area were obtained through the supervised classification of TM/ETM Remote sensing image data in 2007, 2011, 2015, and 2019. The Green Infrastructure (GI) in the study area was obtained by Morphological Spatial Pattern Analysis (MSPA) based on the landscape connectivity index and landscape map theory. The landscape connectivity of the core area was classified and quantitatively evaluated. Then, the potential ecological corridors were extracted and analyzed. The GI pattern change research based on MSPA, landscape connectivity, and map theory provided a new framework for analyzing GI pattern change. Finally, the spatial structure of green infrastructure planning in Zhengzhou central city is obtained, and the optimization strategy of network struc- ture is proposed. The research results can provide references for the planning and optimization of GI networks in the study area and have particular reference significance for the planning and construction of GI networks in other areas. Análisis de los cambios de patrones en las redes de infraestructura verde en la ciudad de Zhengzhou, China, con base en el Análisis Morfológico de los Patrones Espaciales RESUMEN: La infraestructura verde es comúnmente considerada una vía efectiva para mantener la seguridad ecológica regional y su método de construcción juega un papel esencial en su funcionalidad. En este estudio se definió el área central de Zhengzhou como la zona de investigación. Los mapas de uso del suelo en el área de estudio se obtuvieron a través de la clasificación supervisada de la información de imágenes Landsat (TM/ETM) en 2007, 2011, 2015 y 2019. La infraestructura verde en el área de estudio se obtuvo con el Análisis Morfológico de los Patrones Espaciales (del inglés Morphological Spatial Pattern Analysis, MSPA) con base en el índice de conectividad del paisaje y la teoría del mapa del paisaje. La conectividad del paisaje en la zona central del área de estudio se clasificó y se evaluó cuantitativamente. Luego se extrajeron y se analizaron los corredores ecológicos potenciales. La investigación sobre el cambio en el patrón de la infraestructura verde con base en el método MSPA, en la conectividad del paisaje y en la teoría de mapas significó un nuevo marco teórico para analizar el cambio de patrón de este tipo de infraestructura. Finalmente se obtuvo la estructura espacial de la infraestructura verde planeada por la administración de la ciudad de Zhengzhou y se propuso una estrategia de optimización de esta red estructural. Los resultados de la investigación pueden ofrecer referencias para la planeación y optimización de redes de infraestructura verde en el área de estudio y tiene una referenciación particular en estos procesos de planeación y construcción en otras áreas geográficas. Palabras clave: arquitectura del paisaje; Análisis Morfológico de los Patrones Espaciales; infraestructura verde; ciudad de Zhengzhou https://doi.org/10.15446/esrj.v27n3.1056532 mailto:haolj35@ncwu.edu.cn.com https://doi.org/10.15446/esrj.v27n3.1056532 https://doi.org/10.15446/esrj.v27n3.1056532 290 Zhetao Xiao, Donghui Zhao, Lijun Hao 1. Introduction The rapid development and construction of the cities improve their economic efficiency but lead to aggravated destruction of ecological environments. The city’s expansion leads to the fragmentation of ecological networks (Qiu, Chang, & Wang, 2013), and even the retrogressive reduction of ecological networks and the reduction of habitat core areas, which seriously damage the ecological service function of the landscape. The cut-off of ecological corridors leads to a decrease in biological migration channels and affects the development of biodiversity. In a word, the process of urbanization seriously affects the health and integrity of the ecosystem. Green infrastructure (GI), proposed in the 1990s, was initially separated from eco-green environmental network infrastructure and opposed to other conventional infrastructure (grey infrastructure). GI is a network system that contains a variety of natural and restored ecological and landscape elements, according to the definition of the United States Government Joint Expert Group (Benedict, & McMahon, 2012). So far, its connotation and research scale are constantly enriching. Research on various regional scales by MSPA in China and abroad has been continuously carried out and perfected in recent years. As an example, at the national level, there are studies on GI change assessment for the United States territory and forest fragmentation and connectivity throughout Europe. At the urban level, there are studies about the GI development of Virginia and its surrounding states, the studies on GI of the border between Austria and Hungary, and the studies on the construction of related ecological networks in Sichuan and Chongqing, Jiangsu, and Zhejiang of China (Chen, 2015). On the medium and small scale, there are studies about the spatial and temporal changes of ecological network patterns in Guangzhou, Shenzhen, Nanjing, and some counties in China (Zhu, Ren, & Liu, 2019; Chang et al., 2013; Yu et al., 2016; Byrne, Lo, & Jianjun, 2015). Zhengzhou, a national central city, is the core city that plays a leading role in the Central Plains Urban Agglomeration. It is rich in natural resources and runs through the Yellow River. However, during its development, urban diseases become increasingly prominent, including the weakening of urban function, and the poor effect of rain and flood caused by habitat fragmentation. The changes and construction of its green infrastructure should be studied to make it a leader in the development and construction of green infrastructure and to drive the green infrastructure construction of other cities, which is of great significance to the formation of green ecological networks of the Central Plains Urban Agglomeration 2. Applicability of MSPA Introduction to MSPA MSPA (Morphological Spatial Pattern Analysis), a morphological-based spatial pattern analysis, is initially used by scholars at European Union Joint Research Center (JRC) to study forest fragmentation and species conservation (Ostapowicz, 2008). MSPA is an image processing method to analyze the spatial pattern of raster images based on the principle of binary morphology in mathematical morphology. It can classify image pixels according to the corresponding landscape types by accurately recognizing, measuring, and segmenting the images to distinguish the landscape types and structures. Combining the MSPA with the minimum path analysis, the connectivity among the structural landscape elements and elements can be well identified and analyzed, and the potential ecological corridor can be obtained. The emphasis of ecological network research is shifted to the connectivity level of the ecological source site and ecological corridor, which solves the subjectivity of other research methods and makes the constructed ecological network more scientific. Ecological correspondence between MSPA and GI (1) Definition of GI GI is easily understood as the infrastructure related to green space, like engineering facilities. Its more accurate definition is the network of natural areas, open space with internal connectivity, and possible ancillary engineering facilities. This network has the function and value of a natural ecosystem. It provides natural places for human beings and wildlife, such as habitat, clean water sources, and migration channels, constituting the ecological framework to ensure the sustainable development of the environment, society, and economy. At the micro level, GI refers to the patches and corridors of specifically related engineering facilities or green spaces, such as flood control systems, water purification facilities, a secondary forest, a tree, or even a green roof. In the absence of a specific reference, it emphasizes the connectivity of the whole. Specifically, GI is a multi-level system ranging from the land-wide ecological protection network to the street rainwater garden (Liu et al., 2013). At the regional level, GI supports functions of critical ecosystems constituted by national parks, coastlines, major river corridors, long-distance footpaths, and so on. At the urban or community level, GI forms an open space network consisting of urban parks, recreational areas, farmland, community gardens, street landscapes, private gardens, cemeteries, small water bodies and streams, rooftop gardens, etc. (2) MSPA and GI GI is a natural and artificial green space network system composed of “hubs” and “link corridors”. MSPA analyzes the geometric characteristics and connectivity of binary images of any scale and type by dividing it into seven categories that do not overlap, so it can also be used to construct GI networks (Wang & Lin, 2017). First, after it is reclassified using the land cover change data, the GI elements are extracted as “foreground” and other non-GI elements as “background”. Then, through a series of image processing techniques, the foreground is divided into seven categories that do not overlap (Table 1): The “center” is the “hub” of the GI, and the “bridge” is the “corridor” of the GI. Finally, the GI network is constructed according to the “center” and “bridge” of the MSPA to determine the GI elements and patterns, find the potential “hubs” and “corridors”, and realize the interconnected network. Table 1. Landscape ecology meanings for various MSPA elements Type of element Landscape ecology meaning Core area Large green patches and distinguishable nonlinear landscape elements in space. It is the «source» of various ecological processes, providing habitat for wildlife, and plays an important role in the protection of biodiversity and the maintenance of urban ecological environment. Connecting bridge The channel connecting the adjacent core areas and a narrow zone different from the landscape elements on both sides. It has corridor properties and is a channel for the exchange and flow of energy flow and material flow. It has a direct impact on species mobility and landscape connectivity. Marginal zone The transition region between the periphery of the core area and different landscape elements. It has the attributes of landscape boundary in ecology. The adjacent landscape permeates, connects and distinguishes each other through the marginal zone, and has not only characteristics of adjacent landscape but also its own unique components. High heterogeneity. Branch line The passage connecting the core area and the peripheral landscape. It also has corridor properties, but its connectivity is generally poor. Ring zone The internal passage connecting the same core area. It also has corridor properties, and enhances species diffusion and energy flux within the core area. Isolated island Obviously different from the surrounding landscape elements, isolated and broken small patches. The internal species communicate less likely with external species. However, as a small green space in the built area, it plays an important role in the construction of the whole ecological network. Pore The transition zone between the core area and its internal construction land. It also has marginal effects, reflecting the degree of species isolation and the possibility of potential gene variation. 291 Analysis of Green Infrastructure Network Pattern Change in Zhengzhou Central City Based on Morphological Spatial Pattern Analysis 3. Research Areas and Research Methods Research areas Zhengzhou, a core city of the Central Plains Urban Agglomeration, is in central China. It is in the Yellow River’s middle and lower reaches and the Funiu Mountains’ northeast wing to the Huanghuai Plain transition zone. The west is high, the east is low, the middle is high, the northeast is low, and the southeast is low. It has a north-temperate continental monsoon climate, over 100 rivers, and 6 municipal districts. According to the latest approved Zhengzhou Master Planning (2010-2020), Zhengzhou central city includes five districts within the administrative area of Zhengzhou City: Zhongyuan District, Jinshui District, Erqi District, Guancheng District, Huiji District. The total area measures 990 square kilometers. Data source acquisition and preprocessing This study mainly used the remote sensing image data of Zhengzhou city with the Landsat5\Landsat8 satellite TM/ETM format in 2007, 2011, 2015, and 2019, with 30m precision. First, atmospheric and radiation corrections were conducted, and regions of interest were created using ENVI (Xie, Wang, & Luo, 2020). Then, through the supervision classification, visual correction, and field investigation, the remote sensing images were divided into six categories: woodland, cultivated land, grassland, water area, construction land, and unused land. Finally, based on the Zhengzhou Master Planning and the classification of land use types, the six categories were corrected to form the classification map of land use types in Zhengzhou in the four years. Research methods ArcGIS performed mask extraction and re-classification, and the classification map of land use types in Zhengzhou central city was extracted. In graphic data, the four GI elements of woodland, cultivated land, grassland, and water area were extracted as “foreground”, and other non-GI elements of construction land and unused land were extracted as “background”. And 30m x 30m TIFF binary raster data files were generated and imported to the Guidos Toolbox. Based on the evaluation of landscape connectivity of seven types of landscape elements in Zhengzhou in 2019, the classification of core areas and the extraction of potential ecological corridors were carried out, and the GI planning strategy of Zhengzhou central city was put forward. 4. Changes of GI Spatial and Temporal Pattern of Zhengzhou Central City by MSPA The data of MSPA landscape type changes in Zhengzhou central city from 2007 to 2019 were obtained. Table 2 shows that the total GI area of Zhengzhou central city was decreasing year by year, which, however, was slowing down. It decreased by 112.167 km2 from 2007 to 2011, 36.927 km2 from 2011 to 2015, and 52.074 km2 from 2015 to 2019. GI decreased slowest from 2011 to 2015 but quickened in recent years. The above results also indirectly showed that the construction land area in Zhengzhou central city has been increasing yearly. Table 3 shows that the core area was dominant in the landscape type. The area of core area decreased from 449.856 km2 in 2007 to 185.724 km2 in 2019, the ratio to total GI area decreased from 70.68% in 2007 to 42.69% in 2019, and the ratio to the total area of the central city decreased from 45.44% in 2007 to 18.76% in 2019, meaning a significant decrease. This is because Zhengzhou city constantly expands, and its increasing construction land encroaches on other land. Although the isolated islands are not connected, they can play a stepping-stone role of “ecological island hopping” in constructing ecological networks (Wang, Shen, & Jin, 2019) and are likely to form a connecting bridge or core area with the increase of their area. The area of isolated islands increased from 17.424 km2 in 2007 to 40.788 km2 in 2019, the ratio to total GI area increased from 2.73% in 2007 to 9.37% in 2019, and the ratio to the total area of the central city increased from 1.76% in 2007 to 4.12% in 2019, meaning a significant increase. The connecting bridge is vital in constructing ecological networks, such as connecting core areas, material energy exchange, and biological migration. The area of connecting bridge increased from 13.068 km2 in 2007 to 29.997 km2 in 2019, the ratio to total GI area increased from 2.06% in 2007 to 6.9% in 2019, and the ratio to the total area of the central city increased from 1.32% in 2007 to 3.03% in 2019. It showed that the construction of ecological corridors and GI connectivity in Zhengzhou’s central city were on the rise, promoting the overall optimization of ecological network construction in Zhengzhou. In addition, there was almost no change in the ring zone, and the proportion of the pore area was reduced, indicating that the internal integrity of the core area was enhanced. The proportion of the marginal area increased slightly, indicating that the tiny core area increased and the core area fragmentation increased. An increase in the proportion of the area of the branch line indicated an increase in the green belt between construction land and GI. Table 2. The total landscape GI area in Zhengzhou central city from 2007 to 2019 based on MSPA (km²) Year Total GI area(km²) 2007 636.372 2011 524.205 2015 487.278 2019 435.204 Table 3. The changes in various landscape types in Zhengzhou central city from 2007 to 2019 based on MSPA Landscape types Year Area(km²) Ratio to total GI area(%) Ratio to the total area of central city(%) Core area 2007 449.856 70.68 45.44 2011 302.247 57.67 30.53 2015 254.133 52.14 25.67 2019 185.724 42.69 18.76 Isolated island 2007 17.424 2.73 1.76 2011 27.621 5.27 2.79 2015 32.175 6.61 3.25 2019 40.788 9.37 4.12 Pore 2007 41.679 6.55 4.21 2011 27.225 5.19 2.75 2015 19.8 4.05 2 2019 13.563 3.11 1.37 Marginal zone 2007 79.299 12.47 8.01 2011 103.059 19.67 10.41 2015 113.355 23.27 11.45 2019 114.048 26.21 11.52 Ring zone 2007 10.494 1.65 1.06 2011 9.603 1.82 0.97 2015 9.702 2 0.98 2019 9.9 2.26 1 Connecting bridge 2007 13.068 2.06 1.32 2011 23.463 4.47 2.37 2015 21.681 4.45 2.19 2019 29.997 6.9 3.03 Branch line 2007 24.552 3.86 2.48 2011 30.987 5.91 3.13 2015 36.432 7.48 3.68 2019 41.184 9.46 4.16 Figure 1 shows that the construction land area in Zhengzhou central city is rapidly expanding, the peripheral core area is gradually swallowed up, the integrity of the core area is poor, the core patch is increasing, and the GI 292 Zhetao Xiao, Donghui Zhao, Lijun Hao fragmentation is becoming more and more serious. However, depending on the urban ecological construction, some core areas show a trend of converging growth, and the connectivity of some areas is enhanced. The A in Figure 1 is Guxing Town and Guangwu Town in Huiji District, the Yellow River scenic spot, the wetland beach area, and so on. It is mostly rural cultivated land, surrounded by many mountain forests, large forest scenic parks, and the Yellow River Nature Reserve. The urban construction here gradually expanded to the northwest, and plants, schools, and other construction occupied cultivated land. Since the construction of the Yellow River scenic spot railway station in 2015, the railway has split the core area here, and then the landscape fragmentation intensified. The B is Shifu Town and Jalu River of Zhongyuan District of Zhengzhou City, where there are also many cultivated lands and public green spaces such as Shifu Chensha Pool and Zhengzhou Green Valley Quality Culture Park. However, the cultivated land area was reduced due to the construction of numerous residential areas. Still, the core area of landscape along the Jalu River increased. The C is a large number of mountain terraces, as well as Changzhuang Reservoir, Jalu River, South-to-North Water Transfer main trunk canal, and other water bodies. The core area here also has a decreasing trend and fragmentation state, but the process of decreasing is relatively slow. Many water core areas, especially the landscape core areas on both sides of the central trunk canal of the South-to-North Water Transfer Middle Line, are gradually connected. The D is a part of Jinshui District, including the Yellow River Wetland Reserve in the north, Zhengzhou CBD (Central Business District) and Longzi Lake University Park, as well as the Jalu River, Longhu Lake, Dongfeng Canal, Xiong’er River, Weihe River, Longzi Lake, and other waters. There are some cultivated lands along the Yellow River in the north, where the city’s expansion to the north is evident, and the landscape core area needs to be fixed. However, the landscape core area of the river region is also gradually connected as a whole. The E is part of the Guancheng District, where some cultivated land is in Nancao Township. The landscape core area is more seriously broken, and the core area of Qilihe River, Chaohe River, and Diehu Lake is well developed. 2007 2011 2015 2019 293 Analysis of Green Infrastructure Network Pattern Change in Zhengzhou Central City Based on Morphological Spatial Pattern Analysis Figure 1. Temporal and spatial variation in green infrastructure of Zhengzhou central city in 2007, 2011, 2015, and 2019 based on MSPA. 2007 2011 2015 2019 A B C D E 294 Zhetao Xiao, Donghui Zhao, Lijun Hao 5. Classification of Green Infrastructure in Zhengzhou Central City 5.1. Classification of core areas In this study, the possible connectivity index (PC) was selected to evaluate the landscape connectivity of the core area (Gao, Huang, & Li, 2019) and the calculation formula of PC was as follows: PC = i j n i j ijj n L a a p A �� �� ���� �� 1 2 (1) Where “n” is the total number of patches in the landscape, “p*ij” is the probability of direct diffusion of species in the patch “i” and patch “j.” “ai” and “aj” are the areas of the patch “i” and patch “j,” respectively. “AL 2” is the total area of the landscape. Referring to the relevant research, the threshold of patch connectivity distance was set to 500 m, and the connectivity probability was set to 0.5 (Xu et al., 2015). The area of the core area (S) was classified into three levels according to the size of the area (Figure 2): S ≤ 0.2km², 0.2