Journal of Geography, 1998, Volume 53, Supplement: 12- 18.
(Excerpted from doctoral thesis: Safety model in landscape planning with South China as an example. Doctoral theory, master's degree, Graduate School of Design, Harvard University. USA. 1995, changed a lot).
Some points in the landscape play a key role in controlling the horizontal ecological process, and occupying these strategic points in the landscape will bring the advantages of initiative, spatial connection and high efficiency to the ecological process. It is assumed that the horizontal movement of species in the landscape is a competitive process to overcome some resistance. This paper discusses how to distinguish the strategic points of landscape ecology according to the surface characteristics of landscape resistance. This paper discusses three types of landscape resistance surfaces: island type, network type and plateau type. Corresponding to various types of resistance surfaces, the spatial positions of five landscape strategic points are obtained: saddle strategic point, intersection strategic point, central strategic point, edge strategic point and corner strategic point. The article also illustrates the method of identifying landscape strategic points through a case. GIS technology provides strong support for this research.
Keywords: spatial analysis, landscape ecology, landscape pattern, landscape planning
1. Introduction
1。 1 Question: Strategic Point
How to identify some key points in the landscape and control or promote an ecological process by controlling these points, such as the spatial movement of species, the spread of fires and pests, the flow of nutrients and pollutants, etc. These key points are called landscape strategic points (Forman 1995, Yu 1995A-C 1996). An ecological process can be effectively maintained and controlled through the identification, change and management of landscape strategic points.
According to the different origins, it can be divided into two types of landscape ecological strategic points: one is resource-based strategic points, and the other is structural strategic points. Resource-based strategic points directly depend on the resource attributes of the lot, such as geology, soil, hydrology, heat, nutritional conditions and human activities, which determines the biodiversity and rarity of species, thus determining the strategic significance of the lot in biological protection. This paper will discuss the key points of the latter landscape ecological strategy. Its strategy depends on its position in the overall landscape pattern and its influence on the lateral ecological process.
Many scholars have noticed that setting up biological protection patches in some key areas is of great significance to the overall biological protection and the formation of landscape infrastructure (frankel and soul1981; Harris1984; Furman1990; Erwin 199 1)。 In the heterogeneous landscape, some landscape strategic points are obvious, such as the entrance and exit of a basin and the fracture of the corridor (Merriam1984; Forman and godran 1986), the debris with stepping stone function, the intersection of river networks and the junction of valleys and ridges (Harris, 1984, p141-kloc-0/65). The forest patch located in the center of the landscape is more likely to become the habitat of birds than the forest patch located in other areas (Liu et al. 1994). The "land bridge" connecting the islands will have more biological species than other sections. These are all landscape ecological strategic points that can be identified according to experience.
However, the strategic points of landscape ecology are sometimes not directly observed, and they are often potential. The identification of this potential strategic point depends on the understanding and simulation of ecological process dynamics. Up to now, few people have discussed the methodology of distinguishing landscape ecological strategic points, and only recently have they been noticed by a few scholars (Forman 1995, PP 310-317; In 1995a-c, 1996a). Related studies include Knaapen's analysis of the isolation degree of landscape patches by using the minimum cumulative resistance (MCR). According to the surface of MCR, the introduction position of biological protection plaque is determined. Strategic point is an important part of SPS (YU 1995A-C, 1996A-B). Game theory and game defense strategy (von Neumann and Morgan STEM1947; Luce and Raiffa 1957) provide evaluation principles for judging landscape security pattern. Some scholars think that Go and similar board games will help to discuss the key points of landscape strategy (for the discussion of Furman and Yu, see Furman 1995 p.3 16). Weiqi is a game in which black and white players control the chessboard by competing for strategic points in space (Boorman 1969). In order to control the chessboard most effectively, each chess piece must be placed in a key position, and the strategy of these positions follows three principles (Figure 1):
(1) First Mover principle, that is, by controlling a certain point, the initiative of local control is obtained.
(2) the principle of spatial connection, that is, by controlling a certain point, the overall spatial advantage can be obtained.
(3) The principle of high efficiency, that is, by controlling a certain point, the maximum chessboard control can be obtained with the least number of chess pieces.
Figure 1 China Go Basic Space Strategy
Figure 1 strategy of go game
Under certain assumptions, these principles can also be applied to the understanding and discrimination of landscape ecological strategic points. For example, the spatial movement of creatures can be discussed as a competitive control process of landscape, which is essentially a "game" process (Sigmund 1993). Whether organisms can successfully control the landscape depends on how to balance the two aspects of spatial diffusion and maintenance. Maintenance means that individuals must rely on existing populations and communities to survive, and diffusion means that individuals leave the habitats of populations and communities and expand to the outside. The farther an individual leaves his existing habitat, the greater his risk, but the greater his potential contribution to the whole landscape control. At some points in space, when the potential reward brought by species expansion reaches the maximum and the risk borne by individuals reaches the minimum, these spatial positions are strategic points. This can be understood as the Minimax principle (minimax theorem, von Neumann and Morgan1947; Luce and Raiffa 1957). Functionally, these strategic points play a key role in promoting and hindering the ecological process. How to find these landscape strategic points? The surface analysis model of Go strategy and theoretical geography can give us some enlightenment.
1。 2 lateral control process and theoretical geographical model
This paper discusses the identification of landscape ecological strategic points, based on the following assumptions about this process:
(1) The horizontal ecological process of landscape is a competitive control process of landscape.
(2) A certain process must control the landscape by overcoming the landscape resistance.
In the actual ecological process, these two assumptions can basically be satisfied. For example, plants often need to overcome landscape resistance to cover a certain area, and the horizontal movement of disastrous insects, animals crossing the landscape, and spatial migration of populations all have the characteristics of competing to control space and overcome spatial resistance (Johnson1988; Freich et al.1993; Mack 1995； Liebhold et al.1992; Yudeng1996; Williamson1993; Boone and Hunter,1996; Xin Beloff and Wilson1969; Thobroe1981; Pteridium aquilinum 199 1). Many models are used to describe these horizontal motion processes, including gravity model, potential model, diffusion model, stochastic model and so on. (For a detailed overview, see Olson1965; Bartlett1975; Sklar and costanza1990; Chou and Liebhold 1995).
The possibility and dynamics of ecological process covering and controlling landscape can be expressed by resistance or its related concepts, such as accessibility (Arentze et al. 1994), passability (Boone and Hunter 1996), cost distance (for example, ESRI 199 1) and minimum cumulative resistance (minimum cumulative resistance). Yu 1995a-b), landscape resistance (Furman and Godron1986; Forman 1995), and the degree of isolation (isolation, MacArthur and Wilson1967; Beloff and Wilson 1969). All these resistance measures are actually variations or extensions of the concept of distance. In chess board, resistance is idealized as distance, that is, chessboard grid. In island ecology, similar idealized distances are used to describe spatial isolation (MacArthur and Wilson 1967). In the land landscape, the resistance is not only the geometric distance, but also the basement characteristics (Forman, 1995). These resistance measures can be visually expressed by potential surface or trend surface (Warntz1966; ; Jolly and hagt 1968).
Theoretical geographer and regional scientist W. Warntz (1966, 1967) and others have done a lot of research on the general characteristics and spatial analysis of the surface in Harvard University, which is helpful to explore the identification method of landscape ecological strategic points. Werntz Stadium is characterized by peaks, pits, passes and valleys (1966, 1967), as well as ridges and valleys. Quot "Peak" refers to the local maximum of the surface, where the flow disperses, and "concave" refers to the local minimum of the surface, where the flow converges. "Ridge line" is the shunt line connecting two "peaks" and "valley line" is the confluence line connecting two "concave". "Guan" refers to the minimum value on the ridge line and "Saddle" refers to the maximum value on the valley line. It is of guiding significance to determine the strategic point according to the resistance surface of landscape ecological process.
This paper will discuss how to judge the strategic points of landscape ecology according to the surface characteristics of landscape resistance. These strategic points are evaluated by three principles inspired by Go strategy. This paper will focus on the strategic points to control the horizontal movement of species, but this methodology is applicable to the discrimination of other horizontal movements and ecological flow strategic points. The case study is only the explanation of methodology, and its practical application needs further field verification.
2? way
The process of determining landscape ecological strategic points is divided into four steps, of which the contents of the first to third steps have been discussed in previous studies, and this paper will focus on the fourth step:
The first step is to determine the landscape ecological process. The ecological process discussed in this paper is the spatial movement of species, and many scholars have discussed how to choose key species as protection objects (such as Frankel and Soul1981; Armstrong et al.1988; Selm 1988). Other horizontal ecological processes can include the flow of wind, water and nutrients, and the diffusion of disturbances (such as fires and pests).
The second step is to determine the source of the ecological process: one method is to take the habitat as the source according to the specific species (see Selm 1988). Another method is to choose the existing landscape elements, from a big tree to a national park, as the source of ecological process (see Harris1987; North and Harris1986; Noss 199 1)
Thirdly, taking the source of ecological process as the origin, the landscape resistance is calculated and the landscape resistance surface is obtained. The typical resistance surface will be similar to the topographic surface, consisting of peaks, valleys, saddles and ridges, reflecting the dynamics of ecological processes. Different ecological processes may have completely different resistance surfaces. The resistance surface is affected by the spatial distance and surface characteristics, and also depends on the diffusion ability of the process itself.
The fourth step is to determine the strategic points that affect the ecological process according to the resistance surface. According to the morphological characteristics of the resistance surface, it can be divided into three types: island type, network type and plateau type (Figure 2-4). Accordingly, there are many key points of ecological strategy:
Figure 2. Island resistance surface and saddle strategic point
Figure 2. Strategic points of archipelagic resistance surface and saddle point
Figure 3. Network resistance surface and strategic points at intersections.
Figure 3. Network resistance surface and strategic points at intersections
Figure 4. Plateau resistance surface and strategic points of center, edge and corner
Figure 4. Strategic points of plateau resistance surface and center, edge and corner
(1) strategic points of resistance surface and saddle landform of islands
In the island-type resistance surface, the ecological sources with low resistance are scattered in the landscape matrix with high resistance, which is very common in the landscape disturbed by modern people, such as the patches of residual forests scattered in the farmland landscape. Resistance isolines extend concentrically outward around each source. Saddles are formed in the tangent part of the equivalent resistance line centered on each source, and they are the minimum or equilibrium point on the resistance surface. When the diffusion capacity of ecological process exceeds a certain degree, these minimum points and maximum points become the key points connecting different sources, that is, strategic points. The introduction of biological patches at these points will effectively promote and maintain the health and safety of ecological processes and optimize the overall landscape structure. The strategic significance of these points is that they promote the spatial connection between isolated sources and realize the high efficiency of landscape control (Figure 2).
(2) The network resistance surface and the landscape strategic points at the intersection.
In the resistance surface of the network, the low resistance parts are linear and crisscross to form a network. The landscape reflected by this resistance surface includes dendritic water system and farmland shelterbelt system. In addition, on the resistance surface of other types of landscapes, there are also potential corridor networks composed of low resistance areas. Just like the valley and river system on the topographic surface, ecological processes converge or diverge at branch points or intersections, so these points are strategic points to control ecological flow. Its strategic significance lies in controlling the spatial connection and effectively controlling the ecological flow in the landscape. According to the different distributary locations and isocenter points, the landscape strategic points at intersections can form a multi-level hierarchy. As shown in Figure 3, point a 1 is the most important because it controls the flow between multiple ecological sources. Points a2 and a3 are secondary because they control the flow between one source and other adjacent sources. Points a4 and a5 only control one corridor, so they have the least strategic significance. The introduction of ecological protection zones in these strategic landscape points can greatly improve the continuity and integrity of the overall landscape infrastructure.
(3) Plateau-type resistance surface and central, edge and corner strategic points.
The characteristic of plateau resistance surface is that high resistance area is surrounded by low resistance environment. A typical landscape is a landscape formed by selectively cutting forest patches, farmland surrounded by forests or disturbed patches in natural landscapes. The residual landscape with low resistance around becomes the source of biological diffusion to the central high resistance area. If the biological diffusion ability is large enough to overcome the maximum cumulative value of the resistance surface, then the landscape strategic points are located in the center of the high-resistance plateau, because occupying these central parts can provide first-hand, spatial connection and efficient advantages for biological control of the landscape process to the maximum extent, and the risks borne by biological individuals are within the tolerable range. If the biological diffusion ability is not enough to overcome the landscape resistance at one time, for example, the scale of wilderness exceeds the maximum diffusion ability of tree seeds, then the landscape strategic point is located on the edge of the plateau, especially in the corner, which can be vividly described as Go strategy. Limited field observation provides strong support for the existence of the above strategic points in the center, edge and corner landscape (such as Portnoy and Willson1993; Adikot et al.1987; Pearson1993; Foreman1995 p.106-109).
Although the spatial positions of the above-mentioned landscape strategic points are different, they are essentially the same, that is, they provide first-hand, spatially related and efficient advantages for ecological process competition and landscape occupation. Therefore, the landscape changes in these places will have a very strong impact on the landscape ecological process.
3? Identification of Strategic Points in Landscape Ecology: A Case Study
The case study is Danxia Mountain Scenic Area and National Geological Protection Area in Guangdong Province, with a research area of about 300 square kilometers, which consists of typical Danxia landform and vertical and horizontal valleys. Most of the natural vegetation has been destroyed, but there are still some sub-tropical monsoon forest scattered throughout the region. This research is based on a lot of basic work of Sun Yat-sen University and Peking University (Chen Chuankang et al. 1990). GIS data are stored in a 25× 25m2 grid, and the basic data layer includes artificial landscape elements such as topography, vegetation, water system and farmland. The following operations will apply various functions of GIS, including superposition, cost distance, catchment surface and so on. , combined with fragment program. See (1995a) for the detailed calculation process.
(1) ecological process
The ecological process considered in this case is the diffusion and preservation of various organisms, with the aim of protecting the local biodiversity in this area most effectively. Therefore, three groups of widely representative animals are selected as the research objects. , including: medium-sized mammals (civet cats and Cervidae only); Pheasants and amphibians (Ranidae only).
(2) Source
For medium-sized animals and pheasants, the source of spatial diffusion process is the residual subtropical monsoon forest patches, and there are seven patches with different sizes in this area (Figure 5-8), while for amphibians, the source of spatial diffusion is rivers and valleys (Figure 9).
Figure 5. Saddle strategic point determined from the resistance surface of medium-sized mammals.
Fig. 5. Strategic point of saddle point on resistance surface of medium-sized mammals.
Figure 6. Determine saddle strategic point according to pheasant resistance surface.
Figure 6. The strategic point of pheasant saddle point on the resistance surface
Figure 7. The distribution of network low resistance on the resistance surface of medium-sized mammals and the strategic point of intersection.
Figure 7. Strategic point of low resistance network intersection of medium-sized mammals
Figure 8. Distribution of network low resistance on pheasant resistance surface and strategic points of intersection.
Figure 8. Strategic Points of Crosspoint of Pheasant Low Resistance Network
Figure 9. The strategic point at the intersection of the amphibious vehicle resistance surface and the low resistance network and the corner strategic point of the plateau resistance surface.
Figure 9. The pheasant is at the strategic point at the intersection of low resistance network and at the corner of resistance plateau.
(3) Resistance surface
In this study, the minimum cumulative resistance model is used to calculate the landscape resistance surface (see Knaapen et al.1992; At 1995c), the formula is as follows: MCR = fmin ∑ (dij× ri) (I = 1, 2,3 ... n, J- 1, 2,3, ... m).
Dij represents the diffusion distance of species from source J through landscape I, Ri is the resistance of landscape I, and MCR is the minimum cumulative resistance of species from source J to a certain point in space. The function f is unknown, but it reflects the proportional relationship between MCR and variable (Dij×Ri). The resistance Ri of each landscape to species diffusion is determined by the basic characteristics of the landscape and the diffusion ability of the species itself. For medium-sized mammals, vegetation and slope are the main factors. For pheasants in the forest, vegetation type is the main factor. The main factor that determines the diffusion resistance of landscape to amphibians is hydrological conditions. In this case study, hydrological conditions are determined by altitude factors. According to related literature (such as Forman and Godran1986; Selman and Dole 199 1, Liu Chengzhao et al.1962; Wang Song et al. 1962). The resistance of various landscapes to three groups of animals was classified respectively. In principle, if the vegetation composition of the landscape is closer to the original vegetation, the diffusion resistance to all animals will be smaller. The gentler the terrain, the more conducive to the spatial diffusion of medium-sized animals, but for pheasants, the terrain has little effect. Relatively speaking, the lower the altitude, the better the hydrological conditions, and the more conducive to the spread of amphibians. Three resistance surfaces are obtained from the calculation results, which are suitable for three animal groups respectively. These surfaces express the potential possibility and dynamics of the corresponding animal groups in landscape control, and are the basic basis for determining the strategic points of landscape ecology (see, 1995a for details).
(4) Determine the strategic point according to the resistance surface.
The resistance surface of medium-sized animals and pheasants is mainly island-shaped, but the potential low-resistance areas form a dendritic network, which connects adjacent ecological sources and radiates from them (Figure 7-8). The resistance surface of amphibians is mainly reticulate, but there is also a plateau resistance surface formed by the low resistance area of river valley surrounding the high resistance area of Gao Qian (Figure 9). Technically speaking, the key points of landscape strategy of three types of resistance surfaces can be determined according to the methodology part of this paper.
A. Saddle-shaped strategic points: Figure 5-6 shows the saddle-shaped strategic points obtained from the resistant surfaces of medium-sized mammals and pheasants, which are the tangent points of adjacent sources, that is, the ecological impact range of residual south subtropical monsoon forest patches, and are potential springboard to connect adjacent sources.
B. Strategic intersection: For amphibians, the valley system constitutes a low-resistance network, and the intersection of tributaries and tributaries forms a strategic point. On the resistant surfaces of medium-sized mammals and pheasants, low-resistance networks also exist between ecological sources or radiate from sources. According to this, we can identify the landscape strategic points of intersections according to the methods mentioned at the beginning of the article. Figure 7-9 shows some typical intersection strategic points.
C. Angular strategic points: In addition to the network type, there are some plateau resistance surfaces in Figure 9. Considering that amphibians have very limited diffusion ability to overcome resistance, this paper only defines some corners as strategic points (such as valley concave banks). Just to give an example, the diameter of the concave bank arc where the preset strategic point is located is no more than 75 meters (see Figure 9).
By summing up all the above-mentioned landscape strategic points, the overall distribution pattern of landscape ecological strategic points can be obtained, which is of great significance to landscape ecological protection. They are also ecologically sensitive and have a low threshold (Yu, 199 1). By protecting or changing these landscape strategic points, the integrity of the structure and function of landscape ecosystem can be most effectively improved, and the economic cost paid by human beings for protecting the ecosystem can be minimized.
4。 discuss
Assuming that the horizontal process of landscape ecology is a spatial competition and control process to overcome some resistance, the basic principles of maximizing diffusion and minimizing cost and risk should be followed. The strategic points of landscape ecology are the minimum and maximum points or equilibrium points in space, and the landscape change activities at these points will bring abnormal influence to the ecological process. The landscape resistance surface expresses the possibility and dynamics of ecological process control and coverage. The spatial distribution of resistance surface can be shown as island type, network type and plateau type, and correspondingly there are saddle strategic points, intersection strategic points, central strategic points, edge and corner strategic points. Strategic points have multi-level hierarchical characteristics. This study gets some enlightenment from China's strategy model of Go. The three basic principles of Go strategy include firsthand, spatial connection and high efficiency. However, the ecological process is not an idealized black-and-white competition and chessboard control, so the study does not directly use the Go process to simulate the ecological process, but only uses the Go strategy as an analogy to illustrate the methodology, which is explained in the following two points:
First, the "chessboard" of ecological process competition is not a simple grid, but a complex nonlinear resistance surface. The relationship between ecological sources expressed by this resistance surface is not simply reflected in spatial distance, but a comprehensive reflection of distance and basic characteristics of landscape.
Secondly, the black and white pieces in Go are meaningless, that is, they have no weight and rank. However, as a "chess piece" in ecological process, habitat patches (sources) have structure and function, and their shape and hierarchical structure will affect their relationship with adjacent sources. This paper does not discuss this point, but only discusses the strategic points of landscape ecology, regardless of the attributes of the source itself. The case study in this paper is only used as an example of methodology. Due to the limited field observation data, the determined landscape ecological strategic point is not suitable for the actual landscape management and transformation, but it can be used as a thinking mode for further field investigation.