Bird Flight Resistance Analysis and Planning Strategies in Urban Regeneration Areas: A Case Study of a Certain Area in Shenzhen, China

Abstract

At present, the sharp decline in urban biodiversity worldwide is severe. Conducting biological perspective analysis and proposing space construction solutions during urban regeneration can greatly alleviate the contradiction between urban construction and biodiversity conservation. In this study, birds were taken as biological representatives, and a certain area in Shenzhen with strong conservation needs was used as an example. Based on a thorough analysis of bird status, the minimum resistance model was applied to establish a resistance surface describing the real flight movement of the indicator species and to construct a bird conservation pattern from the flight process of birds. The results show that: (1) bird flight resistance is the lowest in the southern green space and northern woodland around the reservoir in the research region, and the resistance is higher in the central part, but the path of least resistance therein has the potential to become a corridor. (2) From the perspective of the community structure of green space vegetation, the sparse woodland and shrubland in the research region have low resistance and high richness, which are the most ideal green space forms for birds; from the perspective of architecture, bird flight resistance shows a negative correlation with building height and a positive correlation with building density. The final urban regeneration design plan is thus derived, and the validity of the method is verified based on the biodiversity index. This study reveals the possibility of applying bird flight resistance analysis based on the minimum resistance model to small and medium-scale urban regeneration areas, and also provides insight into the correlation between flight resistance and spatial design elements, which can assist decision-makers, planners and developers in spatial design and planning from a biological standpoint.

1. Introduction

As the main place for human settlement, cities provide living space, basic survival information and a healthy living environment for a large number of people. Urban regeneration is a significant tool for urban development. It is the process of formulating policies to address urban problems and executing and managing them in a systematic way [1,2]. The earliest authoritative definition was the first urban renewal seminar held in The Hague, Netherlands in August 1958, and it means that people living in cities should repair and transform residential houses, improve the environment, such as streets and parks, and particularly the form or region of land use, so as to create a comfortable living environment. All of these building activities concerning urban improvement can be called urban renewal [3]. In Western cities, many scholars tend to divide the development of urban regeneration into four stages [4,5]: the urban reconstruction stage before the 1960s, which predominantly consisted of demolishing and redeveloping dilapidated urban areas [6]; the urban revitalization stage from the 1960s to 1970s [7], which mainly saw the implementation of neighborhood redevelopment and the upgrading the environment of old urban slums; the urban renewal stage from the 1980s to 1990s, which involved the carrying out of market-oriented redevelopment of old cities, which still focuses on improving the physical space such as urban building form and spatial layout [8,9]; and from the late 1990s onwards, with the development of humanism and sustainable development, urban regeneration came into being. According to Roberts [10], urban regeneration was described as “a comprehensive and integrated vision and action which leads to the resolution of urban problems and which seeks to bring about a lasting improvement in the economic, physical, social and environmental condition of an area that has been subject to change” and since then ecological protection has progressively become one of the important parts of urban regeneration [11]. In the 1980s, China began to rebuild old urban areas [12]. In the 1990s, with the accelerated urbanization, the old city renewal led by real estate development was carried out. Since 2000, it has gradually transformed into a stage of diversified and integrated urban development and urban regeneration similar to that of Western cities [13,14]. In the process of urban regeneration development, the objects are also gradually enriched, from the simple physical space environment to the quality of life, social welfare, economic prospects and policies. Although the names are different in different contexts, it has basically formed a consensus that urban sustainable development is provided through the protection, restoration, reuse, redevelopment and other diversified methods of urban regeneration. Currently, with intensive urban construction and large-scale urban expansion, the ecological land in urban areas has been drastically diminished [15,16], biological processes have been blocked [17], and habitat fragmentation is on the rise [18], resulting in a sharp decline in biodiversity [19,20]. Urbanization has a significant effect on biodiversity, and urban biodiversity conservation has become an important research direction during regeneration [20]. In this study, we will focus on the relationship between the spatial design of urban regeneration and urban biodiversity. (Figure 1)

Urban biodiversity refers to “the diversity and richness of organisms (including genetic variation) and habitats found within and on the margins of human settlements” [21]. The diversity of animal and plant species in a city is one of the important criteria to measure the quality of urban ecosystems. The rich diversity of animals and plants and the stable relationship between organisms are the indicators of ecosystem health, which can be expressed by species richness, diversity, and other indices [22]. As early as the 1960s, Germany, the United States, and Japan, among others, prepared species inventories and mapped the habitat of urban flora and fauna [23]. Up to the 1990s, the research on urban biodiversity mainly focused on plants, and experts paid attention to the relationship of plant diversity and urban green space function by “ecological gardens”, “urban green space systems”, “native plants” and other related studies [24]. Since the beginning of the 21st century, with the development of geographic information technology, the research on animals has gradually increased, forming a multi-disciplinary research situation including ecology, urban science, management science, etc. In the temporal dimension, organisms may exhibit successional characteristics that show differential population distributions at different stages of urban development. Understanding how plants, animals, and other organisms respond to changes in their environment over time is key to successful restoration, rewilding, and habitat management to sustainably provide ecosystem services [25]. In the spatial dimension, the geographical and climatic environments of different cities will have different distributions of plant and animal species. The influence of complex urban environmental factors on the distribution and diversity of urban plants and animals is mainly studied from the landscape and site levels, and how to construct the protection system of plant and animal diversity in the urban environment [26,27]. In the spatial design of urban regeneration, urban planners have been increasingly integrated with theories of landscape ecology [28,29,30] to incorporate ecological environment protection into the urban planning framework [31]. To alleviate habitat fragmentation, scholars often employ the concept of “patch, matrix, corridor” to restore biological processes and enhance urban biodiversity by constructing an ecological network and ecological security pattern within the urban area [32,33]. The core of the method is to identify the matrix with significant impact on regional ecological security as the ecological source, the migration path of important organisms as the corridor, and the areas that should be protected and gently handled during urban development [34,35], which are usually applied in the larger scale of national, regional and urban areas [36].

Nevertheless, with the development of urban regeneration, the ecological land within the urban area is altering in terms of form, structure, and connectivity, and organisms with high mobility or migration capacity are inevitably interacting with the inner city. The regional ecological safety pattern and the existing ecological network cannot guarantee the continuity of biological activities to the maximum extent [37]. Therefore, it is necessary to further explore the biological movement in the inner city and propose a more reasonable space construction solution at the beginning of urban regeneration to protect the biodiversity in the inner city [38,39,40].

As a vital part of biodiversity, birds are sensitive to changes in habitat and to human activities [41], making them an important indicator of the health of urban ecosystems [41] and environmental quality [42,43]. The earliest research on urban birds was to determine the bird species list parks through the field investigation in cities or parks. The detailed list can be distinguished according to seasons and climates to show the diversity of species [44,45]. After the 1980s, bird communities gradually became an independent field of study [46]. On the macroscopic level, many studies usually pay attention to the changes of ecological patterns and species diversity [47]. On the landscape level, many studies discuss how the urban space elements such as parks, green spaces, and water could become the habitat of birds better [48]. In addition, many studies have investigated the distribution of bird communities in different green space structures, such as using the birds’ list from field investigations to explore the influence of tree age, plant height and density of trees and shrubs on bird diversity [26].

In the field of urban planning, improving biodiversity is already one of the goals of urban spatial design. In addition to the above studies focusing on birds and plants, some planners or landscape designers also try to use urban space elements to study the relationship between cities and birds. Previous studies have indicated that urban spatial structure elements such as building height can affect the flight path of birds [49], and the distribution and form of urban buildings can create resistance to the bird flight [50]. However, few existing studies have discussed the spatial design of urban regeneration areas and how to make birds continue to live in the area at the scale of block planning. When planners design some blocks, they usually carry out their plan according to specifications. To improve the diversity index, they will increase the types of plants. Few of them focus on improving the diversity of birds and propose reasonable bird-friendly plans through the procedures of current situation investigation and demonstration. In urban regeneration areas, the architectural form can be changed through spatial renewal so as to reduce the flight resistance of birds and ensure their smooth movement path. The habitat environment can be improved through landscape adjustment to ensure the continued survival of birds in the previous zones. In this study, we will discuss from an urban design perspective how to implement a program to protect bird habitats in a regeneration area and attempt to identify gaps in this area. By employing birds as biological indicator species, and with the assistance of GIS technology, this paper fully considers both bird flight demand and urban regeneration construction needs, analyzes bird flight resistance, and proposes corresponding spatial management strategies to achieve bird conservation and biodiversity maintenance in urban regeneration areas.

This study takes advantage of a rare opportunity to investigate the existing bird habitat in a certain area of Shenzhen prior to urban regeneration, and to propose a space construction solution for urban regeneration with the aim of bird conservation. The main contents of this study are as follows: (1) select suitable methods and models; (2) analyze the flight resistance of representative birds in the research region; (3) propose spatial management strategies for urban regeneration that focus on reducing bird flight resistance; and (4) verify the effectiveness of the methods. The results of this study can provide a biological perspective for decision makers, planners, and developers to carry out the urban regeneration spatial design. Moreover, the theory of landscape ecology and bird resistance analysis are innovatively applied to small and medium-scale urban regeneration areas, and the relationship between biological processes and urban spatial elements are quantitatively analyzed so as to make a valuable contribution to biodiversity conservation in urban regeneration areas.

2. Methods

2.1. Overview

Shenzhen is in the Pearl River Delta region along the southeast coast of Guangdong Province (113°46′ E–114°37′ E, 22°27′ N–22°52′ N), with a total area of 1997.47 km², a built-up area of 927.96 km², and a permanent population of about 17 million. It is a national economic center in China. After massive and rapid expansion, Shenzhen is experiencing an increasing shortage of land resources, and urban land is gradually showing a stock optimization trend. Thus, it is urgent to achieve the transformation of the old city and the improvement of the ecological environment through urban regeneration [51]. At the same time, Shenzhen is an important stopover for migratory birds from East Asia to Australia [52], with more than 100,000 migratory birds inhabiting the city every year [53]. Listed in the Ramsar List of Wetlands of International Importance, Shenzhen Futian Mangrove Nature Reserve is the only national reserve located in an urban center area in China [54], which has a strong interest in bird conservation [55,56].

The research region is in the central urban area of Shenzhen, with Tanglang Mountain in the north and the Futian Mangrove Nature Reserve 1 km away to the south (Figure 2). The research region covers a total area of 4.9 km² and is crossed by the Hongli arterial road and Shennan Avenue in an east-west direction, which divides the research region into three parts: the north part is mainly a reservoir covering an area of 0.26 km², a golf driving range, native forest and sports grounds; the central part is an old city building area with almost no green space; the southern part is an old city building area and a golf course covering an area of 1.36 km². In urban regeneration planning, it is proposed to demolish the existing above-ground area of about 700,000 m² of old city buildings in the research region and upgrade a reservoir and golf course into urban ecological land, with the total above-ground construction area of the research region reaching 3.2 million m². 132 plant and 95 animal species, including 45 bird species, were found on site in the research region, which has some biodiversity value. The goals of urban renewal, to create greater building volume on the one hand and to preserve biodiversity on the other, are in great conflict; and there is an urgent need to balance the competition between biodiversity and building land in urban regeneration.

The topographic data utilized in this study is ASTER GDEM’s 30 M resolution digital elevation data, which can be freely downloaded on the geospatial data cloud http://www.gscloud.cn/ (accessed on 18 March 2022). The high-resolution geographic remote sensing maps are Landsat 8 OLI_TIRS satellite digital deliverables. The current land use data is produced as vector data in ArcGIS10.2 from the image file of the 2016 Shenzhen City Land Resources Survey. By using ArcGIS10.2 software, all the data in this research is eventually unified under the WGS_1984 coordinate system.

2.2. Methodology

The main analytical methods for studying the resistance of bird activities are minimum cumulative resistance, circuit theory, the axial map method, obstacle detection analysis, maximum penetration path, diffusion simulation model, etc. [57]. Among these, two are the most widely employed: minimum cumulative resistance (MCR) and the circuit theory model. In 1992, Knaapen et al. [58] proposed the minimum cumulative resistance model, also known as least cost distance (LCD), least cost path, etc. Its principle is primarily based on the assumption that the species researched will carry out spatial diffusion with the motivation of minimizing the moving resistance or cost between patches on the premise of having complete prior knowledge of the landscape [59]. In the study of bird activities, resistance and cost surface are formed mainly through an expert scoring method or analytic hierarchy process (AHP), and then the path of minimum resistance can be determined through GIS platform simulation. Its advantage is that it can determine the flight situation of birds according to three-dimensional data such as terrain elevation and building height [60].

The circuit theory for path analysis is currently mainly applied to mammals. In the analysis of birds, it is mainly used to build the ecological corridor of bird activities [32]. Its main method is to build a model in circuitscape software and assign different resistance values to different land properties such as woodland, grassland, water and buildings to simulate the connectivity of birds. The most obvious advantage of circuit theory over the minimum cumulative resistance model is that it does not provide a unique corridor solution for the species studied, but the assumptions of this algorithm also limit the situations in which it can be applied. For example, circuit theory is less applicable if the indicator species are migratory birds or flocks rather than native urban birds.

In urban regeneration areas, spatial structural elements such as buildings are hindering factors in the flight process of birds, and three-dimensional analysis is indispensable. In addition, due to the large number of migratory birds in the research region, this study adopted a minimum cumulative resistance model to present the phenomenon more reasonably. After the model analysis of the current situation, the spatial management requirements of urban regeneration are proposed under the premise of protecting the flight and habitat needs of birds. To verify the effectiveness of such management, the current situation and the proposed scheme are then compared to ensure that the bird habitat will not be impaired by development and construction. The research process of this study is shown in Figure 3.

2.3. Bird Flight Resistance Analysis Based on the Minimum Resistance Model

2.3.1. Bird Species Selection

The biological data related to birds and plants in this study were predominantly obtained from on-site surveys, and the research team consisted of all three authors. In the on-site survey, the research team walked along the roads in the research region at a uniform speed of 1.5–2 km/h, equipped with a pair of SWAROVSKI binoculars, a monocular, and a Canon digital SLR camera with a 300 mm telephoto lens. During the survey, the research team observed birds with binoculars, photographed birds and recorded their calls, thus bird species were identified on site by recording their physical characteristics, calls, and flight posture. Record sheets were also filled out to record data on bird species, numbers, and activities. The survey frequency was February, May, July, and November 2018, for a total of four times; all four days were clear and windless in the morning, and the final summary was obtained for the research region bird list. The survey revealed that there were six orders, 21 families and 45 species of birds in the research region. A vast majority of the species here fed on fruits, insects and nectar, including Streptopelia chinensis, Hirundo rustica, and Parus major, which are common or dominant species. Passeriformes had 33 species, accounting for 73.33% of the total surveyed birds. Since a reservoir and the puddles in the southern golf course provide good foraging and resting places, the research region was also rich in waterbirds, mainly Charadriiformes and Ciconiiformes. Among all birds, there are four species under key protection. The national second-class protected animals were Garrulax canorus and Centropus sinensis, and the provincial protected animals were Egretta garzetta and Ardeola bacchus.

In view of the similarity of ecological characteristics and habitat requirements between some species and other taxa, some species or groups are often selected as a “surrogate species” to study the conservation and habitat management of species [61,62,63]. The “indicator species” is a sort of surrogate species, which is an effective tool for biodiversity conservation planning [64]. Therefore, in this study, the four most representative species in the research region were selected as indicator species based on the surveyed species and the requirements of the conservation level, activity frequency, habitat type and biological characteristics [65], as shown in

Table 1. Bird indicator species list.

Species	Resident Type	Region	Conservation	Population Size
Egretta garzetta	Resident bird	Oriental	Provincial protected animal	Relatively small
Garrulax canorus	Resident bird	Oriental	Species listed in the Convention on International Trade in Endangered Species of Wild Fauna and Flora	Relatively large
Hirundo rustica	Summer resident	Wide-distributed	/	Relatively large
Anthus hodgsoni	Winter resident	Palaearctic	/	Relatively small

. These four indicator species cover the three types and three floras occurring in the site, and are of the highest frequency, but are also of conservation significance.

2.3.2. Identification of the “Source”

A “source” is the origin of species dispersal and maintenance, with internal homogeneity and the ability to expand in all directions or to converge on the “source” itself [66,67]. A “source” site, the source of ecological processes, can be not only the natural habitat of an indicator species, but also a landscape unit with a high frequency of activity [68]. Because different bird species congregate at a single foraging site and disperse to multiple foraging sites, the representative sites where indicator species frequently congregate to forage were selected as “sources” for this study.

2.3.3. Determination of Flight Resistance Factors and Coefficients of Birds

Bird flight resistance is affected by both baseline resistance, i.e., the barrier effect of ground cover, and additional resistance, i.e., disturbance from human activities [50]. The design of the resistance value of the basic resistance referred to the experimental results of China [69,70]; the elements of additional resistance usually include noise, vibration, and visual disturbance brought by building height, human traffic flow, vehicle flow, and human behavior. Combining the situation of the research region and the characteristics of the urban regeneration area, we referred to the results of birds research in Shenzhen [71,72] for coefficient design, and then localized and adjusted the resistance factors. Finally, 10 experts in urban planning, landscape design, ecology and birds scored to determine the flight resistance factor and coefficient. In this study, land cover type, building density, building height and road elements, which can be controlled by urban regeneration design, were used as the four resistance factors to analyze the flight resistance of birds in terms of basic resistance and additional resistance (Figure 4).

Then, according to the living habits of the four indicator species, the resistance coefficients are adjusted. Higher values represent greater resistance to bird activities. The final construction of the flight resistance factors and resistance coefficient system for birds in the research region is shown in

Table 2. Flight resistance factors and resistance coefficient for birds.

Resistance Factor			Resistance Coefficient (1–500)
Basic resistance (land cover type)	Woodland (shrubs and trees woodland, shrubbery)		1
	Lawn		50
	Garden		100
	Bare land		300
	Water		400
	Construction land		500
Additional resistance (human disturbance)	Building height	0–50	200
		51–100	100
		>100	50
	Building density	Low-density	50
		Medium density	100
		High-density	150
	Road	Urban roads >30 m wide	90
		Urban roads <30 m wide	60

.

2.3.4. Determination of the Resistance Surface

ArcGIS software was used to reclassify the distribution of various resistance elements into raster data and perform spatial overlay analysis. Combined with the spatial data of “source”, the “cost distance weighting” module is used for calculation, and the distribution map of minimum cumulative resistance surface is obtained. The principle is as follows:

$$MCR=\underset{min}{{\displaystyle \int }}{\displaystyle \sum }_{j=n}^{i=m}\min \left(D_{ij}\times R_i\right)$$

(1)

where ∫ is a monotone increasing function that reflects the positive correlation between the minimum resistance of the species from the source to any point in the space and the distance it travels and the characteristics of the landscape base. Min denotes the minimum cumulative resistance of the evaluated patch for different sources, and ∑ denotes the distance and resistance accumulation across all units between a point i and source j. $D_{ij}$ is the spatial distance of species from source j to point i; $R_i$ is the resistance coefficient of landscape i the movement of the indicator species.

3. Results

3.1. The “Source” of Birds

In the various ecological patches in the research region, such as northern forest, a reservoir, a reservoir hydro-fluctuation belt, and street parks, the details of the habitats of the four indicator species were determined by combining the size, shape, tree cover and height and structural diversity of these patches, as well as the frequency and pattern of bird activity observed in the field. The main activity area of Egretta garzetta was located in the reservoir hydro-fluctuation belt and the bank of a pond in a golf course; the main activity area of Garrulax canorus was the shrubs and trees woodland with vegetation coverage greater than 30% and the water’s edge; the main activity area of Hirundo rustica was the reservoir hydro-fluctuation belt and the grass around the water surface of a pond; the main activity area of Anthus hodgsoni was in the southern golf course and the woodland next to the reservoir. Based on the habitat requirements of the four indicator species, the reservoir hydro-fluctuation belt, the forest land with vegetation coverage of over 30%, the forest land in the golf course and some shrublands were taken as the “sources” of the indicator species, as shown in Figure 5.

3.2. Flight Resistance of Birds

3.2.1. Single Factor Resistance

In the basic resistance analysis, the land cover type was the main basis of evaluation. The construction land with the greatest resistance accounted for nearly 50% of the total research region and had a large aggregation, which would cause greater disturbance to bird flight. Woodland only occupied about 14% of the total research region, and the more concentrated woodland was distributed around the reservoir in the north and in the golf course in the south, which was also the main area of current bird activities. The rest were distributed along the road and the edge of vacant land with complex patch shape edge conditions and high degrees of fragmentation. Grassland, with a high density of patches, accounted for 30% of the total research region, and was the type of land with less resistance and a larger area. The current grassland situation in the north was mostly natural grassland, while the golf course in the south is artificial lawn. However, due to the interweaving of woodland, shrubland and pond in the southern golf course, the micro-ecological environment was good, thus it was still the most important flight and habitat for birds in the study area.

The current building height in the research region was within the range of 0–105 m. The tallest buildings were the northernmost and southernmost residential buildings, and the buildings in the middle of the research region and around the reservoir were mostly low commercial service buildings. Since small birds often fly close to the ground, a single tall building was easy to cause bird strikes, but the number of low buildings were more than a single tall building, so the number of bird strikes was more significant [73,74]; the resistance in the central area increases steeply with strong disturbance as the indicator species traverses the habitat in the north and south.

Since not all areas in the study area were built on, the tracts that contain buildings and had the same land use type were merged or divided, resulting in a research region containing seven areas. We then calculated the ratio of the total base area of the buildings to the occupied land area for the seven areas and concluded that the building density of the seven areas distributed was between 10 and 42%. The resistance grading revealed that the highest building density is in the central buildings of the research region, followed by the residential area in the north. In particular, the residential area on the east side of the north had higher buildings, but its higher density would also have some influence on the activities of birds around the reservoir.

The width of the roads around the research region and the city arterial roads running east-west was greater than 30 m, generating a region with high flight resistance for birds. Its disturbance to birds was mainly reflected in the noise of various cars. There were few internal branch roads in the research region, and the walking space was mostly the space of the square outside the building. Therefore, the internal roads with resistance were urban branch roads, mainly distributed on the west side of the south and the east side of the central. The analysis of each factor is shown in Figure 6.

3.2.2. Comprehensive Resistance

After ArcGIS calculation, the distribution of the minimum cumulative resistance surface in the research region was displayed in Figure 7. The areas with the least ecological resistance in the research region were the northern woodland space around the reservoir and the southern golf course. Due to the complex construction around the northern woodland, bird flight was constrained mainly in the large woodland area on the eastern side of the north, which was also consistent with the information observed during the status study. The vegetation in this area was mostly native tree and shrub vegetation communities with a high degree of crown density, which provided a good habitat for forest birds represented by Garrulax canorus and plays a vital role in the formation of bird habitats in the research region. The northern hydro-fluctuation belt around the reservoir was mainly composed of sparser trees and understory grassland, with similar crown density as the shrubs and trees woodland of the southern golf course, providing flight space for migratory birds such as Anthus hodgsoni and Hirundo rustica. The mudflats of the reservoir, on the other hand, provide space and ample food for waterbirds represented by Egretta garzetta.

The areas with the greatest resistance in the research region were the southern and western parts of the central areas, where buildings are densely established. Although natural elements were diverse, the serious disturbance of artificial elements and high spatial heterogeneity lead to the obstructed flight of birds in the current situation. In addition, Hongli Road, an east-west urban trunk road, crosses the research region, which also increased the flying obstacles of birds.

There was large resistance in the central part of the research region, obstructing the source connectivity between the south and the north, and resulting in poor connectivity of bird activities. However, the central area contained a series of small bare land and square greenery, mainly woodland and shrubland with less ecological resistance, providing fragmented small habitat patches for birds. In the three-dimensional simulation diagram of the resistance surface of the flight resistance pattern (Figure 8), multiple protruding “peaks” and subsiding “valleys” would be formed on the resistance surface, while a concave or gentle patch would appear on the high resistance ridge line of the two “peaks” and the low resistance valley line of the two “valleys”, which was similar to the shape of the “saddle”, indicating that the two patches were close, the landscape resistance between them was small, and the correlation between patches was high, hence there was a great possibility for organisms to migrate between patches. The middle part of the research region had an obvious “saddle” structure, generating a low resistance flight passage for birds.

4. Discussion

4.1. Spatial Planning Strategy for Urban Regeneration

4.1.1. Birds’ Conservation Pattern Based on Flight Resistance

By analyzing the cumulative resistance of the indicator species, it is possible to determine the pattern of bird habitat conservation in the research region based on landscape ecology theory. First, according to the protection degree of core habitat and spatial movement of indicator species, ecological protection zones in the research region can be divided into three grades: low, medium and high. The low-grade zone is suitable for development and construction, the medium grade zone is secondarily suitable, and the high-grade zone is a habitat conservation zone for birds. See Figure 9 for the zoning. The high-security grade zone is the core area for bird activities and an important ecological matrix. The northern part comprises the current woodland, greenery along the urban arterial road and greenery around a reservoir, the central part includes vacant land and its internal greenery, and the southern part includes woodland, shrubland, grassland and water within the golf course, which should be strictly protected. The medium-security grade zone is the buffer zone of the high grade one, including the greenery around the core area, a reservoir, and part of the low-density construction area, which should be limited to its development scale and carry out appropriate habitat restoration to solve the problem of bird flight resistance brought by the current disorderly construction. The low-security grade zone is mainly the current high-density construction area and part of the broader water surface of the reservoir. In the urban regeneration process, new buildings can be mainly laid out in this zone to protect the integrity of bird habitats, while the reservoir can be appropriately adjusted to complement the ecological matrix divided in the north.

Second, the ecological corridors for birds in the research region are determined based on the low resistance flight paths of the indicator species in the three-dimensional analysis. The research of Forman and Godron revealed that the ecological corridor with a width of 12–30.5 m is a good standard for maintaining bird diversity [75], while some researches showed that the appropriate corridor width for protecting bird populations is 200 m [76,77]. Since vegetation corridors can facilitate the migration of plants and animals between habitat fragments [26], a series of 12–200 m wide green spaces are formed in the central part of the research region based on the current low resistance paths to open up the flight paths of bird populations. The reservoir in the north no longer has a water supply function, so the existing hard embankment can be transformed, and waterfront wetlands can be added to reduce flight resistance and increase the suitable habitat for waterbirds to forage and inhabit.

The bird’s conservation pattern provides the research region with distinctive and guiding design principles, and lays the foundation for the urban regeneration spatial design of the research region. To ensure that targeted and implementable birds conservation content can be enforced in the urban regeneration process, specific control requirements are proposed for buildings, green space systems, and other urban regeneration spatial elements.

4.1.2. Planning Strategy of Green Space System

The investigation of the existing situation revealed that the species in the research region consists mainly of forest birds, supplemented by waterbirds. Correlation analysis of bird flight resistance, bird species richness and the green space system can provide a better understanding of the flight preferences and habits of birds in the green space system and convert these data to a percentage (Figure 10). In terms of the land type of the green space system, it is divided into forest, water and lawn, and correlation analysis is conducted with bird flight resistance and bird species richness. It can be seen that the current forest birds have certain flight resistance, but the species richness is also greater; the lawn has less flight resistance and moderate species richness; and the large area of reservoir water surface leads to the greatest flight resistance and lower species richness. Regarding the correlation of vegetation structure, the vegetation structure is divided into five major categories: treeless (T), lawn (L), shrubland (S), sparse woodland with less than 30% crown density (SW), and dense woodland with more than 30% crown density (DW). Correlation analysis with bird flight resistance and bird’s species richness suggests that birds have some flight resistance in dense woodland, but species richness reaches the highest; flight resistance of birds in sparse woodland and shrubland is lower but species richness is higher; the flight resistance of birds in lawn and treeless public space is low and the species richness is also lower.

The analysis above indicates that sparse woodland and shrubland provide more habitat environment for birds since they have more diverse elements of food source plants, roads, and other forest’s edges [78]. Thus, when carrying out urban regeneration planning, firstly, we should protect the matrix of the current bird habitat and maintain the existing vegetation condition. Secondly, we should increase the area of shrubland in many green spaces of single structure, such as the artificial lawn in the southern golf course and the green space of central square, to reduce the flight resistance of birds and increase the richer habitat for birds.

In addition, green space systems in built-up areas are “patches” of habitat for birds, and the current pattern is mostly in the form of lawns planted with scattered trees, with few places for birds to rest. Fernández- Juricic et al. found that the minimum park area required by birds ranged from 10 to 35 hm², and Jokimki et al. found that about 23% of predominantly ground-nesting birds and typical forest birds would avoid parks with an area of less than 0.75 hm² [79,80,81]. Therefore, in the urban regeneration process, green space patches ranging from 0.75–35 hm² are arranged within the building complex to satisfy the stay demand of birds, and the density and abundance of birds are enhanced through landscape measures such as enriching vertical levels and planting food source plants.

4.1.3. Building Planning Strategy

The correlation analysis of building density, building height and current birds in the research region (Figure 11) demonstrates that the current birds in the research region are mostly small birds, which fly at low altitudes and even often skim the ground, showing a negative correlation with building density and a positive correlation with building height. Therefore, the building height and the density should not be too high when it is close to the “source” and the corridor, so as to avoid the obstruction of birds when they land on the trees and water. In the context of urban regeneration, buildings near the northern reservoir and the golf course to the south of the study area should be characterized by low density, and the main high-density buildings should be arranged in the east and west sides of the central area and the west side of the south area far from the source. Therefore, building heights should increase in a gradient from the habitat and ensure that there is sufficient space between new buildings so as to reduce the risk of waterbirds’ foraging flight. When arranging buildings in the middle of the research region, those close to the ecological corridor should be lower and those away from the corridor should be higher.

In summary, birds in the study region were found primarily in the dense forests and grasslands, and bird flight resistance was negatively correlated with the complexity of green space vegetation structure and building height, but positively correlated with building density. The spatial design of urban regeneration in the research region includes the following points: (1) North part: the reservoir embankment is ecologically transformed and wetlands are added, the woodland around the reservoir is maintained as it is, and scrub or trees are added to the grassland and park square. The building design is mainly low-density buildings that do not destroy the original habitat. (2) Central part: a green ecological corridor with a width of 12–200 m is constructed by taking the low-resistance path as the carrier, and green space of park covering an area of 0.75 to 35 hm² is arranged as the patch for bird activity in the high-density built area as far as possible, with rich plant community levels as the best. The building height and the distance to the ecological corridor are in a gradient increasing relationship. (3) South part: the trees and water ponds in the golf course are reserved, the artificial grassland is renovated and scrub or trees are added to form a near-natural landscape green space. The distance between the construction area on the west side and the adjacent course is in a gradient increasing relationship. Finally, a preliminary urban renewal conceptual design can be created for the study area, as shown in Figure 12.

4.2. Verification

Although the spatial management strategy for urban regeneration has been determined through current bird data and simulation analysis, biodiversity analysis is still needed to verify the effectiveness of the management plan and the design method based on bird flight resistance. In this study, the total habitat richness (PR) and diversity index (SHDI) are selected as representative factors to conduct a comparative analysis on the diversity of the park based on the existing studies on birds. During the verification process, firstly, the design plan will be spatially illustrated and data quantified, and then the obtained land use conditions and current land use conditions will be calculated by Fragstats 4.2 software. The conclusions are shown in

Table 3. Bird’s diversity index of the current situation and urban regeneration design plan.

	PR	SHDI
Current situation	44	2.56
Urban regeneration	57	3.10

.

From the verification results, after urban regeneration, although the building density and height indicators increase significantly compared with the status quo, the rationality of the plan can lead to an increase in bird diversity and bird species abundance, which fully demonstrates the effectiveness of the method of applying the minimum resistance model of bird analysis to the space construction of small and medium-scale urban regeneration areas.

4.3. Replication

Our whole thesis is to discuss how to protect birds in urban space design. In the section of introduction, we discussed the development and gaps of “urban regeneration”, “biodiversity” and “birds”. In terms of methodology, we want to find a technical method to learn about the habits of birds and urban spatial elements controlled by the designer. Most of the existing research is carried out by bird experts and planners according to their own research, and few of them will discuss how to integrate the results of bird experts, field research, and space design to achieve better bird conservation. We want to talk about the rationality and necessity of this process, in international community with different birds, can come to different conclusions, but what we want to create is the process approach in space design of urban regeneration. Interest, policy and land development are more concerned in the process of urban regeneration [82]. In the process of space design, biodiversity conservation is usually an indicator [83], which is abstract and difficult to translate into a quantitative solution. Indeed, what we want to emphasize is a process, which is used in spatial design to effectively apply the conclusion of biodiversity conservation. Although bird species differ from region to region, it is still possible to try to use this process to make their practical projects more sustainable. Many studies are on biodiversity conservation or spatial design, but few effectively combine the two to provide ideas for planners and designers.

5. Conclusions

This study takes the contradiction between urban regeneration and biodiversity conservation as an entry point, and adopts birds as the representative to discusses specific measures and methods to achieve biodiversity conservation through reasonable space construction solutions at the beginning of the design of urban regeneration areas. Taking a certain area in Shenzhen as an example, this study conducts a multifaceted study on the current bird situation and proposes a series of design methods to provide a reference for decision makers, planners, developers, and other related workers. The main conclusions of this study are as follows:

(1)

Through the study, the bird flight resistance analysis based on the minimum resistance model is selected as the main method for biodiversity conservation in the research region. By screening indicator species, identifying “sources” and determining bird flight resistance factors and coefficients, we finally obtain the flight resistance surface of birds in the research region and potential corridors from a three-dimensional perspective so as to establish the conservation pattern of birds in the research region.

(2)

The correlation between bird flight resistance, bird species richness and spatial elements of urban regeneration is found out. From the perspective of community structure of green space vegetation, sparse woodland and shrubland in the research region has low resistance and high richness, which are the most ideal green space forms for birds; from the perspective of architecture, bird flight resistance shows a negative correlation with building height and a positive correlation with building density.

(3)

The quantitative design strategies for the spatial elements of urban regeneration from the perspective of birds in the research region is determined. The main strategies are: the north of the research region is dominated by low-density buildings, the reservoir is ecologically transformed and wetlands are added; in the central part, a green ecological corridor with a width of 12–200 m is constructed by taking the low-resistance path as the carrier, and the building height and the distance to the ecological corridor are in a gradient increasing relationship; in the south, the current trees and water ponds are retained, and artificial lawn is transformed to increase shrubland or sparse woodland.

(4)

Based on the biodiversity index, the spatial element planning strategy is determined to lead to an increase in bird richness and species diversity in the research region, and the effectiveness of the method was verified.

For researchers, this study fills in the gap of research with a biological perspective in small and medium-scale urban regeneration areas and provides ideas for future related research. It also offers a complete planning process for decision makers, planners and developers to effectively achieve biodiversity conservation in the research region during future urban regeneration.

However, this study has two main limitations. First, only relevant examples are used as research objects, and the bird species are strictly based on the current investigation results. The correlation between birds and green space systems is locally unique and not generalizable. Second, the main object of this study is the parcels of land for which the total planning target has been set, and the main content of the discussion is how to manage and change the urban spatial elements to achieve biodiversity conservation without evaluating and increasing or decreasing the amount of development and construction. The above issues can be further explored and supplemented in future studies.