Method for calculating effective hydrological connectivity strength index of wetland based on InSAR

Abstract

The invention belongs to the technical field of water resource planning and management, and particularly relates to a method for calculating an effective hydrological connectivity strength index range of a wetland based on an InSAR (Interferometric Synthetic Aperture Radar) technology. The calculation method comprises the following steps: acquiring image data of an area to be calculated; identifying all wetland plaques in the water body area; dividing the wetland plaques into different effective hydrological communication type groups; calculating an elastic force value of each wetland patch; constructing a water level fluctuation-ecological elastic force curve under different effective hydrological communication situations, carrying out segmented fitting on the water level fluctuation to generate a fitting line, calculating the mean value of the elastic force of the wetland plaques under the effective hydrological communication type, generating an elastic force mean value line, and calculating the elastic force mean value line; the two intersection points of the elastic force mean value line and the fitting line are the effective hydrological connection strength index range. According to the method, the wetland plaques of different effective hydrological connection type groups are divided according to the difference of the effective hydrological connection strength, the optimal ecosystem elastic force of the wetland plaques is calculated respectively, and then the effective hydrological connection strength index range is determined.

Description

Technical Field

[0001] The invention belongs to the technical field of water resource planning and management, and in particular relates to a method for calculating a wetland effective hydrological connectivity intensity index range based on InSAR technology. Background Art

[0002] At present, wetland maintenance and restoration has become a very important ecological protection activity. In order to ensure the ecological water demand of wetlands, the government and management departments have implemented a variety of ecological water replenishment or other water level control measures, such as ecological water replenishment implemented in water-deficient lakes and wetlands in the north. However, this measure has problems such as misplaced timing of water replenishment or a single method, which often leads to sudden water level rises; the southern water-rich lakes and wetlands have raised the minimum water level line for water storage and power generation, reducing seasonal fluctuations in water levels. This water replenishment method or water level control that ignores the threshold of hydrological fluctuations causes the water level fluctuations in lakes and wetlands to deviate from the natural rhythm, and the water level and the water demand of the ecosystem are disordered. At the same time, extreme climate events (floods or droughts) caused by global warming, or ecological water replenishment causing sudden rises in water levels in individual wetlands, and human activities such as dams, roads, and reclamation have blocked or blocked the internal water system connectivity of wetlands. The static management methods used to maintain high water levels in the past have reduced the interaction of ecosystem networks, stagnated ecosystem succession, and made single vegetation such as reeds the dominant species. The lack of rhythmic succession has led to reduced biodiversity, increased system vulnerability, reduced stability, and a single ecosystem function, which is not conducive to the long-term optimization of wetland ecosystems. Therefore, the range of water level fluctuations should be taken into account in the setting of ecological water replenishment plans.

[0003] However, due to historical, technical and government reasons, hydrological data for many natural wetlands are missing. This has caused trouble for us in determining the laws of hydrological fluctuations and changes, hindered the quantification of effective hydrological connectivity within wetlands, and thus affected the formulation of wetland protection and restoration policies. Summary of the invention

[0004] The present invention proposes a calculation method for the effective hydrological connectivity strength index of wetland ecosystems based on InSAR. First, a multi-time series relative water level data set is generated in areas without hydrological data based on InSAR technology. Hydrological barriers are identified based on the phase mutation principle of adjacent interference patterns, and the connectivity strength of lake wetlands is judged according to the degree of barriers. Then, different water level fluctuation scenarios are set to obtain effective hydrological connectivity wetland type groups of different strengths. The lagged autocorrelation (AC1) is used to quantify the elasticity of wetland ecosystems under various effective hydrological connectivity strengths. By comparing the elasticity of wetland ecosystems, the effective hydrological connectivity strength range for maintaining the stability of wetland ecosystems is finally determined. This solves the problem in the prior art that the effective hydrological connectivity strength index is inaccurate due to the unclear water level fluctuation process, which leads to inaccurate estimation and affects wetland protection and restoration.

[0005] The technical solution adopted by the present invention to solve the technical problem is:

[0006] A method for calculating the wetland effective hydrological connectivity intensity index based on InSAR, characterized by comprising the following steps:

[0007] Step 1: Select a satellite that can cover the area to be calculated, obtain image data of the area to be calculated at intervals within a set time length, and obtain a multi-time series relative water level data set of the area to be calculated based on the image data;

[0008] Step 2: Identify the water body area of ​​the area to be calculated based on the multi-time series relative water level data set, and identify all wetland patches in the water body area;

[0009] Step 3: Acquire the water level fluctuation data of the wetland patch according to the multi-time series relative water level data set, judge the hydrological effective connectivity strength and its consistency according to the CV value of the water level fluctuation and the double cumulative curve, and divide the wetland patch into different effective hydrological connectivity type groups;

[0010] Step 4: Calculate the elastic force value of each wetland patch;

[0011] Step 5: Construct water level fluctuation-ecological elasticity curves under different effective hydrological connectivity scenarios, perform segmented fitting on the water level fluctuation to generate a fitting line, calculate the mean elasticity of the wetland patches under the effective hydrological connectivity type, and generate an elasticity mean line. The two intersection points of the elasticity mean line and the fitting line are the range of the effective hydrological connectivity intensity index.

[0012] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the area to be calculated can be covered by satellites at different orbital positions, and the number of the satellites is one satellite or more than two;

[0013] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the set time length includes at least two anniversaries;

[0014] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the set time interval for obtaining the image data is 10 to 40 days, and the image data is obtained once every 10 to 40 days when the satellite revisits the area to be calculated.

[0015] Furthermore, in step 2 of the above-mentioned calculation method of the effective hydrological connectivity intensity index of wetlands, the water body threshold ε is determined by the Ostu algorithm in the identified water body area. When SDWI>ε, it is a water body area, otherwise it is a non-water body area. The calculation formula of SDWI is:

[0016] SDWI=ln(10×VV×VH)-8

[0017] Where SDWI is the dual-polarization water index; VV and VH represent the transmission and reception polarization modes of the radar system, VV is vertical transmission and polarization, and VH is vertical transmission and horizontal reception.

[0018] Furthermore, in step 2 of the above-mentioned method for calculating the wetland effective hydrological connectivity intensity index, the range of permanent flooding and the range of periodic flooding are identified by changing the maximum coverage area and the minimum coverage area of ​​the water body in the water body area, and the range of periodic flooding is set as the boundary of the wetland patch.

[0019] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the image data includes phase changes, the water level fluctuation data is relative water level data, and the relative water level height is calculated by the phase shift data between the transmission and reception of the radar signal. The calculation formula of the relative water level height is:

[0020] Where Δh is the relative water level, λ is the incident wavelength of the radar signal, and θ is the satellite incident angle. is the phase change, and d is the difference between the reflected signals at two different times.

[0021] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, a CV value of 0.6-0.75 represents a wetland type ii with slight water level fluctuations, a CV value of 0.75-0.9 represents a wetland type i with moderate water level fluctuations, and a CV value of 0.9-1.05 represents a wetland type iii with severe water level fluctuations.

[0022] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, within a set time length, it includes multiple set time intervals, and water level fluctuation data can be obtained once at each time interval. The sizes of the multiple water level fluctuation data are compared, and the largest value is the instantaneous maximum water level fluctuation.

[0023] Furthermore, in the calculation method of the wetland effective hydrological connectivity intensity index, a set time length includes at least two anniversaries, and the numerical values ​​of the water level fluctuation data of each anniversary are compared, and the difference between the water level fluctuation data with the largest value and the water level fluctuation data with the smallest value is the water level fluctuation value of the anniversary;

[0024] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the water level fluctuation values ​​of all years within the set time length are compared, and the largest value is the maximum water level fluctuation of the year.

[0025] Furthermore, in the calculation method of the above-mentioned wetland effective hydrological connectivity strength index, the elastic force value is the lagged autocorrelation AC1, and the specific calculation formula is: AC1 = e λΔt

[0026] Where Δt is the time step and λ is the recovery rate. The closer λ is to 0, the larger AC1 is and the corresponding elastic force is lower.

[0027] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the AC1 value under each fluctuation amplitude is calculated for each wetland of each effective hydrological connectivity type, so as to draw an AC1-water level fluctuation curve, perform linear fitting on the curve, and find the inflection point of the curve through the threshold judgment formula;

[0028] Furthermore, in the above-mentioned calculation method of the wetland effective hydrological connectivity intensity index, the threshold judgment formula is:

[0029]

[0030] Where t represents water level fluctuation, y represents AC1; β0, β1, and β2 are regression coefficients; α is the turning point assumed by the stepwise regression model, which is determined based on t analysis; when β1 and β2 are found to be different, the α value is the target water level fluctuation threshold point; the three regression coefficients are estimated using least squares linear regression, and the t test is applied to test whether β2 is not equal to 0.

[0031] By means of the above technical solution, the present invention has at least the following advantages:

[0032] 1) The present invention divides wetland patches into different effective hydrological connectivity type groups according to the difference in water level fluctuations, calculates their ecosystem resilience respectively, and then determines the range of the effective hydrological connectivity intensity index.

[0033] 2) The present invention establishes a relationship between the intensity of effective hydrological connectivity and the resilience of the ecosystem, adjusts the key parameters that affect the stability of the wetland ecosystem, and fundamentally improves the self-repairing ability of the wetland, thereby achieving a long-term and economical ecological restoration effect.

[0034] 3) The present invention removes the research restrictions in areas without hydrological data through the latest remote sensing technology, thereby increasing the application breadth of the method and facilitating its promotion in practical applications.

[0035] 4) The present invention realizes the refinement of hydrological regulation by separately calculating different connectivity intensities, which is beneficial to improving the comprehensive effect of ecological restoration.

[0036] 5) The present invention can provide a series of suitable hydrological regulation schemes applicable to different connectivity intensities, ensuring that ecosystems with different effective connectivity intensities maintain the optimal health state, which is more instructive and practical in wetland restoration.

[0037] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The present invention is a schematic diagram of the principle of hydrological barrier and the identification of hydrological barrier when the water level changes in different hydrological seasons;

[0039] Figure 2 The principle of using interferometric synthetic aperture radar double back reflection signals to measure the relative change of wetland water surface;

[0040] Figure 3 Schematic diagram of the wetland patch delineation process

[0041] Figure 4 It is a schematic diagram of phase unwrapping results;

[0042] Figure 5 To determine the fluctuation trend of wetland water level based on double accumulation curves;

[0043] Figure 6 Identify threshold ranges of water level fluctuations for changes in ecosystem resilience;

[0044] Figure 7 Water level fluctuation-ecological resilience curve under different effective hydrological connectivity scenarios. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of the present invention application are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0046] like Figures 1 to 7As shown, this embodiment provides a method for calculating the threshold range of wetland water level fluctuation based on InSAR technology, including the following steps: Step 1: Select a satellite that can cover the area to be calculated, and intermittently obtain image data of the area to be calculated within a set time length, and obtain a multi-time series relative water level data set of the area to be calculated based on the image data; Step 2: Identify the water body area of ​​the area to be calculated based on the image data, and identify all wetland patches in the water body area; Step 3: Judge the fluctuation intensity and consistency based on the CV value and double cumulative curve of the water level fluctuation, and divide the wetland patches into different effective hydrological connectivity intensity type groups; Step 4: Calculate the elastic force value of each wetland patch; Step 5: Construct a water level fluctuation-ecological elastic force curve under different effective hydrological connectivity scenarios, perform segmented fitting of water level fluctuation and ecosystem elastic force to determine the water level fluctuation threshold point, that is, the effective hydrological connectivity intensity, and further confirm the effective hydrological connectivity intensity index range for maintaining ecosystem stability.

[0047] The present invention calculates and obtains a data set including relative water level image information through image data, wherein the image data is a Sentinel-1A or Sentinel-1B C-band vv polarization single-view composite (SLC) image.

[0048] The present invention divides wetland patches into different effective connectivity types according to the difference in water level fluctuations, calculates the segmented fitting water level fluctuation threshold points respectively, and then determines the intensity index range of effective hydrological connectivity to maintain system stability. Ecosystem resilience, as an important criterion for judging the health of an ecological system, is a measure of the ability of an ecosystem to recover from external disturbances. This indicator not only indicates the current health status, but also indicates the long-term recovery capacity of the ecosystem. Therefore, using ecosystem resilience as a standard for wetland restoration is an efficient, long-term and economical indicator. According to the cup-and-ball theory, when the disturbance is at a certain threshold, the ecosystem will achieve optimal resilience, that is, it has the maximum self-regulation ability to maintain the health of the wetland. As a kind of ecological disturbance, water level fluctuations affect the vegetation species and growth status on the one hand, and the material and energy cycle on the other hand. It is an important factor affecting the resilience of the ecosystem and a hydrological regulation method that is easy to operate manually. Therefore, judging the response of ecosystem resilience to hydrological fluctuations is considered to be the key to accurately assessing the health of the ecosystem. And the water level fluctuation amplitude and consistency are divided by CV and double cumulative curve method, and the water level fluctuations of different types of wetlands are divided into wetland type groups i, ii, and iii with different effective hydrological connectivity strengths, which is conducive to analyzing the impact mechanism of changes in effective connectivity strength on wetland ecosystems. Therefore, the present invention takes the intensity index range of effective hydrological connectivity that maintains system stability as the starting point, with the aim of achieving the maximum resilience of the ecosystem. In the past, in the process of calculating ecological resilience, wetlands were usually regarded as a whole and calculated by vegetation index (such as NDVI). However, the basic conditions within the wetland are not the same. The hydrological connectivity strength such as whether the unit wetland patches are connected and whether there are hydrological barriers within the unit wetland patches affects the material hydrological cycle of the wetland, and thus affects the background ecological resilience of the unit wetland patch. The different intensities of effective hydrological connectivity will lead to changes in the vegetation type of the wetland patch, which in turn affects the baseline value of the ecosystem resilience of the wetland type. Therefore, we need to first divide different wetland patches into different wetland type groups i, ii, and iii based on the different effective connectivity intensities. By first classifying and then calculating the wetland elasticity value, we can reflect the impact of different water level fluctuations on the wetland ecosystem, thereby more clearly showing the changes in the ecosystem state under different effective connectivity intensities and determining the intensity index range of effective hydrological connectivity that maintains system stability.

[0049] Changes in water level fluctuations in wetlands will significantly change the structure, function and biodiversity of wetlands, thereby affecting wetland losses. Climate change and human influence have led to changes in hydrological conditions. Under the combined influence of multiple factors, the mechanism of water's effect on wetlands has become increasingly unpredictable. Water level fluctuations will affect the vegetation pattern and characteristics, thereby changing wetland habitat elements. Water level fluctuations will also affect the number of water birds in wetlands and affect the stability of wetland habitats. The amplitude, duration, and frequency of water level fluctuations in wetland ecosystems play a decisive role in the vegetation in wetlands, and wetland vegetation resilience is defined as the ability of a wetland system to recover from external disturbances. Therefore, wetland water level fluctuations can be linked to wetland resilience through the response of vegetation changes. At the same time, changes in wetland water level fluctuations lead to changes in wetland water depth, flooding frequency, and flooding duration, thereby affecting the wetland hydrological connectivity. Identifying and dividing the characteristics of water level changes can quantify the effective hydrological connectivity of the wetland. Using the hydrological threshold as the effective hydrological connectivity intensity of the wetland, wetland type groups with different effective hydrological connectivity intensities are divided under different wetland water level fluctuations. This is conducive to exploring the response mechanism of vegetation elasticity to water level changes in different wetland type groups and identifying the range of effective hydrological connectivity intensity index.

[0050] Specifically, the mean of the elastic force of each wetland type group is used to represent the stable value of the elastic force of the wetland in this type group, and a trend line of the elastic force changing with water level fluctuations is fitted. The intersection of the two represents that the relationship between the wetland elastic force and the water level change will change, which can reflect the transformation of the intrinsic action mechanism of the wetland elastic force due to the change of effective connectivity strength.

[0051] Wetland hydrological connectivity is an important factor affecting wetland habitats. The transfer of materials and energy between wetlands, as well as the water and nutrient transport systems between wetlands and the outside world, all rely on hydrological connectivity. However, in previous studies, hydrological connectivity has been mainly divided into three dimensions: horizontal connectivity, vertical connectivity, and vertical connectivity. This wetland hydrological connectivity framework only considers the impact of water on wetlands in different spatial dimensions, and fails to deeply explore the impact mechanism of changes in hydrological connectivity on wetland habitats under changes in wetland hydrological status. Traditionally, hydrological connectivity is defined at the basin or regional scale, which refers to the exchange of water and water media (such as sediments, energy, and organisms) between rivers or lakes and their floodplains, and between heterogeneous areas within the floodplain. It is assumed that water level fluctuations are the only key indicator affecting the structure and function of wetland ecosystems, and precipitation and periodic floods are only external factors that change water level fluctuations. Based on the interaction mechanism between hydrology and wetland ecosystems, effective hydrological connectivity is defined as the change of hydrological regime (such as flooding depth, duration, frequency and time) under specific wetland geomorphic features, inducing water and water-mediated exchanges (such as sediment, energy and organisms) and accompanying biological and biochemical processes. They play a key control role in the formation of ecological niches of different plant species in wetlands. Considering the cumulative effect and hysteresis effect of hydrological regime on wetland habitats, we use hydrological thresholds to represent the intensity of the impact of effective hydrological connectivity on wetland habitats.

[0052] The phase of the interference pattern generated by the interconnected water surface is a continuous phase change. After the barrier appears, there will be a sudden phase change between two adjacent points. This is the hydrological barrier identified by InSAR theory. Figure 1 .

[0053] The satellite described in the present invention is an earth observation satellite, which uses InSAR technology to generate a multi-time series relative water level data set in areas without hydrological data. Specifically, the satellite described in this embodiment is Sentinel 1. The Sentinel-1 is an earth observation satellite in the Copernicus Program (GMES) of the European Space Agency. It consists of two satellites and carries a C-band synthetic aperture radar. It can provide continuous images (day, night and various weather conditions), and use phase information to extract terrain deformation, and obtain continuous data of different wetland water surfaces. The revisit period of the Sentinel-1 is 12 days. In specific implementation, image data can be obtained once every 12 days. However, due to signal and other reasons, image data may be obtained once every 24 days or 36 days.

[0054] The wetland water level was extracted by using interferometric synthetic aperture radar technology. The relative water level and absolute water level were extracted through the Sentinel-1 dataset. The high spatial resolution vertical ground displacement interferogram was generated based on the pixel-by-pixel SAR phase (i.e., the phase shift between radar signal transmission and reception) observations obtained by interferometric synthetic aperture radar from the same orbital position, different times and the same wetland using Sentinel-1 data. The relative water level difference was calculated based on the phase path length difference, and the identified relative water level and absolute water level were verified against each other.

[0055] The present invention first generates a multi-time series relative water level data set in areas without hydrological data based on InSAR technology, then identifies hydrological barriers based on the principle of phase mutation of adjacent interference patterns, and determines the connectivity strength of lake wetlands according to the degree of barriers. Then, the water level fluctuation intensity and its consistency are obtained by performing CV calculation and double cumulative curve analysis on the water level fluctuations of different wetland patches, and wetland type groups with different effective hydrological connectivity intensities are obtained. The lagged autocorrelation (AC1) is used to quantify the elasticity of the wetland ecosystem under each effective hydrological connectivity intensity, and by comparing the size of the wetland ecosystem elasticity, the intensity index range of the effective hydrological connectivity of the ecosystem under different connectivity intensities is finally determined. The invention can provide a series of suitable hydrological regulation schemes for different connectivity intensities, ensuring that ecosystems with different effective connectivity intensities maintain the best health state, which is more instructive and practical in wetland restoration.

[0056] The InSAR technology, namely Interferometric Synthetic Aperture Radar (InSAR), is a reliable technology for monitoring centimeter-level displacement of the ground surface. It compares pixel-by-pixel SAR phase (i.e., phase shift between radar signal emission and reception) observations obtained from the same orbital position, different times and the same area to generate a vertical ground displacement map with high spatial resolution, called an interference map. Traditional InSAR uses the difference in phase path lengths of two satellites collected from the same orbital position, the same resolution and the same area to generate displacement maps representing the change of solid surfaces over time. These maps are called interference maps, which include visualization of phase differences indicated by coloring, and sometimes contain unstable contours called "stripes", which can be continuous or discontinuous. The water surface of a wetland has different characteristics from the land surface. The present invention uses InSAR technology to detect the water surface displacement map of the water level of a lake wetland to identify the change of the water level of a lake wetland.

[0057] Specifically, the present invention uses InSAR technology to provide a way to obtain water level data, obtain relative water level through satellite phase difference between two periods, and then calculate water level fluctuations, thereby solving the problem of missing hydrological data of natural wetlands due to historical, technical and government attention reasons.

[0058] In specific implementation, if one satellite cannot cover the entire area to be calculated, satellites at different orbital positions can be used to cover the area to be calculated, that is, the number of satellites is one or more than two. Thus, the entire area to be calculated can be covered to obtain relative water level data of the entire area to be calculated.

[0059] The set time length of the present invention includes at least two years, so that the relative water level value of the area to be calculated is obtained each time the satellite covers the area to be calculated, and the instantaneous maximum water level fluctuation is calculated based on the relative water level value, and the maximum fluctuation of the year is calculated within the time length of one year.

[0060] In a specific implementation of the present invention, the set time interval for acquiring image data is 10 to 40 days, and the image data is acquired once every 10 to 40 days when the satellite revisits the area to be calculated. The set time interval for acquiring image data is based on the period of the satellite revisiting the area to be calculated.

[0061] In step 2, the water body area is identified by determining the water body threshold ε through the Ostu algorithm. When SDWI>ε, it is a water body area, otherwise it is a non-water body area. The calculation formula of SDWI is: SDWI=ln(10×VV×VH)-8 (1), where SDWI is the dual-polarization water body index; VV and VH represent the transmission and reception polarization modes of the radar system, VV is vertical transmission and polarization, and VH is vertical transmission and horizontal reception.

[0062] In a specific implementation of the present invention, the range of permanent flooding and the range of periodic flooding are identified by changing the maximum coverage area and the minimum coverage area of ​​the water body in the water body area, and the range of periodic flooding is set as the boundary of the wetland patch.

[0063] Specifically, when the area is identified as a water level, the range of the wetland patch A can be understood by the following formula: A = A max -A min (2), where A is the wetland patch, i.e., the alternating wet-dry zone, and A max is the maximum water coverage area of ​​the target water body identified by formula (1) in a year, A min It is the minimum water coverage area of ​​the target water body identified by formula (1) in one year, that is, the permanent water body.

[0064] In the present invention, the image data includes phase changes, the water level fluctuation data is relative water level data, and the relative water level height is calculated by the phase shift data between the transmission and reception of the radar signal. The calculation formula of the relative water level height is:

[0065] Where Δh is the relative water level, λ is the incident wavelength of the radar signal, and θ is the satellite incident angle. is the phase change, and d is the difference between the reflected signals at two different times.

[0066] Figure 2 The principle of using the double back reflection signal of interferometric synthetic aperture radar to determine the relative change of wetland water surface. The phase change of InSAR technology is mainly due to the double back reflection mechanism of radar signal in water and vegetation. The satellite transmits the radar signal at the same position with an incident angle of θ. In the time interval of satellite return, the water level changes. The red arrow is the direction of the assumed water level change. The water level rises from the dotted line position to the solid line position. SAR1 and SAR2 are the reflection signals of the radar signal at two water level positions respectively. The difference of the reflection signal at two different times is d, which can be calculated by the incident wavelength and the phase change. Finally, d is divided by the cosine value of the incident angle to get the water level change Δh between the two times.

[0067] In a specific embodiment of the present invention, based on the double cumulative curve method, a wetland type group with consistent water level fluctuation process is further screened out from three wetland type groups with different water level fluctuation intensity. The relative water level fluctuation values ​​of two wetlands in the same type of water level fluctuation amplitude are accumulated. If the slope of the generated fitting line does not change, it means that the water level fluctuation time and trend of the two wetlands are consistent, and wetlands with different hydrological fluctuation time and trend in the wetland types with the same water level fluctuation intensity are eliminated. A wetland type group with consistent hydrological fluctuation situation is obtained, which we will define as a wetland type group with the same effective hydrological connectivity intensity in the following text.

[0068] Effective hydrological connectivity is defined as the change of hydrological state (such as flooding depth, duration, frequency and time) under specific wetland geomorphic features, inducing water and water-mediated exchange (such as sediment, energy and organisms) and accompanying biological and biochemical processes. In wetland type groups with the same water level fluctuation intensity and consistency of water level fluctuation process, they have the same effective hydrological connectivity strength. The effective hydrological connectivity strength is represented by the water level fluctuation threshold. Wetland type groups with different effective hydrological connectivity strength have different effective hydrological connectivity strength index ranges. The wetland type groups are divided into three different effective connectivity strengths, namely wetland type group i, wetland type group ii, and wetland type group iii.

[0069] The method for judging the effective hydrological connectivity type according to the CV value of the water level fluctuation and the consistency of the water level fluctuation process is specifically:

[0070] The CV value of 0.6-0.75 is type ii for wetlands with slight water level fluctuations, the CV value of 0.75-0.9 is type i for wetlands with moderate water level fluctuations, and the CV value of 0.9-1.05 is type iii for wetlands with severe water level fluctuations.

[0071] Based on the double cumulative curve method, the relative water level fluctuation values ​​of two wetlands in the same water level fluctuation amplitude type are accumulated. If the slope of the generated fitting line does not change, it means that the water level fluctuation processes of the two wetlands are consistent. The wetlands with different hydrological fluctuation processes in the wetland types with the same water level fluctuation intensity are eliminated.

[0072] The hydrological threshold is used to represent the effective hydrological connectivity intensity. The three different types of wetlands have different effective hydrological connectivity intensities, and are finally divided into wetland types i, ii, and iii with different effective connectivity intensities.

[0073] In a specific implementation of the present invention, a set time length includes multiple set time intervals, and water level fluctuation data can be obtained once for each time interval. The sizes of the multiple water level fluctuation data are compared, and the one with the largest value is the instantaneous maximum water level fluctuation. A set time length includes at least two anniversaries, and the numerical sizes of the water level fluctuation data of each anniversary are compared. The difference between the water level fluctuation data with the largest value and the water level fluctuation data with the smallest value is the water level fluctuation value of the anniversary; the water level fluctuation values ​​of all anniversaries within the set time length are compared, and the one with the largest value is the maximum water level fluctuation of the year. For example, a set time length of two years includes multiple set time intervals. If each time interval (i.e., satellite revisit cycle) is 14 days, then the two-year set time length includes 48 set time intervals.

[0074] In the specific implementation of the present invention, it is necessary to calculate the elastic force, which is: when a system is disturbed from its original equilibrium state, it will recover to its original equilibrium state at a specific rate. The ability of the wetland system to recover from external disturbances is defined as the elastic force of wetland vegetation. The faster the recovery rate, the greater the resilience, which is measured by the lagged autocorrelation (AC1).

[0075] As an important criterion for judging the health of an ecological system, ecosystem resilience is a measure of the ability of an ecosystem to recover from external disturbances. This indicator not only indicates the current health status, but also indicates the long-term resilience of the ecosystem. Therefore, using ecosystem resilience as a standard for wetland restoration is an efficient, long-term and economical indicator. According to the cup-and-ball theory, when the disturbance is at a certain threshold, the ecosystem will reach optimal resilience, that is, it has the maximum self-regulation ability to maintain the health of the wetland. As a kind of ecological disturbance, water level fluctuation affects the vegetation species and growth status on the one hand, and the material and energy cycle on the other hand. It is an important factor affecting the resilience of the ecosystem and a hydrological regulation method that is easy to operate manually. Therefore, judging the response of ecosystem resilience to hydrological fluctuations is considered to be the key to accurately assessing the health status of the ecosystem. Therefore, the present invention takes the intensity index range of the effective hydrological connectivity of the ecosystem under different effective hydrological connectivity intensities as the starting point, and the purpose is to explore the response mechanism of ecosystem resilience.

[0076] The elastic force value is the lag autocorrelation AC1, and the specific calculation formula is: AC1 = e λΔt

[0077] Where Δt is the time step and λ is the recovery rate. The closer λ is to 0, the larger AC1 is and the corresponding elastic force is lower.

[0078] Then, the AC1 value under each fluctuation amplitude is calculated for each type of wetland, so as to draw the AC1-water level fluctuation curve, perform linear fitting on the curve, and find the inflection point (i.e., minimum value) of the curve through the threshold judgment formula. The water level fluctuation value corresponding to this value is the water level fluctuation threshold.

[0079] The threshold determination formula is:

[0080]

[0081] Where t represents water level fluctuation, y represents AC1; β0, β1, and β2 are regression coefficients; α is the turning point assumed by the stepwise regression model, which is determined based on t analysis; when β1 and β2 are found to be different, the α value is the target water level fluctuation threshold point; the three regression coefficients are estimated using least squares linear regression, and the t test is applied to test whether β2 is not equal to 0.

[0082] In order to further illustrate the calculation method of the wetland water level fluctuation threshold range provided by the present invention, the present invention specifically illustrates the implementation steps and effects of the present invention by taking the wetland water level and its connectivity identification of the Momoge wetland, a typical saline-alkali marsh wetland in northern my country, as an example.

[0083] 1. Identification of wetland patches

[0084] (1) Data acquisition: Circle the target study area (Momoge Wetland) on the Google Earth Engine website to download the GRD (Ground Range Detected) data of the Sentinel-1 series data (Sentinel-1) for the target year (2015-2021) in this area. Since Momong is located in the north and has a long freezing period, only data from June to October each year are selected in this case. A total of 76 Sentinel-1A or Sentinel-1B c-band vv polarization single-look composite (SLC) images (i.e., the required data) were obtained from 2015 to 2021.

[0085] (2) Identification of water bodies: Extract VV and VH data from each image in the data set, and use formula (1) to calculate the SDWI of each pixel in each image. Use the Ostu algorithm to determine the water body threshold ε. When SDWI>ε, it is a water body, otherwise it is a non-water body. Among them, the OTSU threshold calculation method is an existing mature threshold calculation method, and its GEE code is as follows:

[0086]

[0087]

[0088] (3) Identification of wetland patches: Figure 3 Schematic diagram of the wetland patch delineation process. a The blue area is the minimum water body range of the wetland, b The blue area is the maximum water body range, and c The colored circle is the wetland patch range. Import the water body results identified in the water body identification into ArcGIS to obtain the water body distribution map, calculate the total water body area in the target area, and find the minimum water body coverage area A within the study time. min ( Figure 3 a) and the largest water body coverage area A max ( Figure 3 b), and the wetland patch ( Figure 3 c). In this case, based on the changes in the maximum and minimum coverage areas of the water body, the permanent and periodic flooding ranges were identified, the periodic flooding range was set as the patch wetland boundary, and 14 different wetland patches were identified.

[0089] From the 14 patches identified in the area to be calculated, two representative patches are selected to specifically illustrate the wetland type classification of the present invention. The two specific patches are wetland patch 10 (Baihe Lake) and wetland patch 03 (Erlongtao River).

[0090] In the time series of the entire growing season, the phase fluctuations of Baihe Lake and Erlongtao River are inconsistent. The phase changes of Erlongtao River are more drastic than those of Baihe Lake, especially in the growing season of 2021. The magnitude and frequency of phase changes are very high. It can be judged that the water level fluctuations between the two are different, and they can be divided into different types for analysis. However, the intensity of phase changes in Baihe Lake (wetland patch 10) and its nearby wetland (wetland patch 09) are similar in the time series of the entire growing season. Based solely on the similarity of the amplitude of water level fluctuations, it cannot fully explain that Baihe Lake and its nearby wetland Erlongtao River are two wetland patches with the same water level changes. We need to further determine whether the water level fluctuation process is consistent to divide the type. Therefore, it is necessary to calculate the CV of the wetland water level fluctuation and the double cumulative curve to determine the consistency of the water level process to quantify the effective hydrological connectivity intensity of the wetland for the division of wetland type groups.

[0091] 2. Calculation of relative water level and water level fluctuation

[0092] The unwrapped phase values ​​of wetland pixels were extracted using InSAR technology, and then the improved minimum cost flow (Delaunay MCF) algorithm was used in the wetland patches. The coherence threshold was set to 0.4 (industry standard). Areas below 0.4 were not unwrapped to obtain the unwrapped phase map ( Figure 4 ), and thus obtained. Based on a certain wetland, a total of 76 Sentinel-1A or Sentinel-1B c-band vv polarization single-view composite (SLC) images from 2015 to 2021 were used to generate a total of 30 relative water level images for the entire Momong wetland from June 2016 to October 2021, with a water level variation range of -0.8 to 1.34 m. The double cumulative curve method was used to determine the accuracy of the results.

[0093] For example, in the wetland pixel of the second phase in June 2016, the extracted phase value is 0.856706, and after unwrapping, the unwrapped phase is 0.395450. Then the phase is used to obtain the relative water level change of 0.001864m.

[0094] 3. Classification of wetland groups with the same water level fluctuation intensity and quantification of effective hydrological connectivity

[0095] (1) First, determine the intensity of the fluctuation based on the CV value of the water level fluctuation

[0096] In the extraction of water level data, wetland patch 11 is affected by the hydrological situation of Taoer River. When the water level rises, it is an open water surface, and when the water level drops, it is a dry land. Therefore, the water level fluctuation data is inaccurate and will not be analyzed and discussed here. Similarly, when wetland patch 12 is Yueliangpao Reservoir, effective pixels can be extracted at the edge of wetland patch 12, but they cannot represent the water level fluctuation of the entire wetland patch. The data that can represent the water level fluctuation of the entire patch in the center of the wetland patch are all low coherence and inaccurate, so they will not be analyzed and discussed here. Therefore, we extracted and calculated the water level fluctuation characteristic values ​​and fluctuation intensity of the 12 wetland patches for classification. The wetland patches in this case are divided into three types: medium fluctuation (i), small fluctuation (ii), and severe fluctuation (iii) (Table 1).

[0097] (2) Based on the double cumulative curve method, wetlands with different water level fluctuation processes in the wetland types with the same water level fluctuation intensity in the first step were eliminated.

[0098] The relative water level fluctuation values ​​of two wetlands in the same type of water level fluctuation are accumulated. If the slope of the generated fitting line does not change, it means that the water level fluctuation time and trend of the two wetlands are consistent. On the contrary, if there is a large difference between the slopes of the generated fitting line before and after, it means that the water level fluctuation time and trend of the two wetlands are heterogeneous, and they are removed from the type. In the figure, wetland 01 is the reference wetland in this type. Therefore, wetland 04 is removed from wetland type i.

[0099] (3) Quantification of effective hydrological connectivity in wetlands

[0100] Effective connectivity is defined as the change in hydrological state (such as flooding depth, duration, frequency and time) under specific wetland geomorphic features, inducing water and water-mediated exchanges (such as sediment, energy and organisms) and accompanying biological and biochemical processes. In wetlands with the same effective hydrological connectivity type, their hydrological state changes are the same, so they have effective hydrological connectivity. The hydrological threshold is used to represent the strength of effective hydrological connectivity. Different types of wetland patches have different effective hydrological connectivity strengths.

[0101] 4. Calculation of elastic force

[0102] The elastic force values ​​of wetland patches with three effective hydrological connectivity types in the entire time series are calculated, where x is the detrended NDVI data, x * Initial state NDVI, λ required, σ given value, elastic force value is negative, the larger the negative value, the stronger the elastic force.

[0103] 5. Determination of effective hydrological connectivity intensity threshold and range

[0104] The AC1 value under each fluctuation amplitude is calculated for each type of wetland patch, so as to draw the AC1-water level fluctuation curve, perform linear fitting on the curve, and find the inflection point (i.e., the minimum value) of the curve through the mathematical method formula (7). The water level fluctuation value corresponding to this value is the water level fluctuation threshold. The water level fluctuation thresholds of type i, ii, and iii wetlands are 15.52, 16.03, and 19.36 cm, respectively, that is, their effective hydrological connectivity intensity thresholds are 15.52, 16.03, and 19.36. Based on the hydrological threshold, as the water level fluctuation amplitude increases, the elastic force of wetland vegetation also changes, from state 1 to state 2. In the process of state transition, there is a hydrological threshold point, and the elastic force on both sides of the threshold point has different trends with water level fluctuations. Linear fit the change trends on both sides to generate the fitting line y f1 y f2 At the same time, the mean elastic force of the wetland is calculated to generate the mean elastic force line. The mean elastic force line is respectively f1 y f2 Intersect at two different intersection points (A, B) and generate two vertical dashed lines. The range between the two blue dashed lines is the effective hydrological connectivity intensity range. The range from the initial red dashed line to the first blue dashed line is state 1, and the range from the second blue dashed line to the final red dashed line is state 2. The black dashed line indicates the threshold point size.

[0105] Figure 6 In order to determine the principle and process of the water level fluctuation threshold range, based on the hydrological threshold, as the water level fluctuation amplitude increases, the elastic force of wetland vegetation also changes from state 1 to state 2. In the process of state transition, there is a hydrological threshold point, and the elastic force on both sides of the threshold point has different trends with water level fluctuation. Linear fitting is performed on the change trends on both sides to generate the fitting line y f1 y f2 At the same time, the mean elastic force of the wetland is calculated to generate the mean elastic force line. The mean elastic force line is respectively f1 y f2 Intersect at two different intersection points (A, B) and generate two vertical dashed lines. The range between the two blue dashed lines is the effective hydrological connectivity intensity range. The range from the initial red dashed line to the first blue dashed line is state 1, and the range from the second blue dashed line to the final red dashed line is state 2. The black dashed line indicates the threshold point size.

[0106] Figure 7The connectivity strength range of wetland type groups with different effective hydrological connectivity strengths was determined, and the AC1 values ​​of the target type wetlands under different water level fluctuation conditions were calculated respectively. The results were linearly fitted, and the target inflection point was obtained. The effective hydrological connectivity strength threshold identification method based on the change of elastic force with the water level fluctuation amplitude was used to identify the effective hydrological connectivity strength threshold ranges of three different types of wetlands.

[0107] When judging the effective hydrological connectivity type, the present invention first judges the fluctuation intensity according to the CV value of the water level fluctuation to determine whether it is a wetland type with the same water level fluctuation intensity. Among the 12 wetland patches in the present invention, patches 01, 02, 03, 04 and 05 are of type i; patches 06, 07, 08, 09 and 10 are of type ii; patches 13 and 14 are of type iii (see Table 1 for details).

[0108] Table 1 Classification of effective hydrological connectivity types of wetland patches

[0109]

[0110] Figure 5 It is the double accumulation result of wetland patches, and the trend of wetland water level fluctuation is judged based on the double accumulation curve. The relative water level fluctuation values ​​of two wetlands of the same type are accumulated. If the slope of the generated fitting line does not change, it means that the water level fluctuation time and trend of the two wetlands are consistent. On the contrary, if there is a large difference between the slopes of the generated fitting line before and after, it means that the water level fluctuation time and trend of the two wetlands are heterogeneous, and they are removed from the type. Wetland 01 in the figure is the reference wetland of this type. Therefore, wetland 04 is removed from type i, and three wetland type groups i, ii, and iii with different effective hydrological connectivity intensity are generated.

[0111] The elasticity AC1 of wetlands with different effective hydrological connectivity types is calculated. AC1 is used to measure the elasticity of wetland ecosystems. The smaller the AC1, the greater the ecosystem elasticity, that is, the healthier the wetland. Figure 7 As shown in the figure, an AC1-water level fluctuation curve is drawn for each type of wetland, and a linear fit is performed on the curve to find the threshold point (i.e., turning point) of the curve. The water level fluctuation value corresponding to this value is the water level fluctuation threshold, which also represents the effective hydrological connectivity intensity. Taking type i and ii wetlands as examples, the effective hydrological connectivity intensities for maintaining system stability are 15.52cm and 16.03cm, respectively. At the same time, the effective hydrological connectivity intensity index range is determined by the intersection of the linear fitting line and the mean line of wetland vegetation elasticity. Taking type i and ii wetlands as examples, the water level fluctuation threshold ranges are 12.26-21.93cm and 13.63-20.17cm, respectively. The effective hydrological connectivity intensity index range indicates the range of state transitions in wetland ecosystems, and can reflect the response of wetland ecosystems to changes in hydrological conditions.

[0112] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention still falls within the scope of the technical solution of the present invention.

Claims

1. The calculation method of the wetland effective hydrological connectivity intensity index based on InSAR is characterized by: The steps include: Step 1: Select a satellite that can cover the area to be calculated, obtain image data of the area to be calculated at intervals within a set time length, and obtain a multi-time series relative water level data set of the area to be calculated based on the image data; Step 2: Identify the water body area of ​​the area to be calculated based on the multi-time series relative water level data set, and identify all wetland patches in the water body area; Step 3: Acquire the water level fluctuation data of the wetland patch according to the multi-time series relative water level data set, judge the hydrological effective connectivity strength and its consistency according to the CV value of the water level fluctuation and the double cumulative curve, and divide the wetland patch into different effective hydrological connectivity type groups; Step 4: Calculate the elastic force value of each wetland patch; Step 5: Construct water level fluctuation-ecological elasticity curves under different effective hydrological connectivity scenarios, perform segmented fitting on the water level fluctuation to generate a fitting line, calculate the mean elasticity of the wetland patches under the effective hydrological connectivity type, and generate an elasticity mean line. The two intersection points of the elasticity mean line and the fitting line are the range of the effective hydrological connectivity intensity index.

2. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: The area to be calculated may be covered by satellites at different orbital positions, and the number of the satellites may be one satellite or more than two satellites; The set time period includes at least two anniversaries; The set time interval for acquiring the image data is 10 to 40 days, and the image data is acquired once every 10 to 40 days when the satellite revisits the area to be calculated.

3. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: In step 2, the water body area is identified by determining the water body threshold ε through the Ostu algorithm. When SDWI>ε, it is a water body area, otherwise it is a non-water body area. The calculation formula of SDWI is: SDWI=ln(10×VV×VH)-8 Where SDWI is the dual-polarization water index; VV and VH represent the transmission and reception polarization modes of the radar system, VV is vertical transmission and polarization, and VH is vertical transmission and horizontal reception.

4. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: In step 2, the range of permanent flooding and the range of periodic flooding are identified by analyzing the changes in the maximum coverage area and the minimum coverage area of ​​the water body in the water body area, and the range of periodic flooding is set as the boundary of the wetland patch.

5. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: The image data includes phase changes, and the water level fluctuation data is relative water level data. The relative water level height is calculated by the phase shift data between the transmission and reception of the radar signal. The calculation formula of the relative water level height is: In the formula, Δh is the relative water level height, λ is the incident wavelength of the radar signal, and θ is the satellite incident angle. is the phase change, and d is the difference between the reflected signals at two different times.

6. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: The CV value of 0.6-0.75 is type ii for wetlands with slight water level fluctuations, the CV value of 0.75-0.9 is type i for wetlands with moderate water level fluctuations, and the CV value of 0.9-1.05 is type iii for wetlands with severe water level fluctuations.

7. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 6, characterized in that: Within a set time length, including multiple set time intervals, water level fluctuation data can be obtained once in each time interval, and the sizes of the multiple water level fluctuation data are compared. The largest value is the instantaneous maximum water level fluctuation.

8. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 7, characterized in that: Within a set time length including at least two anniversaries, the numerical values ​​of the water level fluctuation data of each anniversary are compared, and the difference between the water level fluctuation data with the largest value and the water level fluctuation data with the smallest value is the water level fluctuation value of the anniversary; The water level fluctuation values ​​of all years within the set time length are compared, and the largest value is the maximum water level fluctuation of the year.

9. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: The elastic force value is the lag autocorrelation AC1, and the specific calculation formula is: AC1 = e λΔt In the formula, Δt is the time step, λ is the recovery rate, and the closer λ is to 0, the larger AC1 is, and the corresponding elastic force is lower.

10. The method for calculating the wetland effective hydrological connectivity intensity index according to claim 1, characterized in that: The AC1 value under each fluctuation amplitude is calculated for each wetland with effective hydrological connectivity, so as to draw an AC1-water level fluctuation curve, perform linear fitting on the curve, and find the inflection point of the curve through a threshold judgment formula; The threshold determination formula is: Where t represents water level fluctuation, y represents AC1; β0, β1, and β2 are regression coefficients; α is the turning point assumed by the stepwise regression model, which is determined based on t analysis; when β1 and β2 are found to be different, the α value is the target water level fluctuation threshold point; the three regression coefficients are estimated using least squares linear regression, and the t test is applied to test whether β2 is not equal to 0.

Images