A river section topography evolution identification method based on remote sensing images

Abstract

The application discloses a river section topography evolution identification method based on remote sensing images and relates to the technical field of river topography monitoring.The method comprises the following steps: collecting remote sensing image data and hydrological section measured elevation data of a target river section in a preset time period; identifying a river bank line based on remote sensing images and extracting river channel transverse erosion characteristic parameters; extracting river channel longitudinal section erosion and deposition characteristic parameters based on the measured elevation data; calculating water and sediment condition factors in the corresponding time period, wherein the water and sediment condition factors comprise basic water and sediment factors and composite water and sediment factors; identifying the significant relationship between the transverse and longitudinal characteristic parameters and the water and sediment condition factors by using a Pearson correlation coefficient; and evaluating the fitting degree of the water and sediment condition factors by using a determination coefficient, so as to realize river section topography evolution identification.The application can be widely applied to river evolution analysis, flood control and disaster reduction and river regulation engineering of the Yellow River and other alluvial rivers.

Description

Technical Field

[0001] This invention relates to the field of river topography monitoring technology, and in particular to a method for identifying the evolution of river cross-section topography based on remote sensing images. Background Technology

[0002] The meandering section of the lower Yellow River is characterized by its wide, shallow, and scattered course, intense erosion and sedimentation, and highly variable river morphology. The transverse and longitudinal erosion and sedimentation of the river channel directly impact flood control safety, shoreline stability, and water resource utilization. Traditional river topography monitoring relies on hydrological station cross-section measurements, UAV aerial surveys, and underwater topographic measurements. However, this approach suffers from drawbacks such as high cost, long timeframes, limited coverage, and difficulty in obtaining long-term, comprehensive evolutionary information, failing to meet the need for rapid identification of large-scale, long-sequence river channel evolution.

[0003] Remote sensing technology offers advantages such as large-scale, periodic, and non-contact monitoring, enabling rapid acquisition of information on riverbanks and morphology. Existing remote sensing-based studies on river evolution mostly focus on shoreline changes or overall river morphology analysis, lacking systematic methods for fusing remote sensing images with cross-sectional elevation data, jointly characterizing transverse and longitudinal erosion and sedimentation, and quantitatively analyzing water and sediment driving mechanisms. This makes it difficult to accurately identify the patterns and driving factors of river cross-sectional topographic evolution.

[0004] Therefore, developing a method for identifying river cross-sectional topographic evolution that integrates remote sensing and measured data, quantifies horizontal and vertical erosion characteristics, and identifies water and sediment driving factors is of great engineering value and scientific significance for river management and flood control in the lower reaches of the Yellow River. Summary of the Invention

[0005] The purpose of this invention is to provide a method for identifying the evolution of river cross-section topography based on remote sensing images, aiming to solve or improve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for identifying river cross-sectional topographic evolution based on remote sensing imagery includes: Collect remote sensing image data and measured elevation data of hydrological cross sections for the target river section within a preset time period; Riverbanks are identified based on remote sensing images, and lateral erosion characteristic parameters of the river channel are extracted. Extraction of erosion and sedimentation characteristic parameters of the longitudinal section of the river channel based on measured elevation data; Calculate the water and sediment condition factors within the corresponding time period, wherein the water and sediment condition factors include basic water and sediment factors and composite water and sediment factors; The significance of the lateral and longitudinal characteristic parameters and water and sediment condition factors is identified by using the Pearson correlation coefficient, and the goodness of fit of the water and sediment condition factors is evaluated by using the coefficient of determination, thereby realizing the identification of the evolution of river cross-section topography.

[0007] Furthermore, based on remote sensing imagery, the riverbank is identified, and the specific parameters of lateral erosion in the river channel are extracted as follows: Riverbanks were identified based on remote sensing images. The average daily transverse erosion area of ​​the river channel was extracted by comparing and analyzing two riverbank images. The transverse erosion rate of the river channel was calculated based on the transverse erosion area and the time interval.

[0008] Furthermore, the siltation characteristic parameters of the longitudinal section of the river channel include: riverbank elevation, river channel area, river width, river depth, river phase coefficient, and average elevation of the river channel bottom.

[0009] Furthermore, the water and sediment condition factors for the corresponding time period are calculated as follows: In the formula, Q For runoff, 100 million m³ 3 ; , , The first Daily flow rate, water level, and sediment content (m) 3 / s, m, kg / m 3 ; n d represents the number of days in a flood season / non-flood season. , , These represent the average water level, average flow velocity, and average sediment concentration, in meters (m), m / s, and kg / m³, respectively. 3 ; For the first The water surface area of ​​the day, m 2 By combining the cross-sectional elevation and the first It is calculated from the average water level per day; For scouring strength, m 9 kg -1 s -2 ; The average sediment inflow coefficient is expressed in kg·s / m³. 6 .

[0010] Furthermore, the Pearson correlation coefficient is: In the formula, The Pearson correlation coefficient is used. , These are two sets of sample values ​​for the variables. This represents the number of samples.

[0011] Furthermore, the saliency relation uses P The value is used to represent when P ≤0.05 indicates general correlation, defined as 1 star; when P ≤0.01 indicates a good correlation, defined as a 2-star correlation; when P A correlation of ≤0.001 indicates a significant correlation between the two, defined as a 3-star correlation.

[0012] Furthermore, the modified water index MDNWI was used to perform band calculations on the remote sensing imagery to identify the riverbank. The calculation formula is as follows: in, It is in the green light band. It is in the short infrared band.

[0013] Furthermore, the goodness-of-fit formula is: In the formula, For goodness of fit; , , These are the predicted value, the actual value, and the actual average value, respectively. The number of samples; The closer the value is to 1, the higher the accuracy of the empirical formula.

[0014] The present invention discloses the following technical effects: This invention achieves simultaneous extraction of the transverse and longitudinal erosion and deposition characteristics of river channel topography by fusing remote sensing imagery with measured elevation data from hydrological cross-sections. This solves the problems of traditional monitoring methods, which rely on ground measurements, have limited coverage, and are difficult to conduct long-term continuous observations. The improved water index MDNWI can accurately identify riverbanks and effectively reduce interference from water level and background features in boundary extraction. Through the joint calculation of basic water and sediment factors such as runoff, water level, flow velocity, and sediment concentration with composite water and sediment factors such as scour intensity and sediment inflow coefficient, it can comprehensively reflect the driving effect of water and sediment conditions on river channel topography evolution. The invention also utilizes the Pearson correlation coefficient combined with… P The significance determination of the values ​​can accurately identify the response relationship between lateral erosion, longitudinal erosion, and water and sediment factors, clarify the dominant driving factors of river channel topography evolution during flood season and non-flood season, and improve the accuracy and reliability of evolution pattern identification. The method of this invention has high monitoring efficiency, low cost, and wide coverage, and can realize the quantitative and refined identification of river cross-section topography evolution. It can provide data support and technical basis for river channel management, flood control and disaster reduction, and shoreline protection in the wandering river section of the lower Yellow River. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of lateral erosion in the riverbed. Figure 2 This is a schematic diagram illustrating the calculation of characteristic values ​​for a typical cross-section. Figure 3 A graph showing the relationship between water and sediment conditions over the years; Figure 4 A graph showing the interannual variation of the transverse and longitudinal characteristic parameters of the river channel; Figure 5 A graph showing the fitting relationship between the transverse erosion rate of the river channel and the water and sediment factors; Figure 6 A heatmap showing the correlation between longitudinal characteristic parameters of hydrological station sections and water and sediment conditions during the flood season; Figure 7 A heatmap showing the correlation between longitudinal characteristic parameters of hydrological stations and water and sediment conditions during the non-flood season; Figure 8 A graph showing the fitting relationship between the transverse erosion rate of the river channel and the water and sediment factors; Figure 9 This is a graph showing the relationship between the bottom elevation of the river channel and the sensitivity factor. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The purpose of this invention is to provide a method for identifying the evolution of river cross-section topography based on remote sensing images, aiming to solve or improve at least one of the above-mentioned technical problems.

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] This invention aims to identify the driving factors of transverse and longitudinal erosion in the Yellow River's Huagao section and to construct the relationship between transverse and longitudinal erosion variations. Remote sensing images of the Huagao section from 2000 to 2020, along with data on erosion characteristics at the Huayuankou, Jiahetan, and Gaocun sections, were collected. Pearson correlation coefficients were used to systematically analyze the relationship between transverse and longitudinal erosion characteristics and water and sediment conditions.

[0021] This invention takes the wandering river section from Huayuankou to Gaocun in the lower reaches of the Yellow River as the study area (hereinafter referred to as the "Huagao section"). It combines remote sensing images with cross-sectional elevation data from the Huayuankou, Jiahetan, and Gaocun hydrological stations to conduct a correlation analysis of water and sediment conditions on the lateral erosion rate and longitudinal erosion characteristics of the river section. It comprehensively explores the response relationship between water and sediment conditions and the lateral erosion and longitudinal cross-sectional erosion characteristics of the river channel, identifies the driving factors under the response of water and sediment conditions, and constructs formulas for the changes in the lateral and longitudinal erosion characteristics of the wandering river channel over the past 20 years.

[0022] Water and sediment conditions are the main driving force behind the evolution of river channel topography, and different water and sediment conditions have different effects on the erosion and deposition morphology of the river channel. With the operation of the Xiaolangdi Reservoir, water and sediment conditions have become more controllable, leading to a shift in the downstream river channel morphology from the original random widening and riverbed incision to a stable development of narrow and deep channels. To explore the response relationship between river channel morphology characteristics and water and sediment conditions after the operation of the Xiaolangdi Reservoir, water and sediment condition factors at various hydrological stations were first calculated using both flood and non-flood seasons.

[0023] This invention calculates the runoff at the Huayuankou hydrological station section before and after the flood season based on actual measurement times from 2000 to 2020. Q ), average sediment content ( ), average water level ( ), average flow velocity ( Basic water and sediment condition factors, such as scour intensity, are introduced, taking into account the influence of water and sediment relationships under different combinations on channel morphology evolution. ) and sediment coefficient ( The calculation formula is as follows: (1) In the formula, Q For runoff, 100 million m³ 3 ; , , The first i Daily flow rate, water level, and sediment content (m) 3 / s, m, kg / m 3 ; n d represents the number of days in a flood season / non-flood season. , , These represent the average water level, average flow velocity, and average sediment concentration, in meters (m), m / s, and kg / m³, respectively. 3 ; For the first i The water surface area of ​​the day, m 2 By combining the cross-sectional elevation and the first i It is calculated from the average water level per day; For scouring strength, m 9 kg -1 s -2 ; The average sediment inflow coefficient is expressed in kg·s / m³. 6 .

[0024] Riverbed lateral feature value extraction Using the riverbank erosion rate as a characteristic value of transverse erosion in the river channel, the riverbank was identified based on remote sensing imagery. A comparative analysis of two riverbank images was conducted to extract the average daily transverse erosion area of ​​the river channel. Figure 1 As shown, the red area represents the transverse erosion area of ​​the river channel, and the rate is calculated using the following formula.

[0025] (2) In the formula: It is the rate of riverbank erosion area, in meters. 2 / d; T d represents the number of days between intervals. S The area of ​​lateral erosion in the river channel, i.e., the area of ​​the red region, is in meters. 2 .

[0026] The remote sensing images used in this invention are sourced from the Chinese Academy of Sciences' "Geospatial Data Cloud" (www.gscloud.cn), selecting Landsat 5 TM (2000-2012) and Landsat 8 OLI (2013-2020) satellite data within the range of 2000-2020. Due to the vast amount of remote sensing image data, it is necessary to combine it with ground hydrological information for scientific and effective screening. First, all remote sensing images of the Huagao section within 2000-2020 that were not obscured by cloud cover were selected, and their corresponding imaging time and water level information were statistically analyzed. Considering the difference in resolution between Landsat 5 and Landsat 8 satellite images, to avoid errors caused by resolution, the above information was used to filter and combine two images from the same satellite. The primary screening method utilizes imaging time and water level data. Imaging time is chosen because water and sediment conditions do not change significantly in a short period. Two images taken three months apart show obvious changes in riverbank erosion, which can effectively identify changes in riverbank area. Water level is a key factor affecting shoreline extraction. Different water levels result in significant differences in shoreline extraction. To reduce the impact of water level on shoreline extraction and improve data reliability, the difference between adjacent water levels is used as a screening criterion.

[0027] The results of the remote sensing image screening are shown in Table 1. The average error of the water level before and after screening does not exceed 0.14m, minimizing the error caused by water level in the extraction of riverbank boundaries. ENVI was used to preprocess the screened original remote sensing images, employing geometric correction, radiometric correction, and atmospheric correction to eliminate the influence of atmospheric, water vapor, and radiation conditions on the remote sensing images. Then, band calculation methods were used to brighten the river water, thereby identifying the Huagao section of the riverbank for subsequent calculations. The MDNWI index calculation method is an improved water body index calculation method commonly used in cities, which can effectively eliminate the influence of urban built-up areas on water body brightening. The calculation formula is as follows: (3) In the formula: It is in the green light band; These are short infrared bands, corresponding to bands 2 and 5 of Landsat 5 TM, and bands 3 and 6 of Landsat 8 OLI, respectively.

[0028] Table 1. Remote Sensing Image Selection Parameters

[0029] Riverbed longitudinal feature value extraction The main purpose of the characteristic values ​​of the river channel cross section is to accurately reflect the changes in the morphological characteristics of the river channel, mainly including the riverbank elevation ( H ), river channel area ( S ), River width ( d Basic characteristic parameters such as river depth ( ) and river depth ( ) h=S / dRiver phase coefficient () ζ = d 0.5 / h ), average elevation of the riverbed bottom ( ) and other homogenized cross-sectional erosion characteristic parameters, among which h The average river depth represents the longitudinal erosion characteristic parameter of the cross-section. ζ The river morphology coefficient reflects the overall morphological changes of the river channel, such as its depth and width, shallowness and width. This is a characteristic index of cross-sectional siltation; an increase in the value indicates siltation, while a decrease indicates erosion.

[0030] A comparative analysis was conducted on the riverbed topographic elevation data measured twice a year at the Huayuankou Hydrological Station from 2000 to 2020, such as... Figure 2 As shown in (a), the main changes in the riverbed topography occur between 6720m and 9617m from the riverbank. This is because this is the main route of the river. While there are slight changes in other parts of the levee, these are unrelated to the river and may be related to rainfall and human factors, and are not considered here. Due to the complexity of the riverbed topography, to facilitate the investigation of the response relationship between water and sediment conditions and the longitudinal erosion and deposition characteristics of the cross-section, as shown in (a)... Figure 2 As shown in (b), based on Python, the width of the submerged area at the highest water level during the flood is first used as... d The criteria for judgment are then used, and the low-lying beach elevation is adopted as... H ,statistics H The area enclosed by the river channel elevation, i.e., the shaded area in the figure, is used as... S 断 Finally, calculate separately h , ζ , Feature parameters.

[0031] This invention primarily focuses on the Huayuankou section as the research object, with Jiahetan and Gaocun as secondary verification objects. Data fusion and analysis are performed using cross-sectional characteristics from 2000 to 2004, with each year accounting for approximately 25%. The riverbank elevations of the Huayuankou, Jiahetan, and Gaocun sections differ; therefore, calibration is performed separately to facilitate unified data analysis. Using the 2000 Huayuankou section riverbank elevation as the initial value, calibration values ​​for the Jiahetan and Gaocun sections are calculated.

[0032] (4) In the formula: H 校准 The calibrated riverbank elevations for Jiahetan and Gaocun are shown in meters. H 实际 The elevation of the riverbank is the actual elevation of the riverbed, in meters. a The calibration values ​​for the Jiahetan and Gaocun sections are 19.37m and 30.8m, respectively.

[0033] Data analysis methods With the operation of the Xiaolangdi Reservoir's water and sediment regulation system, the reservoir's sediment retention effect has become significant, altering the inflow of water and sediment. This has led to a significant change in the downstream river channel morphology compared to before, indicating a certain intrinsic relationship between river channel morphology and water and sediment conditions. To investigate the significant correlation between the transverse and longitudinal erosion characteristics of the river channel and water and sediment driving factors, this invention uses the Pearson correlation coefficient formula to conduct a water and sediment factor correlation analysis on the transverse and longitudinal erosion characteristic parameters of the river channel, identifying the main driving factors of transverse and longitudinal erosion. The Pearson correlation coefficient is used to measure the linear correlation between two sets of data X and Y, with a value between -1 and 1. The calculation formula is as follows: (5) In the formula, ρ X,Y The closer to 1, the better. X and Y The stronger the correlation, the more significant the relationship. P The value is used to represent when P ≤0.05 indicates general correlation, defined as 1 star; when P ≤0.01 indicates a good correlation, defined as a 2-star correlation; when P A correlation of ≤0.001 indicates a significant correlation between the two, defined as a 3-star correlation.

[0034] Using goodness of fit R 2 As an indicator of the fit of the formula for the erosion characteristics of a hydrological station section, it is a commonly used indicator to measure the degree of fit of a statistical model to the data. It measures the proportion of the variability of the dependent variable that can be explained by the model, as shown in the following formula.

[0035] (6) In the formula, R 2 For goodness of fit; , , These represent the predicted value, the actual value, and the true average, respectively; n is the number of samples. R 2 The closer the value is to 1, the higher the accuracy of the empirical formula.

[0036] The calculation is performed according to formula (1), and the result is as follows: Figure 3 As shown, the flood season and non-flood season Q The changes are basically consistent, showing a linear upward trend over time, with the flood season within 20 years. Q From 5.32 billion m 3 Increased to 31.05 billion m 3 Non-flood season Q From 6.35 billion m3 Increased to 12.98 billion m 3 ;flood season V Similar to the non-flood season, the two rates remained basically between 0 and 2 m / s between 2000 and 2010, but abruptly changed after 2010. Unaffected by flood season or non-flood season, it gradually increased between 2000 and 2020. It is gradually decreasing. From 6.1m in 2000 9 kg -1 s -2 It grew to 127.4 m in 2020. 9 kg -1 s -2 ,and The elevation dropped from 92.7m to 87.8m, a decrease of 4.9m over 20 years. This may be because the riverbed bottom elevation has been decreasing year by year as the scouring intensity increases, thus causing the water level to drop. and The trend was consistent between the flood season and the non-flood season, both showing a downward trend, with the flood season showing the most significant decline. , Much greater than the non-flood season, flood season From 100×10 -4 kg·s / m 6 Decrease to 20×10 -4 kg·s / m 6 Around, during the non-flood season Then from 60×10 -4 kg·s / m 6 Decrease to 5×10 -4 kg·s / m 6 about.

[0037] Interannual variation analysis of the transverse and longitudinal erosion and deposition characteristic parameters of the river channel. Figure 4 As shown in Figure (a), the transverse erosion rate of the river channel exhibits a decreasing-increasing trend, rising from 4.5 billion m³ between 2002 and 2014. 2 / d decreased to approximately 294 million m 2 / d reached its lowest point, and then showed a slow upward trend between 2014 and 2017, reaching 1.073 billion m³. 2 / d. Figure 4 (b) Figure 4 (c) Figure 4 (d) Figure 4 (e) and Figure 4 (f) shows the interannual variation of the longitudinal characteristic parameters of the river channel, taking the Huayuankou section as an example. HThe elevation decreased over time, from 94.25m to 92.25m, a decrease of 2m at a rate of 0.1m per year. This indicates that between 2000 and 2019, the longitudinal erosion capacity of the river channel exceeded its sedimentation capacity, and the river channel was primarily characterized by erosion. Similarly, h The erosion pattern shows an upward trend, rising from 1.8m in 2000 to 5.1m in 2020, an increase of 3.3m over 20 years. This indicates that as time goes on, the overall erosion pattern of the cross section is developing towards a deeper and narrower direction. S and d The changes are basically the same, indicating S Mainly affected d The impact of changes, and h It's not a big deal. During the flood season, the trend is upward, downward, and then balanced over time, while during the non-flood season, the trend is generally downward. ζ Overall, the trend is considered to be downward, with a pattern of rising first and then falling during the flood season, and a sudden drop in 2010. The overall changes are quite volatile, but during the non-flood season, the overall trend is a gradual decline. This is because the inflow of water and sediment is intense during the flood season, and the river channel morphology is prone to sudden changes, while the inflow of water and sediment is smaller during the non-flood season, and the changes in the river channel morphology are more uniform and stable.

[0038] Based on the Pearson correlation formula, the correlation between water and sediment conditions and the transverse erosion rate of the river section and the longitudinal erosion and deposition characteristics of the hydrological station section were analyzed, and key factors were identified. The remote sensing images used in the transverse erosion treatment span approximately three months, assuming that daily water and sediment conditions are relatively stable within this period, without abrupt extreme values. The correlation analysis of the transverse erosion rate of the river channel is as follows: Figure 5 As shown; however, for the hydrological station cross-section, the actual measurement time interval is six months, during which the daily water and sediment conditions change drastically. In order to fully explore the response relationship between the characteristics of channel morphology changes and incoming water and sediment, it is necessary to consider whether the channel changes are affected by a certain abrupt change in water and sediment conditions. Therefore, the maximum and minimum values ​​of water and sediment conditions are introduced, and the relevant heat map is shown. Figure 6 , Figure 7 As shown.

[0039] Depend on Figure 5 The data shows the rate of lateral erosion of the river channel and Q , , , There is a positive correlation, among which Q , The significance level is 3 stars, indicating the highest level of correlation. Q , The correlation coefficients were 0.89 and 0.84, respectively, while , The significance levels were relatively low, at 2 stars and 1 star respectively, with correlation coefficients of 0.78 and 0.69. This indicates that the transverse erosion of the river channel is mainly affected by the inflow conditions. Although it has some correlation with sediment content, it may not be the main influencing factor. Figure 6 , Figure 7 The channel characteristic parameters and water and sediment inflow conditions during the flood and non-flood seasons show a generally consistent positive and negative correlation, but significant differences exist in terms of statistical significance. During the flood season, the cross-sectional characteristic parameters are mainly related to... Q max , , H max It has a certain correlation ( P ≤0.05), where with Q max , The negative correlation indicates that changes in the characteristics of hydrological station sections during the flood season are mainly driven by water inflow conditions. Q max , With the increase of [value], all characteristic parameters of the hydrological station section decreased, only [value]. h The increase in cross-sectional area indicates a narrower and deeper profile. The characteristic parameters of the river channel morphology during the non-flood season are mainly related to... , , It has a certain correlation ( P ≤0.05), where with , A positive correlation indicates that changes in the characteristics of hydrological station sections during the non-flood season are primarily driven by water and sediment conditions. , With the increase of [value], the characteristic parameters of the hydrological station section all increased, only [value] increased. h The decrease indicates that the cross-sectional shape is wide and shallow.

[0040] based on Figure 5 The analysis results use water level ( ),flow( Q ), sediment content ( ), sediment coefficient ( Fitting analysis was performed on the transverse erosion rate of the river channel, such as... Figure 8 As shown, the lateral erosion rate of the Huagao section channel is positively correlated with both water and sediment conditions, with the highest accuracy observed in fitting the relationship with water level, exhibiting a polynomial correlation. R 2 The value is 0.85, and it shows a linear correlation with flow rate. R 2 The value is 0.79. Lateral erosion rate and sediment inflow conditions. , The correlation is low, its R2 The values ​​were only 0.65 and 0.51, and the data points were relatively scattered compared to the fitted curve, indicating that the confidence interval was large and the correlation was not significant. This may be because when the sediment content of the river channel increases, the elevation of the bottom of the river channel rises, which in turn affects the rise of the river channel water level and affects the transverse erosion of the river channel. Therefore, it is believed that the water level and flow rate are the direct factors of transverse erosion of the river channel, while the sediment content and sediment inflow coefficient are indirect influencing factors.

[0041] Depend on Figure 6 , Figure 7 The heatmap showing the correlation between the driving factors of longitudinal erosion features reveals that, among them Because it is involved in the determination of river channel characteristic parameters, it is not included in the scope of driving factor identification. During the flood season, the elevation of the river channel bottom and... Q , , V min , ξ min It has a good correlation ( P ≤0.01), and V max , S min , , It has general relevance ( P ≤0.05), so adopt Q , , ξ min A fitting analysis was performed on the average elevation of the riverbed; during the non-flood season, the elevation of the riverbed was compared with... , Significant correlation ( P ≤0.001), , ξ min It has a good correlation ( P ≤0.01), and Q min , S max , S min , ξ max It has general relevance ( P ≤0.05), so adopt , , A fitting analysis was performed on the average elevation of the riverbed bottom during the non-flood season.

[0042] The results are as follows Figure 9 As shown, during the flood season, runoff, velocity, sediment inflow coefficient, and longitudinal characteristic parameters of multiple cross sections exhibit negative linear, negative power function, and positive power function relationships, respectively.R 2 The values ​​are 0.51, 0.80, and 0.80. During the non-flood season, sediment concentration, sediment inflow coefficient, scour intensity, and longitudinal characteristic parameters of multiple cross-sections exhibit positive power function correlation, positive logarithmic correlation, and negative power function correlation, respectively. R 2 The correlation coefficients were 0.77, 0.81, and 0.80, respectively. Selecting water and sediment factors with a correlation of 0.8 or higher, the results indicate that longitudinal erosion and sedimentation changes in the river channel during the flood season are mainly related to flow velocity and the daily minimum sediment inflow coefficient. Longitudinal changes in the river channel during the non-flood season are mainly related to scour intensity and sediment inflow coefficient.

[0043] During the flood season, the runoff at Jiahetan and Gaocun deviates significantly from that at Huayuankou in terms of topographic erosion characteristics. The scattered data shows an increasing trend with increasing distance between hydrological stations, indicating that the Huagao section channel is not universally applicable to the driving force of a single water and sediment factor, and the cross-sectional erosion characteristics decrease with increasing distance. During the non-flood season, sediment concentration and sediment inflow coefficient show consistent relationships with multi-section erosion changes on the curves, but sediment concentration exhibits higher dispersion than the sediment inflow coefficient, which explains the lower fitting accuracy. This further proves that a single water and sediment factor is insufficient to drive changes in cross-sectional erosion characteristics at the Huagao section hydrological stations.

[0044] This invention starts with remote sensing images and cross-sectional elevations of multiple hydrological stations, collects transverse and longitudinal erosion and sedimentation characteristic parameters of the river channel, performs correlation analysis based on inflow and sedimentation conditions, identifies driving factors of river channel erosion and sedimentation characteristics, and fits empirical formulas.

[0045] (1) Based on the analysis of riverbed erosion rate under remote sensing images, the lateral erosion rate of the Huagao section riverbed is not significantly related to whether it is the flood season, but mainly related to and Q It has a direct and significant correlation ( P ≤0.001), R 2 The values ​​are 0.85 and 0.79, respectively. , This is an indirect response relationship.

[0046] (2) Based on the analysis of the cross-sectional characteristics of hydrological stations during the flood season and non-flood season over the past 20 years, the cross-sectional morphology during the flood season shows a deep and narrow distribution, and the evolution of the river channel is mainly longitudinal erosion. The cross-sectional morphology during the non-flood season shows a wide and shallow distribution, and the evolution of the river channel is mainly due to the collapse and siltation of the riverbank caused by lateral erosion.

[0047] (3) Based on the correlation analysis of hydrological station cross-sectional characteristics, the longitudinal erosion characteristic parameters of the cross-section have a good correlation with the inflow and sediment conditions, and the driving responses during the flood season and non-flood season are consistent, but... R 2 The values ​​vary considerably. During the flood season, the changes in cross-sectional erosion and sedimentation are mainly determined by the inflow conditions, while during the non-flood season, they are more likely to be determined by both water and sediment conditions.

[0048] (4) Based on the analysis of the relationship between cross-sectional erosion and sedimentation characteristics, the cross-sectional erosion and sedimentation characteristics are not affected by a single water and sediment factor, but rather by the comprehensive driving effect of composite water and sediment factors. The flood season is mainly related to... , ξ min It has a good correlation ( P ≤0.01), and during non-flood seasons, it is the same as , Significant correlation ( P ≤0.001), R 2 All are 0.80 or above.

[0049] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0050] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for identifying the evolution of river cross-section topography based on remote sensing imagery, characterized in that, include: Collect remote sensing image data and measured elevation data of hydrological cross sections for the target river section within a preset time period; Riverbanks are identified based on remote sensing images, and lateral erosion characteristic parameters of the river channel are extracted. Extraction of erosion and sedimentation characteristic parameters of the longitudinal section of the river channel based on measured elevation data; Calculate the water and sediment condition factors within the corresponding time period, wherein the water and sediment condition factors include basic water and sediment factors and composite water and sediment factors; The significance of the lateral and longitudinal characteristic parameters and water and sediment condition factors is identified by using the Pearson correlation coefficient, and the goodness of fit of the water and sediment condition factors is evaluated by using the coefficient of determination, thereby realizing the identification of the evolution of river cross-section topography.

2. The method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, Based on remote sensing imagery, the riverbank identification and extraction of transverse erosion characteristic parameters of the river channel are as follows: Riverbanks were identified based on remote sensing images. The average daily transverse erosion area of ​​the river channel was extracted by comparing and analyzing two riverbank images. The transverse erosion rate of the river channel was calculated based on the transverse erosion area and the time interval.

3. The method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, The siltation characteristic parameters of the longitudinal section of the river channel include: riverbank elevation, river channel area, river width, river depth, river phase coefficient, and average elevation of the river channel bottom.

4. A method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, The specific calculation of water and sediment condition factors within the corresponding time period is as follows: in, Q This refers to runoff; , , The first i Daily flow rate, water level, and sediment content; n The number of days in a flood season / non-flood season; , , These are average water level, average flow velocity, and average sediment concentration, respectively. For the first i The area of ​​water flowing through the sky; For scouring intensity; This represents the average sediment inflow coefficient.

5. The method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, The Pearson correlation coefficient is: In the formula, The Pearson correlation coefficient is used. , These are two sets of sample values ​​for the variables. This represents the number of samples.

6. The method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, The salience relation uses P The value is used to represent when P A value ≤0.05 indicates a general correlation, defined as 1 star; when P A correlation of ≤0.01 indicates a good correlation, which is defined as a 2-star correlation. when P A correlation of ≤0.001 indicates a significant correlation between the two, defined as a 3-star correlation.

7. The method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, The modified water index MDNWI was used to perform band operations on remote sensing images to identify riverbanks. The calculation formula is as follows: in, It is in the green light band. It is in the short infrared band.

8. The method for identifying river cross-sectional topographic evolution based on remote sensing imagery according to claim 1, characterized in that, The goodness-of-fit formula is: In the formula, For goodness of fit; , , These are the predicted value, the actual value, and the actual average value, respectively. The number of samples; The closer the value is to 1, the higher the accuracy of the empirical formula.

Images