A remote sensing monitoring system and method for rapeseed waterlogging disaster

Abstract

The present invention relates to the technical field of waterlogging disasters, and in particular to a remote sensing monitoring system and method for rapeseed waterlogging disasters. The present invention couples a new generation of earth observation satellite observation means, applies historical disaster data and observation data, identifies key sensitive parameters of crop waterlogging disasters, screens rapeseed waterlogging disaster monitoring and evaluation indicators, determines indicator parameters, and constructs a waterlogging disaster monitoring and evaluation model; fully utilizes the respective advantages of ground observation data and satellite remote sensing data, utilizes the advantages of microwave satellite remote sensing data and optical and thermal infrared satellite remote sensing data, and integrates them to form soil moisture content and precipitation data products with higher temporal and spatial resolution; solves the problems that traditional waterlogging disaster monitoring and evaluation means in the prior art are difficult to match the new changes in the current complex underlying surface conditions and difficult to meet the new requirements for monitoring and evaluation accuracy and timeliness at this stage.

Description

Technical Field

[0001] The invention relates to the technical field of waterlogging disasters, and in particular to a remote sensing monitoring system and method for rapeseed waterlogging disasters. Background Art

[0002] In recent years, the frequent occurrence of weather events such as extreme rainfall has made waterlogging gradually become one of the main abiotic stresses limiting crop growth and yield; soil waterlogging has caused significant yield losses to various crops, especially when the soil is uneven and there is too much rain, such problems will become more serious.

[0003] When the soil is waterlogged, the air in the soil pores, especially carbon dioxide, will be replaced by water, causing the carbon dioxide in the soil to decrease rapidly. At the same time, as the oxygen in the soil is exhausted, a large amount of harmful gases in the soil and a significant decrease in the pH value of the rhizosphere soil will occur. These substances and pH will be toxic to plants. When plants are in anoxic conditions, acetaldehyde and other toxic substances will be produced, which will affect the energy supply and root ion transport. At the same time, waterlogging will also affect the root system's absorption and transportation of water. The damage to the underground part's ability to absorb water and nutrients indirectly leads to changes in the aboveground part's functions. Among them, the closure of stomata will inhibit the absorption of carbon dioxide by the leaves and limit the release of carbon dioxide from the plant body to the environment, which will seriously reduce the carbon assimilation efficiency of photosynthetic tissues. The increase in oxygen content and the lack of carbon dioxide content in photosynthetic tissues may lead to enhanced photorespiration, thereby further hindering the net photosynthesis of plants. As the waterlogging continues, the growth and development of crops will be greatly affected. Long-term waterlogging stress accelerates the aging and death of crop roots, resulting in reduced or even total crop yields.

[0004] The soil moisture content and precipitation data obtained by ground meteorological stations are highly accurate, but the ground stations are limited and unevenly distributed, and the observation cost is high; satellite remote sensing data has the advantages of wide observation range, low cost, and high spatial coverage, but the accuracy of satellite remote sensing is limited; the data obtained by different satellites are also different. Microwave satellite remote sensing data has the advantages of short return period and is not affected by cloudy and rainy weather, but the spatial resolution is low; optical and thermal infrared satellite remote sensing data have the advantage of high spatial resolution, but are easily affected by cloudy and rainy weather. Therefore, a remote sensing monitoring system and method for rapeseed waterlogging disaster is urgently needed to solve such problems. Summary of the invention

[0005] To this end, the present invention provides a rapeseed waterlogging disaster remote sensing monitoring system and method to overcome the problems that traditional waterlogging disaster monitoring and assessment methods in the prior art are difficult to match the new changes in the current complex underlying surface conditions and difficult to meet the new requirements for monitoring and assessment accuracy and timeliness at this stage.

[0006] To achieve the above-mentioned object, on the one hand, the present invention provides a rapeseed waterlogging disaster remote sensing monitoring system, comprising: a data acquisition and preprocessing module for data collection and processing;

[0007] Multi-source data fusion module, used for fusion of microwave and optical data, and fusion of ground data and remote sensing data;

[0008] Waterlogging disaster assessment module, used for indicator screening and waterlogging disaster assessment;

[0009] The monitoring and early warning module issues disaster warning information based on the results of waterlogging disaster assessment.

[0010] Furthermore, the data acquisition and preprocessing module includes:

[0011] The satellite remote sensing data acquisition unit uses AMSRE and AMSR2 microwave radiometers to obtain passive microwave remote sensing data of soil moisture content, which has the advantages of short return period and is not affected by cloud and rain weather;

[0012] Use MODIS optical remote sensing data to obtain medium-resolution optical images, including land surface temperature LST, vegetation index NDVI, and reflectivity;

[0013] Ground weather station data acquisition unit,

[0014] Use national ground meteorological stations to collect high-precision meteorological data on rainfall, temperature, wind speed, and relative humidity;

[0015] At the same time, regional automatic weather observation stations are used to supplement national station data to provide more intensive spatial and temporal distribution information;

[0016] The multi-source data fusion module includes:

[0017] The microwave optical data fusion unit aligns remote sensing data of different resolutions and times in space and time, and uses a deep learning model to fuse the data to improve the spatial and temporal resolution of the data;

[0018] The ground data and remote sensing data fusion unit conducts collaborative analysis of ground meteorological data and remote sensing data to improve the overall accuracy of the data; and uses ground observation data to calibrate remote sensing data;

[0019] The waterlogging disaster assessment module determines the most sensitive monitoring indicators to waterlogging disasters through sensitivity analysis, including soil moisture content and chlorophyll content; selects the best assessment indicators and builds a waterlogging disaster monitoring system;

[0020] The reference indicators for the indicator model construction process include:

[0021] Indicators based on meteorological factors: including rainfall and wet damage day index;

[0022] Indicators based on soil oxygen stress: including soil oxygen content and soil hypoxia stress index;

[0023] The monitoring and early warning modules include:

[0024] Data real-time unit, using Internet of Things technology to transmit remote sensing data in real time;

[0025] The early warning system uses waterlogging disaster early warning algorithms based on real-time and historical data, and provides disaster warning information release functions;

[0026] The rapeseed waterlogging disaster remote sensing monitoring system realizes the full process optimization of data acquisition, preprocessing, multi-source data fusion, deep learning model construction, waterlogging disaster assessment, real-time monitoring and early warning, and decision support, effectively improving the accuracy, real-time and practicality of rapeseed waterlogging disaster monitoring.

[0027] On the other hand, the present invention also provides a method for remote sensing monitoring of rapeseed waterlogging disaster, comprising:

[0028] Step S1, water control of waterlogging in the plot, manually adjusting different irrigation water amounts, and controlling the soil moisture in the root zone of the crop planting plot;

[0029] Step S2, model data acquisition, collecting the original data required for model construction, and performing spatiotemporal interpolation and outlier removal on the original data;

[0030] Step S3, waterlogging disaster index screening, performing sensitivity analysis of the indexes at different growth stages of crops, and determining the monitoring and evaluation indexes for waterlogging disasters of rapeseed;

[0031] Step S4, determining the level of the waterlogging disaster indicator model, inverting disaster samples through historical waterlogging disaster monitoring data and experimental data, determining the initial value of drought, and based on the quantitative relationship between the yield reduction rate and the cumulative value of plant protection for waterlogging disaster monitoring and evaluation during the growing season, determining different level thresholds of rapeseed waterlogging disaster, and verifying them, completing the construction of the rapeseed waterlogging disaster monitoring and evaluation model;

[0032] Step S5, waterlogging disaster monitoring and assessment model analysis and application, based on the constructed waterlogging disaster monitoring and assessment model, calculate and analyze the spatiotemporal changes of waterlogging disasters based on meteorological, soil moisture and soil texture data.

[0033] Furthermore, in step S1, irrigation water volume is controlled, and soil moisture in the root zone of the plant planting area is at different inundation levels, so as to analyze the impact mechanism of waterlogging disasters on rapeseed growth and yield; in step S1, a comparative analysis is made on the differences in crop yield losses caused by waterlogging disasters in different growth stages of crops, so as to provide support for in-depth research on the laws of waterlogging disasters at the regional scale, indicator screening, initial value setting of model parameters, parameter correction and result verification.

[0034] Furthermore, the original data used in step S2 include: soil moisture content data inverted by satellite remote sensing, MODIS series data, SRTM DEM data, SoilGrids data, ground meteorological station data and Global Land Data Assimilation System (GLDAS) soil moisture product data.

[0035] Furthermore, the soil moisture content data inverted by satellite remote sensing uses AMSR-E and AMSR-2 passive microwave radiometer data; the spatial downscaling process of soil moisture content and precipitation inverted by satellite remote sensing must rely on the support of relevant surface feature image information provided by satellite remote sensing data with higher spatial resolution.

[0036] Furthermore, the present invention uses MODIS medium-resolution series optical remote sensing images to support spatial downscaling research based on AMSR series observation data; MODIS stands for Moderate-resolution Imaging Spectroradiometer; MODIS imagers are carried on two satellites at the same time;

[0037] MODIS series data include: MYD11 series land surface temperature LST (Land surface temperature) products; MOD / MYD13 series 16-day synthetic vegetation index products; MYD09 series visible light and near infrared, short-wave infrared reflectivity products; MCD12 series annual land use and cover type products; MOD44W annual "land-water" mask products;

[0038] Ground meteorological station data include observation data from national ground meteorological observation stations (hereinafter referred to as national stations) and observation data from regional automatic meteorological observation stations (hereinafter referred to as regional stations).

[0039] Furthermore, the data used for screening waterlogging disaster monitoring and assessment indicators in step S3 include:

[0040] Waterlogging disaster monitoring and assessment indicators based on meteorological factors, waterlogging disaster monitoring and assessment indicators based on groundwater level, waterlogging disaster monitoring and assessment indicators based on soil oxygen stress, and newly constructed waterlogging disaster monitoring indicators based on "crop-environment" multi-factor stress.

[0041] Furthermore, in step S4, the APSIM (Agricultural Production Systems Simulator) crop model and the WOFOST / CGMS (Crop Growth Monitoring System) crop model are used to calculate the soil hypoxia stress characteristics, and the calculation process includes:

[0042] Calculation of soil void moisture content:

[0043] The model calculates the soil void water content (WFPSSW) by obtaining the daily soil surface volume water content (θ), calculates the number of days of continuous waterlogging in the early stage (Dtime), and calculates the impact of hypoxia stress on crop transpiration rate;

[0044] The soil void moisture content is calculated based on the volume moisture content of the surface soil. The formula is:

[0045]

[0046] In the formula, θ is the soil water content in the soil surface layer, BD and SD are dry soil bulk density and soil density, respectively;

[0047] Calculation of the number of days of continuous waterlogging in the early stage:

[0048]

[0049] Where WFPSSW is the soil void water content, Dtime,i is the number of days of waterlogging on the i-th day, and i is calculated from the day when WFPSSW ≥ 0.65;

[0050] Anaerobic stress factor calculation:

[0051] The hypoxic soil moisture value when the soil moisture content exceeds the critical value; the reduction of transpiration rate in the WOFOST / CGMS model occurs after d consecutive days, and the transpiration rate reduction factor under anaerobic conditions; the specific calculation formula of the anaerobic stress factor is:

[0052]

[0053]

[0054] Where, WSFWOFOST is the anaerobic stress factor, WSFmax is the maximum anaerobic stress factor after continuous anaerobic conditions for d days, KWL×dul is the minimum soil moisture content during waterlogging stress; Coefi is the response weight coefficient of rapeseed at different growth stages to waterlogging disasters, ranging from 0 to 1. Waterlogging has little effect during the wintering period, the Coefi value is small, the vegetative growth period becomes larger, and the growth period is the largest, presenting an "S" curve. Therefore, the sigmoid function is used to simulate this feature, and the formula is:

[0055]

[0056] Where i is the number of days from seedling emergence or transplanting.

[0057] Furthermore, in step S4, the threshold level of waterlogging disaster indicators is divided into:

[0058] The regression equation of the most sensitive monitoring indicator of rapeseed was selected to divide the threshold level of disaster indicators, and the relative yield values ​​in the regression equation were assigned 0.95, 0.85 and 0.75 respectively. The indicator thresholds when the waterlogging disaster was under mild, moderate and severe waterlogging stress were obtained by analyzing the regression equation.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention couples a new generation of earth observation satellite observation means, applies historical disaster data and observation data, identifies key sensitive parameters of crop waterlogging disasters, screens rapeseed waterlogging disaster monitoring and evaluation indicators, determines indicator parameters, and constructs a waterlogging disaster monitoring and evaluation model.

[0061] The present invention makes full use of the respective advantages of ground observation data and satellite remote sensing data, and utilizes the advantages of microwave satellite remote sensing data and optical and thermal infrared satellite remote sensing data to fuse them to form soil moisture content and precipitation data products with higher temporal and spatial resolution.

[0062] The present invention adopts artificial intelligence methods, can adaptively learn identification features, simulate complex nonlinear relationships between parameters, explore potential associations between parameters, and fully utilize the advantages of microwave, optical, and thermal infrared satellite remote sensing data.

[0063] It solves the problems that traditional waterlogging disaster monitoring and assessment methods in existing technologies are difficult to match the new changes in current complex underlying surface conditions and are difficult to meet the new demands for monitoring and assessment accuracy and timeliness at this stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of the process of the remote sensing monitoring method for rapeseed waterlogging disaster of the present invention;

[0065] Figure 2 It is a disaster error assessment diagram of the waterlogging disaster monitoring and assessment model of the present invention. DETAILED DESCRIPTION

[0066] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0067] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0068] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0069] Example 1

[0070] See also Figure 1-Figure 2 The present invention provides a rapeseed waterlogging disaster remote sensing monitoring system, comprising:

[0071] Data acquisition and preprocessing module, for data collection and processing;

[0072] Multi-source data fusion module, used for fusion of microwave and optical data, and fusion of ground data and remote sensing data;

[0073] Waterlogging disaster assessment module, used for indicator screening and waterlogging disaster assessment;

[0074] Monitoring and early warning module, which issues disaster warning information based on waterlogging disaster assessment results;

[0075] The data acquisition and preprocessing modules include:

[0076] The satellite remote sensing data acquisition unit uses AMSRE and AMSR2 microwave radiometers to obtain passive microwave remote sensing data of soil moisture content, which has the advantages of short return period and is not affected by cloud and rain weather;

[0077] Use MODI S optical remote sensing data to obtain medium-resolution optical images, including land surface temperature LST, vegetation index NDVI, and reflectivity;

[0078] Ground weather station data acquisition unit,

[0079] Use national ground meteorological stations to collect high-precision meteorological data on rainfall, temperature, wind speed, and relative humidity;

[0080] At the same time, regional automatic weather observation stations are used to supplement national station data to provide more intensive spatial and temporal distribution information;

[0081] The multi-source data fusion module includes:

[0082] The microwave optical data fusion unit aligns remote sensing data of different resolutions and times in space and time, and uses a deep learning model to fuse the data to improve the spatial and temporal resolution of the data;

[0083] The ground data and remote sensing data fusion unit conducts collaborative analysis of ground meteorological data and remote sensing data to improve the overall accuracy of the data; and uses ground observation data to calibrate remote sensing data;

[0084] The waterlogging disaster assessment module determines the most sensitive monitoring indicators to waterlogging disasters through sensitivity analysis, including soil moisture content and chlorophyll content; selects the best assessment indicators and builds a waterlogging disaster monitoring system;

[0085] The reference indicators for the indicator model construction process include:

[0086] Indicators based on meteorological factors: including rainfall and wet damage day index;

[0087] Indicators based on soil oxygen stress: including soil oxygen content and soil hypoxia stress index;

[0088] The monitoring and early warning modules include:

[0089] Data real-time unit, using Internet of Things technology to transmit remote sensing data in real time;

[0090] The early warning system uses waterlogging disaster early warning algorithms based on real-time and historical data, and provides disaster warning information release functions;

[0091] Specifically, by establishing a real-time monitoring and early warning system, real-time monitoring of waterlogging disasters can be carried out; through cloud computing and Internet of Things technology, real-time transmission and processing of remote sensing data can be achieved, and disaster early warning information can be provided in a timely manner; real-time monitoring capabilities can help farmers and agricultural managers to quickly take response measures and reduce losses caused by disasters;

[0092] The rapeseed waterlogging disaster remote sensing monitoring system optimizes the entire process of data acquisition, preprocessing, multi-source data fusion, deep learning model construction, waterlogging disaster assessment, real-time monitoring and early warning, and decision support, effectively improving the accuracy, real-time and practicality of rapeseed waterlogging disaster monitoring;

[0093] The present invention also provides a method for remote sensing monitoring of rapeseed waterlogging disaster, comprising the following steps:

[0094] Step S1, water control of waterlogging in the plot, manually adjusting different irrigation water amounts, and controlling the soil moisture in the root zone of the crop planting plot;

[0095] In step S1, the irrigation water volume is controlled, the soil moisture in the root zone of the plant planting area is at different flooding levels, and the influence mechanism of waterlogging disaster on rapeseed growth and yield is analyzed;

[0096] Step S2, model data acquisition, collecting the original data required for model construction, and performing spatiotemporal interpolation and outlier removal on the original data;

[0097] Specifically, by taking advantage of satellite remote sensing data and ground-based meteorological data, and through multi-source data fusion technology, data utilization can be improved and the errors and uncertainties that may be caused by a single data source can be reduced. Compared with traditional monitoring methods, the multi-source data fusion method has lower costs and a wider coverage, and can conduct efficient disaster monitoring and assessment over a large area.

[0098] The original data used in step S2 include: soil moisture content data inverted by satellite remote sensing, MODIS series data, SRTM DEM data, Soil Grids data, ground meteorological station data and Global Land Data Assimilation System (GLDAS) soil moisture product data;

[0099] Soil moisture content data retrieved from satellite remote sensing, using AMSR-E and AMSR-2 passive microwave radiometer data;

[0100] The spatial reduction process of soil moisture and precipitation inverted by satellite remote sensing must rely on the relevant surface feature image information provided by satellite remote sensing data with higher spatial resolution as support;

[0101] The present invention adopts MODIS medium-resolution series optical remote sensing images to support spatial downscaling research based on AMSR series observation data; MODIS stands for Moderate-resolution Imaging Spectroradiometer; MODIS imagers are carried on two satellites at the same time;

[0102] MODI S series data include: MYD11 series land surface temperature LST (Land surface temperature) products; MOD / MYD13 series 16-day synthetic vegetation index products; MYD09 series visible light and near infrared, short-wave infrared reflectivity products; MCD12 series annual land use and cover type products; MOD44W annual "land-water" mask products;

[0103] Surface meteorological station data, including observation data from national surface meteorological observation stations (hereinafter referred to as national stations) and observation data from regional automatic meteorological observation stations (hereinafter referred to as regional stations);

[0104] Specifically, the remote sensing monitoring method for rapeseed waterlogging disasters has achieved high-precision monitoring of soil moisture and crop waterlogging conditions through the fusion of multi-source data and the application of artificial intelligence algorithms; by fusing microwave, optical and thermal infrared satellite data and combining ground meteorological station data, the temporal and spatial resolution of monitoring data has been greatly improved; the introduction of deep learning models has further improved the accuracy and reliability of data processing;

[0105] Deep learning is the main model building method in this study, and it is compared with traditional machine learning methods such as support vector machines and random forests. To achieve the goal of high-precision soil moisture content and precipitation data products with a temporal frequency of 1d and a spatial resolution higher than 1km×1km, ground meteorological observation data, agricultural meteorological observation data and satellite remote sensing data are used. Artificial intelligence methods such as LSTM, Transformer, SegNet, and U-Net are used to study the multi-source data fusion method of soil moisture content and precipitation, and to build a high-temporal and spatial resolution intelligent inversion model for waterlogging disaster factors and characterization parameters such as precipitation and soil moisture content based on multi-source information.

[0106] Waterlogging is one of the main agricultural meteorological disasters that affect wheat growth in the middle and lower reaches of the Yangtze River. At present, there are two types of characteristic quantities for analyzing the degree and impact of waterlogging on wheat at home and abroad. One is based on the soil groundwater depth index, such as the cumulative excess groundwater depth (SEW30), the cumulative excess surface water depth (SFW), the cumulative comprehensive waterlogging depth (SFEW30), the continuous suppression days of waterlogging (CSD I), etc. This type of characteristic quantity focuses on statistical analysis of the reasons for crop water deficit under waterlogging stress conditions, and cannot timely reflect the degree of crop waterlogging and its impact caused by excessive irrigation. In addition, it is difficult to invert the depth of groundwater level with remote sensing data, so it is rarely used in large-scale crop waterlogging assessment. The other type is based on meteorological elements, such as waterlogging damage day index, wet waterlogging damage day index, etc. This type of characteristic quantity only considers the impact of meteorological conditions, and rarely considers other disaster-pregnant environmental factors (such as soil, topography, and hydrological elements) that are essential in the formation of waterlogging damage, and cannot fully represent the actual degree of crop waterlogging. Waterlogging damage is mainly caused by anaerobic stress or hypoxia stress on the root system of crops, which is mainly reflected in the change of soil oxygen concentration, thereby affecting the normal growth and development of crops or the formation of higher yields. The characteristic value of the degree of crop waterlogging can not only reflect the degree of waterlogging damage to a certain extent, but also reflect the degree of influence of waterlogging damage on crop growth. It is applicable to all types of waterlogging damage and has universal applicability.

[0107] Crop soil hypoxia stress can be calculated by soil surface soil moisture content. At present, remote sensing technology, especially microwave active and passive remote sensing soil moisture monitoring technology, has moved from the ground theoretical test stage to the algorithm research and satellite verification stage to the global soil moisture business monitoring stage, providing technical support for large-scale real-time monitoring of waterlogging based on soil hypoxia stress as a characteristic quantity.

[0108] Step S3, waterlogging disaster index screening, performing sensitivity analysis of the indexes at different growth stages of crops, and determining the monitoring and evaluation indexes for waterlogging disasters of rapeseed;

[0109] The data used for screening waterlogging disaster monitoring and assessment indicators in step S3 include:

[0110] Waterlogging disaster monitoring and evaluation indicators based on meteorological factors, waterlogging disaster monitoring and evaluation indicators based on groundwater level, waterlogging disaster monitoring and evaluation indicators based on soil oxygen stress, and newly constructed waterlogging disaster monitoring indicators based on "crop-environment" multi-factor stress;

[0111] Step S4, determining the level of the waterlogging disaster indicator model, inverting disaster samples through historical waterlogging disaster monitoring data and experimental data, determining the initial value of drought, and based on the quantitative relationship between the yield reduction rate and the cumulative value of plant protection for waterlogging disaster monitoring and evaluation during the growing season, determining different level thresholds of rapeseed waterlogging disaster, and verifying them, completing the construction of the rapeseed waterlogging disaster monitoring and evaluation model;

[0112] The present invention combines the advantages of calculating soil hypoxia stress characteristics in the APSIM (Agricultural Production Systems Simulator) crop model and the WOFOST / CGMS (Crop Growth Monitoring System) crop model, and the calculation process is:

[0113] Calculation of soil void moisture content:

[0114] The model calculates the soil void water content (WFPSSW) by obtaining the daily soil surface volume water content (θ), calculates the number of days of continuous waterlogging in the early stage (Dt ime), and calculates the impact of hypoxia stress on crop transpiration rate;

[0115] The soil void moisture content is calculated based on the volume moisture content of the surface soil. The formula is:

[0116]

[0117] In the formula, θ is the soil water content in the soil surface layer, BD and SD are dry soil bulk density and soil density, respectively;

[0118] Calculation of the number of days of continuous waterlogging in the early stage:

[0119]

[0120] Where WFPSSW is the soil void water content, Dt ime,i is the number of days of waterlogging on the i-th day, and i is calculated from the day when WFPSSW ≥ 0.65;

[0121] Anaerobic stress factor calculation:

[0122] The hypoxic soil moisture value when the soil moisture content exceeds the critical value; the reduction of transpiration rate in the WOFOST / CGMS model occurs after d consecutive days, and the transpiration rate reduction factor under anaerobic conditions; the specific calculation formula of the anaerobic stress factor is:

[0123]

[0124] Where, WSFWOFOST is the anaerobic stress factor, WSFmax is the maximum anaerobic stress factor after continuous anaerobic conditions for d days, KWL×dul is the minimum soil moisture content during waterlogging stress; Coef i is the response weight coefficient of rapeseed at different growth stages to waterlogging disasters, ranging from 0 to 1. Waterlogging has little effect during the wintering period, with a small Coef i value, which increases during the vegetative growth period and is the largest during the reproductive growth period, presenting an "S"-shaped curve. Therefore, the si gmo id function is used to simulate this feature, and the formula is:

[0125]

[0126] In the formula, i is the number of days from seedling emergence or transplanting;

[0127] In step S4, the threshold level of waterlogging disaster indicators is divided into:

[0128] The regression equation of the most sensitive monitoring index of rapeseed was selected to divide the threshold level of disaster index, and the relative yield values ​​in the regression equation were assigned 0.95, 0.85 and 0.75 respectively. The index thresholds when waterlogging disaster was under mild, moderate and severe waterlogging stress were obtained by analyzing the regression equation.

[0129] Specifically, by constructing a three-dimensional dynamic monitoring and evaluation index model, the impact of rapeseed waterlogging disasters can be comprehensively evaluated; the plot waterlogging water control experiment provides multi-party data support, and deeply analyzes the changing mechanism of rapeseed growth and yield under different irrigation water volumes and soil moisture levels;

[0130] Based on these data, the most sensitive monitoring indicators for rapeseed waterlogging disasters were screened out, a waterlogging disaster assessment model was established, and the disasters were graded; the comprehensive assessment model can accurately reflect the severity of waterlogging disasters and the specific impact on crops;

[0131] Step S5, waterlogging disaster monitoring and assessment model analysis and application, based on the constructed waterlogging disaster monitoring and assessment model, calculate and analyze the spatiotemporal changes of waterlogging disasters based on meteorological, soil moisture, and soil texture data;

[0132] Specifically, through deep learning models, we can not only monitor the current waterlogging situation, but also have the ability to predict the development of future disasters; spatiotemporal sequence models such as LSTM (long short-term memory network) can predict the future dynamic changes of waterlogging and provide forward-looking information for agricultural production; combined with a comprehensive analysis platform, it integrates multi-source data and monitoring and evaluation models to provide decision-making support for agricultural management, optimize agricultural measures such as fertilizer addition and drainage management, and improve agricultural production efficiency.

[0133] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A remote sensing monitoring system for rapeseed waterlogging disaster, characterized in that: include: Data acquisition and preprocessing module, for data collection and processing; Multi-source data fusion module, used for fusion of microwave and optical data, and fusion of ground data and remote sensing data; Waterlogging disaster assessment module, used for indicator screening and waterlogging disaster assessment; Monitoring and early warning module, which issues disaster warning information based on waterlogging disaster assessment results; The data acquisition and preprocessing modules include: Satellite remote sensing data acquisition unit, using AMSRE and AMSR2 microwave radiometers to obtain passive microwave remote sensing data of soil moisture content; Use MODIS optical remote sensing data to obtain medium-resolution optical images, including land surface temperature LST, vegetation index NDVI, and reflectivity; Ground weather station data acquisition unit, Use national ground meteorological stations to collect high-precision meteorological data on rainfall, temperature, wind speed, and relative humidity; At the same time, regional automatic weather observation stations are used to supplement national station data; The multi-source data fusion module includes: The microwave optical data fusion unit aligns remote sensing data of different resolutions and times in space and time, and uses a deep learning model for data fusion; The ground data and remote sensing data fusion unit conducts collaborative analysis of ground meteorological data and remote sensing data; and calibrates remote sensing data using ground observation data; The waterlogging disaster assessment module determines the most sensitive monitoring indicators to waterlogging disasters through sensitivity analysis, including soil moisture content and chlorophyll content; selects the best assessment indicators and builds a waterlogging disaster monitoring system; The reference indicators for the indicator model construction process include: Indicators based on meteorological factors: including rainfall and wet damage day index; Indicators based on soil oxygen stress: including soil oxygen content and soil hypoxia stress index; The monitoring and early warning modules include: Data real-time unit, using Internet of Things technology to transmit remote sensing data in real time; The early warning system uses waterlogging disaster warning algorithms to issue warnings based on real-time and historical data, and provides disaster warning information release functions.

2. A remote sensing monitoring method for rapeseed waterlogging disaster, characterized in that: The rapeseed waterlogging disaster remote sensing monitoring system according to claim 1 is used, comprising the following steps: Step S1, water control of waterlogging in the plot, manually adjusting different irrigation water amounts, and controlling the soil moisture in the root zone of the crop planting plot; Step S2, model data acquisition, collecting the original data required for model construction, and performing spatiotemporal interpolation and outlier removal on the original data; Step S3, waterlogging disaster index screening, performing sensitivity analysis of the indexes at different growth stages of crops, and determining the monitoring and evaluation indexes for waterlogging disasters of rapeseed; Step S4, determining the level of the waterlogging disaster indicator model, inverting disaster samples through historical waterlogging disaster monitoring data and experimental data, determining the initial value of drought, and based on the quantitative relationship between the yield reduction rate and the cumulative value of plant protection for waterlogging disaster monitoring and evaluation during the growing season, determining different level thresholds of rapeseed waterlogging disaster, and verifying them, completing the construction of the rapeseed waterlogging disaster monitoring and evaluation model; Step S5, waterlogging disaster monitoring and assessment model analysis and application, based on the constructed waterlogging disaster monitoring and assessment model, calculate and analyze the spatiotemporal changes of waterlogging disasters based on meteorological, soil moisture, and soil texture data; In step S1, the irrigation water volume is controlled, the soil moisture in the root zone of the plant planting area is at different flooding levels, and the influence mechanism of waterlogging disaster on rapeseed growth and yield is analyzed; The original data used in step S2 include: soil moisture content data inverted by satellite remote sensing, MODIS series data, SRTM DEM data, SoilGrids data, ground meteorological station data and global land surface data assimilation system soil moisture product data; Soil moisture content data retrieved from satellite remote sensing, using AMSR-E and AMSR-2 passive microwave radiometer data; MODIS series data include: MYD11 series land surface temperature LST products; MOD / MYD13 series 16-day synthetic vegetation index products; MYD09 series visible light and near infrared, short-wave infrared reflectivity products; MCD12 series annual land use cover type products; MOD44W annual "land-water" mask products; Ground meteorological station data, including observation data from national ground meteorological observation stations and regional automatic meteorological observation stations; The data used for screening waterlogging disaster monitoring and assessment indicators in step S3 include: Waterlogging disaster monitoring and evaluation indicators based on meteorological factors, waterlogging disaster monitoring and evaluation indicators based on groundwater level, waterlogging disaster monitoring and evaluation indicators based on soil oxygen stress, and newly constructed waterlogging disaster monitoring indicators based on "crop-environment" multi-factor stress; In step S4, the APSIM (Agricultural Production Systems Simulator) crop model and the WOFOST / CGMS (Crop Growth Monitoring System) crop model are used to calculate the soil hypoxia stress characteristics. The calculation process includes: Calculation of soil void moisture content: The model calculates the soil void water content (WFPSSW) by obtaining the daily soil surface volume water content (θ), calculates the number of days of continuous waterlogging in the early stage (Dtime), and calculates the impact of hypoxia stress on crop transpiration rate; The soil void moisture content is calculated based on the volume moisture content of the surface soil. The formula is: In the formula, θ is the soil water content in the soil surface layer, BD and SD are dry soil bulk density and soil density, respectively; Calculation of the number of days of continuous waterlogging in the early stage: Where WFPSSW is the soil void water content, Dtime,i is the number of days of waterlogging on the i-th day, and i is calculated from the day when WFPSSW ≥ 0.65; Calculation of anaerobic stress factor: The hypoxic soil moisture value when the soil moisture content exceeds the critical value; the reduction of transpiration rate in the WOFOST / CGMS model occurs after d consecutive days, and the transpiration rate reduction factor under anaerobic conditions; the specific calculation formula of the anaerobic stress factor is: Where, WSFWOFOST is the anaerobic stress factor, WSFmax is the maximum anaerobic stress factor after continuous anaerobic conditions for d days, KWL×dul is the minimum soil moisture content during waterlogging stress; Coefi is the response weight coefficient of rapeseed at different growth stages to waterlogging disasters, ranging from 0 to 1. Waterlogging has little effect on rapeseed during the wintering period, with a small Coefi value, which increases during the vegetative growth period and is the largest during the reproductive growth period, presenting an "S"-shaped curve. Therefore, the sigmoid function is used to simulate this feature, and the formula is: Where i is the number of days from seedling emergence or transplanting.

3. A remote sensing monitoring method for rapeseed waterlogging disaster according to claim 2, characterized in that: In step S4, the threshold level of waterlogging disaster indicators is divided into: The regression equation of the most sensitive monitoring indicator of rapeseed was selected to divide the threshold level of disaster indicators, and the relative yield values ​​in the regression equation were assigned 0.95, 0.85 and 0.75 respectively. The indicator thresholds when the waterlogging disaster was under mild, moderate and severe waterlogging stress were obtained by analyzing the regression equation.

Images