A method for predicting urban flooding disasters based on satellite infrared brightness temperature and precipitation data

Abstract

This invention discloses a method for predicting urban flood disasters based on satellite infrared brightness temperature and precipitation data. The method includes: combining satellite infrared brightness temperature and precipitation data to identify relevant data of complete lifecycle MCSs (Multi-Category System Components); extracting and filtering CCS area evolution data; determining the optimal value of the sensitivity coefficient; fitting the relationship function between CCS area and time, determining the most vigorous development time as the center point, determining the slope critical threshold based on the sensitivity coefficient, and finding boundary points to divide the development, maturity, and dissipation stages; finally, integrating urban underlying surface characteristics and a waterlogging model to quantitatively predict the level of urban flood disasters and push emergency response measures, optimizing the prediction results and response plans through a dynamic feedback mechanism. This invention can accurately capture the essential characteristics of each stage of MCSs, and based on the capture results, achieve full-process automation from meteorological monitoring to flood prediction and emergency response.

Description

Technical Field

[0001] This invention relates to the field of urban flood disaster prediction technology, specifically to an urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data, which is particularly suitable for the automated identification of the life stages of warm-season mesoscale convective systems (MCSs) in the Yangtze-Huaihe River region. Background Technology

[0002] The Meiyu season in the Jianghuai region mainly refers to a continuous and widespread precipitation process that occurs annually from May to July, active in the middle and lower reaches of the Yangtze River and the Huai River basin (110°E–122°E, 28°N–34°N). It is a climatic phenomenon produced by the combined influence of the summer monsoon, the subtropical high, and the Mongolian high. From May to July each year, the warm and humid southwest monsoon and cold and warm fronts repeatedly linger in this region, forming a continuous precipitation belt lasting for several weeks, accompanied by embedded mesoscale convective systems (MCSs). The precipitation brought by MCSs is characterized by high rainfall intensity, drastic short-term variations, and the "train effect" (repeated generation of convective cells along the same path). It can superimpose extreme 1–3 hourly rainfall onto a large-scale stable precipitation background, easily forming localized torrential rain centers in cities, inducing severe urban flooding disasters. This poses a significant threat to urban drainage systems, transportation, municipal infrastructure operation, the safety of residents' lives and property, agricultural production, and water resource management, posing a severe challenge to urban flood control.

[0003] As the core cause of Meiyu frontal storms, the development of macroscopic convective systems (MCSs) typically goes through a developmental stage of aggregation of discrete or linear convective cells, a mature stage of synergistic coexistence of convection and stratiform precipitation, and a dissipation stage of gradual weakening and structural disintegration of cloud clusters and precipitation. Throughout its entire life cycle, MCSs directly determine the intensity, duration, and extent of local precipitation through the superposition of strong convective precipitation and widespread stratiform precipitation. Therefore, accurately identifying the life stage of MCSs and clarifying the precipitation evolution patterns at different stages are crucial prerequisites for achieving accurate prediction of urban flood disasters and optimizing flood control emergency response strategies. This is of great significance for improving the precision of heavy precipitation forecasts and reducing flood damage.

[0004] Existing research has combined satellite infrared brightness temperature data and precipitation data to identify mesoscale convective systems (MCSs). For example, Chinese patent CN120071106B proposes a rapid and efficient method and system for identifying mesoscale convective systems in mid-to-low latitude regions, and Chinese patent CN118519212A proposes a method and device for identifying large-scale convective systems. However, these technical solutions have significant limitations and are difficult to meet the actual needs of urban flood disaster prediction: these studies are mostly based on feature matching of single-frame static data, which can only determine the existence of MCSs and does not fully explore the spatiotemporal dynamic information required for their life stage evolution. They cannot capture the continuous evolution characteristics of cloud area expansion / contraction, intensity enhancement / attenuation, and structural coalescence / disintegration, and are difficult to reflect the dynamic impact of MCSs on urban flooding. At the same time, their core focus is on the overall identification of MCSs, without addressing the occurrence and development of MCSs. The design of differentiated judgment indicators for each stage of the life cycle (maturity-dissipation) makes it difficult to distinguish the essential differences in characteristics between different stages. Furthermore, it fails to consider special scenarios such as the train effect and frequent coalescence of convective cells in the Jianghuai Meiyu region, making automated life cycle segmentation impossible. Manual data analysis is still required, which is not only inefficient but also prone to subjective biases in stage segmentation. It also fails to connect with the core needs of urban flood disaster prediction, and does not integrate key factors such as urban underlying surface characteristics and drainage system capacity. The identification results cannot be directly converted into quantitative indicators of flooding, such as water depth and inundation range, resulting in poor integration with urban flood control emergency response and engineering scheduling decisions, making it difficult to support the formulation of precise disaster prevention measures. Finally, the lack of a dynamic feedback mechanism prevents adjustments to prediction results and response strategies based on real-time evolution of MCSs (Multi-Category System). When facing short-term, drastic changes in heavy rainfall, it is prone to delayed warnings or inappropriate response measures, affecting the effectiveness of urban flood disaster response.

[0005] In summary, existing technologies struggle to automate and accurately segment the lifecycle of mesoscale meteorological systems (MCSs), and the identification results are disconnected from the needs of urban flood disaster prediction and flood control, failing to meet the requirements of precision, timeliness, and practicality in urban disaster prevention and mitigation. Therefore, developing a method capable of accurately identifying the lifecycle of MCSs and directly serving the quantitative prediction and emergency response of urban flood disasters has become an urgent need in the fields of mesoscale meteorology applications and urban disaster prevention engineering. Summary of the Invention

[0006] To address the problems of existing technologies, such as reliance on manual methods for defining the life stages of cloud clusters (MCSs), disconnect from urban flood forecasting needs, and insufficient prediction accuracy and practical adaptability, this invention provides an urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data. By combining the continuous evolution patterns of CCS area (cloud spatial expansion / contraction characteristics) and minimum brightness temperature (cloud intensity enhancement / attenuation characteristics), a differentiated curve slope critical threshold is constructed to replace traditional manual analysis. This achieves automated and precise segmentation of the entire life stage of MCSs—development, maturity, and dissipation. Simultaneously, the stage segmentation results are deeply coupled with urban underlying surface characteristics and waterlogging models to quantitatively output core flood-causing indicators such as water depth and inundation range. Compared to existing technologies that can only determine the existence of MCSs, this method not only accurately captures the essential differences in characteristics of each stage with high stage identification accuracy, effectively avoiding misjudgments caused by single thresholds or empirical values, but also directly connects with urban flood control emergency response needs. It provides a scientific quantitative basis for graded early warning and dynamic scheduling of flood disasters, achieving a closed-loop support from meteorological monitoring to disaster prevention practice.

[0007] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for predicting urban flooding disasters based on satellite infrared brightness temperature and precipitation data, the method comprising the following steps:

[0009] S1. Based on the precipitation climate characteristics and urban flood disaster prediction needs of the study area, determine the CCS (cold cloud cover area) area threshold, duration threshold, and PF (maximum precipitation feature area) value of the study area; collect satellite infrared brightness temperature data and precipitation data of the study area for several consecutive years, and combine them with the determined CCS area threshold, duration threshold, and PF threshold. Use the FlexTRKR method to identify the relevant data of MCSs with complete life cycle that may cause urban flooding and their occurrence time periods, and extract the evolution data of CCS area corresponding to each MCSs-related data segment, filtering out high-frequency fluctuations;

[0010] S2. Analyze the duration of the development, maturity, and dissipation stages of the MCSs samples in the study area, statistically analyze the correlation between each stage and urban flooding, and determine the proportion of the maturity stage in the total life cycle. Combining the proportion of the maturity stage in the total life cycle with the pace of urban flood control, determine the optimal value of the sensitivity coefficient. This sensitivity coefficient is used to represent the degree of attenuation of the area change rate at the boundary of the maturity stage relative to the fastest change rate, and is adapted to the response time of urban drainage facilities.

[0011] S3. Based on the MCSs related data after filtering out high-frequency fluctuations, fit the first relationship function between CCS area and time in the complete life cycle of MCSs. Analyze the first relationship function. If the evolution trend of CCS area shows the characteristics of first increasing and then decreasing, then proceed to step S4; otherwise, proceed to step S5.

[0012] S4. Taking the first derivative of the first relational function yields the slope change equation. By analyzing the maximum point in the first relational function of CCS area and time, the time when MCSs develops most vigorously is determined, and this is set as the center point. Combined with the optimal value of the sensitivity coefficient, the critical thresholds for the first curve slope used to divide the development and maturity stages, and the critical thresholds for the second curve slope used to divide the maturity and dissipation stages are determined. If these can be determined, using the center point as a reference, the slope change equation of the first relational function is expanded forward and backward to find the first time the slope reaches the critical threshold for the first curve slope, and the second curve slope... The boundary points of the critical threshold of the line slope are defined. The time period when the curve slope is between two boundary points is defined as the mature stage of MCSs. The time from the start of MCSs to the start of the mature stage is identified as the development stage. The time from the end of the mature stage to the end of MCSs is identified as the dissipation stage. The life stage division of MCSs is completed and the flood risk level of each stage is marked. At the same time, the functional positioning of each stage is clarified. Among them, the development stage, the mature stage and the dissipation stage correspond to the flood warning preparation period, the high-risk period of flooding and the period of water receding and disposal, respectively. Proceed to step S7; otherwise, proceed to step S6.

[0013] S5. Analyze the evolution trend of the minimum brightness temperature throughout the complete lifecycle of MCSs. If the evolution trend of the minimum brightness temperature shows a characteristic of first decreasing and then increasing, fit the evolution data of the minimum brightness temperature throughout the complete lifecycle of MCSs to obtain the second relationship function of minimum brightness temperature and time sequence. Take the first derivative of the second relationship function to obtain the slope change equation. By analyzing the minimum point in the second relationship function of minimum brightness temperature and time sequence, determine the time when MCSs development is most vigorous and set it as the center point. Combined with the optimal value of the sensitivity coefficient, determine the critical threshold of the slope of the third curve used to divide the development stage and the maturity stage, and the critical threshold of the slope of the fourth curve used to divide the maturity stage and the dissipation stage. If these can be determined, use the center point as the benchmark. From the slope change equation of the second relation function, expand forward and backward to find the boundary point where the slope first reaches the critical threshold of the third curve slope and the critical threshold of the fourth curve slope. Define the time period when the curve slope is between the two boundary points as the maturity stage of MCSs. Identify the time from the start of MCSs to the start of the maturity stage as the development stage, and the time from the end of the maturity stage to the end of MCSs as the dissipation stage. Complete the division of MCSs life stages and mark the flood risk level of each stage. At the same time, clarify the functional positioning of each stage. The development stage, maturity stage and dissipation stage correspond to the flood warning preparation period, the high-risk period for flooding and the period for water receding and disposal, respectively. Proceed to step S7; otherwise, proceed directly to step S6.

[0014] S6. If the current MCSs do not have a complete life cycle and their CCS area, duration, and precipitation intensity have not reached the critical conditions for urban flooding, they are considered to have not developed to a mature stage and the process is terminated. Only the monitoring data of the convective cells are output and labeled as non-flood-causing convective systems, and the process ends.

[0015] S7 standardizes the output of MCSs (Mechanical System Components) for each stage, including start and end times, duration, CCS area evolution rate, hourly precipitation intensity, and quantitative indicators of rain area spatial distribution. Combining urban underlying surface characteristics, and using the per-grid percentile threshold of precipitation in the Jianghuai region during the Meiyu season as a benchmark, it conducts grid-by-grid assessments of the synthetic average precipitation rate for each stage—the flood warning preparation period, the high-risk period for flooding, and the period for handling waterlogging—and classifies the flood disaster warning level for each grid area based on the assessment results. It also uses the urban underlying surface runoff coefficient and drainage capacity to assist in the assessment of net runoff intensity, comprehensively determining the spatial distribution of flood disaster warning levels for each grid area. Based on this, it triggers corresponding levels of meteorological disaster warnings and flood control emergency responses, and pushes emergency response measures adapted to the functional positioning of each stage.

[0016] Furthermore, in step S1, the CCS area threshold is determined by superimposing spatial distribution data of the urban flood-prone core area of ​​the study area to ensure that the rain area corresponding to the CCS area threshold completely covers the urban flood-prone core area and surrounding potentially affected areas; the duration threshold is set in conjunction with the standard procedure duration of urban flood control emergency response to ensure that the duration threshold is adapted to the actual time window requirements of urban flood control emergency response; the PF threshold corresponds to the critical precipitation intensity causing urban flooding, and its value is not lower than the critical rainfall intensity corresponding to the design return period of the urban drainage system in the study area.

[0017] Step S2 further includes:

[0018] The study analyzed the duration of the development, maturation, and dissipation phases of MCSs samples in the research area to determine the allowable proportion of the maturation phase in the total life cycle. The maturation phase includes the phase transition moment when the MCSs cloud expansion rate and minimum Tb cooling rate decrease to 20% of the peak value, and the structural disintegration initiation moment when the cloud contraction rate and minimum Tb heating rate reach 20% of the peak value.

[0019] The first and second relational functions of the MCSs samples in the study area were obtained by fitting. Sensitivity analysis was performed on the sensitivity coefficient η∈[0.05,0.5] to determine its optimal value. The first and second relational functions were processed by the optimal value respectively. The proportion of the mature stage in the total life cycle was found to be within the allowable proportion range, and the two proportions were as close as possible to each other. At the same time, the optimal value was verified to ensure that the identified development stage has sufficient advance warning for flood prevention, the duration of the mature stage matches the effective window of full-load operation of urban drainage facilities, and the timeliness of the dissipation stage identification is adapted to the rhythm requirements of dynamic parameter adjustment of drainage facilities.

[0020] Furthermore, in step S3, a fifth-order polynomial is used to fit the evolution of the CCS area during the complete lifecycle of the MCSs, resulting in the first relationship function between the CCS area and the time interval:

[0021]

[0022] In the formula, , , , , and is the fitting coefficient of the first relational function, and t is the time.

[0023] Step S4 further includes:

[0024] Taking the first derivative of the function relating the area of ​​the CCS to time, we obtain the equation for the slope change:

[0025]

[0026] And find the local maximum point, which satisfies and This maximum point is defined as the center point of the period when the corresponding MCSs are at their strongest. The time variable represents the point of maximum value;

[0027] The critical threshold for the slope of the first curve can be calculated using the following formula. The critical threshold of the slope of the second curve :

[0028]

[0029]

[0030] Where η is the sensitivity coefficient;

[0031] Extend the slope change equation of the first relational function forward and backward to find whether there are critical thresholds for the first curve slope and the second curve slope. If both exist, further search for the boundary points where the slope first reaches the critical thresholds for the first curve slope and the second curve slope. Define the time period when the curve slope is between the two boundary points as the maturity stage of MCSs. Identify the time from the start of MCSs to the start of the maturity stage as the development stage, and the time from the end of the maturity stage to the end of MCSs as the dissipation stage. Otherwise, proceed to step S6.

[0032] Furthermore, in step S4, a fifth-order polynomial is used to fit the evolution of the minimum Tb throughout the complete lifetime of the MCSs, resulting in a second relational function for the minimum brightness temperature and time order:

[0033]

[0034] In the formula, , , , , and The fitting coefficients for the second relational function are denoted as .

[0035] Furthermore, in step S5, the first derivative of the second relational function with respect to the minimum Tb and time is obtained to obtain the slope change equation:

[0036] Satisfaction and The minimum point, which represents the peak of MCSs development, is designated as the center point. This represents the time variable corresponding to the minimum point.

[0037] Further, in step S5, the critical threshold of the slope of the third curve is calculated using the following formula. and the critical threshold of the slope of the fourth curve ;

[0038]

[0039]

[0040] Where η is the sensitivity coefficient;

[0041] Extend the slope change equation of the second relational function forward and backward to find whether there are critical thresholds for the slope of the third curve and the slope of the fourth curve. If both exist, further search for the boundary point where the slope first reaches the critical thresholds for the slope of the third curve and the slope of the fourth curve. Define the time period when the slope of the curve is between the two boundary points as the maturity stage of MCSs. Identify the time from the start of MCSs to the start of the maturity stage as the development stage, and the time from the end of the maturity stage to the end of MCSs as the dissipation stage. Otherwise, proceed to step S6.

[0042] Step S7 further includes:

[0043] Output the start time, end time, duration, CCS area evolution rate, hourly precipitation intensity, and quantitative indicators of spatial distribution of rain areas for each stage of MCSs; construct a meteorological-geographic fusion model based on geographic information system to generate spatial overlay distribution maps of rain areas, urban flood-prone points, and flood control projects for each stage.

[0044] Using the per-grid percentile threshold of precipitation in the Jianghuai region during the Meiyu season as a benchmark, the composite average precipitation rate R at each stage—the flood warning preparation period, the high-risk period for flooding, and the period for handling waterlogging—is evaluated grid-by-grid. The flood disaster warning level L for each region is then determined according to a hierarchical mapping function.

[0045]

[0046] Where R is the composite average precipitation rate for each stage, and P 50 P 60 P 80 P 95 These are the 50th, 60th, 80th, and 95th percentile thresholds of the annual precipitation intensity during the Meiyu season at each grid point. When urban underlying surface data is available, the net runoff intensity q=φ·R-Q_drain is assessed by combining the urban underlying surface runoff coefficient φ and the drainage capacity Q_drain of the drainage system, and the warning level of each grid point is comprehensively corrected.

[0047] Trigger the corresponding level of flood control emergency response and push out emergency response measures that are appropriate to the functional positioning of each stage; specifically, in the development stage, push out instructions for the pre-positioning of flood control materials, the intensification of monitoring of flood-prone areas and the standby of rescue teams; in the mature stage, push out instructions for the full-load operation of drainage pumping stations, the transfer of people in flood-prone areas, traffic control and the scheduling of flood control projects; in the dissipation stage, push out instructions for the dynamic adjustment of drainage facilities, the investigation of post-disaster water accumulation and the repair of municipal facilities.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] First, the urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data of the present invention abandons the single variable identification mode, and instead takes CCS area as the core, supplemented by minimum Tb to achieve dual protection. It can realize the automatic identification of MCSs life stages, eliminate the subjective bias of manually dividing MCSs life stages to a certain extent, and contribute to the forecasting and early warning of heavy precipitation.

[0050] Secondly, the urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data of this invention customizes basic thresholds such as CCS area and duration for the precipitation climate characteristics of the study area. At the same time, it determines the slope critical threshold based on the actual evolution characteristics of MCS cloud clusters merging / disintegration, rather than using a fixed ratio. This adapts to special scenarios such as the train effect and frequent convective cell merging in the Meiyu region of the Yangtze River and Huai River, and can directly support the refined forecasting of heavy precipitation from the Meiyu front, providing practical technical support for urban flood disaster prediction.

[0051] Third, the urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data of the present invention can accurately identify each life stage of MCSs, and clarify the core quantitative characteristics such as precipitation intensity, duration and rain area range corresponding to different stages. It can establish a strong correlation between stage evolution and flood disaster, provide quantitative basis for predicting the time window and intensity of local heavy precipitation, effectively reduce the forecast error of extreme rainfall, and directly meet the decision-making needs of actual scenarios such as urban flood control. It can provide a scientific basis for resource allocation, measure activation and risk management, and improve the accuracy and timeliness of disaster prevention decisions.

[0052] Fourth, the urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data of the present invention achieves automated identification through function fitting and slope threshold analysis based on existing satellite infrared brightness temperature and precipitation data, without the need for additional observation equipment. Compared with traditional manual analysis methods, this method significantly reduces the manpower and time costs of data processing and analysis while effectively improving the efficiency of stage identification. It supports batch data processing and real-time result output, meeting the timeliness and large-scale application requirements of urban flood disaster prediction.

[0053] Fifth, the urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data of this invention deeply couples MCSs stage identification with urban underlying surface characteristics and waterlogging models, quantifying and outputting core flood-causing indicators such as water depth and inundation range. Through a dynamic feedback mechanism, it continuously optimizes prediction results and emergency response measures, forming a closed-loop process encompassing stage identification, flood prediction, tiered early warning, emergency response, and iterative optimization. This closed-loop support model efficiently transforms meteorological data into practical disaster prevention instructions, helping to shift flood control measures from experience-driven to data-driven, significantly improving the effectiveness of urban flood disaster response, and effectively reducing disaster losses. Attached Figure Description

[0054] Figure 1 This is a flowchart of the urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data of the present invention;

[0055] Figure 2 The distribution of the MCSs of this invention from 2001 to 2021 is shown (a total of 272 cases), and the dashed line is the fitted straight line;

[0056] Figure 3 The present invention uses a Meiyu MCS event on June 22, 2020 as an example to divide the life stages. Figures (a) and (c) show the changes in CCS area and minimum Tb after normalization of the life cycle. The blue solid line is the average CCS area change curve, the red solid line is the average minimum Tb change curve, the blue and red shaded areas are the standard deviation range, and the gray dashed line is the distribution of extreme points. (b) shows the CCS area change during an MCS process that occurred on June 22, 2020. The blue dots are the original CCS area, the red solid line is the fitted curve, the yellow, red, and blue shaded areas represent the development, maturity, and dissipation stages, and the black dashed line represents the stage transition point. (d) shows the slope change of the fitted curve. The red solid line represents the slope, the yellow and blue dashed lines represent the positive and negative thresholds of the slope, respectively, and the black dashed line represents the stage transition point.

[0057] Figure 4 The probability density distribution of the duration of the development, maturation and dissipation phases of 269 MCS events was statistically analyzed for this invention.

[0058] Figure 5 Figure 1 shows the total precipitation caused by MCSs in the Yangtze-Huaihe River Basin during the Meiyu season from 2001 to 2021, and the distribution of total precipitation in the three life stages. Figure 2(a) shows the total precipitation caused by MCSs; Figure 3(b) shows the total precipitation caused by MCSs in the development stage; Figure 4(c) shows the total precipitation caused by MCSs in the maturity stage; and Figure 5(d) shows the total precipitation caused by MCSs in the dissipation stage.

[0059] Figure 6The quantitative prediction map of urban flood disaster obtained in this invention uses the MCS event on June 22, 2020 as an example, and calculates the grid-by-grid point P in the Jianghuai region using IMERG precipitation data from the Meiyu season of 2001 to 2021. 50 P 60 P 80 P 95 Based on the percentile threshold spatial field, the spatial distribution of flood warnings in the three stages was obtained. Figure (a) shows the development stage, Figure (b) shows the maturity stage, and Figure (c) shows the dissipation stage. Detailed Implementation

[0060] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0061] This invention discloses a method for predicting urban flooding disasters based on satellite infrared brightness temperature and precipitation data. The method includes the following steps:

[0062] S1. Based on the precipitation climate characteristics and urban flood disaster prediction needs of the study area, the CCS area threshold, duration threshold, and PF threshold of the study area are determined. Satellite infrared brightness temperature data and precipitation data of the study area for several consecutive years are collected. Combined with the determined CCS area threshold, duration threshold, and PF threshold, the FlexTRKR method is used to identify the MCSs related data with complete life cycle and their occurrence time. The evolution data of CCS area corresponding to each MCSs related data segment is extracted, and high-frequency fluctuations are filtered out.

[0063] S2. Analyze the duration of the development, maturity, and dissipation stages of the MCSs samples in the study area, statistically analyze the correlation between each stage and urban flooding, and determine the proportion range of the maturity stage in the total life cycle. Combine the proportion range of the maturity stage in the total life cycle with the correlation with urban flooding to determine the optimal value of the sensitivity coefficient. This sensitivity coefficient is used to represent the degree of attenuation of the area change rate at the boundary of the maturity stage relative to the fastest change rate, and is adapted to the response time of urban drainage facilities. The optimal value of the sensitivity coefficient must ensure that the identified development stage has sufficient early warning lead time, the maturity stage duration matches the effective window of full-load operation of drainage pumping stations, and the identification time of the dissipation stage is adapted to the rhythm requirements of dynamic parameter adjustment of drainage facilities.

[0064] S3. Based on the MCSs-related data after filtering out high-frequency fluctuations, a fifth-order polynomial is fitted to the evolution of CCS area during the complete life cycle of MCSs to obtain the first relationship function between CCS area and time during the complete life cycle of MCSs. The first relationship function is analyzed. If the evolution trend of CCS area shows a first increase followed by a decrease, then proceed to step S4; otherwise, proceed to step S5.

[0065] S4. Calculate the first derivative of the first relational function to obtain the slope change equation. Analyze the maximum points in the first relational function to determine the most vigorous development time of MCSs, and set these as the center point. Combine the optimal value of the sensitivity coefficient to determine the critical threshold of the first curve slope (used to divide the development stage and the mature stage) and the critical threshold of the second curve slope (used to divide the mature stage and the dissipation stage). If these can be determined, using the center point as the benchmark, expand forward and backward from the slope change equation of the first relational function to find the boundary points where the slope first reaches the critical thresholds of the first and second curve slopes. Define the period when the curve slope is between the two boundary points as the mature stage of MCSs (high-risk period for flooding). Identify the time from the start of MCSs to the start of the mature stage as the development stage (preparatory period for flood warning), and the time from the end of the mature stage to the end of MCSs as the dissipation stage (period for water receding and handling). Complete the life stage division and label the flood risk level of each stage, then proceed to step S7; otherwise, proceed to step S5.

[0066] S5. Analyze the evolution trend of the minimum brightness temperature throughout the complete lifecycle of MCSs. If it shows a characteristic of first decreasing and then increasing, fit the evolution data of the minimum brightness temperature throughout the complete lifecycle of MCSs to obtain the second relationship function of minimum brightness temperature and time. Take the first derivative of the second relationship function to obtain the slope change equation. By analyzing the minimum point, determine the time when MCSs development is most vigorous and set it as the center point. Combined with the optimal value of the sensitivity coefficient, determine the critical threshold of the slope of the third curve (used to divide the development stage and the maturity stage) and the critical threshold of the slope of the fourth curve (used to divide the maturity stage and the dissipation stage). If the stage can be determined, using the center point as the benchmark, expand forward and backward from the slope change equation of the second relationship function to find the boundary point where the slope first reaches the critical threshold of the third curve slope and the critical threshold of the fourth curve slope. Define the time period when the curve slope is between the two boundary points as the mature stage of MCSs (high risk period of flooding), identify the time from the start of MCSs to the start of the mature stage as the development stage (preparatory period for flood warning), and identify the time from the end of the mature stage to the end of MCSs as the dissipation stage (period of water receding and disposal). Proceed to step S7; otherwise, proceed directly to step S6.

[0067] S6. If the current MCSs do not have a complete life cycle, and their CCS area, duration, and precipitation intensity have not reached the critical conditions for urban flooding, they are regarded as non-flood-causing convective systems that terminated before reaching maturity. Only the monitoring data of the convective cells are output and labeled as non-flood-causing convective systems, and the process ends.

[0068] S7 standardizes the output of quantitative indicators such as the start and end times, duration, CCS area evolution rate, hourly precipitation intensity, and spatial distribution of rain areas for each stage of MCSs. Combining the characteristics of the urban underlying surface, and using the per-grid percentile threshold of precipitation in the Jianghuai region during the Meiyu season as a benchmark, it evaluates the composite average precipitation rate for each stage of the flood warning preparation period, the high-risk period for flooding, and the period for handling waterlogging at each grid point, and analyzes the flood disaster warning level for each grid point. For example, areas with a composite average precipitation rate at each stage that is not lower than P50 and lower than P60 during the plum rain season at the grid point are designated as blue warning areas; areas with a composite average precipitation rate not lower than P60 and lower than P80 are designated as yellow warning areas; areas with a composite average precipitation rate not lower than P80 and lower than P95 are designated as orange warning areas; and areas with a composite average precipitation rate not lower than P95 are designated as red warning areas. Furthermore, the net runoff intensity is assessed by combining the urban underlying surface runoff coefficient and drainage capacity to comprehensively determine the spatial distribution of flood disaster warning levels for each region. Based on this, corresponding meteorological disaster warnings and flood control emergency responses are triggered, and emergency response measures adapted to the functional positioning of each stage are pushed out. For example, during the development stage, instructions are pushed out for the pre-positioning of flood control materials, increased monitoring of flood-prone areas, and standby of rescue teams; during the mature stage, instructions are pushed out for full-load operation of drainage pumping stations, evacuation of people from flood-prone areas, traffic control, and scheduling of flood control projects; and during the dissipation stage, instructions are pushed out for dynamic adjustment of drainage facilities, post-disaster waterlogging investigation, and repair of municipal facilities.

[0069] This invention considers that the life cycle of MCSs can generally be divided into three stages: development, maturity, and dissipation. The cloud and precipitation characteristics of each stage are significantly different and directly related to the risk of urban flooding. Based on the evolution characteristics of the cold cloud cover region (CCS) and minimum brightness temperature (Tb) of MCSs throughout their entire life cycle, an automated identification method for the life stages of MCSs, which is widely applicable to mid-latitude regions, is designed. The identification results are deeply coupled with urban underlying surface characteristics and waterlogging models to achieve full-process automation from meteorological monitoring to urban flood disaster prediction and emergency response.

[0070] Step 1: FlexTRKR identifies MCSs. Based on the precipitation climate characteristics of the study area and the needs of urban flood disaster prediction, the identification criteria are flexibly modified, such as CCS area threshold, duration threshold, and precipitation characteristic area PF threshold. The data used in this invention mainly comes from the NCEP / CPC global merged infrared brightness temperature dataset, covering the time range of 2001-2021 with a horizontal resolution of 4km, and the latest version of precipitation data product IMERGV07 released by NASA's Global Precipitation Measurement Mission (GPM), covering the time range of 2001-2021 with a horizontal resolution of 0.1°×0.1° and a temporal resolution of 0.5h. Based on the FlexTRKR method for tracking and identifying MCSs, this invention determines the life stage of MCSs by combining the evolution of CCS area during the complete life cycle of the MCSs with the evolution of the minimum Tb.

[0071] Step 2: Analyze the duration of the development, maturity, and dissipation phases of the MCSs samples in the study area to determine the proportion of the maturity phase in the total life cycle; combine the proportion of the maturity phase in the total life cycle to determine the optimal value of the sensitivity coefficient, which is used to represent the degree of decay of the area change rate at the boundary of the maturity phase relative to the fastest change rate.

[0072] This invention requires the determination of the critical thresholds for the first and third curve slopes used to divide the development and maturity stages of MCSs, as well as the critical thresholds for the second and fourth curve slopes used to divide the maturity and dissipation stages, through sensitivity coefficients.

[0073] The complete life cycle of MCSs exhibits a nonlinear evolution characteristic. The cloud area initially expands rapidly, with rapid vertical development, corresponding to a rapid decrease in the minimum brightness temperature at the cloud top. At the mature stage, the system's organization reaches its peak, the cloud area remains near its maximum and relatively stable, the expansion rate slows, and the cloud top height stabilizes. Then, the system structure disintegrates, the cloud area shrinks rapidly, and the minimum brightness temperature at the cloud top rises rapidly. Therefore, the boundary of the mature stage corresponds to a transition region on the slope change equation curve where the slope is near zero (the extreme point) and the change slows significantly. Based on this principle, this invention defines the curve slope threshold as the characteristic proportion of the extreme value of the first derivative, using it as the sensitivity coefficient η. Its physical meaning is the degree of attenuation of the area change rate at the mature stage boundary relative to the fastest change rate, reflecting the sensitivity of the identification system in transitioning from "rapid evolution" to "relative stability." Simultaneously, it is necessary to ensure that the identified development stages allow sufficient lead time for flood warnings, and the duration of the mature stage matches the effective window for the full-load operation of urban drainage pumping stations. The selection process for η includes: analyzing the duration of the development, maturation, and dissipation phases of MCSs samples in the analysis region, determining the allowable proportion of the mature phase in the total lifecycle, for example, the mature phase of MCSs during the Jianghuai Meiyu season typically accounts for 15% to 30% of the total lifecycle; performing sensitivity analysis on η∈[0.05,0.5] to determine its optimal value, and processing the first and second relational functions with this optimal value, ensuring that the proportion of the mature phase in the total lifecycle is within the allowable proportion range, and that the two identified proportions are as close as possible to each other. For example, regarding data related to MCSs during the Jianghuai Meiyu season, it was found that when η≤0.1, the identified mature phase is too short for the first relational function, making it difficult to reflect the stable maintenance characteristics of the system; when η≥0.3, the identified mature phase is too long for the second relational function. Based on the above analysis, for MCSs during the Meiyu season in the Yangtze River and Huai River basins, in this embodiment, η=0.2 is determined as the optimal value of the sensitivity coefficient. This value can effectively capture the stage transition when the cloud expansion rate and the minimum Tb cooling rate of MCSs decrease to 20% of the peak value. At this time, the system has basically completed the rapid aggregation process and entered the organized maintenance stage. At the same time, it can also accurately identify the structural disintegration start time when the cloud contraction rate and the minimum Tb warming rate reach 20% of the peak value. The proportion of the identified mature stage in the total life cycle is within the allowable range and is compatible with the response time of urban drainage facilities.

[0074] Step 3: Determine the development, maturity, and decline stages of MCSs based on the evolution of CCS. This specifically includes:

[0075] Step A1: To smooth the evolution of CCS and filter out high-frequency fluctuations, a fifth-order polynomial fit is performed on the evolution of CCS area throughout the complete lifecycle of MCSs. Where a0, a1, ..., a5 are fitting coefficients, and t is the time interval (the duration from 0 to the MCSs). If the evolution trend of the CCS area shows an initial increase followed by a decrease, proceed to step A2; otherwise, proceed to step four.

[0076] Step A2: Determine the time when MCSs develops most vigorously, and take the first derivative of the fifth-order polynomial of CCS and time to obtain the slope change equation. And solve for the maximum point (which must satisfy) and This point corresponds to the peak of MCS development and is designated as the center point.

[0077] Step A3: Determine the critical threshold for the slope of the first curve and the critical threshold for the slope of the second curve for different development states of MCSs. and Using the characteristic proportion η of the first derivative extremum as the threshold standard, on the slope change equation curve, with the center point as the reference, the search is conducted forward and backward to find the slope that first reaches its maximum value. and The boundary point; the slope of the curve is located at... and The time period between these points is defined as the maturity stage of MCSs (high-risk period for flooding). After sensitivity testing, η in this embodiment is set to 0.2. If either the critical threshold for the slope of the first curve or the critical threshold for the slope of the second curve is not found in the slope change equation of the first relational function, then proceed to step five.

[0078] Step A4: Identify the time from the start of MCSs to the start of the maturity stage as the development stage (flood warning preparation period), and the time from the end of the maturity stage to the end of MCSs as the dissipation stage (water receding and disposal period).

[0079] Step Four: This step is a method for non-typical MCSs development stages. For individual MCSs whose CCS does not exhibit a characteristic of increasing followed by decreasing throughout their lifecycle, making it impossible to identify the complete lifecycle, Tb data will be used for identification. The identification process is the same as the method described above, but the minimum point is set as the center point. Specifically, it includes:

[0080] Step B1: Perform fifth-order polynomial fitting on the evolution of the minimum Tb during the complete lifecycle of MCSs. , where b0, b1, ..., b5 are the fitting coefficients.

[0081] Step B2: Take the first derivative of the second relationship function with respect to the minimum Tb and time to obtain the slope change equation, and obtain the minimum point (which must satisfy...). and This point represents the peak of MCS development and is designated as the center point.

[0082] Step B3: Determine the peak development period of MCSs, and determine the critical thresholds for the slope of the third curve and the slope of the fourth curve for different development states of MCSs. and Using the characteristic proportion η of the first derivative extremum as the threshold standard, on the slope change equation curve, with the center point as the reference, the search is conducted forward and backward to find the slope that first reaches its maximum value. and The boundary point, the slope of the curve is located at and The time period between these points is defined as the maturity stage of MCSs (high-risk period for flooding). After sensitivity testing, η is set to 0.2 in this embodiment. If either the critical threshold for the slope of the third curve or the critical threshold for the slope of the fourth curve is not found in the slope change equation of the second relationship function, then proceed to step five.

[0083] Step B4: The time from the start of MCSs to the start of the maturity stage is identified as the development stage (flood warning preparation period), and the time from the end of the maturity stage to the end of MCSs is identified as the dissipation stage (water receding and disposal period).

[0084] Step 5: Handling of Non-Flood-Causing Convective Systems. If the three life stages are still not identified after Steps 3 and 4, these MCSs are usually short-lived or weak in intensity, and have not undergone a complete process from development to maturity and dissipation. Furthermore, their CCS area, duration, and precipitation intensity have not reached the critical conditions for urban flooding. They are considered as weak convective cells and are regarded as non-flood-causing convective systems that terminated before reaching maturity, thus ending the process.

[0085] Step Six: Quantitative Prediction and Emergency Response Push for Urban Flooding. Based on the life stage division results completed in Step Three or Four, and using the multi-year grid-by-grid percentile threshold of precipitation in the Jianghuai region during the Meiyu season as a benchmark, the composite average precipitation rate R of each stage is evaluated grid-by-grid. The spatial distribution of flood disaster warning levels for each region is determined according to the hierarchical mapping function L=f(R,P): P 50 ≤R <P 60 Blue alert, P 60 ≤R <P 80 Yellow alert, P 80 ≤R <P 95 Orange alert level, R≥P 95 A red alert is issued; when urban underlying surface data is available, the net runoff intensity is adjusted by combining the runoff coefficient and drainage capacity; the corresponding level of flood control emergency response is triggered, emergency response measures adapted to the functional positioning of each stage are pushed, and the process ends.

[0086] Example

[0087] like Figure 3 As shown in (a) and (c), during the complete lifecycle of MCSs, the CCS area shows a trend of first increasing and then decreasing, while Tb shows the opposite trend of first decreasing and then increasing. Moreover, the evolution of the two variables is relatively gentle near the extreme point. Based on this characteristic, the implementation process of this example is as follows:

[0088] Step S1: Identify MCSs with a complete lifecycle that may cause urban flooding.

[0089] Step S101: Determine the research area as the Meiyu season in the Jianghuai region (110°E-122°E, 28°N-34°N), lasting 495 days. Set the MCSs identification threshold to: require the system's CCS area to exceed 6×10⁻⁶. 4 km 2 After verification by superimposing the data with the spatial distribution of urban built-up areas and historically flood-prone points in the study area, the range of the rain area corresponding to this threshold can effectively cover the main waterlogging risk areas in the city; the duration exceeds 6 hours, which is consistent with the duration characteristics of MCSs and can reserve sufficient lead time for the activation of urban flood control emergency response; the duration of the longest precipitation characteristic area (PF, precipitation intensity ≥3mm / h) in CCS with a length of more than 100km exceeds 4 hours.

[0090] Step S102: During the study, a total of 272 MCSs that could potentially cause urban flooding were identified in the Jianghuai region.

[0091] Step S2: Use CCS area to identify the MCS that occurred on June 22, 2020.

[0092] Step S201: The evolution of CCS usually has large fluctuations, so smoothing is required. A fifth-order polynomial is used to fit the evolution of CCS area in this MCS. If the evolution trend of CCS area shows an increase followed by a decrease, proceed to step S202; otherwise, proceed to step S3.

[0093] Step S202: Take the first derivative of the fifth-order polynomial of CCS and time series to obtain the slope change equation, and find the maximum point (the position where the first derivative is zero and the second derivative is less than zero). This maximum point corresponds to the peak time series of MCS development, and is used as the center point. Figure 3 As can be seen from (b) and (d) in the data, the strongest point of MCS development occurred in the 6th time period.

[0094] Step S203: Determine the slope thresholds of the first and second curves of the fitted equation, with the feature ratio η set to 0.2 for this specific case. Approximately 4095km 2 ·h -1 , -4095km 2 ·h-1 On the first derivative curve, using the center point as a reference, extend the curve forward and backward to find the slope that first reaches its maximum value. and The boundary point; the slope of the curve is located at... and The time period between these points is defined as the maturity stage of MCS, such as... Figure 3 (d) The slope equation curves reached the positive and negative slope thresholds at the 5th and 10th time intervals, respectively. Therefore, the duration of the MCS maturity stage from the 5th to the 10th time interval was 5 hours.

[0095] Step S204: The development stage (times 0 to 5) before the maturity stage is identified, lasting 5 hours; the dissipation stage (times 10 to 21) after the maturity stage is identified, lasting 11 hours.

[0096] Step S3: Method for Atypical MCS Development Stages. For individual MCSs whose CCS does not exhibit a pattern of increasing followed by decreasing throughout their lifecycle and whose complete lifecycle cannot be identified, Tb data will be used for identification. The identification process is the same as the method described above, except that the minimum point is set as the center point.

[0097] Step S4: After running the above identification process, it was found that only 3 out of all 272 MCSs events failed to identify a complete life cycle. The other 269 cases all had complete development, maturity, and dissipation stages, and all were identified using CCS area. The duration of the MCSs in the three stages was statistically analyzed. Figure 4 ) and the resulting precipitation ( Figure 5 (b), (c), and (d)) found that MCSs had the shortest duration (concentrated in 3-4 hours) during the mature stage (high-risk period for flooding) but caused the most precipitation. The duration and precipitation of the development stage were slightly higher than those of the dissipation stage.

[0098] Step S5: Quantitative prediction and emergency response push for urban flooding disasters. Based on the MCS life stage division results on June 22, 2020, the grid-by-grid P of the Jianghuai region was calculated using IMERG precipitation data from the Meiyu season of 2001 to 2021. 50 P 60 P 80 P 95 Based on the percentile threshold spatial field, the composite average precipitation rate R for each stage—the flood warning preparation period, the high-risk period for flooding, and the period for handling waterlogging—is evaluated grid-by-grid. The spatial distribution of flood disaster warning levels for each region is determined according to the hierarchical mapping function L=f(R,P): P 50 ≤R <P 60 It has been determined to be a blue alert, P 60 ≤R <P 80It has been determined to be a yellow alert, P 80 ≤R <P 95 An orange alert is issued if R ≥ P. 95 It was classified as a red alert. Taking the MCS incident on June 22, 2020 as an example (…), Figure 6 During the development stage (hours 0-5), the overall average precipitation rate of the core grid points in the rain area is low, with orange alerts predominating and red alerts reaching high values ​​in some areas. The rain area is distributed in a southwest-northeast band. During the mature stage (hours 5-9), precipitation intensity increases significantly, with the widest coverage of orange to red alerts, and the high-risk area for flooding expands northeastward. During the dissipation stage (hours 9-21), precipitation intensity gradually weakens, the rain area shrinks and shifts eastward, and the alert level is mainly orange, with a significant reduction in the area under red alerts. The corresponding level of flood control emergency response is triggered, and emergency response measures adapted to the functional positioning of each stage are pushed out: During the development stage, instructions are pushed out for the pre-positioning of flood control materials, increased monitoring of flood-prone areas, and standby of rescue teams; during the mature stage, instructions are pushed out for full-load operation of drainage pumping stations, evacuation of people from flood-prone areas, traffic control, and the opening and closing of flood control projects; during the dissipation stage, instructions are pushed out for dynamic adjustment of drainage facilities, post-disaster waterlogging investigation, and repair of municipal facilities.

[0099] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0100] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for predicting urban flood disaster based on satellite infrared brightness temperature and precipitation data, characterized in that, The method Includes the following steps: S1. Based on the precipitation climate characteristics and urban flood disaster prediction needs of the study area, the CCS area threshold, duration threshold, and PF threshold of the study area are determined. Satellite infrared brightness temperature data and precipitation data of the study area for several consecutive years are collected. Combined with the determined CCS area threshold, duration threshold, and PF threshold, the FlexTRKR method is used to identify the relevant data of MCSs with complete life cycle that may cause urban flooding and their occurrence time periods. The evolution data of CCS area corresponding to each MCSs-related data segment is extracted, and high-frequency fluctuations are filtered out. S2. Analyze the duration of the development, maturity, and dissipation stages of the MCSs samples in the study area, statistically analyze the correlation between each stage and urban flooding, and determine the proportion of the maturity stage in the total life cycle. Combining the proportion of the maturity stage in the total life cycle with the pace of urban flood control, determine the optimal value of the sensitivity coefficient. This sensitivity coefficient is used to represent the degree of attenuation of the area change rate at the boundary of the maturity stage relative to the fastest change rate, and is adapted to the response time of urban drainage facilities. S3. Based on the MCSs related data after filtering out high-frequency fluctuations, fit the first relationship function between CCS area and time in the complete life cycle of MCSs. Analyze the first relationship function. If the evolution trend of CCS area shows the characteristics of first increasing and then decreasing, then proceed to step S4; otherwise, proceed to step S5. S4. Taking the first derivative of the first relational function yields the slope change equation. By analyzing the maximum point in the first relational function of CCS area and time, the time when MCSs develops most vigorously is determined, and this is set as the center point. Combined with the optimal value of the sensitivity coefficient, the critical thresholds for the first curve slope used to divide the development and maturity stages, and the critical thresholds for the second curve slope used to divide the maturity and dissipation stages are determined. If these can be determined, using the center point as a reference, the slope change equation of the first relational function is expanded forward and backward to find the first time the slope reaches the critical threshold for the first curve slope, and the second curve slope... The boundary points of the critical threshold of the line slope are defined. The time period when the curve slope is between two boundary points is defined as the mature stage of MCSs. The time from the start of MCSs to the start of the mature stage is identified as the development stage. The time from the end of the mature stage to the end of MCSs is identified as the dissipation stage. The life stage division of MCSs is completed and the flood risk level of each stage is marked. At the same time, the functional positioning of each stage is clarified. Among them, the development stage, the mature stage and the dissipation stage correspond to the flood warning preparation period, the high-risk period of flooding and the period of water receding and disposal, respectively. Proceed to step S7; otherwise, proceed to step S6. S5. Analyze the evolution trend of the minimum brightness temperature throughout the complete lifecycle of MCSs. If the evolution trend of the minimum brightness temperature shows a characteristic of first decreasing and then increasing, fit the evolution data of the minimum brightness temperature throughout the complete lifecycle of MCSs to obtain the second relationship function of minimum brightness temperature and time sequence. Take the first derivative of the second relationship function to obtain the slope change equation. By analyzing the minimum point in the second relationship function of minimum brightness temperature and time sequence, determine the time when MCSs development is most vigorous and set it as the center point. Combined with the optimal value of the sensitivity coefficient, determine the critical threshold of the slope of the third curve used to divide the development stage and the maturity stage, and the critical threshold of the slope of the fourth curve used to divide the maturity stage and the dissipation stage. If these can be determined, use the center point as the benchmark. From the slope change equation of the second relation function, expand forward and backward to find the boundary point where the slope first reaches the critical threshold of the third curve slope and the critical threshold of the fourth curve slope. Define the time period when the curve slope is between the two boundary points as the maturity stage of MCSs. Identify the time from the start of MCSs to the start of the maturity stage as the development stage, and the time from the end of the maturity stage to the end of MCSs as the dissipation stage. Complete the division of MCSs life stages and mark the flood risk level of each stage. At the same time, clarify the functional positioning of each stage. The development stage, maturity stage and dissipation stage correspond to the flood warning preparation period, the high-risk period for flooding and the period for water receding and disposal, respectively. Proceed to step S7; otherwise, proceed directly to step S6. S6. If the current MCSs do not have a complete life cycle and their CCS area, duration, and precipitation intensity have not reached the critical conditions for urban flooding, they are considered to have not developed to a mature stage and the process is terminated. Only the monitoring data of the convective cells are output and labeled as non-flood-causing convective systems, and the process ends. S7 standardizes the output of MCSs (Mechanical System Components) for each stage, including start and end times, duration, CCS area evolution rate, hourly precipitation intensity, and quantitative indicators of rain area spatial distribution. Combining urban underlying surface characteristics, and using the per-grid percentile threshold of precipitation in the Jianghuai region during the Meiyu season as a benchmark, it conducts grid-by-grid assessments of the synthetic average precipitation rate for each stage—the flood warning preparation period, the high-risk period for flooding, and the period for handling waterlogging—and classifies the flood disaster warning level for each grid area based on the assessment results. It also uses the urban underlying surface runoff coefficient and drainage capacity to assist in the assessment of net runoff intensity, comprehensively determining the spatial distribution of flood disaster warning levels for each grid area. Based on this, it triggers corresponding levels of meteorological disaster warnings and flood control emergency responses, and pushes emergency response measures adapted to the functional positioning of each stage.

2. The method for predicting urban flood disaster based on satellite infrared brightness temperature and precipitation data according to claim 1, characterized in that, In step S1, the CCS area threshold is determined by overlaying spatial distribution data of the urban flood-prone core area of ​​the study region to ensure that the rain area corresponding to the CCS area threshold completely covers the urban flood-prone core area and surrounding potentially affected areas; the duration threshold is set in conjunction with the standard process duration of urban flood control emergency response to ensure that the duration threshold is adapted to the actual time window requirements of urban flood control emergency response. The PF threshold corresponds to the critical rainfall intensity that causes urban flooding, and its value is not lower than the critical rainfall intensity corresponding to the design return period of the urban drainage system in the study area. 3.The method for predicting urban flood disaster based on satellite infrared brightness temperature and precipitation data according to claim 1, characterized in that, Step S2 further includes: The study analyzed the duration of the development, maturation, and dissipation phases of MCSs samples in the research area to determine the allowable proportion of the maturation phase in the total life cycle. The maturation phase includes the phase transition moment when the MCSs cloud expansion rate and minimum Tb cooling rate decrease to 20% of the peak value, and the structural disintegration initiation moment when the cloud contraction rate and minimum Tb heating rate reach 20% of the peak value. The first and second relational functions of the MCSs samples in the study area were obtained by fitting. Sensitivity analysis was performed on the sensitivity coefficient η∈[0.05,0.5] to determine its optimal value. The first and second relational functions were processed by the optimal value respectively. The proportion of the mature stage in the total life cycle was found to be within the allowable proportion range, and the two proportions were as close as possible to each other. At the same time, the optimal value was verified to ensure that the identified development stage has sufficient advance warning for flood prevention, the duration of the mature stage matches the effective window of full-load operation of urban drainage facilities, and the timeliness of the dissipation stage identification is adapted to the rhythm requirements of dynamic parameter adjustment of drainage facilities.

4. The method for predicting urban flood disaster based on satellite infrared brightness temperature and precipitation data according to claim 1, characterized in that, In step S3, a fifth-order polynomial is used to fit the evolution of the CCS area during the complete lifecycle of the MCSs, resulting in the first relationship function between the CCS area and the time interval: ； wherein , , , , and are fitting coefficients of the first relationship function, and t is the time.

5. The method for predicting urban flood disaster based on satellite infrared brightness temperature and precipitation data according to claim 4, characterized in that, Step S4 further includes: Taking the first derivative of the function relating the area of ​​the CCS to time, we obtain the equation for the slope change: ； and solve the maximum point, which satisfies and define the maximum point as the center point of the strongest time of the corresponding MCSs development, denotes the time variable corresponding to the maximum point. The first curve slope critical threshold value and the second curve slope critical threshold value are solved according to the following formulas and : ； ； Where η is the sensitivity coefficient; Extend the slope change equation of the first relational function forward and backward to find whether there are critical thresholds for the first curve slope and the second curve slope. If both exist, further search for the boundary points where the slope first reaches the critical thresholds for the first curve slope and the second curve slope. Define the time period when the curve slope is between the two boundary points as the maturity stage of MCSs. Identify the time from the start of MCSs to the start of the maturity stage as the development stage, and the time from the end of the maturity stage to the end of MCSs as the dissipation stage. Otherwise, proceed to step S6. 6.The method for predicting urban flood disaster based on satellite infrared brightness temperature and precipitation data according to claim 1, characterized in that, In step S5, a fifth-order polynomial is used to fit the evolution of the minimum Tb over the entire lifetime of the MCSs, resulting in a second relational function for the minimum brightness temperature and time order: ； In the formula, , , , , and The fitting coefficients for the second relational function are denoted as .

7. The urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data according to claim 6, characterized in that, In step S5, the first derivative of the second relationship function with respect to minimum Tb and time is obtained to obtain the slope change equation: ; Satisfaction and The minimum point, which represents the peak of MCSs development, is designated as the center point. This represents the time variable corresponding to the minimum point.

8. The urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data according to claim 1, characterized in that, In step S5, the critical threshold of the slope of the third curve is calculated using the following formula. and the critical threshold of the slope of the fourth curve ; ； ； Where η is the sensitivity coefficient; Extend the slope change equation of the second relational function forward and backward to find whether there are critical thresholds for the slope of the third curve and the slope of the fourth curve. If both exist, further search for the boundary point where the slope first reaches the critical thresholds for the slope of the third curve and the slope of the fourth curve. Define the time period when the slope of the curve is between the two boundary points as the maturity stage of MCSs. Identify the time from the start of MCSs to the start of the maturity stage as the development stage, and the time from the end of the maturity stage to the end of MCSs as the dissipation stage. Otherwise, proceed to step S6.

9. The urban flood disaster prediction method based on satellite infrared brightness temperature and precipitation data according to claim 1, characterized in that, Step S7 further includes: Output the start time, end time, duration, CCS area evolution rate, hourly precipitation intensity, and quantitative indicators of spatial distribution of rain areas for each stage of MCSs; construct a meteorological-geographic fusion model based on geographic information system to generate spatial overlay distribution maps of rain areas, urban flood-prone points, and flood control projects for each stage. Using the per-grid percentile threshold of precipitation in the Jianghuai region during the Meiyu season as a benchmark, the composite average precipitation rate R at each stage—the flood warning preparation period, the high-risk period for flooding, and the period for handling waterlogging—is evaluated grid-by-grid. The flood disaster warning level L for each region is then determined according to a hierarchical mapping function. ； where R is the average rainfall rate of each stage of synthesis, P 50 , P 60 , P 80 , P 95 are the 50th, 60th, 80th, and 95th percentile threshold values of the rainfall intensity during the Meiyu period for each grid, respectively; when urban underlying surface data is available, the net runoff intensity q = φ·R - Q_drain is evaluated with the aid of the urban underlying surface runoff coefficient φ and the drainage system drainage capacity Q_drain, and the warning levels of each grid are comprehensively corrected. Trigger the corresponding level of flood control emergency response and push out emergency response measures that are appropriate to the functional positioning of each stage; specifically, in the development stage, push out instructions for the pre-positioning of flood control materials, the intensification of monitoring of flood-prone areas and the standby of rescue teams; in the mature stage, push out instructions for the full-load operation of drainage pumping stations, the transfer of people in flood-prone areas, traffic control and the scheduling of flood control projects; in the dissipation stage, push out instructions for the dynamic adjustment of drainage facilities, the investigation of post-disaster water accumulation and the repair of municipal facilities.

Images