Wetland ecology monitoring method

Abstract

The invention relates to the technical field of wetland ecology, in particular to a wetland ecology monitoring method which comprises the following steps: 1) image acquisition; 2) image preprocessing: preprocessing the collected image; 3) image classification: selecting different ground feature classification samples, extracting features of spectral brightness, texture and space structure of similar ground features in different wavebands to create a training classifier, and further dividing the whole image to obtain a ground feature classification map; 4) constructing monitoring indexes including a water eutrophication NDVI, a black and odorous water recognition index BOI, an enhanced vegetation index EVI, an optimized soil adjustment vegetation index OSAVI and a normalized difference red edge index NDRE; by constructing the monitoring indexes and the ecological index formula, the wetland ecology is systematically monitored and judged, the monitoring method is stable and reliable, and a scientific judgment basis can be provided for protection and restoration of the wetland.

Description

Technical Field

[0001] The present invention relates to the technical field of wetland ecology, and in particular to a method for monitoring wetland ecology. Background Art

[0002] Jiangxi Poyang Lake Nanji International Important Wetland (hereinafter referred to as "Nanji Wetland") is one of the first members of the Yangtze River Wetland Protection Network and a member of the East Asia-Australasia Flyway Wetland Network. In 2020, it became the 64th internationally important wetland in China. Nanji Wetland is located in Jiangxi Province, the first batch of national ecological civilization pilot areas. It is located in the southern part of the main lake area of Poyang Lake, the world's rare migratory bird habitat and China's largest freshwater lake. It is the delta front formed by the confluence of the north branch, middle branch and south branch of the Ganjiang River into the open waters of Poyang Lake. Poyang Lake receives the water from the five major rivers of the Ganjiang River, Fuhe River, Xinjiang River, Raohe River and Xiuhe River, as well as the inflowing water from the Boyang River, Zhangtian River, Tongjin River, etc. After regulation and storage, it flows into the Yangtze River through Hukou. It is a water-passing, throughput and seasonal connected lake, and is an important part of the Yangtze River Basin ecosystem and one of the most important wetland groups in the Yangtze River. During the flood season, Nanji Wetland is connected with the waters of Poyang Lake, and only Nanshan Island and Jishan Island are exposed. During the dry season, in addition to Nanshan Island, Jishan Island and the roads connecting the two islands and the outside world, more than 30 dish-shaped lakes are formed in Nanji Wetland, presenting a unique scene of "becoming a lake during the flood season and a sandbank during the dry season".

[0003] The plant species in Nanji Wetland are rich, the flora composition is complex, the types are diverse, and it has an obvious transitional nature of the north-south plant convergence. Although wetland plants have the "brand" of zonality, the azonal (non-zonality) characteristics are obvious. According to the latest scientific investigation report of Nanji Wetland, there are 7 phyla, 11 classes, 27 orders, 59 families and 154 genera of phytoplankton in Nanji Wetland. There are 80 families, 205 genera and 313 species of vascular plants, including 7 families, 7 genera and 8 species of ferns, accounting for 8.75%, 3.41% and 2.56% of the total respectively; 73 families, 198 genera and 305 species of angiosperms, accounting for 91.25%, 96.59% and 97.44% of the families, genera and species of vascular plants respectively; the wildlife and habitat resources are extremely rich, and it plays a huge role in regulating the water level of the Yangtze River, conserving water sources, improving the local climate and maintaining the ecological balance of the surrounding areas. Therefore, it is of great significance to monitor or maintain the ecological environment of the wetland in real time. At present, the monitoring methods of Poyang Lake Nanji Wetland are mainly sample plot hydrology and environmental monitoring, that is, monitoring towers are set up in different sections, and ecological data such as water bodies and vegetation are regularly extracted to monitor the overall wetland, lacking a systematic monitoring and judgment method. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a wetland ecological monitoring method in view of the deficiencies of the prior art. By constructing monitoring indicators and ecological index formulas, the wetland ecology can be systematically monitored and judged. The monitoring method is stable and reliable, and can provide a scientific basis for the protection and restoration of wetlands.

[0005] To achieve the above object, the present invention adopts the following technical solutions: A wetland ecological monitoring method, comprising the following steps: 1) Image acquisition: Using a CW-15 fixed-wing unmanned aerial vehicle and a DJI M300 multi-rotor unmanned aerial vehicle equipped with an MSK-8 industrial multi-spectral camera to collect multi-spectral images of the sample plot; 2) Image preprocessing: Preprocessing the collected images; 3) Image classification: Selecting different ground object classification samples, extracting the spectral brightness, texture, and spatial structure features of the same type of ground objects in different bands to create a training classifier, and then dividing the entire image to obtain a ground object classification map; 4) Constructing monitoring indicators: including water eutrophication NDVI, black and odorous water body identification index BOI, enhanced vegetation index EVI, optimized soil-adjusted vegetation index OSAVI, and normalized difference red-edge index NDRE; 5) Determining weights: The weight of water eutrophication NDVI is ω 1 , the weight of the black and odorous water body identification index BOI is ω 2 , the weight of the enhanced vegetation index EVI is ω 3 , the weight of the optimized soil-adjusted vegetation index OSAVI is ω 4 , the weight of the normalized difference red-edge index NDRE is ω 5 , and ω 1 +ω 2 +ω 3 +ω 4 +ω 5 = 1; 6) Constructing an ecological index formula: Normalizing the mean values of each index to make them in the same order of magnitude. Let the normalized water eutrophication NDVI be χ 1 , the black and odorous water body identification index BOI be χ 2 , the enhanced vegetation index EVI be χ 3 , the optimized soil-adjusted vegetation index OSAVI be χ 4 , and the normalized difference red-edge index NDRE be χ 5 ; The ecological index formula EI = ω 3 χ 3 +ω 4 χ 4 +ω 5 χ 5 -ω1 χ 1 -ω 2 χ 2 ; When EI ≥ 0.8, the wetland ecological environment is excellent; when EI is between 0.6 - 0.8, it is good; when EI is between 0.3 - 0.6, it is medium; when EI ≤ 0.3, the ecological environment is poor and wetland ecological restoration is required.

[0006] Furthermore, 2.1 Water eutrophication NDVI: In the water environment, NDVI = (NIR - Red) / (NIR + Red), where NIR and Red are the reflectances of the near-infrared band and the red band respectively; 2.2 Black and odorous water body identification index BOI = ((ρ ｇ -ρ r ) / ρ ｇ +ρ r +ρ b ); ρ r 、ρ ｇ 、ρ ｂ represent the reflectances of the red band, the green band and the blue band in the water body respectively; 2.3 Enhanced vegetation index EVI = 2.5*(NIR – RED) / ((NIR + 6*RED – 7.5*BLUE)+1), where NIR, RED, and BLUE are the reflectances of the near-infrared band, the red band, and the blue band; 2.4 Optimized soil-adjusted vegetation index OSAVI = (NIR - Red) / (NIR + Red + 0.16), where NIR and Red are the reflectances of the near-infrared band and the red band respectively; 2.5 Normalized difference red-edge index NDRE = (NIR - RedEdge) / (NIR + RedEdge), where RedEdge is the reflectance of the red-edge band.

[0007] Furthermore, the weights are determined by the expert scoring method, and the ω 1 、ω 2 、ω 3 、ω 4 、ω 5 are 0.154, 0.056, 0.712, 0.032, and 0.046 respectively.

[0008] Furthermore, the land cover classification samples are water body, forest swamp, sedge, wild rice, reed, inland beach, cultivated land, and construction land respectively.

[0009] The beneficial effects of the present invention are: 1. The present invention discloses a wetland ecological monitoring method for monitoring the ecology of Nanjishan Wetland in Poyang Lake. Since there are rich plant species in Nanjishan Wetland, with a large proportion of forest swamps, sedges and reed maces, and there are more than 20 dish-shaped lakes in the water system, among which there are 7 dish-shaped lakes in the experimental area, namely, Zhanbei Lake, Sanniwan, Zengbei Lake, Baisha Lake, Nanshen Lake, Fengwei Lake, and Shannan Lake, and the water systems are connected; the rural roads in Nanjishan Township are connected to the outside world, and people live concentrated on Nanshan Island and Jishan Island. Historically, it has mainly engaged in fishery production, and the production and life are extremely closely related to Nanjishan Wetland. Therefore, the water body is disturbed relatively frequently; in addition, the proportion of inland tidal flats is also relatively high.

[0010] Based on the actual situation of the wetland, this application determines that the key ecological indicators of this wetland ecology include vegetation monitoring indicators, water body monitoring indicators, and land degradation assessment indicators. And because there are concentrated populations in each dish-shaped lake, which has a great impact on the water body ecology, the water body monitoring indicator is a negative impact. The specific ecological indicators are water body eutrophication NDVI, black and odorous water body identification indicator BOI, enhanced vegetation index EVI, normalized difference red edge index NDRE, and optimized soil-adjusted vegetation index OSAVI.

[0011] 2. According to the area and coverage rate of the wetland ecological classification samples, as well as the specific ecological situation of Nanjishan Wetland, the weight of each ecological indicator is determined by expert scoring, and according to the positive or negative impact of each weight on the ecology, an ecological index formula EI = ω 3 χ 3 + ω 4 χ 4 +ω 5 χ 5 -ω 1 χ 1 -ω 2 χ 2 ; Finally, through actual calculation, the ecological index of Nanjishan Wetland in Poyang Lake at a certain period is calculated. According to the ecological index, it is determined that the wetland ecological environment is good, which is consistent with the monitoring situation of the quadrats in the wetland, indicating that the monitoring method of this application is stable and effective and can be used for the ecological monitoring of Nanjishan Wetland in Poyang Lake. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a picture of a radiometric calibration panel; Figure 2 is the preprocessed image; Figure 3 is the classification result map; Figure 4 is the water body eutrophication distribution map; Figure 5 is Figure 4 the partial enlarged view in Figure 6 is the spectral curve schematic diagram of general water body and black and odorous water body; Figure 7 It is a schematic diagram of the spatial distribution of black and smelly water bodies; Figure 8 It is a distribution map of EVI of vegetation in Poyang Lake Wetland; Figure 9 It is a distribution map of OSAVI of vegetation in Poyang Lake Wetland; Figure 10 It is a distribution map of NDRE of vegetation in Poyang Lake Wetland. Specific implementation manners

[0013] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0014] The monitoring system of this application is used for the Nanjishan International Important Wetland in Poyang Lake. The monitoring time is September 2024. This wetland ecological monitoring method includes the following steps: 1) Image acquisition: Due to the special terrain of the wetland, after manual investigation, the UAV flight path and shooting parameters are designed. The vertical and horizontal CW-15 UAV is used, equipped with an MSK-8 industrial multi-spectral aerial camera, with one RGB channel and five multi-spectral monochromatic channels built-in. The data collected this time includes five single bands of green (560nm), red 1 (625nm), red 2 (670nm), red edge (700nm) and near-infrared (850nm), as well as the visible light (RGB) band. The focal length of the camera is 25mm, the image size is 6000×4000 pixels, and the sensor size is 23.5mm×15.6mm; the flight time period is from 10:00 to 15:30 at noon. The flight path is automatically generated by the vertical and horizontal UAV ground station software. The flight altitude is about 1200 meters, the forward overlap is 60%, and the side overlap is 50%. A total of 9 flight sorties are made. The radiation calibration board includes four reflectivities of black, dark gray, light gray, and white, with a size of 120*30cm. As Figure 1 shown, when collecting radiation calibration data, ensure that the target is located in the middle of the image and the target accounts for more than 50%.

[0015] Table 1 Technical specification parameters of the vertical and horizontal CW-15 fixed-wing UAV

[0016] Table 2 Technical specification parameters of the DJI M300 multi-rotor UAV

[0017] 2) Image preprocessing: Based on the ENVI remote sensing image processing platform, after defining the coordinate system for the collected images, radiometric calibration, atmospheric and geometric correction are carried out. The single-band images after correction are fused to form raster image data containing 8 bands, namely red, green, blue, red edge, and near-infrared. Then, through image mosaicking and cropping, synthesis is performed to obtain the preprocessed image. See Figure 2 ; 3) Image classification: Taking Figure 2 as a reference, classification samples are selected from the fused multi-band image data. First, six major categories are selected: "water body", "forest swamp", "marsh grassland", "inland beach", "cultivated land", and "construction land". Marsh grassland mainly includes submerged plant communities, Carex and Phalaris arundinacea communities on low-lying shoals, Polygonum communities, Zizania latifolia communities, and Phragmites australis and Miscanthus lutarioriparius communities on high-lying shoals. Submerged vegetation is submerged in the lake and overlaps with the water body area in UAV images, making it difficult to distinguish. Therefore, it is not separately classified during classification and is included in the water body. Vegetations such as Carex, Zizania latifolia, Polygonum, Miscanthus lutarioriparius, and Phragmites australis are the most important dominant vegetation types in Poyang Lake Nanjishan International Important Wetland in autumn and winter, showing irregular strip-shaped or patchy distributions with large areas. The Zizania latifolia community and the Polygonum community are significantly confused in autumn and winter. Other hygrophytic vegetations (such as Phalaris arundinacea) have smaller distribution areas in autumn and winter and are often mixed with the dominant communities, making it difficult to accurately interpret. Considering the above, in this application, the wetland marsh grassland in Poyang Lake Nanjishan International Important Wetland is divided into three types: Carex (mid- and low-lying shoal plant communities mainly composed of Carex, including Phalaris arundinacea and Artemisia selengensis), Zizania latifolia (including Zizania latifolia and Polygonum communities), and Phragmites australis and Miscanthus lutarioriparius (high-lying shoal plant communities mainly composed of Phragmites australis and Miscanthus lutarioriparius). In summary, a total of 8 types of samples, namely "water body", "forest swamp", "Carex", "Zizania latifolia", "Phragmites australis and Miscanthus lutarioriparius", "inland beach", "cultivated land", and "construction land", are selected. Based on the ENVI remote sensing image processing platform, the spectral brightness, texture, and spatial structure characteristics of the same type of ground objects in different bands are extracted to create a training classifier, and then the entire image is divided to obtain a land cover classification map. See Figure 3 , and the areas and coverage rates of the 8 classification samples are calculated. According to the classification results, the water body in Poyang Lake Nanjishan International Important Wetland accounts for about 14.63%, continuously supplying water to the wetland; vegetation (forest swamp, Carex, and Phragmites australis and Miscanthus lutarioriparius) accounts for about 55.37%, among which the forest swamp accounts for about 0.26%, Carex accounts for about 24.89%, Phragmites australis and Miscanthus lutarioriparius account for about 3.65%, and the Zizania latifolia community accounts for about 26.57%; the inland beach accounts for about 28.86%; since the data acquisition period is the dry season, the water level drops and the beach is exposed; construction land and cultivated land are interdependent and are mainly distributed in Nanjishan Township, accounting for 0.69% and 0.45% respectively; 4) Monitoring index construction: Based on the 8 sample categories of this wetland, the monitoring indexes of this wetland are divided into water body monitoring indexes, vegetation monitoring indexes and soil degradation monitoring indexes, specifically including: water body eutrophication NDVI, black and odorous water body identification index BOI; enhanced vegetation index EVI, optimized soil-adjusted vegetation index OSAVI and normalized difference red edge index NDRE; 4.1 Water body eutrophication NDVI: In the water body environment, NDVI = (NIR - Red) / (NIR + Red), where NIR and Red are the reflectivities of the near-infrared band and the red band respectively; for the area where the NDVI value is greater than -0.1, it is the cyanobacteria bloom area, and there is a water body eutrophication problem. Among them, -0.1 ≤ NDVI ≤ 0.2 represents mild eutrophication, indicating that the cyanobacteria bloom coverage within the pixel is 0 - 30%; 0.2 < NDVI ≤ 0.4 represents moderate eutrophication, indicating that the cyanobacteria bloom coverage within the pixel is 31% - 80%; NDVI > 0.4 represents severe eutrophication, indicating that the cyanobacteria bloom coverage within the pixel is 81% - 100%. The distribution map is shown in Figure 4 and 5 As shown, water body eutrophication mainly exists in lakes with green or blackish water, near construction land, and at the edges of inland beaches and water bodies. Judging from the image that the water body is green or blackish may be due to the presence of submerged vegetation in the water area and low water level, resulting in a relatively high calculated NDVI value; the main reason for eutrophication near construction land is mainly caused by human activities, such as agricultural drainage and domestic wastewater discharge, which are likely to discharge excessive nutrients such as nitrogen and phosphorus into the nearby water area.

[0018] 4.2 Black and odorous water body identification index BOI. A black and odorous water body refers to an organic pollution in which the water environment suffers beyond its self-purification ability. The aerobic decomposition of organic matter makes the oxygen consumption rate in the water body greater than the reoxygenation rate, resulting in water body hypoxia, incomplete degradation of organic matter, slow speed, and the generation of odorous substances such as hydrogen sulfide, ammonia, and mercaptan during the anaerobic biodegradation process. At the same time, black substances are formed, causing the water body to become black and odorous, which is a serious water pollution phenomenon. According to the spectral characteristics, there are obvious differences in the spectral information performance between black water bodies and normal water bodies. As Figure 6 shown, the spectral difference can be used to understand the spatial distribution of black and odorous water bodies faster and more comprehensively.

[0019] This application uses the different spectral reflectance characteristics of general water body and black and odorous water body samples to enhance the spectral difference between the two and distinguish urban black and odorous water bodies from general water bodies. Since the change of the general water body between the green band and the red band is relatively fast and the change of the black and odorous water body is not obvious, based on this difference, the black and odorous water body identification index BOI is selected, and the threshold of BOI is 0.07; BOI = ((ρ ｇ -ρ r ) / ρ ｇ +ρ r +ρb ); ρ r , ρ ｇ , ρ ｂ represent the reflectance values of the red, green, and blue light bands in the water body, respectively. When the BOI index threshold is 0.07, the black and odorous water bodies in the test area can be extracted, as shown in Figure 7 .

[0020] 4.3 Enhanced Vegetation Index: While considering the atmospheric influence and soil background, EVI can also better reflect the condition of high-density vegetation and is more suitable for monitoring high-vegetation coverage areas such as wetlands in this application. The specific calculation formula is as follows: EVI = 2.5 * (NIR – RED) / ((NIR + 6 * RED – 7.5 * BLUE) + 1), The near-infrared band can reflect the chlorophyll content and leaf area index of vegetation; the red band can reflect the chlorophyll content of vegetation and the absorption of leaves; the blue band can reflect atmospheric scattering and the reflection on the vegetation surface. EVI < 0: When the EVI value is less than 0, it indicates that the growth condition of the vegetation wetland is extremely poor. Maybe there is almost no vegetation coverage in this area, or the vegetation has suffered serious damage resulting in wetland degradation; 0 ≤ EVI < 0.2: This interval indicates that the growth condition of the wetland vegetation is not good. Maybe factors such as low soil fertility, insufficient water, and poor light conditions in this area limit the normal growth of vegetation. The vegetation is in a relatively sparse or growth-inhibited state, and both the chlorophyll content and leaf area index are at a low level, leading to a low EVI value.

[0021] 0.2 ≤ EVI < 0.5: It means that the growth condition of the wetland vegetation is average, in a normal growth but not lush state, with a medium vegetation coverage, and all growth indicators are at a conventional level. The reflection and absorption of the near-infrared, red, and blue light bands are relatively stable, and the calculated EVI value is in this interval. 0.5 ≤ EVI < 0.8: It indicates that the growth condition of the wetland vegetation is good, with a high vegetation coverage, a strong growth trend, a rich chlorophyll content, a large leaf area index, and the vegetation can fully carry out photosynthesis. The combination of each band makes the EVI value at a relatively high level. EVI ≥ 0.8: It shows that the growth condition of the wetland vegetation is extremely excellent, the vegetation grows extremely lushly, the ecosystem is relatively stable and healthy, and all indicators show that the vegetation is in an excellent growth state.

[0022] Through the above band combination, the growth state of vegetation can be evaluated more accurately, and the result is as shown in Figure 8 , and the growth condition of the wetland vegetation is excellent.

[0023] 4.4 Optimized Soil Adjusted Vegetation Index: OSAVI is an index designed to reduce the influence of soil background. It is particularly suitable for areas with low vegetation coverage and is commonly used in desertification monitoring and land degradation assessment. The specific calculation is as follows: OSAVI = (NIR - Red) / (NIR + Red + 0.16) OSAVI < 0.2: This numerical range indicates that the vegetation growth condition is poor. There may be significant soil background interference, low vegetation coverage, and vegetation growth is limited by factors such as insufficient soil fertility and water scarcity. The reflectance of the near-infrared band of the vegetation is low, and the reflectance of the red band is relatively high, resulting in a low OSAVI value. 0.2 ≤ OSAVI < 0.4: It means that the vegetation growth condition is average, in a stage of normal growth but not lush enough, and the vegetation coverage is moderate. 0.4 ≤ OSAVI < 0.6: It indicates that the vegetation growth condition is good, with high vegetation coverage, vigorous vegetation growth, and considerable leaf area index and chlorophyll content. The reflectance of the near-infrared band is high, and the reflectance of the red band is low, and the difference between the two increases, resulting in an increase in the OSAVI value. OSAVI ≥ 0.6: It shows that the vegetation growth condition is very excellent, mostly seen in areas with high vegetation coverage, extremely lush vegetation growth, a stable and healthy ecosystem, and the reflectance of the near-infrared band is significantly higher than that of the red band, with a large difference between the two, resulting in a high OSAVI value.

[0024] The calculation results are as Figure 9 shown. The OSAVI indices of the Nanjishan International Important Wetland in Poyang Lake are all above 0, and there is no land degradation situation yet.

[0025] 4.5 NDRE (Normalized Difference Red Edge Index): NDRE utilizes the characteristics of the red edge band and is mainly used to evaluate the health status and chlorophyll content of vegetation. The red edge band is very sensitive to the physiological state of vegetation, so NDRE is particularly suitable for monitoring the nutritional status and pests and diseases of crops. The specific calculation formula is as follows: NDRE = (NIR - RedEdge) / (NIR + RedEdge) NDRE < 0: It indicates that the vegetation is in a very unhealthy state, possibly suffering from severe pests and diseases, extreme drought or flood disasters, resulting in a sharp decline in the chlorophyll content of the vegetation, and even possible vegetation death or near-death. At this time, the reflectance of the near-infrared band is low, and the reflectance of the red edge band is relatively high, and the difference between the two is negative, resulting in an NDRE value less than 0.

[0026] 0 ≤ NDRE < 0.2: This indicates that the vegetation health condition is poor, and there may be problems such as nutrient deficiency, mild pests and diseases, etc. The chlorophyll content of the vegetation is lower than the normal level, the reflection ability of near-infrared light is weakened, and the reflectance in the red-edge band changes relatively little, resulting in a small difference between the two, and the NDRE value is in this range. This situation is common in crops under certain environmental stress or areas where vegetation grows poorly.

[0027] 0.2 ≤ NDRE < 0.4: This means that the vegetation health condition is average, in a state of normal growth but with some potential risks.

[0028] 0.4 ≤ NDRE < 0.6: This indicates that the vegetation health condition is good, with a normal growth trend, sufficient chlorophyll content, and can effectively carry out photosynthesis. At this time, the reflectance in the near-infrared band is significantly higher than that in the red-edge band, and the difference between the two is large, which is common in areas of vigorously growing crops or natural vegetation.

[0029] NDRE ≥ 0.6: This indicates that the vegetation health condition is excellent, and it is mostly seen in areas of extremely lush, undisturbed, and nutrient-rich vegetation. The physiological state of the vegetation is good, and its reflection ability of near-infrared light is very strong. The calculation result of the red-edge band reflection is as Figure 10 shown. The NDRE distribution in the range of Nanchang International Important Wetland in Poyang Lake is relatively concentrated, and the ecological environment is good.

[0030] 5) Determine the weights. The weight of NDVI for water eutrophication is ω 1 ; the weight of the black and odorous water body identification index BOI is ω 2 ; the weight of the enhanced vegetation index EVI is ω 3 ; the weight of the optimized soil-adjusted vegetation index OSAVI is ω 4 ; the weight of the normalized difference red-edge index NDRE is ω 5 ; and ω 1 + ω 2 + ω 3 + ω 4 + ω 5 = 1; the weights are determined by the expert scoring method, and ω 1 , ω 2 , ω 3 , ω 4 , ω 5 are 0.154, 0.056, 0.712, 0.032, 0.046 respectively; 6) Construct the ecological index formula: Normalize the mean values of each index to make them in the same magnitude. Let the normalized NDVI for water eutrophication be χ 1 , the black and odorous water body identification index BOI be χ 2 , and the enhanced vegetation index EVI be χ 3The optimized soil-adjusted vegetation index OSAVI is χ 4 The normalized difference red-edge index NDRE is χ 5 The ecological index EI formula is: EI = ω 3 χ 3 + ω 4 χ 4 + ω 5 χ 5 - ω 1 χ 1 - ω 2 χ 2 When EI ≥ 0.8, the wetland ecological environment is relatively superior; when EI is between 0.6 - 0.8, it is good; when EI is between 0.3 - 0.6, it is medium; when EI ≤ 0.3, the ecological environment is poor and wetland ecological restoration is required.

[0031] After normalizing the mean values of the above-mentioned indicators of this application (using the ENVI remote sensing image processing platform), they are 0.084, 0.152, 0.894, 0.691, and 0.402 respectively. After calculation, the monthly ecological index EI of the Poyang Lake wetland in this application = 0.712χ 3 + 0.032χ 4 + 0.046χ 5 - 0.154χ 1 - 0.056χ 2 = 0.656, and the wetland ecological environment is good. In addition, the mean values of the indicators for each of the 12 months of each year can be calculated to obtain the annual ecological index.

[0032] The applicant has arranged quadrat survey areas in wetland water bodies, green spaces, and tidal flats respectively. The area of each plot is set to 1 * 1 m². The quadrat record data includes the names of plant communities, species composition, canopy density or coverage, landform, altitude, slope, aspect, slope position, soil sampling and monitoring, and water quality monitoring. The overall situation of the quadrat survey is consistent with the above ecological index data, that is, the ecological condition of this wetland is good, indicating that the ecological monitoring method of this application has high reliability.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Any other modifications or equivalent replacements made by those of ordinary skill in the art to the technical solutions of the present invention should be covered within the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solutions of the present invention.

Claims

1. A wetland ecological monitoring method, characterized in that: The following steps are involved: 1) Image acquisition: The multispectral images of the sample sites were collected using Zongheng CW-15 fixed-wing drone and DJI M300 multi-rotor drone equipped with MSK-8 industrial-grade multispectral camera; 2) Image preprocessing: preprocess the collected images; 3) Image classification: Select different ground object classification samples, extract the spectral brightness, texture, and spatial structure characteristics of similar ground objects in different bands to create a training classifier, and then divide the entire image to obtain a ground object classification map; 4) Construct monitoring indicators: including eutrophication NDVI, black and odorous water identification index BOI, enhanced vegetation index EVI, optimized soil adjustment vegetation index OSAVI and normalized difference red edge index NDRE; 5) Determine the weights: the weight of eutrophication NDVI is ω1, the weight of black and odorous water identification index BOI is ω2, the weight of enhanced vegetation index EVI is ω3, the weight of optimized soil adjusted vegetation index OSAVI is ω4, the weight of normalized difference red edge index NDRE is ω5, and ω1+ω2+ω3+ω4+ω5=1; 6) Construct the ecological index formula: standardize the mean of each index to make it at the same level, set the standardized eutrophication NDVI of water body as χ1, the black and odorous water body identification index BOI as χ2, the enhanced vegetation index EVI as χ3, the optimized soil adjustment vegetation index OSAVI as χ4, and the normalized difference red edge index NDRE as χ5; the ecological index formula EI=ω3χ3+ω4χ4+ω5χ5-ω1χ1-ω2χ2; When EI ≥ 0.8, the wetland ecological environment is excellent; EI between 0.6-0.8 is good, 0.3-0.6 is medium, and when EI ≤ 0.3, the ecological environment is poor and wetland ecological restoration is needed.

2. A wetland ecological monitoring method according to claim 1, characterized in that: 2.1 NDVI of water eutrophication: In the water environment, NDVI = (NIR-Red) / (NIR+Red), where NIR and Red are the reflectances of the near-infrared band and the red light band respectively; 2.2 Black and odorous water identification index BOI = ((ρ ｇ -ρ r ) / ρ ｇ +ρ r +ρ b );ρ r , ｇ , ｂ Respectively represent the reflectance of red light band, green light band and blue light band in water body; 2.3 Enhanced Vegetation Index EVI = 2.5*(NIR–RED) / ((NIR+6*RED–7.5*BLUE)+1), where NIR, RED, and BLUE are the reflectances of the near-infrared band, the red band, and the blue band; 2.4 Optimized soil adjusted vegetation index OSAVI = (NIR-Red) / (NIR+Red+0.16), where NIR and Red are the reflectances of the near infrared band and the red light band, respectively; 2.5 Normalized difference red edge index NDRE = (NIR-RedEdge) / (NIR+RedEdge), RedEdge is the reflectivity of the red edge band.

3. A wetland ecological monitoring method according to claim 1, characterized in that: The weights are determined by expert scoring, and ω1, ω2, ω3, ω4, and ω5 are 0.154, 0.056, 0.712, 0.032, and 0.046, respectively.

4. A wetland ecological monitoring method according to claim 1, characterized in that: The land feature classification samples are water bodies, forest swamps, mosses, foxtail millet, reed, inland tidal flats, cultivated land, and construction land.

Images