Mangrove forest phenology and physiological trait seasonal rhythm monitoring method
Through the CCDC-HANTS algorithm combining satellite remote sensing images and ground measured data, the problem of large-scale monitoring of mangroves is solved, and the seasonal trajectory fitting and phenological stability detection of mangrove physiological traits is realized, providing mangrove health status accurate assessment and scientific protection basis.
Patent Information
- Application Number
- CN202411146996.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-12-03
AI Technical Summary
The existing technology is difficult to effectively monitor the changes in physiological characteristics of different populations of mangroves and their response mechanisms to hydrological-climatic variables on a large scale. The traditional method cannot systematically analyze the spatial distinction and phenological patterns of mangroves, limiting mangroves. Accurate assessment of health status.
Using the CCDC-HANTS dual phenology algorithm combined with satellite remote sensing images and ground measurement data, the seasonal trajectory of the physiological traits of mangroves is fitted through the remote sensing spectral index time series, seasonal characteristic parameters are extracted, and physiological traits in the phenology process of mangroves are analyzed. and phenological stability is detected to identify the growing season nodes and characteristics of mangroves.
Large-scale spatiotemporal differentiation monitoring of mangrove phenological parameters is realized, which makes up for the limitations of traditional methods, provides accurate assessment of mangrove health status and scientific protection basis, and can analyze the seasonality between different mangrove populations and physiological traits. difference.
Smart Images

Figure CN119067798A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of coastal wetland vegetation phenology monitoring, and particularly relates to a seasonal rhythm detection algorithm for mangroves and their key physiological traits, and combines spectral index time series, change monitoring algorithms and ground measured data to achieve accurate and efficient monitoring of the health status of mangroves in coastal wetlands. Background Art
[0002] Mangrove ecosystems, as one of the wetland types with the richest biodiversity and unique ecological functions on Earth, play a key role in promoting global sustainable development goals (SDGs), achieving carbon cycle processes, and mitigating global climate change. However, these ecosystems are facing severe challenges from sea level rise and extreme climate events induced by global warming. These factors have a profound impact on the spatial distribution and biological cycle rhythm of mangroves, triggering dual changes in ecological pattern and functional traits. Physiological parameters such as nitrogen, chlorophyll and water content are important indicators of the growth and health of mangrove populations. Analyzing the variation pattern of these physiological parameters has become an important way to study the mutual feedback mechanism of vegetation to environmental changes. Current research focuses on analyzing the correlation pattern between the spatial distribution of mangroves and hydrological-climatic elements, but there is still a lack of in-depth and detailed analysis and quantitative evaluation of the changes in physiological characteristics of different mangrove populations and the response mechanism of their biological phenology to hydrological-climatic variables, which restricts the high-quality realization of global sustainable development goals.
[0003] Phenological rhythm is the core content of spatiotemporal dynamic monitoring of mangroves. Traditional studies mostly rely on ground station deployment and field sampling technology to capture the spatial variation of mangroves and seasonal changes in vegetation functional traits at the regional scale. However, this method is limited by geographical location and observation scale, and it is difficult to systematically analyze the spatial differentiation and phenological patterns of mangroves at a broader landscape level. It can only provide local or point-like ecological information. Compared with traditional methods, the application of remote sensing technology in recent years has provided a new perspective for monitoring the spatial distribution and phenological rhythm of large-scale mangroves. At present, the spatiotemporal differentiation characteristics of mangrove phenological parameters are still unclear. The realization of mangrove phenology based on remote sensing technology can reduce the cost and time required for ground measurement, further explore the general laws of mangrove phenological characteristics, and analyze the stability of mangrove phenological rhythm. In addition, the precise monitoring of mangrove phenology provides a valuable reference scientific basis for evaluating the complex relationship between mangrove phenological changes and environmental factors.
[0004] The typical manifestation of mangrove phenological characteristics is the rhythm of its physiological characteristics. Physiological characteristics include leaf area index, canopy chlorophyll content and other physiological parameters, which are indicators of mangrove health. Some scholars have compared the phenological differences of mangrove populations by investigating the phenological characteristics of flowering and fruiting of different mangrove populations on the ground. Other studies have inverted single-phase or multi-phase mangrove physiological parameters based on remote sensing technology. These studies are limited to regional scales and are not conducive to extracting the phenological trajectory of these parameters, and thus cannot analyze the spatial distribution of phenological characteristics on a large scale and compare the differences in phenological characteristics of different populations. In order to make up for the shortcomings of the above methods, using different remote sensing spectral vegetation indices or indicators calculated by empirical formulas to characterize mangrove physiological parameters has become one of the main ways to monitor its phenology. For example, NDVI and gNDVI are used to indicate mangrove canopy chlorophyll, and NDWI is used to indicate vegetation canopy water content. These spectral index time series are used to fit phenological trajectories and extract phenological characteristics using methods such as CCDC, BFAST and HANTS. However, although the above studies have achieved the time-series monitoring of large-scale physiological parameters of mangroves, there are still two defects: (1) These studies mainly focus on the analysis of phenological characteristics at the mangrove and non-mangrove levels, and the heterogeneity of phenological characteristics among different mangrove populations needs to be analyzed in detail. (2) Most scholars use a single indicator to extract the phenological characteristics of mangroves. This method ignores the physiological complexity of mangroves and is not enough to fully explain the rhythm of mangrove physiological characteristics.
[0005] Therefore, improvement is urgently needed. Summary of the invention
[0006] In order to solve the above problems, the primary purpose of the present invention is to provide a method for monitoring the phenological rhythms of key physiological traits at the scale of mangrove species. This method combines satellite remote sensing images, remote sensing spectral indices and ground-measured data, and uses the CCDC-HANTS dual phenological algorithm to fit the seasonal change trajectory of mangrove physiological traits at stable sample points or sample areas, extract seasonal characteristic parameters, and analyze the spatiotemporal attributes of these parameters. At the same time, the seasonal time difference of different physiological traits in the same mangrove species with the change of mangrove phenology is explored, the different patterns of different physiological traits in the mangrove phenological process are analyzed, and the phenological differences of different mangrove species and the seasonal differences between their physiological traits are further explored.
[0007] Another purpose of the present invention is to monitor the phenological stability detection and phenological characteristic differentiation of mangroves in large-scale regions. Based on the seasonal breakpoint detection module of the CCDC algorithm, it is identified whether there are seasonal breakpoints (i.e., multiple phenological trajectories) on the time series fitting trajectory, and whether the mangroves have grown stably in the past five years. On this basis, the stable phenological trajectory of mangroves is fitted, and the phenological parameter information such as the start (SOS), peak (POS), end (EOS) and amplitude (Amplitude) of the mangrove growing season is extracted, and the spatiotemporal differentiation of these phenological parameters in a large-scale spatial range is analyzed.
[0008] To achieve the above object, the present invention provides the following technical solutions:
[0009] A method for monitoring the seasonal rhythm of mangrove phenology and physiological characteristics, the method comprising the following steps:
[0010] Step (1): remote sensing image preprocessing;
[0011] Based on the Google Earth Engine platform, Sentinel-2 and Landsat 5 / 7 / 8 remote sensing images were used, and the study area and time for monitoring were selected. These images were preprocessed, including radiation normalization, removal of cloud, snow and shadow pixels.
[0012] Step (2): Generate the original time series;
[0013] The original time series of each band of the remote sensing image is constructed based on the loop function map.
[0014] Step (3): Remove abnormal observation data and reconstruct the time series;
[0015] The tide covers the mangrove vegetation canopy, resulting in abnormal observations of water pixels and semi-submerged vegetation pixels in the time series. The improved normalized difference water index (MNDWI) is used to remove water pixels. The water adjusted vegetation index (WAVI) and the median synthesis method of the interannual spectral index are used to weaken the outliers of semi-submerged vegetation pixels in the time series of that year.
[0016] Step (4): Calculate the remote sensing spectral index indicating the physiological characteristics of mangroves;
[0017] The remote sensing spectral indices calculated to indicate the physiological characteristics of mangroves mainly include: Enhanced Vegetation Index (EVI)-leaf area index, Normalized Difference Water Index (NDMI)-canopy water content, Green Difference Vegetation Index (GDVI)-nitrogen requirement and Normalized Difference Vegetation Index (NDVI)-chlorophyll content.
[0018] Step (5): Determine phenological stability based on seasonal breakpoint detection;
[0019] The continuous change detection classification (CCDC) algorithm is used to fit the mangrove phenology trajectory in the study area. At the same time, the CCDC algorithm diagnoses whether there are breakpoints in the original time series based on the input bands, that is, whether vegetation succession occurs at the image point in the time series, and judges the stability of the mangrove phenology at the image point.
[0020] Step (6): Extraction of mangrove phenological parameters;
[0021] 25% of the amplitude in the phenological trajectory was used as the threshold to capture the start of the growing season (SOS) and the end of the growing season (EOS) of mangroves, and the time of the peak of the growing season (POS) was the time corresponding to the peak of the phenological trajectory.
[0022] Step (7): spatial characteristic analysis of phenological parameters;
[0023] Analyze the differences in the delay and advance of the start, end and peak of the mangrove growing season in different protected areas, as well as the trends of these parameters with spatial variation, such as changes from coastal to inland and from upstream to downstream of rivers. Combined with the actual growth process of mangroves, understand the spatial patterns of mangrove life cycle events (such as germination, flowering, maturity, leaf fall, etc.) and their relationship with environmental factors. Statistically analyze the frequency distribution of the start, end and peak of the mangrove growing season in different months. Compare the spatial characteristics of phenological parameters at different scales (Sentinel 2 and Landsat images), and analyze the impact of scale effect on the extraction of mangrove phenological parameters.
[0024] Step (8): Fitting the phenological trajectory of physiological traits of dominant species in mangroves based on stable measured points;
[0025] Step (9): Extraction of seasonal characteristic parameters of physiological traits of dominant species in mangroves;
[0026] Based on 25% of the amplitude of the seasonal rhythm trajectory, the rhythm growth point and decline point of the physiological trait are captured, and the rhythm peak value is the maximum value of the physiological trait content. Based on the mangrove species determined by the ground-measured sample points or sample areas, the coordinates of the sample points or sample areas in the image are determined, and the rhythm growth points, decline points and rhythm peak values of the seasonal changes of the physiological traits (canopy water content, canopy chlorophyll, nitrogen demand and leaf area index) of specific mangrove species in these areas are extracted.
[0027] Step (10): seasonal difference analysis of physiological traits of dominant species in mangroves;
[0028] Based on the seasonal rhythm of physiological traits of dominant mangrove species, the time of occurrence of key events of different physiological traits of the same tree species was explored, the time difference of the same event of physiological traits was compared, and the rhythmic pattern of physiological traits in the mangrove phenological process was analyzed. On this basis, the rhythmic characteristics of physiological traits of different mangrove species were compared, and the interspecific differences in seasonal rhythms of physiological traits of mangroves were analyzed.
[0029] Furthermore, in step (3), the normalized difference water index (MNDWI) is used to remove water pixels, and the formula is as follows:
[0030] T 1,y =F1(T 0,y , MND WI y >0)
[0031] MNDWT=(GREEN-SWIR) / (GREEN+SWIR)
[0032] Where y represents the year, T 1,y is the time series after removing the abnormal observations caused by water pixels, T 0,y is the original time series, F1 represents the operation of removing water pixels, GREEN and SWIR represent the green band and shortwave infrared band of the remote sensing image, respectively.
[0033] The water-adjusted vegetation index (WAVI) and the median synthesis of the interannual spectral index are used to weaken the outliers of the semi-submerged vegetation pixels in the annual time series. The formula is as follows:
[0034] T 2,y =F2(T 1,y ,k×median(WA VI y ) <WA VI y )
[0035] WAVI=1.5×(NIR-BLUE) / (NIR+BLUE+0.5)
[0036] Where T 2,y It is the time series after weakening the abnormal observation values caused by semi-submerged vegetation pixels. F2 represents the operation of removing some semi-submerged vegetation pixels. k represents the weakening coefficient. The larger the k value is, the higher the degree of weakening is. The possibility that the reconstructed time series includes the observation values of semi-submerged vegetation pixels is lower. After repeated experiments, the value range of k is determined to be 0.5-0.7. NIR and BLUE represent the near-infrared band and blue band of remote sensing images, respectively.
[0037] Furthermore, in step (4), the formulas of EVI, NDMI and GDVI are as follows:
[0038] EVI=2.5×(NIR-RED) / (NIR+6×RED-7.5×BLUE+1)
[0039] NDMI=(NIR-SWIR1) / (NIR+SWIR1)
[0040] GDVI=NIR-Green
[0041] NDVI=(NIR-RED) / (NIR+RED)
[0042] Where RED represents the red band of the remote sensing image.
[0043] Furthermore, in step (5), in order to accurately monitor the phenological changes of mangroves, the bands input by CCDC include RED, GREEN, BLUE, NIR, SWIR1, SWRI2, EVI, NDMI, GDVI and NDVI. The formula of the CCDC algorithm is as follows:
[0044]
[0045] Where i is the i-th band in the satellite time series image, t is the t-th day of the year, P(i, t) represents the fitting value of the i-th band on the t-th day of the year, and a and b represent the phenological fitting coefficients, respectively.
[0046] Furthermore, in step (6), 25% of the amplitude in the phenological trajectory is used as the threshold for capturing the SOS of the start of the growing season and the EOS of the end of the growing season of the mangroves. The time of the peak of the growing season POS is the time corresponding to the peak of the phenological trajectory, and the formula is as follows:
[0047] Amplitude i =max[P(i)]-min[P(i)]
[0048] T SOS =F3{[P(i)==0.25Amplitude i ] min}
[0049] T EOS =F3{[P(i)==0.25Amplitude i ] max}
[0050] T POS =F3{[P(i)==Amplitude i ]}
[0051] Where F3 represents the time T corresponding to the phenological fitting value.
[0052] In step (8), the dominant species of mangroves are determined based on the ground measured sample points, and the seasonal variation trajectory of the physiological parameters of each mangrove species is accurately fitted based on the HANTS algorithm, and the formula is as follows:
[0053] VI(T(t),n,l,k)=A(T(t),n,l,k)+B(T(t),n,l,k)+C
[0054]
[0055]
[0056] Where VI represents the fitting curve of seasonal changes of mangrove physiological parameters, A and B represent the cosine and sine harmonics of the fitting physiological parameter phenological curve, and C represents a constant; a m,j and b m,j is the fitting coefficient, T is the original time series of physiological parameters, n is the number of harmonics selected when fitting the phenological curve, l represents the first mangrove community; k m represents the frequency of the mth harmonic.
[0057] Compared with the prior art, the present invention is beneficial in that:
[0058] Based on remote sensing images, the present invention monitors the stability of mangrove phenological changes on a large spatial scale, extracts phenological parameters such as the start, end, peak and amplitude of the growing season of mangroves in the stable period, and makes up for the defects of small scope and high cost of ground field phenological observation. At the same time, obtaining the spatiotemporal differentiation of large-scale mangrove phenological parameters can provide a basis for formulating more accurate scientific protection of mangroves and promoting the sustainable development of mangroves.
[0059] At the same time, the present invention monitors the stability of mangrove phenology, further combines ground measured data to obtain the seasonal change patterns of physiological traits of different dominant mangrove species, analyzes the differences in seasonal rhythms between mangrove populations and between physiological traits, and compares the differences in physiological trait rhythms of the same mangrove dominant species in different regions, and further analyzes the health status of mangrove ecosystems, including but not limited to the early, delayed changes and disturbances in the seasonal rhythms of mangrove physiological traits caused by biological invasions, pests and diseases, climate change and human impacts. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 A flow chart of a method for monitoring the seasonal rhythm of mangrove phenology and physiological traits provided in an embodiment of the present invention.
[0061] Figure 2 This is a spatial distribution map of mangrove phenological parameters provided by an embodiment of the present invention.
[0062] Figure 3 A seasonal rhythm diagram of physiological parameters of dominant mangrove species provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The principles and features of the present invention are described below in conjunction with the drawings. The examples are only used to explain the present invention and are not used to limit the scope of the present invention.
[0064] The implementation steps of the present invention are described below respectively.
[0065] Figure 1 FIG. 1 is a flow chart of the method for monitoring seasonal rhythm of mangrove phenology and its physiological traits implemented by the present invention. As shown in the figure, the method includes the following steps:
[0066] Step (1): remote sensing image preprocessing;
[0067] Based on the Google Earth Engine platform, Sentinel-2 and Landsat 5 / 7 / 8 remote sensing images were used, and the study area and time for monitoring were selected. These images were preprocessed, including radiation normalization, removal of cloud, snow and shadow pixels.
[0068] Step (2): Generate the original time series;
[0069] The original time series of each band of the remote sensing image is constructed based on the loop function map.
[0070] Step (3): Remove abnormal observation data and reconstruct the time series;
[0071] The tide covers the mangrove vegetation canopy, causing abnormal observations in the time series containing water pixels and semi-submerged vegetation pixels. The modified normalized difference water index (MNDWI) is used to remove water pixels. The formula 1.1-1.2 is as follows:
[0072] T 1,y =F1(T 0,y , MND WI y >0) (1.1)
[0073] MNDWI=(GREEN-SWTR) / (GREEN+SWIR) (1.2)
[0074] Where y represents the year, T 1,yis the time series after removing the abnormal observations caused by water pixels, T 0,y is the original time series, F1 represents the operation of removing water pixels, GREEN and SWIR represent the green band and shortwave infrared band of the remote sensing image, respectively.
[0075] The water-adjusted vegetation index (WAVI) and the median synthesis of the interannual spectral index are used to weaken the outliers of the semi-submerged vegetation pixels in the annual time series. Formulas 1.3-1.4 are as follows:
[0076] F 2,y =F2(T 1,y ,k×median(WA VI y ) <WA VI y ) (1.3)
[0077] WAVI=1.5×(NIR-BLUE) / (NIR+BLUE+0.5) (1.4)
[0078] Where T 2,y It is the time series after weakening the abnormal observation values caused by semi-submerged vegetation pixels. F2 represents the operation of removing some semi-submerged vegetation pixels. k represents the weakening coefficient. The larger the k value is, the higher the degree of weakening is. The possibility that the reconstructed time series includes the observation values of semi-submerged vegetation pixels is lower. After repeated experiments, the value range of k is determined to be 0.5-0.7. NIR and BLUE represent the near-infrared band and blue band of remote sensing images, respectively.
[0079] Step (4): Calculate the remote sensing spectral index indicating the physiological characteristics of mangroves;
[0080] The remote sensing spectral indices used to calculate the physiological characteristics of mangroves mainly include: enhanced vegetation index (EVI)-leaf area index, normalized difference moisture index (NDMI)-canopy moisture content, green difference vegetation index (GDVI)-nitrogen demand and normalized difference vegetation index (NDVI)-chlorophyll content. The formulas 1.5-1.8 are as follows:
[0081] EVI=2.5×(NIR-RED) / (NIR+6×RED-7.5×BLUE+1) (1.5)
[0082] NDMI=(NIR-SWIR1) / (NIR+SWIR1) (1.6)
[0083] GDVI=NIR-Green (1.7)
[0084] NDVI=(NIR-RED) / (NIR+RED) (1.8)
[0085] Where RED represents the red band of the remote sensing image.
[0086] Step (5): Determine phenological stability based on seasonal breakpoint detection;
[0087] The continuous change detection classification (CCDC) algorithm is used to fit the mangrove phenology trajectory in the study area. At the same time, the CCDC algorithm diagnoses whether there are breakpoints in the original time series based on the input bands, that is, whether vegetation succession occurs at the image point in the time series, and determines the stability of the mangrove phenology at the image point. In order to accurately monitor the changes in mangrove phenology, the CCDC input bands include RED, GREEN, BLUE, NIR, SWIR1, SWRI2, EVI, NDMI, GDVI and NDVI. The formula 1.9 of the CCDC algorithm is as follows:
[0088]
[0089] Where i is the i-th band in the satellite time series image, t is the t-th day of the year, P(i, t) represents the fitting value of the i-th band on the t-th day of the year, and a and b represent the phenological fitting coefficients, respectively.
[0090] Step (6): Extraction of mangrove phenological parameters;
[0091] 25% of the amplitude in the phenological trajectory is used as the threshold for capturing the start of the growing season (SOS) and the end of the growing season (EOS) of mangroves. The time of the peak of the growing season (POS) is the time corresponding to the peak of the phenological trajectory. Formulas 1.10-1.13 are as follows:
[0092] Amplitude i =max[P(i)]-min[P(i)] (1.10)
[0093] T SOS =F3{[P(i)==0.25Amplitude i ] min} (1.11)
[0094] T EOS =F3{[P(i)==0.25Amplitude i ] max} (1.12)
[0095] T POS =F3{[P(i)==Amplitude i ]} (1.13)
[0096] Where F3 represents the time T corresponding to the phenological fitting value.
[0097] Step (7): spatial characteristic analysis of phenological parameters;
[0098] Analyze the differences in the delay and advance of the start, end and peak of the mangrove growing season in different protected areas, as well as the trends of these parameters with spatial variation, such as changes from coastal to inland and from upstream to downstream of rivers. Combined with the actual growth process of mangroves, understand the spatial patterns of mangrove life cycle events (such as germination, flowering, maturity, leaf fall, etc.) and their relationship with environmental factors. Statistically analyze the frequency distribution of the start, end and peak of the mangrove growing season in different months. Compare the spatial characteristics of phenological parameters at different scales (Sentinel 2 and Landsat images), and analyze the impact of scale effect on the extraction of mangrove phenological parameters.
[0099] Step (8): Fitting the phenological trajectory of physiological traits of dominant species in mangroves based on stable measured points;
[0100] The dominant mangrove species were determined based on ground-measured sampling points, and the seasonal variation trajectory of the physiological parameters of each mangrove species was accurately fitted based on the HANTS algorithm. The formulas 1.14-1.15 are as follows:
[0101] VI(T(t),n,l,k)=A(T(t),n,l,k)+B(T(t),n,l,k)+C (1.14)
[0102]
[0103] Where VI represents the fitting curve of seasonal changes of mangrove physiological parameters, A and B represent the cosine and sine harmonics of the fitting physiological parameter phenological curve, and C represents a constant; a m,j and b m,j is the fitting coefficient, T is the original time series of physiological parameters, n is the number of harmonics selected when fitting the phenological curve, l represents the lth mangrove community; k m represents the frequency of the mth harmonic.
[0104] Step (9): Extraction of seasonal characteristic parameters of physiological traits of dominant species in mangroves;
[0105] Based on 25% of the amplitude of the seasonal rhythm trajectory, the rhythm growth point and decline point of the physiological trait are captured, and the rhythm peak value is the maximum value of the physiological trait content. Based on the mangrove species determined by the ground-measured sample points or sample areas, the coordinates of the sample points or sample areas in the image are determined, and the rhythm growth points, decline points and rhythm peak values of the seasonal changes of the physiological traits (canopy water content, canopy chlorophyll, nitrogen demand and leaf area index) of specific mangrove species in these areas are extracted.
[0106] Step (10): seasonal difference analysis of physiological traits of dominant species in mangroves;
[0107] Based on the seasonal rhythm of physiological traits of dominant mangrove species, the time of occurrence of key events of different physiological traits of the same tree species was explored, the time difference of the same event of physiological traits was compared, and the rhythmic pattern of physiological traits in the mangrove phenological process was analyzed. On this basis, the rhythmic characteristics of physiological traits of different mangrove species were compared, and the interspecific differences in seasonal rhythms of physiological traits of mangroves were analyzed.
[0108] It can be seen that the spatial distribution of mangrove phenological parameters is as follows: Figure 2 As shown in Figure 1, SOS is the start of the growing season, EOS is the end of the growing season, and POS is the peak of the growing season. Figure 3 This is a seasonal rhythm diagram of physiological parameters of dominant mangrove species. In the figure, SOS represents the beginning of the growing season, EOS represents the end of the growing season, and POS represents the peak of the growing season. As shown in the figure, mangrove phenology can be accurately identified; and the seasonal change trajectories of the four physiological traits of different mangrove species can be obtained by combining ground-measured data, and the seasonal patterns of different physiological traits can be explored, thereby realizing accurate monitoring of the rhythm of physiological traits at the species scale.
[0109] The present invention monitors the stability of mangrove phenology based on intensive long-term images within the year, realizes large-scale monitoring of mangrove phenological events, and draws a spatiotemporal differentiation map of mangrove phenological characteristics. This method improves the accuracy of CCDC algorithm in identifying mangrove phenology by inputting multiple bands or spectral indices for identifying phenological characteristics; proposes CCDC-HANTS dual phenology algorithm, and combines it with ground measured data to obtain the seasonal change trajectories of four physiological traits of different mangrove species, explore the seasonal patterns of different physiological traits, and realize accurate monitoring of the rhythm of physiological traits at the species scale.
[0110] In summary, the advantages of the present invention are as follows:
[0111] 1. By inputting the original bands and characteristic bands sensitive to mangrove phenology, the CCDC algorithm is used to monitor the succession of vegetation populations in mangrove areas, analyze the stability of mangrove phenology, fit the vegetation phenology curve, and analyze the spatiotemporal heterogeneity of mangrove phenological characteristics.
[0112] 2. The CCDC-HANTS dual phenology algorithm is proposed to capture the seasonal rhythm characteristics of different physiological traits of dominant mangrove species, evaluate the differences in rhythm time and change rate of different physiological traits of the same dominant species, analyze the seasonal patterns of each physiological trait, and further explore the differences in phenological patterns among mangrove species, and analyze the relationship between the phenological patterns of mangroves and the seasonal patterns of their physiological traits.
[0113] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for monitoring the seasonal rhythm of mangrove phenology and physiological traits, Features The steps include: Step (1): remote sensing image preprocessing; Based on the Google Earth Engine platform, Sentinel-2 and Landsat 5 / 7 / 8 remote sensing images were used, and the study area and time for monitoring were selected to pre-process the images; Step (2): Generate the original time series: Construct the original time series of each band of remote sensing image based on the loop function map; Step (3): Remove abnormal observation data and reconstruct the time series; When reconstructing the time series, the tide covers the mangrove vegetation canopy, resulting in abnormal observations of water pixels and semi-submerged vegetation pixels in the time series. The improved normalized difference water index is used to remove water pixels; the water-adjusted vegetation index and the median synthesis method of the interannual spectral index are used to weaken the abnormal values of semi-submerged vegetation pixels in the time series of that year; Step (4): Calculate the remote sensing spectral index indicating the physiological characteristics of mangroves; The remote sensing spectral indices calculated to indicate the physiological characteristics of mangroves mainly include: Enhanced Vegetation Index EVI...Leaf Area Index, Normalized Difference Water Index NDMI...Canopy Water Content, Green Difference Vegetation Index GDVI...Nitrogen Demand and Normalized Difference Vegetation Index NDVI...Chlorophyll Content; Step (5): Determine phenological stability based on seasonal breakpoint detection; The continuous change detection classification (CCDC) algorithm was used to fit the mangrove phenological trajectory in the study area. At the same time, the CCDC algorithm diagnosed whether there were breakpoints in the original time series based on the input bands. Step (6): Extraction of mangrove phenological parameters; 25% of the amplitude in the phenological trajectory was used as the threshold for capturing the start SOS and end EOS of the growing season of mangroves, and the time of the peak of the growing season POS was the time corresponding to the peak of the phenological trajectory; Step (7): spatial characteristic analysis of phenological parameters; Analyze the differences in the delay and advance of the start, end and peak of the mangrove growing season in different protected areas, as well as the trend of these parameters with spatial variation. Combined with the actual growth process of mangroves, understand the spatial pattern of mangrove life cycle events and their relationship with environmental factors. Statistically calculate the frequency distribution of the start, end and peak of the mangrove growing season in different months, compare the spatial characteristics of phenological parameters at different scales, and analyze the impact of scale effect on the extraction of mangrove phenological parameters. Step (8): Fitting the phenological trajectory of physiological traits of dominant species in mangroves based on stable measured points; The dominant species of mangroves were determined based on ground-measured sampling points, and the seasonal variation trajectory of the physiological parameters of each mangrove species was accurately fitted based on the HANTS algorithm. The formula is as follows: VI(T(t),n,l,k)=A(T(t),n,l,k)+B(T(t),n,l,k)+C Where VI represents the fitting curve of seasonal changes of mangrove physiological parameters, A and B represent the cosine and sine harmonics of the fitting physiological parameter phenological curve, and C represents a constant; a m,j and b m,j is the fitting coefficient, T is the original time series of physiological parameters, n is the number of harmonics selected when fitting the phenological curve, l represents the first mangrove community; k m represents the frequency of the mth harmonic; Step (9): Extraction of seasonal characteristic parameters of physiological traits of dominant species in mangroves; Based on 25% of the amplitude of the seasonal rhythm trajectory, the rhythm growth point and decay point of the physiological trait are captured, and the rhythm peak value is the maximum value of the physiological trait content; based on the mangrove species determined by the ground measured sample points or sample areas, the coordinates of the sample points or sample areas in the image are determined, and the rhythm growth points, decay points and rhythm peak values of the seasonal changes of the physiological traits of specific mangrove species in these areas are extracted; Step (10): seasonal difference analysis of physiological traits of dominant species in mangroves; Based on the extracted seasonal rhythms of physiological traits of dominant mangrove species, the time of occurrence of key events of different physiological traits of the same tree species was explored, the time difference of the same event of physiological traits was compared, and the rhythmic pattern of physiological traits in the mangrove phenological process was analyzed. On this basis, the rhythmic characteristics of physiological traits of different mangrove species were compared, and the inter-specific differences in the seasonal rhythms of mangrove physiological traits were identified.
2. The method for monitoring the seasonal rhythm of mangrove phenology and physiological characteristics according to claim 1, Features In step (3), the improved normalized difference water index MNDWI is used to remove water pixels, and its formula is as follows: T 1,y =F 1 (T 0,y ,MNDWI y >0) MNDWI=(GREEN-SWIR1) / (GREEN+SWIR1) Where y represents the year, T 1,y is the time series after removing the abnormal observations caused by water pixels, T 0,y is the original time series, F 1 represents the operation of removing water pixels, GREEN and SWIR represent the green band and short-wave infrared band of remote sensing images, respectively; The water-adjusted vegetation index (WAVI) and the median synthesis of the interannual spectral index are used to weaken the outliers of semi-submerged vegetation pixels in the annual time series. The formula is as follows: T 2,y =F 2 (T 1,y ,k×median(WAVI y )<WAVI y ) WAVI=1.5×(NIR-BLUE) / (NIR+BLUE+0.5) Where T 2,y is the time series after weakening the abnormal observations caused by semi-submerged vegetation pixels, F 2 It represents the operation of removing some semi-submerged vegetation pixels, k represents the attenuation coefficient, the larger the k value is, the higher the degree of attenuation is, and the lower the possibility that the reconstructed time series includes the observation values of semi-submerged vegetation pixels is. After repeated experiments, the value range of k is determined to be 0.5-0.7, NIR and BLUE represent the near-infrared band and blue band of the remote sensing image, respectively.
3. The method for monitoring seasonal rhythm of mangrove phenology and physiological characteristics according to claim 1, Features In step (4), the formulas for calculating EVI, NDMI, GDVI, and NDVI are as follows: EVI=2.5×(NIR-RED) / (NIR+6×RED-7.5×BLUE+1) NDMI=(NIR-SWIR1) / (NIR+SWIR1) GDVI=NIR-Green NDVI=(NIR-RED) / (NIR+RED) Where RED represents the red band of the remote sensing image.
4. The method for monitoring the seasonal rhythm of mangrove phenology and physiological characteristics according to claim 3, Features In step (5), the bands input by CCDC include RED, GREEN, BLUE, NIR, SWIR1, SWRI2, EVI, NDMI, GDVI and NDVI, and the formula of the CCDC algorithm is as follows: Where i is the i-th band in the satellite time series image, t is the t-th day of the year, P(i, t) represents the fitting value of the i-th band on the t-th day of the year, and a and b represent the phenological fitting coefficients, respectively.
5. The method for monitoring the seasonal rhythm of mangrove phenology and physiological characteristics according to claim 1, Features In step (6), 25% of the amplitude in the phenological trajectory is used as the threshold for capturing the start of the growing season SOS and the end of the growing season EOS of the mangrove forest. The time of the peak of the growing season POS is the time corresponding to the peak of the phenological trajectory, and the formula is as follows: Amplitude i =max[P(i)]-min[P(i)] T SOS =F 3 {[P(i)==0.25Amplitude i ] min } T EOS =F 3 {[P(i)==0.25Amplitude i ] max } T POS =F 3 {[P(i)==Amplitude i ]} Where F 3 Indicates the time T corresponding to the phenological fitting value.
6. The method for monitoring the seasonal rhythm of mangrove phenology and physiological characteristics according to claim 1, Features In the step (7), the spatial characteristics of phenological parameters of images at different scales, including Sentinel 2 and Landsat images, are compared to analyze the impact of scale effect on the extraction of mangrove phenological parameters.
