Tree-shaped river network multi-scale longitudinal connectivity dynamic evaluation method considering man-made interference

Abstract

The invention discloses a dynamic evaluation method for multi-scale longitudinal function connectivity of a tree-shaped river network fused with man-made interference. A river network topology adjacency matrix is constructed, an SCS-CN model is coupled to simulate natural runoff, and a reservoir regulation and storage coefficient and a sluice opening probability are introduced to correct artificial regulation and control influences; calculating river flow resistance by combining a Manning formula, and constructing a communication weight factor taking a runoff-resistance synergistic effect as a core; and based on a path product method and a node averaging method, longitudinal connectivity step-by-step evaluation from the river reach to the sub-drainage basin and the drainage basin scale is realized. The method breaks through the limitation that traditional structure connectivity evaluation is static and functional connectivity depends on measured data, can dynamically respond to rainfall events and project scheduling, is suitable for river network connectivity monitoring and management under multiple time scales such as day, month and year, and has a good popularization prospect.

Description

Technical Field

[0001] This invention relates to the fields of hydrology and hydraulic engineering, and specifically to a dynamic evaluation method for the multi-scale longitudinal connectivity of tree-like river networks that takes into account human interference. Background Technology

[0002] Dendritic river networks, a typical water system form in plains areas, bear the weight of dense populations and high-intensity economic activities. Their connectivity directly impacts water resource utilization, ecological security, and flood risk. However, under intense human intervention, the widespread construction of water conservancy projects (such as dams, reservoirs, and pumping stations) and the frequent application of scheduling rules have caused multi-dimensional disturbances to the river network: spatially, dams fragment the river network; temporally, reservoir scheduling disrupts natural runoff rhythms; and in terms of water volume, diversion and drainage projects alter local water distribution and weaken natural hydraulic gradients. These changes result in a highly dynamic and nonlinear evolution of river network connectivity.

[0003] Current mainstream connectivity assessment methods are mainly divided into two categories: structural and functional. Structural methods (such as node connectivity rates or landscape pattern indices based on graph theory) can characterize the static topological relationships of river networks, but they cannot reflect the material transport efficiency driven by actual hydrological processes. Functional methods (such as flow-weighted network transport models or hydraulic resistance models), while dedicated to quantifying flux transport, still have certain limitations: the flow-weighted method is highly dependent on measured hydrological data and is difficult to apply to river sections without monitoring stations; traditional resistance models, because they do not couple the chain-driven process of "rainfall-runoff-engineering regulation" (e.g., ignoring the dynamic impact of sluice gate opening and closing probabilities), are difficult to capture connectivity changes under human interference. Crucially, existing methods generally lack the ability to dynamically simulate multiple time scales (day / month / year), resulting in insufficient applicability when dealing with complex regulation scenarios. Therefore, there is an urgent need to construct a method for quantifying the longitudinal functional connectivity of tree-like river networks that can simultaneously integrate natural hydrological rhythms and human scheduling rules, reduce data dependence, and support multi-scale dynamic assessment. Summary of the Invention

[0004] Objective: To address the shortcomings of existing connectivity assessment methods that generally neglect the dynamic nature of hydrological processes and the complexity of human regulation, this invention aims to develop a dynamic assessment method for the longitudinal functional connectivity of tree-like river networks that can simultaneously couple natural hydrological rhythms and engineering scheduling rules. By integrating the runoff curve method (SCS-CN), the Manning resistance formula, and probabilistic dam control rules, a connectivity assessment method driven by the synergistic interaction of rainfall, runoff, regulation, and resistance is constructed.

[0005] Technical Solution: To achieve the above objectives, this invention provides a dynamic evaluation method for the multi-scale longitudinal connectivity of a tree-like river network considering human interference, comprising the following steps: (1) River network topology generalization. River network vector data for the study area was collected, including river channels, reservoirs, and sluice gates. The river network in the study area was generalized into a topological structure according to the following rules: river channels were generalized as edges; reservoirs, lakes, sluice gates, and river confluences were generalized as points; estuaries, river mouths, and river sources were ignored. Node numbers were... V 1, V 2, ..., V n-1 , V n And define the adjacency matrix to construct the adjacency matrix M∈R n×n , ( n (where M is the total number of nodes), the adjacency matrix M is a directed graph representation, and water flow is allowed only if the nodes can flow from the graph. i Flow direction j hour, M ( i , j =1 otherwise 0, ensuring path calculation conforms to the hydraulic flow direction. Construct a connectivity factor, representing the efficiency of the river's flow during transport, defined as the weight of each edge. W 1. W 2、…、 W m ( m (Total number of edges).

[0006] (2) Natural runoff simulation and anthropogenic regulation correction. Land use, soil type, rainfall, and measured runoff data were collected for the study area. An SCS-CN model was constructed to calculate the runoff process of each river segment in the river network under typical rainfall scenarios. The calculation formula is as follows: ; ; ; In the formula: Surface runoff (mm); P The total precipitation for this event is (mm). Sa Storage parameters (mm); CN The number of runoff curves is related to soil characteristics and land use patterns. Ia It is the initial value (mm). Sa Value follows CN The value changes as it changes in space.

[0007] Rivers contain sluice gates and reservoirs, and their hydrological and hydraulic conditions are subject to significant human interference; therefore, runoff must take these factors into account. For example, reservoirs regulate flow through storage capacity, while sluice gates regulate flow through gate width and opening probability. The calculation formula is: ; In the formula, This refers to the reservoir regulation coefficient. svol It is the storage capacity of the reservoir. A It is the drainage area; This is the sluice gate regulation coefficient. B The width of the gate. B 0 represents the width of the river channel. Prob This represents the probability of opening the gate.

[0008] As a preferred option, in step (3), the probability of opening the gate is calculated from the perspective of probability statistics based on the control water level or rainfall scheduling rules, and the probability of meteorological and hydrological events occurring is the probability of opening the gate.

[0009] (3) Calculation of channel resistance characteristics. The confluence analysis fully considers the confluence process of the river network and uses channel resistance characteristics as the representation. The channel resistance is calculated based on the Manning formula, and the calculation formula is as follows: ; In the formula, Rs The resistance to water flow in the river section (mm); The length of the river segment (km); The roughness coefficient is dimensionless. The cross-sectional area of ​​the river section (m²) 2 ); R The hydraulic radius (m); S It is a hydraulic gradient.

[0010] (4) Constructing the connectivity weighting factor. Based on the above SCS-CN distributed flow generation model, Manning resistance equation, and artificial adjustment coefficient ( C 1. C 2) Construct the connectivity factor as In the formula, To pass through the river section n The vertical connectivity represents the efficiency of water flow through this river segment. Output the connectivity weight vector for all nodes. W =[ W 1, W 2,…, W m ] T The connectivity factor is normalized and ranges from (0,1). The larger the value, the higher the efficiency of water flow through the river section and the stronger the connectivity.

[0011] (5) Calculation of longitudinal connectivity of tree-like river networks at different spatial scales. Based on the adjacency matrix and weight vector, the longitudinal connectivity of the path from one node to another is calculated using the following formula: ; In the formula, For the river section j to the river section kLongitudinal connectivity along the path is a connection j and k The product of the local longitudinal connectivity of all river segments along the path represents the connection probability between a pair of river segments.

[0012] The average vertical connectivity of a given node to all other nodes is called the connectivity of that node. The formula is as follows: ; In the formula, For river section k The vertical connectivity of the nodes is calculated as pointing to the river segment. k It is the average connectivity of the paths. Therefore, it represents the probability that a single river segment connects to all other river segments in the river network.

[0013] The average connectivity of all nodes is used as the overall connectivity of the river network. The calculation formula is as follows: ; In the formula, HCI For the longitudinal connectivity of the river network, it represents the average probability that any river segment connects to all other river segments in the river network.

[0014] Beneficial effects: Compared with the prior art, the present invention, using the above technical solution, has the following technical effects: 1. Defining connectivity with "flow efficiency" as the core, based on runoff generation and confluence mechanisms, and integrating runoff driving force and river resistance, is more in line with the essence of hydrology.

[0015] 2. By coupling the SCS-CN runoff model with reservoir / sluice gate scheduling rules, connectivity assessment can respond to different rainfall scenarios and artificial control strategies, supporting flexible assessments from single river segments to the entire basin, and from daily to annual scales, to meet different management needs.

[0016] 3. The algorithm can be modularly designed and integrated into a GIS platform or hydrological management system. Attached Figure Description

[0017] Figure 1 It is a river system map of the study area; Figure 2 It is an SCS-CN model construction; Figure 3 It is the spatial dynamics of river connectivity; Figure 4 It is the spatial dynamics of connectivity at the sub-basin scale; Figure 5 It is the temporal dynamics of watershed-scale connectivity. Detailed Implementation

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] This embodiment takes the Qinhuai River Basin as an example to conduct a longitudinal functional connectivity assessment and analysis of the tree-like river network. The study first collected maps of the study area, including basic data such as river channels, reservoirs, and sluice gates. Figure 1 Then, the SCS-CN model was constructed to simulate runoff, the Manning formula was used to calculate channel resistance, and then the channel connectivity weight factor was constructed. Finally, based on the river network adjacency matrix and connectivity weight vector, and taking into account the cascading process of the tree-like river network, the longitudinal connectivity of the river network at different scales was calculated.

[0020] The SCS-CN model, according to Figure 2 The flowchart illustrates the steps involved in the study, which included collecting data on land use, soil characteristics, and typical rainfall events in the Qinhuai River basin, and constructing the SCS-CN model. The equations for calculating runoff using the SCS model are as follows: Q =( P -0.2 S ) 2 / ( P +0.8 S ),in: Q Runoff measured in millimeters. P Rainfall depth is measured in millimeters. S The maximum potential water storage capacity is measured in millimeters. S The relationship with the number of curves is as follows: S =25400 / CN -254. Where: CN The curve number, a dimensionless parameter, reflects the ability to generate surface runoff. Preferably, the CN value from Chapter 9 of the National Engineering Handbook is used, and the curve number is determined by combining land cover and soil data. The runoff curve method typically uses the curve numbers listed in Table 1. Figure 2 bd respectively illustrates the spatial distribution of land use, soil characteristics, and CN values ​​in the Qinhuai River Basin. If there are sluice gates or reservoirs controlling the river channel, the runoff needs to be multiplied by a regulation or storage coefficient. Preferably, the sluice gate regulation coefficient is calculated from the gate width and opening probability, extrapolated based on the control water level or rainfall scheduling rules; the probability of a meteorological and hydrological event occurring is the sluice gate opening probability. The reservoir storage coefficient is calculated from the reservoir capacity ratio. The calculation formula is: ; In the formula, This refers to the reservoir regulation coefficient. svol It is the storage capacity of the reservoir. A It is the drainage area; This is the sluice gate regulation coefficient. B The width of the gate. B 0 represents the width of the river channel. Prob This represents the probability of opening the gate.

[0021] Table 1. Lookup table for common curve number values

[0022] The river resistance is calculated using the Manning formula. The data used in the Manning formula calculation are the river cross-section, water level, riverbed slope, river width, and riverbed roughness, obtained through field measurements and consultation of hydrological yearbooks. The Manning formula is as follows: ; In the formula, Rs The resistance to water flow in the river section (mm); The length of the river segment (km); The roughness coefficient is dimensionless. The cross-sectional area of ​​the river section (m²) 2 ); R The hydraulic radius (m); S It is a hydraulic gradient.

[0023] The connectivity weighting factor is calculated from runoff and resistance, and the calculation formula is as follows: ; In the formula, For river section n The dynamic water flow connectivity represents the efficiency of water flow through this section of the river.

[0024] The adjacency matrix is ​​extracted after river network generalization according to the following rules: the source, outlet, and confluence of rivers are generalized as network nodes, and the rivers are generalized as network edges, with connectivity weight factors assigned to the edge vector file.

[0025] The longitudinal connectivity of the river network at different scales, including longitudinal connectivity at the path, river segment, and river network scales, are calculated using the following formulas: ; In the formula, For the river section j to the river section k Dynamic flow connectivity along the path is a measure of the connection j and k The product of the local water flow connectivity of all river segments along the path represents the connection probability between a pair of river segments.

[0026] ; In the formula, For river section k The dynamic water flow connectivity of the nodes is calculated as a direction pointing to the river segment. kIt is the average connectivity of the paths. Therefore, it represents the probability that a single river segment connects to all other river segments in the river network.

[0027] ; In the formula, HCI N represents the dynamic flow connectivity of the river network, which is the average probability that any segment of the river connects to all other segments in the river network. The value of N is n.

[0028] The results of the longitudinal connectivity calculation of the tree-like river network in the Qinhuai River Basin are as follows: The spatial distribution of the average connectivity at the river section scale in the Qinhuai River Basin from January to December 2017 is shown in the figure. Figure 3 As shown, the spatial distribution of average connectivity at the sub-basin scale is as follows: Figure 4 As shown, the monthly average connectivity change trend at the watershed scale is as follows: Figure 5 As shown, the connectivity results obtained directly reflect the "transmission accessibility" of water resources within the basin. High connectivity areas mean smooth water flow and efficient water exchange, while low connectivity areas may experience water "stagnation" or "interruption" due to dam blockage or excessive resistance. This data can guide inter-regional water transfer, adjust the probability of sluice gate opening based on connectivity data (such as increasing the opening frequency after rainfall), or optimize reservoir regulation coefficients (increasing the discharge flow) to ensure irrigation water delivery and identify water waste points.

[0029] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic evaluation of the longitudinal connectivity of a tree-like river network at multiple scales, considering human interference, characterized in that, Includes the following steps: (1) The actual tree-like river network is generalized into a graph theory network model according to the following rules: the river channel is generalized as an edge, and the reservoirs, lakes, sluice gates and river confluences are generalized as nodes, with the nodes numbered as follows: V 1, V 2, ..., V n-1 , V n And define the adjacency matrix to construct the adjacency matrix M∈R n×n , n Let M be the total number of nodes, and let M be the adjacency matrix representing a directed graph. Water flow is allowed only if the nodes can pass through the graph. i Flow direction j hour, M ( i , j =1 otherwise 0, construct the connectivity factor, representing the efficiency of the river's flow during transport, defined as the weight of each edge. W 1. W 2、…、 W m , m The total number of edges; (2) Construct the SCS-CN model for the study area and calculate the potential runoff process at each node of the river network under typical rainfall scenarios. The calculation formula is as follows: ； ； ； In the formula: Surface runoff; P This represents the total precipitation of the event; Sa To store parameters; CN The number of runoff curves is related to soil characteristics and land use patterns. Ia It is the initial value. Sa Value follows CN The value changes as it changes in space; (3) Define the flow resistance in the river channel as the residence time. Substitute the cross-sectional area of ​​the river, the hydraulic radius, the riverbed slope, and the roughness into the Manning formula to calculate the flow resistance of each river channel, which is used as the weight of the network edge. The calculation formula is as follows: ； In the formula, Rs denoted as the resistance to water flow in the river section; L is the length of the river section; n is the roughness coefficient, which is dimensionless. The cross-sectional area of ​​the river section; R The hydraulic radius; S For hydraulic gradient; (4) Construct connectivity factors based on runoff and channel resistance. In the formula, Rs represents the water flow resistance of the river section. To pass through the river section n The vertical connectivity represents the efficiency of water flow through this river segment, and the output is the connectivity weight vector of all nodes. W =[ W 1, W 2,…, W m ] T ; (5) Calculate the vertical connectivity of the path from one node to another based on the adjacency matrix and weight vector. The calculation formula is as follows: ； In the formula, For the river section j to the river section k Longitudinal connectivity along the path is a connection j and k The product of the local longitudinal connectivity of all river segments along the path represents the connection probability between a pair of river segments; The average vertical connectivity of a given node to all other nodes is called the connectivity of that node. The formula is as follows: ； In the formula, For river section k The vertical connectivity of the nodes is calculated as pointing to the river segment. k The average connectivity of the paths; The average connectivity of all nodes is used as the overall connectivity of the river network. The calculation formula is as follows: ； In the formula, HCI For the longitudinal connectivity of the river network, it represents the average probability that any river segment connects to all other river segments in the river network.

2. The method for dynamic evaluation of multi-scale longitudinal connectivity of tree-like river networks considering human interference as described in claim 1, characterized in that: In step (2), if there are water conservancy facilities, the runoff is multiplied by an adjustment coefficient for correction. The calculation formula is as follows: ； In the formula, This refers to the reservoir regulation coefficient. svol It is the storage capacity of the reservoir. A It is the drainage area; This is the sluice gate regulation coefficient. B The width of the gate. B 0 represents the width of the river channel. Prob This represents the probability of opening the gate.

3. The method for dynamic evaluation of multi-scale longitudinal connectivity of tree-like river networks considering human interference as described in claim 2, characterized in that: In step (4), .

Images