Risk assessment method and device for sea-derived landing flood and storage medium

Abstract

The invention relates to the field of carbon neutralization, and discloses a risk assessment method and device for sea-derived landing flood and a storage medium, and the method comprises the steps: collecting data; identifying a sea-source landing flood event; performing spatio-temporal feature quantification on the sea-derived landing flood; attribution and influence of sea-derived landing flood; carrying out migration and landing analysis and mechanism of sea-derived landing flood; the method has the beneficial effects that the response to artificial climate change is clarified, the physical mechanism of the method is analyzed, the understanding of the sea-source excessive land-direction water vapor transport cascade effect is deepened, and a scientific basis is provided for making a sea-source landing flood disaster prevention and reduction decision.

Description

Technical Field

[0001] The present invention relates to the field of carbon neutrality, and in particular to a risk assessment method, device and storage medium for sea-originated landfall floods. Background Art

[0002] Flood disasters are the most serious meteorological disasters in the world today, causing a large number of casualties and economic losses every year. The surplus of available water in the atmosphere (precipitation minus evaporation) is the basis for the formation of land floods. In the context of anthropogenic climate change, warming caused by anthropogenic emissions is changing the atmospheric moisture content and transport patterns on a global scale. According to the Clausius-Clapeyron equation, the atmospheric water holding capacity is expected to increase with rising temperatures, which may lead to a significant increase in global runoff extremes. On the other hand, the enhanced thermal difference between land and sea caused by anthropogenic warming is driving more ocean water vapor to be transported to land, which means that flood events caused by ocean water vapor may increase. However, the role of climate change caused by human activities in sea-originated landfall flood events has not yet attracted the attention of the academic community, and the physical mechanism of sea-originated landfall flood events has not been studied in detail. Summary of the invention

[0003] The purpose of the present invention is to propose a risk assessment method, device and storage medium for sea-originated landfall floods, so as to solve the technical problem that the physical mechanism of sea-originated landfall floods has not been studied in detail.

[0004] The present invention provides a risk assessment method for sea-sourced landfall floods, comprising the following steps:

[0005] Step S1: Data collection; collect precipitation, evaporation, sea surface temperature, sea level pressure, meridional wind speed, zonal wind speed, specific humidity and surface pressure data from the Weather Forecast Center reanalysis dataset ERA5, the sixth phase of the International Coupled Comparison Program CM IP6, and the large sample set of system models CanESM5; collect land water storage anomaly data from the machine learning-based land water storage dataset GTWS-MLrec; collect mask files for land on a global scale;

[0006] Step S2: Identification of sea-sourced landfall flood events; combining the data obtained in step S1, using the available water kernel density estimation to perform standardization, selecting grid points that are higher than the preset value of the standardized index as flood grid points, and obtaining the flood space-time cube through the Lagrangian three-dimensional tracking method, and dividing the flood into land-sourced and sea-sourced flood space-time cubes;

[0007] Step S3: quantify the spatiotemporal characteristics of sea-sourced landfall floods; combine the spatiotemporal cubes of land-sourced and sea-sourced floods obtained in step S2 to calculate the characteristics of land-sourced and sea-sourced flood events; identify land flood events and quantify their characteristics in combination with the process of step S2, and identify flood events caused by land-sourced and sea-sourced atmospheric floods;

[0008] Step S4: Attribution and impact of sea-sourced landfall floods; Combined with the spatiotemporal characteristics of sea-sourced landfall flood events obtained in step S3, quantify the relative changes of sea-sourced landfall flood characteristics under three different external forcing scenarios of anthropogenic emissions, greenhouse gas emissions, and anthropogenic aerosol emissions relative to the historical climate average, analyze the impact of different human activities on the characteristics of sea-sourced landfall floods in different climate zones, and calculate the signal-to-noise ratio to evaluate the relative contributions of internal climate variability and forced response under external forcing;

[0009] Step S5: Analysis and mechanism of migration and landing of sea-sourced landfall floods; Combined with the sea-sourced landfall flood events obtained in step S3, an atmospheric circulation physical model is established to analyze the large-scale environmental variables, atmospheric circulation and water vapor transport conditions of the migration and landing of sea-sourced landfall flood events, analyze the seasonal migration path and landing process of sea-sourced landfall floods, and estimate the strengthening process and physical mechanism of sea-sourced landfall floods in the future;

[0010] Step S6: Comprehensive risk assessment of sea-originated landfall floods; Combined with the grid characteristics of sea-originated landfall floods obtained in step S3, future predictions of sea-originated landfall flood event characteristics are made, and the relative changes and signal-to-noise ratios of sea-originated landfall flood event characteristics under different climate zones in the future are explored; a water vapor transport model is established to identify the atmospheric physical driving factors of sea-originated landfall floods; combined with a machine learning clustering algorithm, risk assessments of sea-originated landfall floods in different climate zones are made.

[0011] A storage medium stores instructions and data for implementing a risk assessment method for sea-originated landfall floods.

[0012] A risk assessment device for sea-sourced landfall floods comprises: a processor and a storage medium; the processor loads and executes instructions and data in the storage medium to implement a risk assessment method for sea-sourced landfall floods.

[0013] The beneficial effects provided by the present invention are:

[0014] (1) The present invention provides a method for identifying sea-sourced landfall floods, which helps the academic community deepen its understanding of the landward transport of excess water vapor from the sea. The present invention identifies the events of sea-sourced landfall floods, clarifies the changes and impacts of sea-sourced landfall floods in time and space, and identifies the cascade effect of sea-sourced landfall floods and land floods, providing a scientific basis for the formulation of flood prevention and control policies and measures.

[0015] (2) The present invention quantifies the characteristics of sea-borne landfall floods under different anthropogenic emission scenarios, identifies the impact of human activities on sea-borne landfall flood events in different climate regions, and provides theoretical support for decision makers to make decisions on adapting to and mitigating the effects of climate change on sea-borne landfall floods.

[0016] (3) The present invention analyzes the large-scale environmental variables, atmospheric circulation and water vapor transport conditions of the migration and landing of sea-originated landfall flood events, clarifies the migration paths of sea-originated landfall flood events in history and the future, and the seasonal evolution laws of sea-originated landfall flood events, providing theoretical support for the prediction of land floods. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic flow chart of the method of the present invention;

[0018] Figure 2 It is a schematic diagram of the three-dimensional evolution and climate characteristics of sea-originated landfall floods and the land floods they propagate;

[0019] Figure 3 It is a schematic diagram of the frequency of sea-originated landfall flood generation, migration and landfall;

[0020] Figure 4 It is a schematic diagram of the relative changes in the characteristics of sea-sourced landfall flood events compared with the historical climate average under different anthropogenic emissions;

[0021] Figure 5 It is a schematic diagram of the propagation path and mechanism of sea-originated landfall floods from the North Atlantic and the eastern North Pacific to the middle and high latitudes of the Northern Hemisphere;

[0022] Figure 6 This is a schematic diagram of the relationship between sea temperature and available water based on maximum covariance analysis;

[0023] Figure 7 It is a schematic diagram of the future projections of the intensification of the propagation process of sea-originated landfall floods from the North Atlantic and the eastern North Pacific to the mid- and high-latitudes of the Northern Hemisphere;

[0024] Figure 8 This is a schematic diagram of the seasonal landfall process of future sea-originated landfall floods from the northern Atlantic Ocean to the northern Eurasian continent;

[0025] Fig. 9 This is a schematic diagram of the relative changes in the characteristics of sea-originated landfall flood events compared to the historical climate average under future high emission scenarios;

[0026] Fig.10 This is a schematic diagram of the contribution of atmospheric physical driving factors during sea-originated landfall floods under future high emission scenarios;

[0027] Fig.11It is a working schematic diagram of the hardware device of the present invention. DETAILED DESCRIPTION

[0028] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0029] Before formally describing the present invention, the scheme of the present invention is first generally described for easy understanding.

[0030] Please refer to Figure 1 , Figure 1 It is a schematic diagram of the process of the present invention.

[0031] The present invention provides a risk assessment method for sea-sourced landfall floods, comprising the following steps:

[0032] Step S1: Data collection; Collect precipitation, evaporation, sea surface temperature, sea level pressure, meridional wind speed, zonal wind speed, specific humidity and surface pressure data from the fifth generation reanalysis dataset of the European Centre for Medium-Range Weather Forecasts (ERA5), the sixth phase of the International Coupled Comparison Program (CM IP6), and the Canadian Earth System Model Version 5 Large Sample Collection (CanESM5); Collect land water storage anomaly data from the land water storage dataset reconstructed based on machine learning (GTWS-MLrec); Collect mask files for land on a global scale;

[0033] Step S2: Identification of sea-sourced landfall flood events; Combined with the data obtained in step S1, the available water volume (precipitation and evaporation difference) kernel density estimation is used for standardization, and grid points with a standardization index higher than 0.8 are selected as flood grid points. The flood space-time cube is obtained by the Lagrangian three-dimensional tracking method, and the flood is divided into land-source and sea-source flood space-time cubes;

[0034] In step S2, the flood index is calculated by kernel density estimation using the atmospheric available water (precipitation and evaporation difference) of each mode of historical climate model experiment, natural forcing experiment, greenhouse gas forcing experiment, anthropogenic aerosol forcing experiment and future socio-economic development path simulation experiment, as shown in the following formula:

[0035]

[0036] In the formula, SWI is the flood index, t is the month, is the empirical distribution function obtained by kernel density estimation, WA is the available water in the atmosphere;

[0037] The flood index of the natural forcing test and the future social and economic development path simulation test adopts the empirical distribution function obtained by kernel density estimation of historical climate model test data to ensure climate stability. In the present invention, the kernel density estimation adopts the Gaussian kernel function.

[0038] For a single grid point, when SWI is greater than 0.8, the grid point is defined as a flood grid point; adjacent flood grid points in a single layer are merged into two-dimensional continuous flood patches; any spatially continuous three-dimensional flood patches are merged into a single three-dimensional flood event if the overlapping area is greater than 100,000 square kilometers in continuous time steps; three-dimensional flood events last at least 3 days and have an area greater than 100,000 square kilometers.

[0039] The land mask file of step S1 is used to set the minimum landing area, and the three-dimensional flood events originating from the ocean and with a landing area exceeding the threshold are identified as sea-sourced landing flood events. The GTWS-MLrec data obtained in step S1 is used to standardize the land water storage anomaly, identify land flood events that are continuous in time and space, set the minimum overlapping area, and identify the sea-sourced landing flood-land flood cascade process. In the present invention, the minimum landing area is set to 100,000 km 2 , the minimum overlap area is 100000km 2 .

[0040] Step S3: Quantify the spatiotemporal characteristics of sea-sourced landfall floods; combine the spatiotemporal cubes of land-sourced and sea-sourced floods obtained in step S2 to calculate the characteristics of land-sourced and sea-sourced flood events; identify land flood events and quantify their characteristics in combination with the process of step S2, and identify flood events caused by land-sourced and sea-sourced atmospheric floods.

[0041] The temporal, spatial and attribute characteristics of sea-originated landfall flood events are duration, area and intensity, respectively. Duration refers to the duration of a sea-originated landfall flood event, area refers to the total area of ​​spatial grids affected by the continuous process of a sea-originated landfall flood event, and intensity refers to the weighted average of the grid area of ​​available water anomalies of all grid points in a sea-originated landfall flood event.

[0042] Step S4: Attribution and impact of sea-originated landfall floods; Combined with the spatiotemporal characteristics of sea-originated landfall flood events obtained in step S3, quantify the relative changes of sea-originated landfall flood characteristics relative to the historical climate average under three different external forcing scenarios of anthropogenic emissions, greenhouse gas emissions, and anthropogenic aerosol emissions, analyze the impact of different human activities on the characteristics of sea-originated landfall floods in different climate zones, and calculate the signal-to-noise ratio to evaluate the relative contributions of internal climate variability and forced response under external forcing.

[0043] In step S4, different climate zones are defined, and the world is divided into 21 climate zones according to the climate reference zones provided by the Intergovernmental Panel on Climate Change. Spatial grid points in different climate zones are selected, and area-weighted average is performed to calculate the relative changes of the sea-sourced landfall flood characteristics (I) between given time periods under three different external forcings of anthropogenic emissions, greenhouse gas emissions, and anthropogenic aerosol emissions relative to the historical climate average, as shown in the following formula:

[0044]

[0045] Furthermore, the signal-to-noise ratio is calculated to assess the relative contribution of internal climate variability and the forced response under external forcing, as follows:

[0046]

[0047] The signal is It is the forced response of the ensemble mean characteristics of sea-sourced landfall floods based on 32 models of CM IP6 and 50 ensembles of CanESM5. The noise is defined as the inter-member standard deviation σ(ΔI 周期,外强迫 ). The absolute value of SNR is greater than 1 (less than 1), which means that the effect of external forcing in CMIP6 (CanESM5) is stronger (weaker) than the combined uncertainty of climate internal variability and model formation differences (climate internal variability), that is, the anthropogenic climate change signal does (does not) appear.

[0048] Step S5: Analysis and mechanism of migration and landing of sea-sourced landfall floods; Combined with the sea-sourced landfall flood events obtained in step S3, an atmospheric circulation physical model is established to analyze the large-scale environmental variables, atmospheric circulation and water vapor transport conditions of the migration and landing of sea-sourced landfall flood events, analyze the seasonal migration path and landing process of sea-sourced landfall floods, and estimate the strengthening process and physical mechanism of sea-sourced landfall floods in the future;

[0049] Combined with the initial data set, an atmospheric circulation physical model was established to analyze the large-scale environmental variables, atmospheric circulation, and water vapor transport conditions of the migration and landing of sea-sourced landfall flood events. The large-scale environmental variables of the water vapor transport simulation process of the atmospheric circulation physical model use sea surface temperature and sea level pressure as high / low pressure system measurement indicators, and the maximum covariance analysis method is used to explore the impact of sea surface temperature anomalies on soil moisture changes on all data sets. The maximum covariance analysis method is as follows:

[0050]

[0051] Where X and Y are the spatial matrices of the two data sets, respectively, and N is the time span. m and q refer to the number of grid points in the X and Y spatial matrices, respectively. x,yis the covariance matrix. U and V are the spatial modes of X and Y respectively. PC x,m and PC y,q is the time series of unit m in X and unit q in Y;

[0052] The atmospheric circulation and water vapor transport simulation process of the atmospheric circulation physical model is as follows:

[0053]

[0054] In the formula, is (u, v), u is the zonal wind, v is the longitudinal wind, q is the relative humidity, is the water vapor flux;

[0055] Using the obtained migration and landing water vapor transport conditions, the migration path and landing process of sea-originated landfall floods are analyzed;

[0056] Step S6: Comprehensive risk assessment of sea-originated landfall floods; Combined with the grid characteristics of sea-originated landfall floods obtained in step S3, future predictions of sea-originated landfall flood event characteristics are made, and the relative changes and signal-to-noise ratios of sea-originated landfall flood event characteristics under different climate zones in the future are explored; a water vapor transport model is established to identify the atmospheric physical driving factors of sea-originated landfall floods; combined with a machine learning clustering algorithm, risk assessments of sea-originated landfall floods in different climate zones are made.

[0057] In step S6, based on the grid characteristics of sea-originated landfall flood events in each model: frequency, duration, area and intensity, the future relative changes of sea-originated landfall flood event characteristics under different scenarios are quantified.

[0058] In step S6, a water vapor transport model is established to identify the atmospheric physical driving factors of sea-originated landfall floods, as follows:

[0059]

[0060] In the formula, and Represent the monthly average values ​​of specific humidity and wind speed, and The specific humidity and wind speed of the historical climate model test base period, and represents the difference between the average specific humidity and average wind speed before and after the landing of the sea-sourced landfall flood event in the historical climate model test and the benchmark period of the historical climate model test, ps is the surface pressure, pt is the top of the atmosphere pressure, δ is the deviation operator, is the gradient operator, d is the integral differential and p is the air pressure. All the above parameters constitute the atmospheric physical driving factors;

[0061] Based on the grid indicators of sea-originated landfall flood events in each model: frequency, duration, intensity, area and atmospheric physical driving factors, the self-organizing map and K-means clustering models are used to obtain the comprehensive risks of different grid points.

[0062] As an implementation case, the present invention takes the global period from 1961 to 2020 as an example to further describe the present invention. The implementation case is used to illustrate the present invention, but this case is not used to limit the scope of application of the present invention. It is also applicable to different regions and other time periods.

[0063] The implementation flow chart of the method of the present invention for identifying sea-sourced landfall flood events and analyzing physical mechanisms is as follows: Figure 1 As shown, the specific steps are as follows:

[0064] (1) Collection of test data;

[0065] In this implementation case, the fifth generation reanalysis dataset of the European Center for Medium-Range Weather Forecasts (ERA5), 32 climate models of the sixth phase of the International Coupled Comparison Program (CM IP6), and 50 sample sets of the Canadian Earth System Model Version 5 Large Sample Set (CanESM5) were collected for precipitation, evaporation, sea surface temperature, sea level pressure, meridional wind speed, zonal wind speed, specific humidity, and surface pressure data; the land water storage anomaly data of the land water storage reconstruction dataset based on machine learning (GTWS-MLrec) were collected; the above data spans from 1921 to 2100. The meteorological data are interpolated using the bilinear interpolation method with a spatial resolution of 1.5°×1.5°.

[0066] (2) Identification of sea-originated landfall flood events;

[0067] This implementation case is global in scope and combines historical climate model experiments with future socio-economic development path simulation experiments on a time scale to obtain meteorological data from 1921 to 2100. For the reanalysis data and climate model experiment data from 1961 to 2020, the available atmospheric water (the difference between precipitation and evaporation) is used to calculate flood indicators through kernel density estimation, as shown in the following formula:

[0068]

[0069] In the formula, SWI is the flood index, t is the month, is the empirical distribution function obtained by kernel density estimation, WA is the available water in the atmosphere;

[0070] Furthermore, the empirical distribution function of the reanalysis data and climate model test data from 1961 to 2020 is used to calculate flood indicators using the spliced ​​data of the historical climate model test from 1921 to 2100, the future socio-economic development path simulation test, and the natural forcing test data. The flood spatial raster data of each time step is thresholded, the flood state threshold is set, and the spatial raster data is converted into binary raster data of 1 (flood) and 0 (non-flood). In this implementation case, the flood state threshold is defined as 0.8.

[0071] The Lagrangian 3D tracking method is used to identify 3D flood events. The minimum landing area is set, and 3D flood events originating from the ocean with a landing area exceeding the threshold are identified as sea-sourced landfalling flood events. Step S1 is used to obtain GTWS-MLrec data, and the land water storage anomaly is standardized. Similar to the previous steps, spatiotemporally continuous land flood events are identified, and the minimum overlapping area is set to identify the sea-sourced landfalling flood-land flood cascade process. The spatiotemporal evolution of a sea-sourced landfalling flood-land flood cascade process is shown in Figure 2. Figure 2 In this case, the minimum landing area is set to 100,000 km 2 , the minimum overlap area is 100000km 2 .

[0072] (3) Quantification of spatiotemporal characteristics of sea-originated landfall floods;

[0073] This implementation case quantifies the three-dimensional event characteristics of sea-originated landfall floods from historical climate model experiments, natural forcing experiments, greenhouse gas experiments, anthropogenic aerosol experiments, and future socioeconomic development path simulation experiments obtained through experimental data from 32 international coupled comparison programs in the sixth phase of CMIP6 and 50 Canadian Earth System Model version 5 large sample sets (CanESM5). The temporal, spatial, and attribute characteristics of sea-originated landfall flood events are duration, projected area, and intensity. Duration is the duration of a sea-originated landfall flood event, projected area is the total area of ​​spatial grids affected by the continuous process of a sea-originated landfall flood event, and intensity is the weighted average of the grid area of ​​available water anomalies of all grid points in a sea-originated landfall flood event. Land flood events are calculated in the same way. Figure 2 and Figure 3 As shown in the data, the duration (+133.3%) and area (+649.0%) of global sea-sourced landfall floods (582 events) were significantly higher than those of land-sourced atmospheric floods (471 events). Compared with other regions, transatlantic and North Pacific sea-sourced landfall floods originating from the Atlantic and North Pacific occurred more frequently, lasted longer, were more intensive, and had a wider range in the mid- and high-latitudes of the Northern Hemisphere (above 48°N), South America, and Australia.

[0074] (4) Attribution and impacts of sea-originated landfall floods;

[0075] This implementation case uses the grid index of the spatiotemporal characteristics of sea-sourced landfall floods, and divides the world into 21 climate zones according to the climate reference zones provided by the Intergovernmental Panel on Climate Change. Spatial grid points in different climate zones are selected, and area-weighted averages are performed to calculate the relative changes in the sea-sourced landfall flood characteristics (I) between given periods under three different external forcings: anthropogenic emissions, greenhouse gas emissions, and anthropogenic aerosol emissions, relative to the historical climate average, as shown in the following formula:

[0076]

[0077] Furthermore, the signal-to-noise ratio is calculated to assess the relative contribution of internal climate variability and the forced response under external forcing, as follows:

[0078]

[0079] The signal is It is the forced response of the ensemble mean characteristics of sea-sourced landfall floods based on 32 models of CM IP6 and 50 ensembles of CanESM5. The noise is defined as the inter-member standard deviation σ(ΔI 周期,外强迫 ). The absolute value of SNR is greater than 1 (less than 1), which means that the effect of external forcing in CMIP6 (CanESM5) is stronger (weaker) than the combined uncertainty of climate internal variability and model formation differences (climate internal variability), that is, the anthropogenic climate change signal does (does not) appear.

[0080] like Figure 4 As shown in Figure 2, compared with the historical climate average (1981-2010), spatially, the frequency, duration, intensity and area of ​​sea-originated landfall floods increased in 44.7%, 45.4%, 45.0% and 60.4% of the regions under anthropogenic emissions in CMIP6 and CanESM5, respectively, during 1981-2010. This increase is caused by the continued global greenhouse gas emissions. The warming caused by greenhouse gases brings more water vapor and stronger land-sea thermal differences, creating favorable conditions for the generation and landing of sea-originated landfall floods. In terms of time, under anthropogenic emissions (mainly greenhouse gas emissions), the relative changes in the characteristics of sea-originated landfall floods in the middle and high latitudes of the Northern Hemisphere have changed from negative to positive values ​​since the last two decades of the 20th century with the increase in anthropogenic emissions. However, the anthropogenic climate change signals in these changes are not captured in the historical period, probably due to the offsetting effect of regional cooling caused by anthropogenic aerosol emissions.

[0081] (5) Analysis and mechanism of migration and landing of sea-originated landfall floods;

[0082] Combined with the initial data set, an atmospheric circulation physical model was established to analyze the large-scale environmental variables, atmospheric circulation, and water vapor transport conditions of the migration and landing of sea-sourced landfall flood events. The large-scale environmental variables of the water vapor transport simulation process of the atmospheric circulation physical model use sea surface temperature and sea level pressure as high / low pressure system measurement indicators, and the maximum covariance analysis method is used to explore the impact of sea surface temperature anomalies on soil moisture changes on all data sets. The maximum covariance analysis method is as follows:

[0083]

[0084] Where X and Y are the spatial matrices of the two data sets, respectively, and N is the time span. m and q refer to the number of grid points in the X and Y spatial matrices, respectively. x,y is the covariance matrix. U and V are the spatial modes of X and Y respectively. PC x,m and PC y,q is the time series of unit m in X and unit q in Y;

[0085] The atmospheric circulation and water vapor transport simulation process of the atmospheric circulation physical model is as follows:

[0086]

[0087] In the formula, is (u, v), u is the zonal wind, v is the longitudinal wind, q is the relative humidity, is the water vapor flux;

[0088] Using the obtained migration and landing water vapor transport conditions, the migration path and landing process of sea-originated landfall floods are analyzed;

[0089] The migration and landing water vapor transport conditions obtained are used to further analyze the migration path and landing process of sea-originated landfall floods. Figure 5 As shown, two main landfall pathways of sea-originated landfall floods from the Atlantic and Pacific Oceans to mid- and high-latitudes in the Northern Hemisphere were identified during 1961-2020. The first pathway originates in the North Atlantic and reaches northern Eurasia through seasonally related water vapor transport. Specifically, from December to February of the following year, the water vapor surplus originates in the northwest Atlantic and migrates northeastward to Europe. By March, the eastward water vapor surplus makes landfall in Europe and finally reaches Northeast Asia in August. The second landfall pathway of sea-originated landfall floods originates in the northeastern Pacific and migrates to western North America through a shift in the circulation pattern.

[0090] like Figure 5As shown in Figure 1, changes in the North Atlantic sea surface temperature play an important role in the migration of the available water (water vapor) surplus to the land on the first landing path. The long-term warming of the North Atlantic sea surface temperature is conducive to the generation of a large amount of water vapor surplus in low-latitude areas in winter, which is then transported by northerly winds and ocean currents ( Figure 6 ). Influenced by the long-distance eastward water vapor transport of the westerly belt, this water vapor surplus then landed over the northern Eurasian continent and spread inland. Figure 5 As shown in Figure 2, in the second path, the water vapor surplus generated near the equator and migrating northward is hindered by the abnormal high pressure pattern over the northeastern Pacific (i.e., anticyclonic anomaly, water vapor divergence, and positive sea level pressure anomaly). In contrast, during the post-landfall period, the high pressure pattern over the northeastern Pacific turns into a low pressure pattern ( Figure 5 D) Changes in circulation patterns caused the excess water vapor to migrate landward and eventually land in western North America ( Figure 5 C) This migration process may be affected by the Pacific Ocean surface temperature. The warming tropical Pacific will cause the subtropical jet stream to extend eastward, thus generating an abnormal cyclonic circulation near the deeper Aleutian low pressure.

[0091] like Figure 6 As shown in the first path, the North Atlantic winter sea surface temperature will warm at a rate 2.3 times the historical warming rate in the next 10 years ( Figure 7 D). In addition, the extreme atmospheric moisture surplus in the North Atlantic in winter and the extreme atmospheric moisture surplus in northern Eurasia from March to August will both increase significantly ( Figure 7 B and D). The relationship between winter SST, the North Atlantic winter extreme atmospheric moisture surplus, and the March-August extreme atmospheric moisture surplus over northern Eurasia is also predicted to be stronger. Along this landfall path, not only increased SST and extreme atmospheric moisture surplus, but also a stronger connection between them will enhance the seasonally correlated water vapor transport from the North Atlantic and to northern Eurasia under anthropogenic warming ( Figure 7 A and Figure 8 ), leading to more severe and widespread sea-originated landfall floods in the future.

[0092] In the second landfall path, under the anthropogenic climate change scenario, the difference in sea level pressure anomaly before and after landfall will further increase ( Figure 7 E), indicating that this change will further intensify in the future. The enhanced transformation of the circulation pattern will lead to the northeastern Pacific ( Figure 7 and Figure 8 ) and an intensification of the moisture surplus, resulting in more marine-derived landfalling floods in western North America.

[0093] (6) Comprehensive risk assessment of sea-originated landfall floods;

[0094] The physical model is constructed from multiple dimensions from the perspective of physical factors, and the physical factors are split:

[0095]

[0096] In the formula, and Represent the monthly average values ​​of specific humidity and wind speed, and The specific humidity and wind speed of the historical climate model test base period, and represents the difference between the average specific humidity and average wind speed before and after the landing of the sea-sourced landfall flood event in the historical climate model test and the benchmark period of the historical climate model test, ps is the surface pressure, pt is the top of the atmosphere pressure, δ is the deviation operator, is the gradient operator, d is the integral differential and p is the air pressure.

[0097] like Fig. 9 and Fig.10 As shown in Figure 2, the thermodynamic component is the dominant driving force in the mid- and high-latitudes of the Northern Hemisphere (76.8% in the mid- and high-latitudes of the Northern Hemisphere). This thermodynamic effect can be further attributed to the convergence (advection) of 72.9% (27.1%) of the thermodynamically dominant regions. Thermodynamically, the increase in ocean water vapor caused by the increase in sea surface temperature and the enhanced water holding capacity of the atmosphere due to anthropogenic climate change promotes the transport of more ocean water vapor to the mid- and high-latitudes of the Northern Hemisphere, thereby enhancing the atmospheric moisture surplus in these regions.

[0098] Based on the grid characteristics of sea-originated landfall flood events in each model: frequency, duration, intensity and area, self-organizing mapping and K-means clustering models are used to obtain the comprehensive risks of different grids. This implementation case uses the grid characteristics of sea-originated landfall flood events: frequency, duration, area and intensity, namely the historical simulation from 1981 to 2010 and the future forecast from 2071 to 2100. Unsupervised clustering is performed based on self-organizing mapping and K-means clustering models to obtain the comprehensive risks of different grids, and the risk levels are divided into three levels. The data of the self-organizing mapping and K-means clustering models in this implementation case adopts the integrated average grid index deviation percentage of sea-originated landfall flood events from historical climate model experiments and future socio-economic development path simulation experiments. Fig. 9As shown, during the period 2071-2100, areas with high risk levels are expected to experience extreme sea-origin landfall floods with a frequency of 0.3 times / year, a duration of 0.69 months / year, an intensity of 0.65 / year, and an area range of 116,000 square kilometers / year. Globally, 61.1% of the land area will experience high risks that did not exist in historical climate periods under future high emission scenarios. Regionally, more than half of the areas in 62% of climate regions are at high levels of risk. The historical and predicted spatial patterns of risk assessment are consistent with the patterns identified in the scale of sea-origin landfall floods.

[0099] See also Fig.11 , Fig.11 4 is a schematic diagram of the working of the hardware device of an embodiment of the present invention, wherein the hardware device specifically comprises: a risk assessment device 401 for sea-origin landfall floods, a processor 402 and a storage medium 403.

[0100] A risk assessment device 401 for sea-origin landfall flood: The risk assessment device 401 for sea-origin landfall flood implements the risk assessment method for sea-origin landfall flood.

[0101] Processor 402: The processor 402 loads and executes the instructions and data in the storage medium 403 to implement the risk assessment method for sea-originated landfall floods.

[0102] Storage medium 403: The storage medium 403 stores instructions and data; the storage medium 403 is used to implement the risk assessment method for sea-originated landfall floods.

[0103] The beneficial effects of the present invention are:

[0104] (1) The present invention provides a method for identifying sea-sourced landfall floods, which helps the academic community deepen its understanding of the landward transport of excess water vapor from the sea. The present invention identifies the events of sea-sourced landfall floods, clarifies the changes and impacts of sea-sourced landfall floods in time and space, and identifies the cascade effect of sea-sourced landfall floods and land floods, providing a scientific basis for the formulation of flood prevention and control policies and measures.

[0105] (2) The present invention quantifies the characteristics of sea-borne landfall floods under different anthropogenic emission scenarios, identifies the impact of human activities on sea-borne landfall flood events in different climate regions, and provides theoretical support for decision makers to make decisions on adapting to and mitigating the effects of climate change on sea-borne landfall floods.

[0106] (3) The present invention analyzes the large-scale environmental variables, atmospheric circulation and water vapor transport conditions of the migration and landing of sea-originated landfall flood events, clarifies the migration paths of sea-originated landfall flood events in history and the future, and the seasonal evolution laws of sea-originated landfall flood events, providing theoretical support for the prediction of land floods.

[0107] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A risk assessment method for sea-origin landfall floods, characterized by: The method comprises the following steps: Step S1: Data collection; collect precipitation, evaporation, sea surface temperature, sea level pressure, meridional wind speed, zonal wind speed, specific humidity and surface pressure data from the Weather Forecast Center reanalysis dataset ERA5, the sixth phase of the International Coupled Intercomparison Program CMIP6, and the large sample set of system models CanESM5; collect land water storage anomaly data from the machine learning-based land water storage dataset GTWS-MLrec; collect mask files for land on a global scale; Step S2: Identification of sea-sourced landfall flood events; combining the data obtained in step S1, using the available water kernel density estimation to perform standardization, selecting grid points that are higher than the preset value of the standardized index as flood grid points, and obtaining the flood space-time cube through the Lagrangian three-dimensional tracking method, and dividing the flood into land-sourced and sea-sourced flood space-time cubes; Step S3: Quantification of spatiotemporal characteristics of sea-originated landfall floods; Combine the land-source and sea-source flood space-time cubes obtained in step S2 to calculate the land-source and sea-source flood event characteristics; Combined with the process of step S2, land flood events are identified and their characteristics are quantified to identify flood events caused by land-based and sea-based atmospheric floods; Step S4: Attribution and impact of sea-sourced landfall floods; Combined with the spatiotemporal characteristics of sea-sourced landfall flood events obtained in step S3, quantify the relative changes of sea-sourced landfall flood characteristics under three different external forcing scenarios of anthropogenic emissions, greenhouse gas emissions, and anthropogenic aerosol emissions relative to the historical climate average, analyze the impact of different human activities on the characteristics of sea-sourced landfall floods in different climate zones, and calculate the signal-to-noise ratio to evaluate the relative contributions of internal climate variability and forced response under external forcing; Step S5: Analysis and mechanism of migration and landing of sea-sourced landfall floods; Combined with the sea-sourced landfall flood events obtained in step S3, an atmospheric circulation physical model is established to analyze the large-scale environmental variables, atmospheric circulation and water vapor transport conditions of the migration and landing of sea-sourced landfall flood events, analyze the seasonal migration path and landing process of sea-sourced landfall floods, and estimate the strengthening process and physical mechanism of sea-sourced landfall floods in the future; Step S6: Comprehensive risk assessment of sea-originated landfall floods; Combined with the grid characteristics of sea-originated landfall floods obtained in step S3, future predictions of sea-originated landfall flood event characteristics are made, and the relative changes and signal-to-noise ratios of sea-originated landfall flood event characteristics under different climate zones in the future are explored; a water vapor transport model is established to identify the atmospheric physical driving factors of sea-originated landfall floods; combined with a machine learning clustering algorithm, risk assessments of sea-originated landfall floods in different climate zones are made.

2. A risk assessment method for sea-sourced landfall floods as claimed in claim 1, characterized in that: Step S2 is specifically as follows: S21: Using the atmospheric available water from historical climate model experiments, natural forcing experiments, greenhouse gas forcing experiments, anthropogenic aerosol forcing experiments, and future socioeconomic development path simulation experiments, the flood index is calculated by kernel density estimation, as shown in the following formula: In the formula, SWI is the flood index, t is the month, is the empirical distribution function obtained by kernel density estimation, WA is the available water in the atmosphere; S22: The Lagrangian tracking method identifies regional flood events in three dimensions as follows: For a single grid point, when SWI is greater than 0.8, the grid point is defined as a flood grid point; adjacent flood grid points in a single layer are merged into two-dimensional spatially continuous flood patches; any spatially continuous three-dimensional flood patches are merged into a single three-dimensional flood event if the overlapping area is greater than the preset square kilometers in continuous time steps; the three-dimensional flood event lasts for at least the preset number of days and the area is greater than the preset square kilometers; S23: Using the land mask file of step S1, set the minimum landing area, identify the three-dimensional flood events originating from the ocean and with a landing area exceeding the threshold as sea-originated landfall flood events, use step S1 to obtain the GTWS-MLrec data, standardize the land water storage anomaly, use steps S21-S22 to identify land flood events that are continuous in time and space, set the minimum overlapping area, and identify the sea-originated landfall flood-land flood cascade process.

3. A risk assessment method for sea-sourced landfall floods as claimed in claim 2, characterized in that: Step S3 is specifically as follows: The temporal, spatial and attribute characteristics of sea-originated landfall flood events are duration, area and intensity, respectively. Duration refers to the duration of a sea-originated landfall flood event, area refers to the total area of ​​spatial grids affected by the continuous process of a sea-originated landfall flood event, and intensity refers to the weighted average of the grid area of ​​available water anomalies of all grid points in a sea-originated landfall flood event.

4. A risk assessment method for sea-sourced landfall floods as claimed in claim 3, characterized in that: Step S4 is specifically as follows: S41: Define different climate zones and divide the world into multiple climate zones based on the climate reference zones provided by the Intergovernmental Panel on Climate Change; S42: Attribution and impacts of sea-originated landfall floods. Select spatial grid points in different climate zones and perform area-weighted averaging. Calculate the relative changes in the characteristics of sea-originated landfall floods under three different external forcing scenarios: anthropogenic emissions, greenhouse gas emissions, and anthropogenic aerosol emissions, relative to the historical climate average. Analyze the impact of human activities on the characteristics of sea-originated landfall floods, and calculate the signal-to-noise ratio to evaluate the relative contributions of internal climate variability and forced response under external forcing.

5. A risk assessment method for sea-sourced landfall floods as claimed in claim 4, characterized in that: Step S5 is specifically as follows: Combined with the initial data set, an atmospheric circulation physical model was established to analyze the large-scale environmental variables, atmospheric circulation and water vapor transport conditions of the migration and landing of sea-sourced landfall flood events. The large-scale environmental variables of the water vapor transport simulation process of the atmospheric circulation physical model use sea surface temperature and sea level pressure as high / low pressure system measurement indicators, and the maximum covariance analysis method is used to explore the impact of the North Atlantic sea surface temperature anomaly on the change of available water volume in the high latitudes of the northern hemisphere of Eurasia on all data sets; the maximum covariance analysis method is as follows: Where X and Y are the spatial matrices of two data sets, respectively, and N is the time span; m and q refer to the number of grid points in the X and Y spatial matrices, respectively; C x,y is the covariance matrix; U and V are the spatial modes of X and Y respectively; PC x,m and PC y,q is the time series of unit m in X and unit q in Y; The atmospheric circulation and water vapor transport simulation process of the atmospheric circulation physical model is as follows: In the formula, is (u, v), u is the zonal wind, v is the longitudinal wind, q is the relative humidity, is the water vapor flux; Using the obtained migration and landing water vapor transport conditions, the migration path and landing process of sea-originated landfall floods are analyzed.

6. A risk assessment method for sea-origin landfall floods as claimed in claim 5, characterized in that: In step S6, a water vapor transport model is established to identify the atmospheric physical driving factors of sea-originated landfall floods, as follows: In the formula, and Represent the monthly average values ​​of specific humidity and wind speed, and The specific humidity and wind speed of the historical climate model test base period, and represents the difference between the average specific humidity and average wind speed before and after the landing of the sea-sourced landfall flood event in the historical climate model test and the benchmark period of the historical climate model test, ps is the surface pressure, pt is the top of the atmosphere pressure, δ is the deviation operator, is the gradient operator, d is the integral differential and p is the air pressure. All the above parameters constitute the atmospheric physical driving factors; Based on the grid indicators of sea-originated landfall flood events in each model: frequency, duration, intensity, area and atmospheric physical driving factors, the self-organizing map and K-means clustering models are used to obtain the comprehensive risks of different grid points.

7. A storage medium, characterized in that: The storage medium stores instructions and data for implementing a risk assessment method for sea-originated landfall floods as described in any one of claims 1 to 6.

8. A risk assessment device for sea-origin floods, characterized in that: include: Processor and storage medium; the processor loads and executes instructions and data in the storage medium to implement a risk assessment method for sea-originated landing floods as described in any one of claims 1 to 6.

Images