A method and system for predicting rapid dry-hot-to-humid composite extreme weather

Abstract

The present invention discloses a method and system for predicting rapid dry-heat-to-humid composite extreme weather, which relates to the field of meteorological and hydrological technology. An objective multi-fractal detrended fluctuation analysis method is used to determine event threshold data, historical composite extreme weather data, and future composite extreme weather data, which is not affected by people's subjective consciousness. Based on the event threshold data, historical event characteristic change data, future event characteristic change data, dependency assessment data, and future potential risk assessment data corresponding to the historical composite extreme weather data and the future composite extreme weather data are calculated. The extreme weather prediction data of the detection area is constructed using the historical event characteristic change data, the future event characteristic change data, the dependency assessment data, and the future potential risk assessment data. The present invention fully considers the interaction and dependency relationship between the values of various meteorological elements in the composite extreme event, and considers the occurrence characteristics and potential risks of the event under the future scenario mode, thereby improving the accuracy of prediction for high-risk rapid dry-heat-to-humid composite extreme weather.

Description

Technical Field

[0001] The present invention relates to the field of meteorological and hydrological technology, and in particular to a method and system for predicting rapid dry-hot-to-humid composite extreme weather. Background Art

[0002] In recent years, extreme weather events have become increasingly frequent due to global warming. These events typically do not occur in isolation, but rather combine to form complex events, including dry-hot events, wet-hot events, and dry-wet events. These complex events are often more harmful than individual complex extreme events and result from the simultaneous or sequential occurrence of multiple complex extreme events, influenced by various factors. Traditional research on complex extreme events often focuses on a single complex extreme event or risk, with less attention paid to the rapid transitions between these complex extreme events. The cumulative hazards of these rapidly transitioning complex extreme events can increase risk, and insufficient attention can lead to underestimation of risk. Therefore, it is crucial to consider the interactions and dependencies between the values of these meteorological components in complex extreme events. Hot, dry, or wet conditions can harm ecosystems and endanger human well-being. During warm seasons, due to reduced rainfall, dry conditions often occur simultaneously with extreme heat. This combination can lead to dry-hot complex events. Growing evidence shows that extreme rainfall events are closely associated with preceding dry and hot weather, which often leads to subsequent floods. Abnormal land evaporation and surface heat flux caused by preceding dry and hot events can promote water vapor convergence, increasing the likelihood of heavy rainfall following droughts. In addition, enhanced thermal forcing caused by short, intense dry and hot periods, combined with sudden moisture accumulation, can lead to unstable atmospheric temperatures, which can cause convective rainfall. This mechanism potentially explains the observation that heavy rainfall events tend to occur shortly after dry and hot events, which can then trigger floods, known as wet events. This combination defines a continuous dry-hot to wet (DHW) composite extreme event. The transition from extreme drought to extreme wetness is crucial for restoring regional water resources. However, very dry topsoil can form a hard crust, leading to surface runoff and flooding. It can also lead to other hazards such as landslides and mudslides.

[0003] Generally speaking, when weather conditions deviate significantly from their average, they can be considered unlikely events. From a statistical perspective, such events can be called composite extreme events. To identify composite extreme events, most existing studies use established percentiles as composite extreme event thresholds, such as the 95th or 99th percentile of precipitation. However, the use of these established percentiles as percentile thresholds for composite extreme events is significantly influenced by human subjectivity. Existing methods ignore the inherent variability of event sequences, leading to both overestimation and underestimation of composite extreme events. Therefore, using established percentiles in traditional methods to determine composite extreme event thresholds is inappropriate in many cases.

[0004] Furthermore, initial disasters not only exacerbate the impact of subsequent disasters, but also increase the vulnerability of affected communities due to the alternating occurrence of disasters. In the future, extreme dry, hot, and wet events will become more frequent in many land areas around the world. Furthermore, due to anthropogenic climate change, previously rare combined extreme dry and hot weather has gradually become more common since 2000. These factors have further led to the poor accuracy of existing forecasting methods for rapid dry-hot-to-wet combined extreme weather. Summary of the Invention

[0005] The present invention provides a method and system for predicting rapid dry-heat-to-humidity composite extreme weather, which are used to improve the accuracy of composite extreme weather prediction.

[0006] The present invention provides a rapid dry-heat-to-humidity composite extreme weather forecasting method, comprising:

[0007] Obtain climate model data corresponding to the detection area, use bilinear interpolation and equidistant cumulative distribution function methods to downscale the climate model data for bias correction, and use multifractal detrended fluctuation analysis to screen event thresholds and composite extreme events on the corrected climate model data to generate event threshold data, historical composite extreme weather data, and future composite extreme weather data;

[0008] Based on the event threshold data, characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data are calculated respectively to generate historical event characteristic change data and future event characteristic change data;

[0009] Performing a dependency assessment based on the historical composite extreme weather data and the future composite extreme weather data to generate dependency assessment data;

[0010] Based on the population and urban expansion data corresponding to the detection area, the future potential risk assessment of the dry-hot-to-humid composite extreme event is performed to generate future potential risk assessment data;

[0011] The extreme weather prediction data corresponding to the detection area is constructed using the historical event characteristic change data, the future event characteristic change data, the dependency assessment data and the future potential risk assessment data.

[0012] Optionally, the steps of using a bilinear interpolation method and an equidistant cumulative distribution function method to perform downscaling bias correction on the climate model data, and using a multifractal detrended fluctuation analysis method to perform event threshold and composite extreme event screening on the corrected climate model data, and generating event threshold data, historical composite extreme weather data, and future composite extreme weather data, include:

[0013] Using a bilinear interpolation method and an equidistant cumulative distribution function method to perform data downscaling bias correction and data partitioning on the climate model data, to generate target historical climate model data and target future climate model data;

[0014] Using a multifractal detrended fluctuation analysis method to perform event threshold screening on the target historical climate model data and the target future climate model data to generate event threshold data;

[0015] According to the event threshold data and the preset dry-heat-to-humid composite extreme event definition data, the target historical climate model data and the target future climate model data are respectively screened for dry-heat-to-humid composite extreme events to generate historical composite extreme weather data and future composite extreme weather data.

[0016] Optionally, the step of using a bilinear interpolation method and an equidistant cumulative distribution function method to perform data downscaling bias correction and data partitioning on the climate model data to generate target historical climate model data and target future climate model data includes:

[0017] downscaling the initial preset meteorological forcing dataset and the climate model data using a bilinear interpolation method to generate a target preset meteorological forcing dataset and initial climate model data;

[0018] Dividing the initial climate model data according to preset year intervals to generate intermediate historical climate model data and intermediate future climate model data;

[0019] The cumulative distribution function corresponding to the target preset meteorological forcing data set is derived by using an equidistant cumulative distribution function method to generate a cumulative distribution function;

[0020] The intermediate historical climate model data and the intermediate future climate model data are bias-corrected using the cumulative distribution function to generate target historical climate model data and target future climate model data.

[0021] Optionally, the step of performing event threshold screening on the target historical climate model data and the target future climate model data using a multifractal detrended fluctuation analysis method to generate event threshold data includes:

[0022] Detrending the target historical climate model data and the target future climate model data to construct a time series;

[0023] respectively calculating the fluctuation amplitudes of the time series at multiple preset time scales to generate multiple multifractal features;

[0024] Dividing the sequence corresponding to the multifractal feature according to preset intervals to generate multiple divided sequences;

[0025] When the long-term correlation coefficient corresponding to the division sequence has a preset coefficient, the division interval corresponding to the preset coefficient is used as event threshold data.

[0026] Optionally, the event threshold data includes a drought event determination threshold, a heat event determination threshold, and a wet event determination threshold; and the step of calculating, based on the event threshold data, characteristic change data of each dry-heat-to-wet composite extreme event in the historical composite extreme weather data and the future composite extreme weather data to generate the historical event characteristic change data and the future event characteristic change data comprises:

[0027] The total number of dry-hot-to-humid composite extreme events in the historical composite extreme weather data is used as the frequency of historical composite extreme events;

[0028] The total number of dry-hot-to-humid composite extreme events in the future composite extreme weather data is used as the frequency of future composite extreme events;

[0029] Calculating the sum of the duration of the dry and hot period and the duration of the wet period in the historical composite extreme weather data to generate the duration of the historical composite extreme event;

[0030] Calculating the sum of the duration of the dry and hot period and the duration of the wet period in the future composite extreme weather data to generate the duration of the future composite extreme event;

[0031] Substituting the meteorological data in the historical composite extreme weather data and the future composite extreme weather data, the drought event determination threshold, the heat event determination threshold, and the wet event determination threshold into a preset time intensity calculation formula, respectively, to calculate the historical composite extreme event intensity and the future composite extreme event intensity;

[0032] The preset time intensity calculation formula is:

[0033] ;

[0034] in, is the event intensity; is the precipitation on day i; is the maximum temperature on day i; is the runoff volume of the jth section; Determine the precipitation corresponding to the threshold for drought events; Determine the maximum temperature corresponding to the threshold for the heat event; Determine the runoff volume corresponding to the threshold for the wet event; is the duration of dry heat; is the duration of wet events in the composite extreme weather data; establishing thresholds for drought events; Determine thresholds for thermal events; Determine thresholds for wet events;

[0035] Constructing historical event characteristic change data corresponding to the historical composite extreme weather data by using the historical composite extreme event frequency, the historical composite extreme event duration, and the historical composite extreme event intensity;

[0036] The future event characteristic change data corresponding to the future composite extreme weather data are constructed by using the future composite extreme event frequency, the future composite extreme event duration and the future composite extreme event intensity.

[0037] Optionally, the step of performing dependency assessment based on the historical composite extreme weather data and the future composite extreme weather data to generate dependency assessment data includes:

[0038] Calculating the runoff depth corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset hydrodynamic model to generate a plurality of runoff indices;

[0039] Calculating the temperatures corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset temperature probability density function to generate a plurality of temperature indices;

[0040] Calculating the precipitation corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset precipitation probability density function to generate a plurality of precipitation indices;

[0041] Substituting the runoff index, the temperature index, the precipitation index, and the event threshold data into a preset correlation probability formula respectively to calculate a plurality of correlation probabilities;

[0042] The preset correlation probability formula is:

[0043] ;

[0044] Where P is the probability of occurrence of DHW events; D is the precipitation index; T is the temperature index; R is the runoff index; C is the connection function; d is the dry event threshold data; t is the heat event threshold data; r is the wet event threshold data; u is the conditional distribution function;

[0045] Substituting the conditional probability and the single probability corresponding to the relevant probability into a preset probability multiplication factor formula to calculate the probability multiplication factor corresponding to the dry-hot-to-humid composite extreme event;

[0046] The preset probability multiplication factor formula is:

[0047] ;

[0048] in, is the probability multiplication factor; is the conditional probability of three events occurring in sequence; is the single probability of a single event occurring;

[0049] Using all of the described probability multiplication factors, dependency estimation data are constructed.

[0050] Optionally, the population and urban expansion data include the population area, urban area, and total number of grid cells corresponding to each grid; and the step of performing future potential risk assessments on the dry-heat-to-humidity combined extreme event based on the population and urban expansion data corresponding to the detection area to generate future potential risk assessment data includes:

[0051] Substituting the grid event frequency, population area, urban area, and total number of grid cells corresponding to the dry-heat-to-humidity composite extreme event into a preset risk assessment formula, respectively, to calculate a risk assessment value corresponding to the dry-heat-to-humidity composite extreme event;

[0052] The preset risk assessment formula is:

[0053] ;

[0054] in, For urban areas or populations exposed to combined extreme events of dry-heat-to-humidity, that is, risk assessment values; is the grid event frequency of the w-th grid in the corresponding time period and scenario; is the population area or urban area of grid w; n is the total number of grid cells;

[0055] All of the risk assessment values are used to construct future potential risk assessment data.

[0056] The present invention also provides a rapid dry-heat-to-humidity composite extreme weather prediction system, comprising:

[0057] An event data generation module is used to obtain climate model data corresponding to the detection area, downscale the climate model data using a bilinear interpolation method and an equidistant cumulative distribution function method, and use a multifractal detrended fluctuation analysis method to screen event thresholds and composite extreme events on the corrected climate model data to generate event threshold data, historical composite extreme weather data, and future composite extreme weather data;

[0058] An event characteristic change data generation module is used to calculate the characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data based on the event threshold data, and generate historical event characteristic change data and future event characteristic change data;

[0059] A dependency evaluation data generation module, configured to perform dependency evaluation based on the historical composite extreme weather data and the future composite extreme weather data to generate dependency evaluation data;

[0060] a future potential risk assessment data generation module, configured to perform future potential risk assessments on the dry-heat-to-humidity composite extreme event based on the population and urban expansion data corresponding to the detection area, and generate future potential risk assessment data;

[0061] The extreme weather prediction data construction module is used to use the historical event characteristic change data, the future event characteristic change data, the dependency assessment data and the future potential risk assessment data to construct the extreme weather prediction data corresponding to the detection area.

[0062] The present invention also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of implementing any of the above-mentioned methods for rapid dry-heat-to-humid composite extreme weather prediction.

[0063] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements any of the above-mentioned methods for rapid dry-heat-to-humidity composite extreme weather prediction.

[0064] It can be seen from the above technical solutions that the present invention has the following advantages:

[0065] This invention addresses the coarse spatial resolution and accuracy issues of global climate models (GCMs) by downscaling climate model data using bilinear interpolation and equidistant cumulative distribution function methods. It also employs an objective multifractal detrended fluctuation analysis method to determine event thresholds, which are unaffected by subjective influences. This method calculates historical event characteristic change data, future event characteristic change data, dependency assessment data, and future potential risk assessment data. Compared to existing prediction methods that are subject to subjective influences on temporal and spatial thresholds for composite extreme events and fail to consider the interactions and dependencies between composite extreme events, this application fully considers the interactions and dependencies between the values of various meteorological elements in composite extreme events, while also considering the occurrence characteristics and potential risks of such events under future scenario models. This improves the accuracy of predictions for high-risk, rapid dry-hot-to-humid composite extreme weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0067] Figure 1 A flowchart of the steps of a rapid dry-heat-to-humidity composite extreme weather prediction method provided in Example 1 of the present invention;

[0068] Figure 2 A flowchart of the steps of a rapid dry-heat-to-humidity composite extreme weather prediction method provided in Example 2 of the present invention;

[0069] Figure 3 Schematic diagram of the frequency (ac), duration (df) and intensity (gi) of DHW events in the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0070] Figure 4 The probability density diagram of the frequency of DHW events in the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0071] Figure 5 The probability density map of the duration of DHW events in the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0072] Figure 6The probability density map of the intensity of DHW events in the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0073] Figure 7 Schematic diagram of the exposure of the Chinese population to DHW events in the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0074] Figure 8 A time series diagram of the exposure of the Chinese population to DHW events under the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0075] Figure 9 A time series diagram of the exposure of urban areas to DHW events under the historical period, SSP245 and SSP585 scenarios provided in Example 2 of the present invention;

[0076] Figure 10 This is a structural block diagram of a rapid dry-heat-to-humidity composite extreme weather prediction system provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0077] While current research has primarily focused on dry, hot, and humid events, exploration of abrupt transitions from dry, hot, and humid to wet events has been limited. This is particularly true in future scenario simulations, which lack bias-corrected CMIP6 meteorological data to predict the characteristics of such events in countries with rapid urbanization and population growth. For countries with rapid urbanization and population growth, close attention should be paid to historical trends and future projections of such continuous DHW events. Assessing locations and populations at high risk from dry, hot, and humid events is crucial for supporting regional disaster prevention and mitigation decisions.

[0078] Existing extreme weather prediction methods do not fully consider this rapid dry-hot-humid transition event. In other words, the prediction of rapid dry-hot-humid combined extreme weather lacks the following technical details:

[0079] Insufficient consideration has been given to the appropriate setting of temporal and spatial thresholds for such extreme events;

[0080] The interactions and dependencies between these extreme events are not considered;

[0081] The occurrence characteristics and potential risks of such events in future scenario models are not taken into account.

[0082] Therefore, an embodiment of the present invention provides a method and system for predicting rapid dry-hot-to-humid composite extreme weather, which is used to solve the technical problem that the existing prediction methods are affected by people's subjective consciousness in setting the time and space thresholds of composite extreme events, and do not consider the interactions and dependencies of composite extreme events, resulting in low accuracy of prediction results.

[0083] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions 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 embodiments described below 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 making creative work are within the scope of protection of the present invention.

[0084] Example 1

[0085] See also Figure 1 , Figure 1 This is a flowchart of the steps of a method for rapid dry-heat-to-humidity composite extreme weather prediction provided in Example 1 of the present invention.

[0086] A method for rapid dry-hot-to-humidity composite extreme weather forecasting provided in Example 1 of the present invention includes:

[0087] Step 101: Acquire climate model data corresponding to the detection area, use bilinear interpolation method and equidistant cumulative distribution function method to downscale the climate model data for bias correction, and use multifractal detrended fluctuation analysis method to screen event thresholds and composite extreme events on the corrected climate model data to generate event threshold data, historical composite extreme weather data, and future composite extreme weather data.

[0088] In an embodiment of the present invention, climate model data corresponding to the detection area is obtained, and the climate model data includes historical climate data of the detection area, data of six CMIP6 models of SSP245 and SSP585. A bilinear interpolation method and an equidistant cumulative distribution function method are used to perform data downscaling bias correction and data partitioning on the climate model data to generate target historical climate model data and target future climate model data. A multifractal detrended fluctuation analysis method is used to perform event threshold screening on the target historical climate model data and the target future climate model data to generate event threshold data. According to the event threshold data and the preset dry-heat-to-humid composite extreme event definition data, the target historical climate model data and the target future climate model data are respectively screened for dry-heat-to-humid composite extreme events to generate historical composite extreme weather data and future composite extreme weather data.

[0089] Step 102: Calculate the characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data based on the event threshold data to generate historical event characteristic change data and future event characteristic change data.

[0090] In an embodiment of the present invention, the total number of dry-hot-to-humid composite extreme events in historical composite extreme weather data is used as the historical composite extreme event frequency. The total number of dry-hot-to-humid composite extreme events in future composite extreme weather data is used as the future composite extreme event frequency. The sum of the duration of the dry-hot period and the duration of the wet period in the historical composite extreme weather data is calculated to generate the historical composite extreme event duration. The sum of the duration of the dry-hot period and the duration of the wet period in the future composite extreme weather data is calculated to generate the future composite extreme event duration. Meteorological data, drought event determination thresholds, heat event determination thresholds, and wet event determination thresholds in the historical composite extreme weather data and future composite extreme weather data are substituted into preset time intensity calculation formulas to calculate the historical composite extreme event intensity and the future composite extreme event intensity. The historical composite extreme event frequency, historical composite extreme event duration, and historical composite extreme event intensity are used to construct historical event characteristic change data corresponding to the historical composite extreme weather data. The future composite extreme event frequency, future composite extreme event duration, and future composite extreme event intensity are used to construct future event characteristic change data corresponding to the future composite extreme weather data.

[0091] Step 103: Perform dependency assessment based on historical composite extreme weather data and future composite extreme weather data to generate dependency assessment data.

[0092] In an embodiment of the present invention, a preset hydrodynamic model is used to calculate the runoff depth corresponding to historical composite extreme weather data and future composite extreme weather data, and multiple runoff indices are generated. A preset temperature probability density function is used to calculate the temperature corresponding to historical composite extreme weather data and future composite extreme weather data, and multiple temperature indices are generated. A preset precipitation probability density function is used to calculate the precipitation corresponding to historical composite extreme weather data and future composite extreme weather data, and multiple precipitation indices are generated. The runoff index, temperature index, precipitation index, and event threshold data are respectively substituted into the preset correlation probability formula to calculate multiple correlation probabilities. The conditional probability and single probability corresponding to the correlation probability are substituted into the preset probability multiplication factor formula to calculate the probability multiplication factor corresponding to the dry-hot-to-wet composite extreme event.

[0093] Step 104 : Based on the population and urban expansion data corresponding to the detection area, a future potential risk assessment is performed on the dry-hot-to-humid composite extreme event to generate future potential risk assessment data.

[0094] In this embodiment of the present invention, population and urban expansion data includes the population area, urban area, and total number of grid cells corresponding to each grid. The grid event frequency, population area, urban area, and total number of grid cells corresponding to the combined dry-heat-to-humidity extreme event are substituted into a preset risk assessment formula to calculate a risk assessment value for the combined dry-heat-to-humidity extreme event. All risk assessment values are used to construct future potential risk assessment data.

[0095] Step 105: Use historical event feature change data, future event feature change data, dependency assessment data, and future potential risk assessment data to construct extreme weather forecast data corresponding to the detection area.

[0096] In this embodiment of the present invention, after calculating historical and future event characteristic change data corresponding to DHW events in the detection area for historical periods and future scenarios, as well as calculating dependency assessment data and future potential risk assessment data for each DHW event in the detection area, all of this calculated data is used to construct extreme weather forecast data for the detection area. This enables the prediction of rapid, large-scale, combined extreme weather events, such as dry-hot to wet weather, in both space and time.

[0097] In an embodiment of the present invention, climate model data corresponding to a detection area is obtained, and the climate model data is downscaled and bias-corrected using a bilinear interpolation method and an equidistant cumulative distribution function method. The corrected climate model data is then screened for event thresholds and composite extreme events using a multifractal detrended fluctuation analysis method, generating event threshold data, historical composite extreme weather data, and future composite extreme weather data. Based on the event threshold data, characteristic change data for each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data are calculated to generate historical event characteristic change data and future event characteristic change data. A dependency assessment is performed based on the historical composite extreme weather data and the future composite extreme weather data to generate dependency assessment data. Based on the population and urban expansion data corresponding to the detection area, a future potential risk assessment is performed for the dry-hot-to-humid composite extreme event to generate future potential risk assessment data. Extreme weather forecast data corresponding to the detection area is constructed using the historical event characteristic change data, the future event characteristic change data, the dependency assessment data, and the future potential risk assessment data. By using bilinear interpolation and equidistant cumulative distribution function methods to downscale climate model data, the authors address the coarse spatial resolution and accuracy issues of Global Climate Models (GCMs). The objective multifractal detrended fluctuation analysis method is used to determine event thresholds, overcoming the problem of inaccurate threshold settings due to regional or other factors, and avoiding the influence of subjective consciousness on the determination process.

[0098] Compared with existing prediction methods, the time and space threshold settings of compound extreme events are affected by people's subjective consciousness and do not take into account the interactions and dependencies of compound extreme events, resulting in low accuracy of prediction results. In this embodiment, by calculating historical event feature change data, future event feature change data, dependency assessment data and future potential risk assessment data, the interactions and dependencies of compound extreme events are fully considered, as well as the occurrence characteristics and potential risks of such events under future scenario models, thereby effectively improving the accuracy of prediction results.

[0099] Example 2

[0100] See also Figure 2 , Figure 2 This is a flowchart of the steps of a method for rapid dry-heat-to-humidity composite extreme weather prediction provided in Example 2 of the present invention.

[0101] Another method for rapid dry-heat-to-humidity composite extreme weather prediction provided by Example 2 of the present invention includes:

[0102] Step 201: Acquire climate model data corresponding to the detection area, use bilinear interpolation method and equidistant cumulative distribution function method to perform data downscaling bias correction and data partitioning on the climate model data, and generate target historical climate model data and target future climate model data.

[0103] Furthermore, step 201 may include the following sub-steps S11-S14:

[0104] S11. Use a bilinear interpolation method to downscale the initial preset meteorological forcing dataset and climate model data to generate a target preset meteorological forcing dataset and initial climate model data.

[0105] S12. Divide the initial climate model data according to preset year intervals to generate intermediate historical climate model data and intermediate future climate model data.

[0106] S13. Using an equidistant cumulative distribution function method, deriving a cumulative distribution function corresponding to a target preset meteorological forcing data set, and generating a cumulative distribution function.

[0107] S14. Using the cumulative distribution function corresponding to the target preset meteorological forcing data set, bias correction is performed on the intermediate historical climate model data and the intermediate future climate model data to generate target historical climate model data and target future climate model data.

[0108] The initial preset meteorological forcing dataset is the high-resolution and high-precision China Meteorological Forcing Dataset (CMFD) (0.1° / 3h). The target preset meteorological forcing dataset is the China Meteorological Forcing Dataset (CMFD) with adjusted resolution (0.25° / 3h).

[0109] The preset year range is usually set to 1979-2020 as the historical period and 2021-2100 as the future period.

[0110] In an embodiment of the present invention, the detection of DHW events is not limited to history, but further simulations are also made for the future. In order to solve the problems of coarse spatial resolution and low precision of multiple global climate models (GCMs), an equidistant cumulative statistical downscaling method, i.e., an equidistant cumulative distribution function method (EDCDFm), is adopted. First, the spatial resolution of the high-resolution and high-precision China Meteorological Forcing Dataset (CMFD) (0.1° / 3h), i.e., the initial preset meteorological forcing data set and climate model data, is adjusted by bilinear interpolation method, and the resolution corresponding to these data is adjusted from 0.1° to 0.25°, respectively, to obtain the target preset meteorological forcing data set and the initial climate model data. The initial climate model data is divided into intermediate historical climate model data and intermediate future climate model data according to 1979-2020 as the historical period and 2021-2100 as the future period. Then, the cumulative distribution function derived from the corresponding CMFD grid sequence is used, i.e., the equidistant cumulative distribution function method is used to derive the cumulative distribution function corresponding to the target preset meteorological forcing data set. This cumulative distribution function was used to adjust the daily maximum and minimum temperatures, and daily precipitation, of each GCM on a grid-based basis. This cumulative distribution function was then used to perform bias correction on the intermediate historical and future climate model data, generating target historical and future climate model data. This method was applied to data from six CMIP6 models from SSP245 and SSP585, effectively retaining the capture of extreme values of climate elements and thus improving the accuracy of model downscaling and simulation. The target historical climate model data included climate element values corresponding to the historical climate data of the test area and climate element values for the historical period under the SSP245 and SSP585 scenarios. The target future climate model data included climate element values for the future period under the SSP245 and SSP585 scenarios. Parameters were estimated using the 1979–2020 series for the historical period and the 2021–2100 series for the future period. The formulas for calculating climate element values for the historical and future periods are shown below, respectively:

[0111] ;

[0112] ;

[0113] in, are the values of climate elements in the historical period after bias correction; is the bias-corrected value of the climate element for the future period; is the quantile function obeyed by the measured data; is the quantile function that the pattern data obeys; is the cumulative distribution function for the historical period; is the cumulative distribution function for the future period; is the meteorological element value in the historical period; is the meteorological element value for the future period.

[0114] Step 202: Use a multifractal detrended fluctuation analysis method to perform event threshold screening on the target historical climate model data and the target future climate model data to generate event threshold data.

[0115] Furthermore, step 202 may include the following sub-steps S21-S24:

[0116] S21. Detrend the target historical climate model data and the target future climate model data to construct a time series.

[0117] S22. Calculate the fluctuation amplitudes of the time series at multiple preset time scales respectively to generate multiple multifractal features.

[0118] S23. Divide the sequences corresponding to the multifractal features according to preset intervals to generate multiple divided sequences.

[0119] S24. When the long-term correlation coefficient corresponding to the divided sequence has a preset coefficient, the divided interval corresponding to the preset coefficient is used as event threshold data.

[0120] In an embodiment of the present invention, when determining the threshold, 1979-2020 is used as the historical period, and 2021-2100 is used as the future period. The trend fluctuation analysis (MF-DFA) method is used to extract the optimal threshold for each grid corresponding to the detection area. For determining the threshold of extreme events, most existing studies use some established precipitation percentiles as the threshold of extreme precipitation, such as the 95% or 99% percentile. However, the method of determining extreme events by percentile thresholds is greatly influenced by people's subjective consciousness and ignores the changing characteristics of the precipitation series itself. Therefore, the multifractal detrended fluctuation analysis (MF-DFA) method is used in the invention to extract event threshold data, and the event threshold data includes the extreme precipitation threshold, extreme temperature threshold, extreme runoff threshold, drought event determination threshold, heat event determination threshold and wet event determination threshold of each grid corresponding to the detection area. MF-DFA can analyze the multifractal characteristics at different scales in the time series, and considers different weights in the fluctuation analysis to capture more complex fluctuation characteristics, with good objectivity. The basic steps are as follows:

[0121] 1. Detrend the target historical climate model data and target future climate model data to eliminate the trend component in the time series.

[0122] 2. Calculate the detrended fluctuation amplitude at different time scales and analyze the multifractal characteristics of the fluctuation characteristics. Specifically, calculate the fluctuation amplitude of the time series at multiple preset time scales to generate multiple multifractal characteristics. Then, when different intervals are set, the long-term correlation coefficient changes. This means that the series corresponding to the multifractal characteristics is partitioned according to the preset intervals, generating multiple partitioned sequences.

[0123] 3. However, when a certain threshold is reached, the long-term correlation coefficient no longer changes. The first corresponding value that does not change is the extreme event threshold, which is ultimately used to determine the extreme event threshold. When the long-term correlation coefficient corresponding to the partition sequence has a preset coefficient, the partition interval corresponding to the preset coefficient is used as the event threshold data, where the preset coefficient refers to the long-term correlation coefficient that no longer changes.

[0124] For example, the specific extraction process of extreme precipitation threshold is as follows:

[0125] For a sequence of length N { , k=1, 2,…, N}, the MF-DFA method can be divided into five steps.

[0126] (1) Create a new sequence:

[0127] ;

[0128] in, is a time series; is the kth climate model data; is the mean of the climate model data; i is the time, i=1, 2,…, N.

[0129] (2) The new time series Divide into non-overlapping equal-length subintervals of length s, and divide the sequence of length N into = int subintervals (N / s). Since the sequence length N is not necessarily divisible by the subinterval length s, in order to ensure that the original sequence information is not lost, we can generate another group starting from the end of the sequence, so we can get a total of 2 subinterval of .

[0130] (3) For each subinterval v(v=1, 2, ..., 2 ) data to perform polynomial regression fitting and obtain the local trend function , which can be a first-order, second-order or higher-order polynomial, respectively denoted as DFA1, DFA2, etc. In each subinterval, the trend is eliminated and the mean square error is The calculation is as follows:

[0131] ;

[0132] (4) Determine the q-order wave function of the complete time series using the following formula: , where q can be any non-zero real number. When q=0, the wave function is .

[0133] ;

[0134] (5) By analyzing the double logarithmic coordinate graph and relationship, in which is the generalized Hurst index. The scaling index of the wave function can be determined That is, the multifractal characteristics, the scaling exponent of the wave function It can be determined by the following formula.

[0135] ;

[0136] in, In the threshold calculation for this patent, a second-order polynomial is used to perform regression fitting on the data of each sub-interval v (v = 1, 2, ..., 2 ) and q = 2. This study does not discuss the meaning of the physical DFA index (i.e., the long-term correlation coefficient), but uses it as a parameter to measure physical properties, discuss the impact of extreme values on the system DFA index, and use this as a basis to determine the threshold of extreme events. The core idea is that when based on different variable intervals J, the long-term correlation coefficient However, when J reaches a certain threshold, No longer changes, the first J means is the threshold value of extreme variables.

[0137] Step 203 : Screen the target historical climate model data and the target future climate model data for dry-heat-to-humidity composite extreme events according to the event threshold data and the preset dry-heat-to-humidity composite extreme event definition data, and generate historical composite extreme weather data and future composite extreme weather data.

[0138] In an embodiment of the present invention, the preset dry-heat-to-humid composite extreme event definition data is defined as continuous dry-heat-to-humidity, which is defined as a dry-heat extreme duration of no less than 3 days in the three-dimensional space of latitude × longitude × time, followed by a flood event exceeding a certain threshold within 5 days. A dry-heat-to-humid composite extreme event with a total duration (≥7 days) and covering a certain area (≥1 grid) is considered a dry-heat-to-humid composite extreme event. According to the preset dry-heat-to-humid composite extreme event definition data, a spatiotemporal continuous event tracking (SCET) method using the implemented 3D connected component (CC3D) labeling algorithm is used to track three-dimensional (3D) continuous DHW events from gridded daily temperature, precipitation, and runoff, and these values are stored in a three-dimensional 3D array. Using a long-term series of parameters (precipitation, temperature, and runoff) from 1979 to 2020 as a benchmark, a 15-day window is set. If the daily temperature, precipitation, and runoff at a grid point exceeds the corresponding extreme precipitation threshold, extreme temperature threshold, and extreme runoff threshold, the grid cell in the three-dimensional array (i.e., the latitude × longitude × time three-dimensional grid point) is first marked as "1." If it does not exceed the threshold, it is marked as "0." This filtering method allows for the simultaneous tracking of continuous DHW events in space and time. The corresponding filtering formula for DHW events is as follows:

[0139] ;

[0140] in, is the Heaviside function, x>0, =1, x≤0, =0; is the number of the i-th day of the total number of wet events; is the number of the jth day in the total number of dry and hot events; is the total number of wet events; is the total number of dry heat events; =3 represents the duration of dry heat, which means that the dry heat time lasts for at least 3 days; window The duration of the wet event in the event has a length of 5, indicating that the flood event occurs at least 3 days after the dry and hot event.

[0141] The DHW events from 1979 to 2020 were used to construct historical composite extreme weather data, while the DHW events from 2021 to 2100 were used to construct future composite extreme weather data.

[0142] Step 204 : Calculate the characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data based on the event threshold data to generate historical event characteristic change data and future event characteristic change data.

[0143] Furthermore, the event threshold data includes a drought event determination threshold, a heat event determination threshold, and a wet event determination threshold. Step 204 may include the following sub-steps S31-S37:

[0144] S31. The total number of dry-hot-to-humid composite extreme events in the historical composite extreme weather data is taken as the frequency of historical composite extreme events.

[0145] S32. The total number of dry-hot-to-humid composite extreme events in the future composite extreme weather data is used as the frequency of future composite extreme events.

[0146] S33. Calculate the sum of the duration of the dry and hot period and the duration of the wet period in the historical composite extreme weather data to generate the duration of the historical composite extreme event.

[0147] S34. Calculate the sum of the duration of the dry and hot period and the duration of the wet period in the future composite extreme weather data to generate the duration of the future composite extreme event.

[0148] S35. Substitute the meteorological data, drought event determination threshold, heat event determination threshold, and wet event determination threshold in the historical composite extreme weather data and the future composite extreme weather data into the preset time intensity calculation formula to calculate the historical composite extreme event intensity and the future composite extreme event intensity.

[0149] S36. Use the historical composite extreme event frequency, historical composite extreme event duration, and historical composite extreme event intensity to construct historical event characteristic change data corresponding to historical composite extreme weather data.

[0150] S37. Use the frequency, duration and intensity of future composite extreme events to construct future event characteristic change data corresponding to future composite extreme weather data.

[0151] In the embodiment of the present invention, the total number of historical and future DHW events is defined as the historical and future frequencies, and the historical composite extreme event frequency and the future composite extreme event frequency are obtained. The historical composite extreme event frequency and the future composite extreme event frequency are as follows: Figure 3 As shown in (a), (b) and (c) in the figure, the frequency of composite extreme events corresponding to the detection area includes the frequency of composite extreme events under the historical scenario, SSP245 scenario and SSP585 scenario in the detection area. The corresponding probability density maps are shown in Figure 4 shown.

[0152] The sum of the duration of the dry and hot period and the duration of the wet period is defined as the total duration, and the duration of historical composite extreme events and the duration of future composite extreme events are obtained. The duration of historical composite extreme events and the duration of future composite extreme events are as follows: Figure 3 As shown in (d), (e) and (f) in the figure, the duration of the composite extreme event corresponding to the detection area includes the duration of the composite extreme event under the historical scenario, SSP245 scenario and SSP585 scenario in the detection area. The corresponding probability density maps are shown in Figure 5 shown.

[0153] By substituting the meteorological data, drought event determination threshold, heat event determination threshold and wet event determination threshold in the historical composite extreme weather data and future composite extreme weather data into the preset time intensity calculation formula, the historical composite extreme event intensity and future composite extreme event intensity are calculated. Figure 3 As shown in (g), (h) and (i), the composite extreme event intensity corresponding to the detection area includes the composite extreme event intensity under the historical scenario, SSP245 scenario and SSP585 scenario in the detection area. The corresponding probability density maps are shown in Figure 6 shown.

[0154] The preset time intensity calculation formula is:

[0155] ;

[0156] in, is the event intensity; D is the daily precipitation; T is the daily maximum temperature; R is the runoff; is the precipitation on day i; is the maximum temperature on day i; is the runoff volume of the jth section; Determine the precipitation corresponding to the threshold for drought events; Determine the maximum temperature corresponding to the threshold for the heat event; Determine the runoff volume corresponding to the threshold for the wet event; is the duration of dry heat, =3 represents the duration of dry heat, which means that the dry heat lasted for at least 3 days in the early stage; is the duration of wet events in the composite extreme weather data, and the window is the duration of the wet event in the event, with a length of 5, indicating that the flood event occurred after the dry and hot event lasted for at least 3 days; establishing thresholds for drought events; Determine thresholds for thermal events; Determines the threshold value for the wet event.

[0157] Finally, the calculated historical composite extreme event frequency, duration, and intensity are used to construct the historical event characteristic change data corresponding to the historical composite extreme weather data. The calculated future composite extreme event frequency, duration, and intensity are used to construct the future event characteristic change data corresponding to the future composite extreme weather data.

[0158] Step 205: Perform dependency assessment based on historical composite extreme weather data and future composite extreme weather data to generate dependency assessment data.

[0159] Furthermore, step 205 may include the following sub-steps S41-S46:

[0160] S41. Use a preset hydrodynamic model to calculate the runoff depth corresponding to historical composite extreme weather data and future composite extreme weather data, and generate multiple runoff indices.

[0161] S42. Calculate the temperatures corresponding to historical composite extreme weather data and future composite extreme weather data using a preset temperature probability density function to generate multiple temperature indices.

[0162] S43. Calculate the precipitation corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset precipitation probability density function, and generate multiple precipitation indices.

[0163] S44. Substitute the runoff index, temperature index, precipitation index and event threshold data into the preset correlation probability formula respectively to calculate and obtain multiple correlation probabilities.

[0164] S45. Substitute the conditional probability and the single probability corresponding to the relevant probability into the preset probability multiplication factor formula to calculate the probability multiplication factor corresponding to the dry-hot-to-humid composite extreme event.

[0165] S46. Use all probability multiplication factors to construct dependency assessment data.

[0166] The default hydrodynamic model is the global hydrodynamic model CaMa-Flood model. The default temperature probability density function is the four-parameter Beta function. The default precipitation probability density function is the mixed Gamma distribution function.

[0167] In this embodiment of the present invention, after downscaling bias correction, the VIC hydrological model driven by six GCM data was used to simulate daily and grid-by-grid runoff depths in China from 1979 to 2100. This represents a runoff generation model, with runoff generation occurring directly at each grid. A preset hydrodynamic model was used to calculate the runoff depth corresponding to historical and future composite extreme weather data, yielding a runoff index. Specifically, the global hydrodynamic model (CaMa-Flood) was used to capture runoff data within a specific river network to calculate the runoff depth for each grid, yielding the corresponding runoff index. The calculation formula for runoff depth is R = QΔt / 1000A, where R is the runoff depth, Δt (s) is the time period, Q is the average flow rate, and A is the drainage area. The CaMa-Flood model, a global hydrodynamic model, captures runoff data within a specific river network, representing a confluence model. The previous model is grid data, with water now flowing downwards and converging into the river channel. A four-parameter Beta function was used to calculate the temperature corresponding to each grid in the historical and future composite extreme weather data, yielding the corresponding temperature index. The preset precipitation probability density function is used to calculate the precipitation corresponding to each grid in the historical composite extreme weather data and future composite extreme weather data, and the corresponding precipitation index is generated.

[0168] The probability of wet events under the combined dry and hot conditions is calculated using the multivariate distribution established by the Copula model. The runoff index, temperature index, precipitation index, and event threshold data are substituted into the preset correlation probability formula to calculate multiple correlation probabilities. The correlation probability can be expressed as:

[0169] ;

[0170] Where P is the probability of occurrence of DHW events; D is the precipitation index; T is the temperature index; R is the runoff index; C is the link function, copula function; d is the dry event threshold data; t is the heat event threshold data; r is the wet event threshold data; u is the conditional distribution function;

[0171] The probability multiplication factor (PMF) is used as a measure to quantify the strength of the dependence between different variables. The conditional probability and the single probability corresponding to the relevant probability are substituted into the preset probability multiplication factor formula to calculate the probability multiplication factor corresponding to the dry-hot-to-humid composite extreme event.

[0172] The default probability multiplication factor formula is:

[0173] ;

[0174] in, is the probability multiplication factor; is the conditional probability of the dry event, hot event and wet event occurring in sequence; The single probability of a single event occurring.

[0175] For example, if we consider a scenario where dry represents an extreme drought with rainfall below the 10th percentile, and “wet” represents an extreme drought with runoff levels exceeding the 90th percentile, then the PMF index for this dry-hot composite event with high runoff can be calculated as follows:

[0176] ;

[0177] The PMF is a parameter ranging from 0 to infinity. When two hazards are independent, the PMF is equal to 1. When the hazards are positively correlated, the PMF is greater than 1, and its value increases as the strength of the correlation increases. Conversely, for negatively correlated hazards, the PMF ranges from 0 to 1.

[0178] Using all probability multiplication factors, we construct dependency assessment data to calculate the interdependent changes in precipitation, temperature, and runoff during these DHW events. For example, if there is a dry and hot period in the early stage, what is the probability of flooding in the later stage? We also examine whether it is related to the hot or dry events.

[0179] Step 206: Based on the population and urban expansion data corresponding to the detection area, perform future potential risk assessments on the dry-hot-to-humid composite extreme event to generate future potential risk assessment data.

[0180] Furthermore, step 206 may include the following sub-steps S51-S52:

[0181] S51. Substitute the grid event frequency, population area, urban area and total number of grid cells corresponding to the dry-heat-to-humidity composite extreme event into the preset risk assessment formula to calculate the risk assessment value corresponding to the dry-heat-to-humidity composite extreme event.

[0182] S52. Use all risk assessment values to construct future potential risk assessment data.

[0183] In an embodiment of the present invention, the DHW-exposed population and urban area in the present invention represent the number of individuals living in areas susceptible to potential damage caused by DHW events. It is an estimate of the severity of the potential impact that a DHW event may have on a specific area. Exposure to DHW events is calculated as shown in a preset risk assessment formula, which consists of two main components: the frequency of DHW events and the population size (P) or urban area (U). The grid event frequency, population area, urban area, and total number of grid cells corresponding to the dry-heat-to-humidity composite extreme event are substituted into the preset risk assessment formula to calculate the risk assessment value corresponding to the dry-heat-to-humidity composite extreme event, and all risk assessment values are used to construct future potential risk assessment data.

[0184] The default risk assessment formula is:

[0185] ;

[0186] in, For urban areas or populations exposed to combined extreme events of dry-heat-to-humidity, that is, risk assessment values; is the grid event frequency of the w-th grid in the corresponding time period and scenario; is the population area or urban area of grid w; n is the total number of grid cells.

[0187] Step 207: construct extreme weather forecast data corresponding to the detection area using historical event feature change data, future event feature change data, dependency assessment data, and future potential risk assessment data.

[0188] In the embodiment of the present invention, the specific implementation process of step 207 is similar to that of step 105 and will not be repeated here.

[0189] In the embodiment of the present invention, the present invention predicts large-scale, rapid occurrence of dry-hot-to-humid composite events in space and time, rather than just a single extreme event. Based on the SSP245 and SSP585 climate models that have been downscaled after bias correction, an objective extreme weather threshold determination method is used to detect the frequency, duration and intensity change characteristics of DHW in the Chinese region in the historical period (1979-2014) and future scenarios (2015-2100). Daily precipitation and maximum temperature data of the calibrated multi-model ensemble generated by the CMIP6 model, as well as runoff data jointly simulated by the VIC and CaMa-Flood models, are used. The interdependent changes in precipitation, temperature and runoff in these DHW events are explored. Based on dynamic historical and future population and urban expansion data, the driving factors and potential risks of such DHW events are assessed.

[0190] In a validation of historical bias-correction results, at a 0.25° grid scale, the average absolute errors of the three variables for single-model data for monthly precipitation, monthly average daily maximum temperature, and monthly average daily minimum temperature after bias correction from 1985 to 2014 ranged from 0.4 to 4.6 mm / month, 0.06 to 0.1°C / month, and 0.06 to 0.08°C / month. Further validation using runoff simulations revealed that the Nash efficiency coefficient (NSE) values for most watersheds indicated that the simulated monthly runoff accuracy ranged from 0.7 to 0.9. This indicates that the bias-correction significantly improved the accuracy of monthly precipitation, monthly average daily maximum temperature, and monthly average daily minimum temperature. The bias-correction method was validated using historical events, demonstrating its reliability.

[0191] like Figures 3 to 9 As shown, the results show that the spatiotemporal characteristics of DHW events show obvious regional patterns. Compared with the historical period of 1979-2014, the frequency of future DHW events in the entire region decreased by 22% and 25% under the two forecast scenarios (2015-2100), respectively. Although the total duration decreased by about 15 days, the average intensity increased by 7 to 11.4 times. It is worth noting that the frequency and duration of DHW events in the northwest region decreased. However, the frequency, duration and intensity of DHW in the middle and lower plains of the Yangtze River Basin and the Yellow River Basin increased significantly. In addition, the possibility of short-term high-intensity DHW events is amplified and is expected to occur in the Yangtze River and Yellow River Basins around 2023, 2038 and 2058. As Figures 8 and 9 As shown in Figure 3, the main drivers of the increased exposure risk are the increased frequency of DHW events in the plain areas of the middle and lower reaches of the Yangtze and Yellow River basins, and the simultaneous growth of the population in these areas.

[0192] Preferably, the invention can also consider the following points based on existing research:

[0193] Current research uses precipitation, air temperature, and runoff to calculate and characterize dryness, heat, and humidity. This can be replaced with parameters such as soil moisture, evapotranspiration, surface temperature, drought index, and flood index to characterize the transition from dryness to humidity. This means improving the parameters for dryness, heat, and humidity.

[0194] The current time scale of events is daily, but the fast conversion process can also be characterized by hourly and minutely scales, and can also be reflected to a certain extent on the monthly scale. That is, by changing the time scale for calculation, the results are more precise, which is also the core content of the present invention.

[0195] Example 3

[0196] See also Figure 10 , Figure 10 This is a structural block diagram of a rapid dry-heat-to-humidity composite extreme weather prediction system provided in Example 3 of the present invention.

[0197] A rapid dry-heat-to-humidity composite extreme weather forecasting system provided in Example 3 of the present invention includes:

[0198] The event data generation module 1001 is used to obtain the climate model data corresponding to the detection area, use the bilinear interpolation method and the equidistant cumulative distribution function method to downscale the climate model data for bias correction, and use the multifractal detrended fluctuation analysis method to screen the event thresholds and composite extreme events of the corrected climate model data to generate event threshold data, historical composite extreme weather data and future composite extreme weather data.

[0199] The event characteristic change data generation module 1002 is used to calculate the characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data based on the event threshold data, and generate historical event characteristic change data and future event characteristic change data.

[0200] The dependency evaluation data generating module 1003 is configured to perform dependency evaluation based on historical composite extreme weather data and future composite extreme weather data to generate dependency evaluation data.

[0201] The future potential risk assessment data generation module 1004 is used to perform future potential risk assessments on the dry-hot-to-humid composite extreme event based on the population and urban expansion data corresponding to the detection area, and generate future potential risk assessment data.

[0202] The extreme weather prediction data construction module 1005 is used to construct extreme weather prediction data corresponding to the detection area using historical event feature change data, future event feature change data, dependency assessment data and future potential risk assessment data.

[0203] Optionally, the event data generating module 1001 includes:

[0204] The climate model data generation module is used to use the bilinear interpolation method and the equidistant cumulative distribution function method to perform data downscaling bias correction and data partitioning on the climate model data, and generate target historical climate model data and target future climate model data.

[0205] The event threshold data generation module is used to use the multifractal detrended fluctuation analysis method to perform event threshold screening on the target historical climate model data and the target future climate model data to generate event threshold data.

[0206] The composite extreme weather data generation module is used to screen the target historical climate model data and the target future climate model data for dry-heat-to-humid composite extreme events according to the event threshold data and the preset dry-heat-to-humid composite extreme event definition data, and generate historical composite extreme weather data and future composite extreme weather data.

[0207] Optionally, the future climate model data generation module may perform the following steps:

[0208] The initial preset meteorological forcing dataset and climate model data are downscaled using the bilinear interpolation method to generate the target preset meteorological forcing dataset and initial climate model data.

[0209] Divide the initial climate model data into data according to preset year intervals to generate intermediate historical climate model data and intermediate future climate model data;

[0210] The equidistant cumulative distribution function method is used to derive the cumulative distribution function corresponding to the target preset meteorological forcing data set to generate a cumulative distribution function;

[0211] The cumulative distribution function is used to perform bias correction on the intermediate historical climate model data and the intermediate future climate model data to generate the target historical climate model data and the target future climate model data.

[0212] Optionally, the event threshold data generation module may perform the following steps:

[0213] Detrending the target historical climate model data and the target future climate model data to construct a time series;

[0214] Calculate the fluctuation amplitude of the time series at multiple preset time scales and generate multiple multifractal features;

[0215] Divide the sequences corresponding to the multifractal features according to preset intervals to generate multiple divided sequences;

[0216] When the long-term correlation coefficient corresponding to the divided sequence has a preset coefficient, the divided interval corresponding to the preset coefficient is used as event threshold data.

[0217] Optionally, the event threshold data includes a drought event determination threshold, a heat event determination threshold, and a wet event determination threshold. The event characteristic change data generation module 1002 may perform the following steps:

[0218] The total number of dry-hot-to-humid composite extreme events in the historical composite extreme weather data is taken as the frequency of historical composite extreme events;

[0219] The total number of dry-hot-to-humid composite extreme events in future composite extreme weather data is used as the frequency of future composite extreme events;

[0220] Calculate the sum of the duration of dry and hot periods and the duration of wet periods in historical composite extreme weather data to generate the duration of historical composite extreme events;

[0221] Calculate the sum of the duration of dry and hot periods and the duration of wet periods in future composite extreme weather data to generate the duration of future composite extreme events;

[0222] Substitute meteorological data, drought event determination threshold, heat event determination threshold, and wet event determination threshold in historical composite extreme weather data and future composite extreme weather data into the preset time intensity calculation formula to calculate the historical composite extreme event intensity and future composite extreme event intensity;

[0223] The preset time intensity calculation formula is:

[0224] ;

[0225] in, is the event intensity; is the precipitation on day i; is the maximum temperature on day i; is the runoff volume of the jth section; Determine the precipitation corresponding to the threshold for drought events; Determine the maximum temperature corresponding to the threshold for the heat event; Determine the runoff volume corresponding to the threshold for the wet event; is the duration of dry heat; is the duration of wet events in the composite extreme weather data; establishing thresholds for drought events; Determine thresholds for thermal events; Determine thresholds for wet events;

[0226] The historical event characteristic change data corresponding to the historical composite extreme weather data are constructed using the historical composite extreme event frequency, historical composite extreme event duration and historical composite extreme event intensity;

[0227] The frequency, duration and intensity of future composite extreme events are used to construct the future event characteristic change data corresponding to the future composite extreme weather data.

[0228] Optionally, the dependency evaluation data generating module 1003 may perform the following steps:

[0229] A preset hydrodynamic model is used to calculate the runoff depth corresponding to historical and future composite extreme weather data, generating multiple runoff indices.

[0230] The preset temperature probability density function is used to calculate the temperature corresponding to historical composite extreme weather data and future composite extreme weather data to generate multiple temperature indices;

[0231] The preset precipitation probability density function is used to calculate the precipitation corresponding to historical composite extreme weather data and future composite extreme weather data, and multiple precipitation indices are generated;

[0232] Substitute the runoff index, temperature index, precipitation index and event threshold data into the preset correlation probability formula respectively to calculate multiple correlation probabilities;

[0233] The default correlation probability formula is:

[0234] ;

[0235] Where P is the probability of occurrence of DHW events; D is the precipitation index; T is the temperature index; R is the runoff index; C is the connection function; d is the dry event threshold data; t is the heat event threshold data; r is the wet event threshold data; u is the conditional distribution function;

[0236] Substitute the conditional probability and single probability corresponding to the relevant probability into the preset probability multiplication factor formula to calculate the probability multiplication factor corresponding to the dry-hot-to-humid composite extreme event;

[0237] The default probability multiplication factor formula is:

[0238] ;

[0239] in, is the probability multiplication factor; is the conditional probability of three events occurring in sequence; is the single probability of a single event occurring;

[0240] All probability multiplication factors are used to construct dependency assessment data.

[0241] Optionally, the future potential risk assessment data generation module 1004 may perform the following steps:

[0242] Substitute the grid event frequency, population area, urban area, and total number of grid cells corresponding to the dry-heat-to-humidity composite extreme event into the preset risk assessment formula to calculate the risk assessment value corresponding to the dry-heat-to-humidity composite extreme event;

[0243] The default risk assessment formula is:

[0244] ;

[0245] in, For urban areas or populations exposed to combined extreme events of dry-heat-to-humidity, that is, risk assessment values; is the grid event frequency of the w-th grid in the corresponding time period and scenario; is the population area or urban area of grid w; n is the total number of grid cells;

[0246] Use all risk assessment values to construct future potential risk assessment data.

[0247] An embodiment of the present invention also provides an electronic device, which includes: a memory and a processor, wherein a computer program is stored in the memory; when the computer program is executed by the processor, the processor executes the rapid dry-hot-to-humid composite extreme weather prediction method as described in any of the above embodiments.

[0248] The memory may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, a hard disk, or ROM. The memory has storage space for program code for executing any of the method steps described above. For example, the storage space for program code may include individual program codes for implementing various steps in the method described above. These program codes may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. The program code may be compressed, for example, in a suitable format. When executed by a computing device, these codes cause the computing device to execute the various steps of the rapid dry-hot-to-humidity combined extreme weather prediction method described above.

[0249] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for predicting rapid dry-heat-to-humid composite extreme weather as described in any of the above embodiments is implemented.

[0250] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0251] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0252] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0253] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0254] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0255] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A rapid dry-hot-to-humid composite extreme weather forecasting method, characterized in that: include: Obtain climate model data corresponding to the detection area, use bilinear interpolation and equidistant cumulative distribution function methods to downscale the climate model data for bias correction, and use multifractal detrended fluctuation analysis to screen event thresholds and composite extreme events on the corrected climate model data to generate event threshold data, historical composite extreme weather data, and future composite extreme weather data; Based on the event threshold data, characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data are calculated respectively to generate historical event characteristic change data and future event characteristic change data; Performing a dependency assessment based on the historical composite extreme weather data and the future composite extreme weather data to generate dependency assessment data; Based on the population and urban expansion data corresponding to the detection area, the future potential risk assessment of the dry-hot-to-humid composite extreme event is performed to generate future potential risk assessment data; constructing extreme weather forecast data corresponding to the detection area using the historical event characteristic change data, the future event characteristic change data, the dependency assessment data, and the future potential risk assessment data; The event threshold data includes a drought event determination threshold, a heat event determination threshold, and a wet event determination threshold; the step of calculating the characteristic change data of each dry-heat-to-wet composite extreme event in the historical composite extreme weather data and the future composite extreme weather data based on the event threshold data to generate the historical event characteristic change data and the future event characteristic change data includes: The total number of dry-hot-to-humid composite extreme events in the historical composite extreme weather data is used as the frequency of historical composite extreme events; The total number of dry-hot-to-humid composite extreme events in the future composite extreme weather data is used as the frequency of future composite extreme events; Calculating the sum of the duration of the dry and hot period and the duration of the wet period in the historical composite extreme weather data to generate the duration of the historical composite extreme event; Calculating the sum of the duration of the dry and hot period and the duration of the wet period in the future composite extreme weather data to generate the duration of the future composite extreme event; Substituting the meteorological data in the historical composite extreme weather data and the future composite extreme weather data, the drought event determination threshold, the heat event determination threshold, and the wet event determination threshold into a preset time intensity calculation formula, respectively, to calculate the historical composite extreme event intensity and the future composite extreme event intensity; The preset time intensity calculation formula is: Where DHWI is the event intensity; D i is the precipitation on the i-th day; T is the maximum temperature on the i-th day; R j is the runoff volume of the jth section; D dryth Determine the precipitation corresponding to the threshold for drought events; T hotth The maximum temperature corresponding to the threshold value for the heat event is determined; R wetth is the runoff corresponding to the threshold value for wet events; λ is the duration of dry and hot weather; ΔT is the duration of wet events in the composite extreme weather data; dryth is the threshold value for drought events; hotth is the threshold value for heat events; wetth is the threshold value for wet events; Constructing historical event characteristic change data corresponding to the historical composite extreme weather data by using the historical composite extreme event frequency, the historical composite extreme event duration, and the historical composite extreme event intensity; The future event characteristic change data corresponding to the future composite extreme weather data are constructed by using the future composite extreme event frequency, the future composite extreme event duration and the future composite extreme event intensity.

2. The rapid dry-heat-to-humidity composite extreme weather prediction method according to claim 1 is characterized in that: The steps of using a bilinear interpolation method and an equidistant cumulative distribution function method to perform downscaling bias correction on the climate model data, and using a multifractal detrended fluctuation analysis method to perform event threshold and composite extreme event screening on the corrected climate model data, and generating event threshold data, historical composite extreme weather data, and future composite extreme weather data, include: Using a bilinear interpolation method and an equidistant cumulative distribution function method to perform data downscaling bias correction and data partitioning on the climate model data, to generate target historical climate model data and target future climate model data; Using a multifractal detrended fluctuation analysis method to perform event threshold screening on the target historical climate model data and the target future climate model data to generate event threshold data; According to the event threshold data and the preset dry-heat-to-humid composite extreme event definition data, the target historical climate model data and the target future climate model data are respectively screened for dry-heat-to-humid composite extreme events to generate historical composite extreme weather data and future composite extreme weather data.

3. The rapid dry-heat-to-humidity composite extreme weather prediction method according to claim 2, characterized in that: The step of using a bilinear interpolation method and an equidistant cumulative distribution function method to perform data downscaling bias correction and data partitioning on the climate model data to generate target historical climate model data and target future climate model data includes: downscaling the initial preset meteorological forcing dataset and the climate model data using a bilinear interpolation method to generate a target preset meteorological forcing dataset and initial climate model data; Dividing the initial climate model data according to preset year intervals to generate intermediate historical climate model data and intermediate future climate model data; The cumulative distribution function corresponding to the target preset meteorological forcing data set is derived by using an equidistant cumulative distribution function method to generate a cumulative distribution function; The intermediate historical climate model data and the intermediate future climate model data are bias-corrected using the cumulative distribution function to generate target historical climate model data and target future climate model data.

4. The rapid dry-heat-to-humidity composite extreme weather prediction method according to claim 2, characterized in that: The step of using a multifractal detrended fluctuation analysis method to perform event threshold screening on the target historical climate model data and the target future climate model data to generate event threshold data includes: Detrending the target historical climate model data and the target future climate model data to construct a time series; respectively calculating the fluctuation amplitudes of the time series at multiple preset time scales to generate multiple multifractal features; Dividing the sequences corresponding to the multifractal features according to preset intervals to generate multiple divided sequences; When the long-term correlation coefficient corresponding to the division sequence has a preset coefficient, the division interval corresponding to the preset coefficient is used as event threshold data.

5. The rapid dry-heat-to-humidity composite extreme weather prediction method according to claim 1, characterized in that: The step of performing dependency assessment based on the historical composite extreme weather data and the future composite extreme weather data to generate dependency assessment data includes: Calculating the runoff depth corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset hydrodynamic model to generate a plurality of runoff indices; Calculating the temperatures corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset temperature probability density function to generate a plurality of temperature indices; Calculating the precipitation corresponding to the historical composite extreme weather data and the future composite extreme weather data using a preset precipitation probability density function to generate a plurality of precipitation indices; Substituting the runoff index, the temperature index, the precipitation index, and the event threshold data into a preset correlation probability formula respectively to calculate a plurality of correlation probabilities; The preset correlation probability formula is: Where P is the probability of occurrence of DHW events; D is the precipitation index; T is the temperature index; R is the runoff index; C is the connection function; d is the dry event threshold data; t is the heat event threshold data; r is the wet event threshold data; u is the conditional distribution function; Substituting the conditional probability and the single probability corresponding to the relevant probability into the preset probability multiplication factor formula, the probability multiplication factor corresponding to the dry-heat-to-humidity composite extreme event is calculated; The preset probability multiplication factor formula is: Among them, PMF is the probability multiplication factor; P f is the conditional probability of three events occurring in sequence; t f is the single probability of a single event occurring; Using all of the described probability multiplication factors, dependency estimation data are constructed.

6. The rapid dry-heat-to-humidity composite extreme weather prediction method according to claim 1, characterized in that: The population and urban expansion data include the population area, urban area, and total number of grid cells corresponding to each grid; and the step of performing a future potential risk assessment for the dry-heat-to-humidity composite extreme event based on the population and urban expansion data corresponding to the detection area to generate future potential risk assessment data includes: Substituting the grid event frequency, population area, urban area, and total number of grid cells corresponding to the dry-heat-to-humidity composite extreme event into a preset risk assessment formula, respectively, to calculate a risk assessment value corresponding to the dry-heat-to-humidity composite extreme event; The preset risk assessment formula is: Puexp is the urban area or population exposed to the combined extreme event of dry-heat-humidity transition, i.e. the risk assessment value; DHPfre w is the grid event frequency of the wth grid in the corresponding period and scenario; PU w is the population area or urban area of grid w; n is the total number of grid cells; All of the risk assessment values are used to construct future potential risk assessment data.

7. A rapid dry-heat-to-humidity composite extreme weather forecasting system, characterized in that: include: An event data generation module is used to obtain climate model data corresponding to the detection area, downscale the climate model data using a bilinear interpolation method and an equidistant cumulative distribution function method, and use a multifractal detrended fluctuation analysis method to screen event thresholds and composite extreme events on the corrected climate model data to generate event threshold data, historical composite extreme weather data, and future composite extreme weather data; An event characteristic change data generation module is used to calculate the characteristic change data of each dry-hot-to-humid composite extreme event in the historical composite extreme weather data and the future composite extreme weather data based on the event threshold data, and generate historical event characteristic change data and future event characteristic change data; a dependency evaluation data generating module, configured to perform dependency evaluation based on the historical composite extreme weather data and the future composite extreme weather data, and generate dependency evaluation data; a future potential risk assessment data generation module, configured to perform future potential risk assessments on the dry-heat-to-humidity composite extreme event based on the population and urban expansion data corresponding to the detection area, and generate future potential risk assessment data; an extreme weather prediction data construction module, configured to construct extreme weather prediction data corresponding to the detection area using the historical event characteristic change data, the future event characteristic change data, the dependency assessment data, and the future potential risk assessment data; The event threshold data includes a drought event determination threshold, a heat event determination threshold, and a wet event determination threshold; The event characteristic change data generation module is specifically configured to use the total number of dry-hot-to-humid composite extreme events in the historical composite extreme weather data as the historical composite extreme event frequency; use the total number of dry-hot-to-humid composite extreme events in the future composite extreme weather data as the future composite extreme event frequency; calculate the sum of the duration of the dry-hot period and the duration of the wet period in the historical composite extreme weather data to generate the duration of the historical composite extreme event; Calculating the sum of the duration of the dry and hot period and the duration of the wet period in the future composite extreme weather data to generate the duration of the future composite extreme event; Substituting the meteorological data, the drought event determination threshold, the heat event determination threshold, and the wet event determination threshold in the historical composite extreme weather data and the future composite extreme weather data into a preset time intensity calculation formula, respectively, to calculate the historical composite extreme event intensity and the future composite extreme event intensity; using the historical composite extreme event frequency, the historical composite extreme event duration, and the historical composite extreme event intensity, to construct historical event characteristic change data corresponding to the historical composite extreme weather data; Using the future composite extreme event frequency, the future composite extreme event duration, and the future composite extreme event intensity, constructing future event characteristic change data corresponding to the future composite extreme weather data; The preset time intensity calculation formula is: Where DHWI is the event intensity; D i is the precipitation on the i-th day; T is the maximum temperature on the i-th day; R j is the runoff volume of the jth section; D dryth Determine the precipitation corresponding to the threshold for drought events; T hotth The maximum temperature corresponding to the threshold value for the heat event is determined; R wetth is the runoff corresponding to the threshold for wet events; λ is the duration of dry and hot weather; ΔT is the duration of wet events in the composite extreme weather data; dryth is the threshold for drought events; hotth is the threshold for heat events; wetth is the threshold for wet events.

8. An electronic device, characterized in that: It includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the rapid dry-hot-to-humid composite extreme weather prediction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the rapid dry-heat-to-humidity composite extreme weather prediction method according to any one of claims 1 to 6 is implemented.

Images