CN115793099A - Identification method of landfall tropical cyclone intensity attenuation and its impact on rainfall assessment - Google Patents

Abstract

The invention discloses a method for identifying the strength attenuation of a tropical cyclone during landing and evaluating the influence of the tropical cyclone on rainfall. According to the method, kaplan-DeMaria exponential decay fitting is carried out on the intensity within 24 hours after the tropical cyclone logs in to obtain the intensity decay time scale of the tropical cyclone, time change characteristic analysis is carried out on the intensity decay time scale of the tropical cyclone logging in, factors influencing the intensity decay of the tropical cyclone logging in are explored, the relative contribution of the influencing factors to the intensity decay of the tropical cyclone logging in is quantitatively evaluated, and the influence of the intensity decay change of the tropical cyclone logging in the tropical cyclone is finally discussed by combining three aspects of sea surface temperature rise, tropical cyclone path transfer and large-scale environmental characteristics to clarify the physical mechanism of the intensity decay change of the tropical cyclone logging in the tropical cyclone. The invention has the beneficial effects that: further improving the weather service guarantee capability of city safety and providing scientific basis for disaster prevention and reduction, tropical cyclone disaster resistance and government decision.

Description

Identification method of landfall tropical cyclone intensity attenuation and its impact on rainfall assessment

technical field

The invention relates to the field of atmospheric science and technology, in particular to a method for identifying the intensity attenuation of a landing tropical cyclone and its impact on rainfall.

Background technique

Tropical cyclones are cyclonic eddies that occur over tropical and subtropical oceans. The Northwest Pacific is the region with the most tropical cyclone occurrences in the world, with an average of 30 tropical cyclones each year, accounting for about 33% of the total number of tropical cyclones in the world. my country is located in the western part of the Pacific Northwest, and nearly half of the land (from Liaoning in the north to Hainan in the south, and as far west as 100°E west of the southwest) can be affected by tropical cyclones. The average number of tropical cyclones that land in my country every year is 7, and the occurrence time is usually from May to October. Tropical cyclones are often accompanied by disastrous weather such as strong winds, heavy rains, and storm surges when they land. The heavy rains and subsequent floods caused by tropical cyclones often cause huge socio-economic losses and casualties in the affected areas.

Due to land obstruction, frictional energy consumption, and being far away from water vapor sources, tropical cyclones decay rapidly after landfall. Li and Chakraborty's research found that compared with 50 years ago, the intensity of tropical cyclones in the North Atlantic decreased from 75% to 50% one day after landfall, and the intensity decreased more slowly. Zhu et al. and Song et al. found that the attenuation of tropical cyclone landfall in the continental United States and continental Asia has slowed down over the past few decades. Some studies have demonstrated a power-law relationship between tropical cyclone landfall intensity and economic losses, thus slowing down the intensity decay of tropical cyclones after landfall may increase the potential disasters in inland areas. However, the research on how the intensity attenuation of landfalling tropical cyclones affects tropical cyclone rainfall is still relatively scarce. Based on this, it is of great significance to study the changes in the time scale (τ) of the attenuation of the intensity of tropical cyclones landed on the mainland, which is of great significance for the prediction of tropical cyclone rainfall and the estimation of destructive potential. Provide scientific basis for resisting tropical cyclone disasters and government decision-making.

Contents of the invention

The purpose of this method is to propose a method for identifying the intensity attenuation of landfalling tropical cyclones and assessing their impact on rainfall in view of the lack of research methods on the impact of landfalling tropical cyclone intensity decay rate on tropical cyclone rainfall.

Landing tropical cyclone intensity attenuation identification of the present invention and its method for assessing the impact on rainfall comprise the following steps:

Step S1: data collection; collect the measured best track data of tropical cyclones, global precipitation data and meteorological reanalysis data;

Step S2: Screening of landfalling tropical cyclones; combined with the best path data of tropical cyclones obtained in step S1, screening all landfalling tropical cyclones, and selecting landfalling tropical cyclones satisfying the principle according to the screening principle of landfalling tropical cyclones;

Step S3: Identification of the time scale of the attenuation of the intensity of the landfalling tropical cyclone; combined with the landfalling tropical cyclone obtained in step S2 that satisfies the principle, perform Kaplan-DeMaria exponential decay fitting on the intensity of each tropical cyclone within 24 hours after landfall to identify the corresponding landfalling tropical cyclone Cyclone intensity decay time scale τ, and exclude log-in tropical cyclone events greater than 2 times the standard deviation of the mean value of τ;

Step S4: Identify the characteristics of the attenuation change of the intensity of the landfalling tropical cyclone; combine the τ values of the landfalling tropical cyclones obtained in Step S3, count the time series of the τ values of the landfalling tropical cyclones, and analyze the probability of the τ value of the landfalling tropical cyclones in the two stages before and after the research period Density curves, and identify the temporal variation characteristics of landfalling tropical cyclones τ;

Step S5: Identification of factors affecting the attenuation of landfalling tropical cyclone intensity; combined with the time series of landfalling tropical cyclone τ values obtained in step S4, the correlation analysis between factors that may affect the attenuation of landfalling tropical cyclones and this time series is carried out to identify the factors affecting landfalling tropical cyclones. All factors of change in cyclone intensity attenuation;

Step S6: Quantification of the relative contribution of factors affecting the attenuation change of the intensity of the landfalling tropical cyclone; combined with all the factors that affect the attenuation change of the intensity of the landfalling tropical cyclone obtained in step S5, quantify the relative contribution of each factor to the attenuation change of the intensity of the landfalling tropical cyclone according to the relative contribution calculation method contribute;

Step S7: Analysis of the physical mechanism of the attenuation change of the intensity of the landfalling tropical cyclone; combining the characteristics of the attenuation change of the intensity of the landfalling tropical cyclone obtained in step S4 with all the factors affecting the attenuation change of the intensity of the landfalling tropical cyclone obtained in step S5, the transition from the track of the tropical cyclone and the large-scale Two aspects of environmental characteristics clarify the physical mechanism of the attenuation change of the intensity of landfalling tropical cyclones;

Step S8: Identification of landfall tropical cyclone rainfall and its quantitative indicators; combined with the landfall tropical cyclone obtained in step S2, based on the objective weather map analysis method to identify the landfall tropical cyclone rainfall field, and then identify the landfall tropical cyclone rainfall index.

Step S9: Analysis of the impact of landfall tropical cyclone intensity attenuation changes on tropical cyclone rainfall; combined with the τ value of landfall tropical cyclone obtained in step S3 and the landfall tropical cyclone rainfall index obtained in step S8, the change of tropical cyclone rainfall index is counted, and the tropical cyclone rainfall index is calculated. Correlation analysis between cyclone rainfall index and landfall tropical cyclone τ value, and then analyze the impact of landfall tropical cyclone intensity attenuation changes on tropical cyclone rainfall.

The beneficial effects that the present invention provides are:

(1) The present invention proposes a kind of identification method of landfall tropical cyclone intensity decay speed situation, has identified the long-term change trend of landfall tropical cyclone intensity decay speed, has analyzed the physical mechanism of landfall tropical cyclone intensity decay speed change, which is important for landfall tropical cyclone intensity decay speed. Estimating the destructive potential of a cyclone is of great importance.

(2) The present invention has analyzed the influence of the attenuation change of the intensity of the landfalling tropical cyclone on the temporal and spatial process of the rainfall by the difference of the tropical cyclone rainfall under different landfalling tropical cyclone intensity attenuation intervals, and has clarified that the attenuation of the intensity of the landing tropical cyclone slows down the amplification of the tropical cyclone rainfall effect.

Description of drawings

Figure 1 is a flow chart of the implementation of the identification method of landfall tropical cyclone intensity attenuation and its impact on rainfall assessment;

Figure 2 shows the characteristics of the attenuation velocity of the tropical cyclones landing in China from 1967 to 2018 and the correlation analysis between the sea surface temperature in the Northwest Pacific Ocean and the time scale (τ) of the attenuation of the intensity of the tropical cyclones landing in China;

Figure 3 shows the longitude of landfall center, latitude of landfall center, movement speed within 24 hours after landfall, movement speed of vertical coastline within 24 hours after landfall, landfall intensity and its relationship with τ value of 143 tropical cyclones from 1967 to 2018;

Figure 4 is the change curve of the generation position and landfall position of the landfalling tropical cyclone from 1967 to 2018;

Figure 5 is an analysis of the generation, development and large-scale environmental differences of tropical cyclones that landed in East China and South China from 1967 to 2018;

Figure 6 is the spatial distribution of large-scale environmental variable differences in the longitude of the landfalling tropical cyclone center in the maximum 10 years and the minimum 10 years;

Figure 7 shows the temporal and spatial evolution of rainfall caused by tropical cyclone Ken and the change curve of tropical cyclone rainfall indicators (Pmean and Pmax) within 48 hours from 18:00 on August 3 to 18:00 on August 5, 1989;

Figure 8 is a boxplot analysis of the change rate of tropical cyclone rainfall indicators caused by landfalling tropical cyclones from 1979 to 2018 with the corresponding τ quantile of landfalling tropical cyclones;

Figure 9 shows the time variation of tropical cyclone rainfall indicators within the unrestricted rainfall range caused by landfalling tropical cyclones from 1979 to 2018 and the analysis of their response to the corresponding annual τ mean value;

Figure 10 shows the time variation of tropical cyclone rainfall indicators within the limited rainfall range caused by landfalling tropical cyclones from 1979 to 2018 and the analysis of their response to the corresponding annual τ mean value.

Detailed ways

In order to make the purpose, technical solution and advantages of the present invention clearer, the embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

The present invention first explains the relevant basic concepts and the core points of the application as follows, and then elaborates on the technical solution of the present invention.

Please refer to Fig. 1, Fig. 1 is a schematic flow chart of the method of the present invention; an enhanced hybrid differential evolution method for interplanetary detection orbit design, including the following:

Step S1: data collection; collect the measured best track data of tropical cyclones, global precipitation data and meteorological reanalysis data;

Step S2: Screening of landfalling tropical cyclones; combined with the optimal path data of tropical cyclones obtained in step S1, firstly screen all landfalling tropical cyclones, and further select the appropriate landfalling tropical cyclone according to the screening principle of landfalling tropical cyclones;

Step S3: Identify the time scale of the intensity decay of the landfalling tropical cyclone; combine the landfalling tropical cyclone obtained in step S2, perform Kaplan-DeMaria exponential decay fitting on the intensity of each tropical cyclone within 24 hours after landfall to identify the corresponding landfalling tropical cyclone intensity decay time scale (τ), and further exclude events greater than 2 times the standard deviation of the mean value of τ;

Step S4: Identify the characteristics of the attenuation change of the intensity of the landfalling tropical cyclone; combine the τ values of the landfalling tropical cyclones obtained in Step S3, count the time series of the τ values of the landfalling tropical cyclones, and analyze the probability of the τ value of the landfalling tropical cyclones in the two stages before and after the research period Density curves, and identify the temporal variation characteristics of landfalling tropical cyclones τ;

Step S5: Identification of factors affecting the attenuation of landfalling tropical cyclone intensity; combined with the time series of landfalling tropical cyclone τ values obtained in step S4, the correlation analysis between factors that may affect the attenuation of landfalling tropical cyclones and this time series is carried out to identify the factors affecting landfalling tropical cyclones. All factors of change in cyclone intensity attenuation;

Step S6: Quantification of the relative contribution of factors affecting the attenuation change of the intensity of the landfalling tropical cyclone; combined with all the factors that affect the attenuation change of the intensity of the landfalling tropical cyclone obtained in step S5, quantify the relative contribution of each factor to the attenuation change of the intensity of the landfalling tropical cyclone according to the relative contribution calculation method contribute;

Step S7: Analysis of the physical mechanism of the attenuation change of the intensity of the landfalling tropical cyclone; combining the characteristics of the attenuation change of the intensity of the landfalling tropical cyclone obtained in step S4 with all the factors affecting the attenuation change of the intensity of the landfalling tropical cyclone obtained in step S5, the transition from the track of the tropical cyclone and the large-scale Two aspects of environmental characteristics clarify the physical mechanism of the attenuation change of the intensity of landfalling tropical cyclones;

Step S8: Identification of landfall tropical cyclone rainfall and its quantitative indicators; combined with the landfall tropical cyclone obtained in step S2, based on the objective weather map analysis method to identify the landfall tropical cyclone rainfall field, and then identify the landfall tropical cyclone rainfall index.

Step S9: Analysis of the impact of landfall tropical cyclone intensity attenuation changes on tropical cyclone rainfall; combined with the τ value of landfall tropical cyclone obtained in step S3 and the landfall tropical cyclone rainfall index obtained in step S8, the change of tropical cyclone rainfall index is counted, and the tropical cyclone rainfall index is calculated. Correlation analysis between cyclone rainfall index and landfall tropical cyclone τ value, and then analyze the impact of landfall tropical cyclone intensity attenuation changes on tropical cyclone rainfall.

In the method of the present invention, in the step S1, the optimal path data of the tropical cyclone includes the cyclone center position and the maximum wind speed near the center of the tropical cyclone every 6 hours; the global precipitation data is a grid with high temporal and spatial resolution. The point precipitation data has a time resolution of 3 hours and a spatial resolution of 0.1°×0.1°; the meteorological reanalysis data includes meridional wind, zonal wind, relative vorticity, vertical velocity, relative humidity, specific humidity, Whole-layer water vapor, meridional water vapor flux, zonal water vapor flux, soil moisture, and sea surface temperature.

In the step S2, the principle of screening tropical cyclones for landfall includes four steps: 1. The intensity of the tropical cyclone at a position before landing should reach the level of a severe tropical storm and above (intensity ≥ 24.5m/s); 2. The tropical cyclone should be at least Stay for 24 hours, that is, the tropical cyclone has at least 4 consecutive position records of 6 hours on land; ③The intensity of the tropical cyclone cannot increase after landfall; ④The tropical cyclone has one position before landfall and four positions after landfall Temperate denaturation and temperate transition cannot occur.

In the step S3, the formula for the Kaplan-DeMaria exponential decay of the intensity of the tropical cyclone within 24 hours after landfall is as follows:

In the formula, V(t) is the intensity of the landfalling tropical cyclone, and t is the time within 24 hours after the tropical cyclone lands (because the time of the tropical cyclone is divided into 6 hours, so t=t ₁ , t ₂ , t ₃ , t ₄ , V(t ₁ ) is the intensity of the first location of the landfalling tropical cyclone, and τ is the time scale of the intensity decay of the landing tropical cyclone. Therefore, the fitting line of ln(V(t)/V(t ₁ )) at the four locations The slope is -1/τ. The larger the value of τ, the slower the intensity of the landfalling tropical cyclone decays.

In the step S4, the time-varying feature of the time scale of the intensity decay of the landfalling tropical cyclone is the variation trend calculated by the least square method, and the slope obtained by fitting the time series of the research variables by the least square method is the variation trend. The least squares fitting formula for the slope is as follows:

为研究时期年份的均值，y_i为研究时期中研究变量的第i年的值，

为研究时期内研究变量的均值。In the formula, b is the slope, n is the total number of years in the research period, x _i is the i-th year in the research period,

is the mean value of the years in the research period, y _i is the value of the i-th year of the research variable in the research period,

is the mean value of the research variable during the study period.

In the step S5, the factors that may affect the attenuation of the landfalling tropical cyclone include the location of the tropical cyclone landing center (including the longitude and latitude of the landing center), the landing intensity, the moving speed within 24 hours after landing, and the offshore sea surface temperature, and then use the spearman correlation The correlation between the aforementioned factors and the time series of landfalling tropical cyclone τ values was analyzed by coefficient, and then the significance of the obtained correlation was tested by using the t test. Finally, all factors affecting the attenuation of landfalling tropical cyclones were screened according to the magnitude and significance of the correlation.

In the step S6, the relative contribution calculation method includes three steps:

(1) Firstly, the study period is divided into two periods before and after, and the average value of landfalling tropical cyclone τ in these two periods is counted, and then the increment N _a of the average value of landfalling tropical cyclone intensity τ in the latter period relative to the previous period is calculated (this increment The amount is the increment under the joint influence of all factors); ② When calculating the relative contribution of a certain factor, it is assumed that other factors do not change in the later period, that is, only the influence of this factor on the attenuation of landfalling tropical cyclones is considered in the latter period, Calculate the mean value of the landfalling tropical cyclone τ in the latter period, and make a difference between the mean value and the mean value of the previous period obtained in ① to obtain the increment N ₁ of the mean value of the landfalling tropical cyclone τ under the influence of this factor alone; The percentage (N ₁ /N _a ·100%) obtained by dividing the increment of the mean value of landfalling tropical cyclones τ by the increment under the joint influence of all factors is the relative contribution of this factor to the intensity attenuation of landfalling tropical cyclones.

In the step S7, the shift of the tropical cyclone track is obtained by studying the long-term change trend of the tropical cyclone generation location and landing location, and then dividing the tropical cyclone landing location into two regions, and performing the generation and development of these two regions and the large-scale environmental difference Analysis, the research variables include the generation location of the landfalling tropical cyclone (generation longitude and latitude), duration at sea, landfall intensity, moving speed within 24 hours after landfall, and specific humidity at 500hPa at the landfall point. In addition, it is necessary to combine the spatial difference distribution of large-scale environmental variables to clarify the physical mechanism of the attenuation change of the intensity of landfalling tropical cyclones.

In the step S8, there are six rainfall indicators for landfalling tropical cyclones: ① grid point rainfall average value, that is, the average value of rainfall at each grid point within a unit time within the rainfall range caused by tropical cyclones; ② grid point rainfall maximum value, that is, tropical cyclone The maximum value of the grid point rainfall per unit time within the rainfall range; ③ The average value of the total grid point rainfall within the unrestricted rainfall range, that is, the rainfall of all grid points that have experienced rainfall within the unrestricted rainfall range within a fixed period of time after the tropical cyclone landed The average value of the total amount; ④The maximum value of the total rainfall of the grid points within the unrestricted rainfall range, that is, the maximum value of the total rainfall of all grid points that have experienced rainfall within the unrestricted rainfall range within a fixed period of time after the tropical cyclone landed; ⑤Restricted rainfall The average value of total rainfall at grid points within the range, that is, the average value of total rainfall at all grid points that have experienced rainfall within the limited rainfall range within a fixed period of time after tropical cyclone landfall; ⑥ The maximum value of total rainfall at grid points within the limited rainfall range, namely The maximum value of the total rainfall of all grid points that have experienced rainfall within the limited rainfall range within a fixed period of time after the tropical cyclone makes landfall.

It should be noted that the unlimited range of rainfall means the range composed of all the grid points that have experienced rainfall in a certain period of time when the tropical cyclone landed, that is to say, these grid points are not subject to the requirement that there must be rainfall per unit time in this period. limit. The opposite is to limit the range of rainfall, which means that the tropical cyclone lands in the range of all grid points that have rainfall at each unit time within a certain period of time, that is to say, these grid points have rainfall at each unit time during this period . Therefore, the number of rainfall grid points in the unrestricted range is large, and the number of rainfall grid points in the restricted range is small.

In the step S9, the changes in the tropical cyclone rainfall index include: 1. The τ values of all landfalling tropical cyclones are divided into 10 quantile intervals (0-10th, 10-20th, ..., 90-100th) from small to large The change of the grid point rainfall mean value and the interval mean value of the grid point rainfall maximum value change rate within a fixed period of time; ② The long-term change trend of the grid point total rainfall average value and the grid point total rainfall maximum value obtained by the least square method; The correlation analysis between the tropical cyclone rainfall index and the landing tropical cyclone τ value in step S9 is to use the spearman correlation coefficient to analyze the correlation between the mean value of the quantile range of the landing tropical cyclone τ value and the mean value of the rainfall change rate in the corresponding interval, and the grid point rainfall The correlation between the time series of total mean value (maximum value of total rainfall at grid points) and the time series of landfall tropical cyclone τ, and the significance of the obtained correlation was tested by t test.

Embodiment Taking the tropical cyclone (tropical cyclone, TC) that landed in China from 1967 to 2018 as an example, the technical solution of the identification method of intensity attenuation of landfalling tropical cyclone and its impact on rainfall assessment method of the present invention will be further described. The examples are used to illustrate the present invention, but not to limit the scope of application of the present invention, and it is equally applicable to subregions or time periods.

Landing tropical cyclone intensity attenuation identification of the present invention's method and the implementation flowchart of the evaluation method of rainfall impact thereof are as shown in Figure 1, and concrete steps are as follows:

(1) Collection of basic data;

In this embodiment, the tropical cyclone best track data set of IBTrACS version 4 (International Best Track Archive for Climate Stewardship version 4) was collected, including the cyclone center position, the maximum wind speed near the center of the tropical cyclone every 3 hours, and the time of the data set Covers 1967-2018. It should be noted that when analyzing the intensity attenuation changes of landfalling tropical cyclones, the 6-hour track data from the IBTrACS dataset from 1967 to 2018 was selected for research; 3-hour track data to establish the relationship between tropical cyclones and rainfall; collected precipitation variables in the data set driven by surface meteorological elements in China, the time series is from 1979 to 2018, the time resolution is 3 hours, and the horizontal spatial resolution is 0.1°× 0.1°; the ERA5 reanalysis data from the European Center for Medium-Range Weather Forecasting (ECMWF) was collected. Meteorological reanalysis data include meridional wind, zonal wind, relative vorticity, vertical velocity, relative humidity, specific humidity, whole layer water vapor, Meridional water vapor flux, zonal water vapor flux, soil moisture, and sea surface temperature. Table 1 is the selected data information:

Table 1 main data information

(2) Screening of landfalling tropical cyclones;

In this example, the tropical cyclone optimal track data of IBTrACS version 4 is used to screen the tropical cyclones that landed in China during 1967-2018. The specific screening principles for landfall tropical cyclones include four steps: Reaching the level of severe tropical storm and above (intensity ≥ 24.5 m/s); ② The tropical cyclone stays on land for at least 24 hours, that is, the tropical cyclone has at least 4 consecutive position records of 6 hours on land; ③ After the tropical cyclone lands The intensity of the tropical cyclone cannot increase; ④The tropical cyclone cannot change from extratropical zone to extratropical zone at one position before landfall and at four positions after landfall. According to the above four conditions, a total of 150 tropical cyclones were selected.

(3) Identification of the time scale of intensity decay of landfalling tropical cyclones;

In this embodiment, based on the 150 tropical cyclones that landed in China in 1967-2018 obtained in embodiment (2), the intensity of each tropical cyclone within 24 hours after landing is carried out by Kaplan-DeMaria exponential fitting, and the Kaplan-DeMaria exponential formula The formula for attenuation is as follows:

In the formula, V(t) is the intensity of the landfalling tropical cyclone, and t is the time within 24 hours after the tropical cyclone lands (because the time of the tropical cyclone is divided into 6 hours, so t=t ₁ , t ₂ , t ₃ , t ₄ ), V(t ₁ ) is the intensity of the first location of the landfalling tropical cyclone, and τ is the time scale of the intensity decay of the landfalling tropical cyclone. Therefore, the slope of the fitted line at the four locations is -1/τ. The larger the value of τ, the slower the intensity of the landfalling tropical cyclone decays.

Next, 7 events with anomalously large decay timescales (τ) were excluded from the 150 tropical cyclones (i.e. events larger than 2 standard deviations of the mean value of τ were excluded). In the end, a total of 143 tropical cyclone landfall events were used as the research objects of the impact of landfall tropical cyclone intensity decay on rainfall.

(4) Identification of the intensity attenuation and variation characteristics of landfalling tropical cyclones;

In this embodiment, based on the τ values of 143 tropical cyclones that landed in China in 1967-2018 obtained in the embodiment (3), firstly, 1992 was used as a boundary, and the research period 1967-2018 was divided into two periods of equal length (1967 -1992 and 1993-2018), the 143 tropical cyclone tracks screened in these two periods are shown in Figure 2a. From 1967 to 1992 and from 1993 to 2018, the coverage intervals of the attenuation time scales (i.e., τ) of landfalling tropical cyclones are relatively large, indicating that the intensity attenuation of various landfalling tropical cyclones varies greatly, implying that there are multiple factors that have an impact on the attenuation of a single tropical cyclone. impact (Fig. 2b). Compared with 1967-2018, the probability density curve of τ value shifted significantly to the right in 1993-2018, and the high value of τ occurred more frequently during this period, indicating that the attenuation rate of landfalling tropical cyclones slowed down.

Furthermore, the time series of landfalling tropical cyclone τ values were counted, and the change trend of τ value was calculated according to the least square method. This change trend is the change characteristic of the intensity decay speed of 143 tropical cyclones that landed in China from 1967 to 2018. As shown in Figure 2c, from the perspective of the moving average τ value change from 1967 to 2018, the τ value showed a significant upward trend at a rate of 1.8h/decade in the past 1967-2018 (p<0.01), increasing from 22h to 32h, the increase rate is 45%.

(5) Identification of factors affecting the intensity attenuation change of landfalling tropical cyclones;

In this embodiment, based on the time series of tropical cyclone τ values that landed in China in 1967-2018 obtained in Example (4), the location of the tropical cyclone landing center (including the longitude and latitude of the landing center), the intensity of the landing, and the post-landing time series of the tropical cyclone were analyzed. The moving speed within 24 hours and the long-term change trend of sea surface temperature, and the correlation between the above factors and the time series of landing tropical cyclone τ value were analyzed by spearman correlation coefficient, and then the t test was used to test the significance of the correlation obtained, and finally All factors affecting the attenuation of landfalling tropical cyclones were screened according to the magnitude of their correlation and their significance.

The results show that from 1967 to 2018, the longitude of the tropical cyclone's landfall center showed a significant increasing trend at a rate of 0.23 degrees/decade (p<0.05), and showed a significant positive correlation with τ (r=0.34, p<0.05; Fig. 3a-b); the latitude of the landfall center only shows a weak pole shift trend at 0.07degrees/decade, and shows a weak positive correlation with τ (r=0.19, p=0.18; Fig. 3c-d); the intensity of tropical cyclone landfall is The rate of 0.40(m/s)/decade showed a significant increasing trend (p<0.05), and showed a significant positive correlation with τ (r=0.27, p<0.10; Figure 3i-j); within 24 hours after landing The moving speed of , and the moving speed vertical to the coastline (V _t and V _t sin∝) have a weak positive correlation with τ (the correlation coefficients r are all low and have not passed the significance test (Fig. 3e-h); In the past 52 years, it showed a significant upward trend at a rate of 0.11 K/decade (p<0.01), and had a significant positive correlation with the value of τ (r=0.49, p<0.01; Figure 2d, f). In conclusion, The order of factors affecting the intensity attenuation of landfalling tropical cyclones is sea surface temperature, center position of tropical cyclone landfall, and landfall intensity, while the moving speed of tropical cyclones within 24 hours after landfall has a weak influence on the intensity attenuation of landfalling tropical cyclones.

(6) Quantification of the relative contribution of factors affecting the intensity attenuation change of landfalling tropical cyclones;

In this embodiment, the strength order of the influencing factors of the intensity attenuation of the landfalling tropical cyclone obtained based on the embodiment (5) is sea surface temperature, the center position of the tropical cyclone landfall, and the landfall intensity. Then, according to the relative contribution calculation method, the relative contribution of each factor to the intensity attenuation change of the landfalling tropical cyclone is quantified. Considering that the τ change in this study is highly correlated with the sea surface temperature and the landfall location (landfall longitude and latitude), refer to Li and Chakraborty The calculation method of τ first calculates the relative contribution of sea surface temperature increase and landfall position change to τ. The specific process is as follows (see Table 2 for the following values):

① From 1967 to 1992, the average value of τ in East China (South China) was 33.29 (25.53) h, and the proportion of landfalling tropical cyclone events was 46.15% (53.85%). At 34.58 (26.27) h, the proportion of landfalling tropical cyclone events was 53.95% (46.05%), and the mean values of τ in both regions were increasing. From 1967 to 1992 (1993 to 2018), the mean value of τ for landfalling tropical cyclone events in China was 29.11 (30.76) h, and the value of τ increased by 1.65 h in the two periods before and after.

②If the proportion of landfalling tropical cyclone events remains unchanged (at this time, the increase of sea surface temperature is used as a control variable), the mean value of τ for landfalling tropical cyclone events in China from 1993 to 2018 is 30.11h (46.15%×34.58+53.85% ×26.27=30.11), the increase of sea surface temperature leads to an increment of τ of 1h (30.11-29.11=1h), so the relative contribution of sea surface temperature to τ is 60.6% (1/1.65=60.6%).

③If there is no change in sea surface temperature between 1967-1992 and 1993-2018 (at this time, the position change of landfalling tropical cyclones is used as a control variable), the mean value of τ for China's landfalling tropical cyclone events in 1993-2018 is 29.72h ( 53.95%×33.29+46.05%×25.53=29.72), the change in the position of the landfalling tropical cyclone causes the increment of τ to be 0.61h (29.72-29.11=0.61), so the relative contribution of the change in the position of the landfalling tropical cyclone to τ is 37.0% (0.61 /1.65=37.0%). At the same time, it can be obtained that the relative contribution of the increase in the intensity of landfalling tropical cyclones to τ is 2.4% (100%-60.6%-37.0%=2.4%).

To sum up, in this example, the relative contributions of sea surface temperature increase, landfall position change and landfall tropical cyclone intensity increase to τ are 60.6%, 37.0% and 2.4%, respectively.

Table 2 Comparison of intensity decay time scales (τ) of tropical cyclones landed in East China and South China

(7) Analysis of the physical mechanism of the intensity attenuation change of the landfalling tropical cyclone;

In this embodiment, based on the characteristics of the attenuation change of the intensity of the landfalling tropical cyclone obtained in embodiment (4) and all factors affecting the attenuation of the intensity of the landfalling tropical cyclone obtained in the embodiment (5), from the track transfer of the tropical cyclone and the large-scale environmental characteristics Two aspects clarify the physical mechanism of the intensity attenuation change of landfalling tropical cyclones.

First, the track shift of tropical cyclones is obtained by studying the long-term change trends of tropical cyclone generation locations and landfall locations. Figure 4 shows that from 1967 to 2018, the landfall longitude and latitude of tropical cyclones increased by 0.23degrees/decade and 0.15degrees, respectively. The rate of /decade showed a significant increasing trend (p<0.05 and p<0.10); the longitude of tropical cyclones showed a significant increasing trend at a rate of 0.60degrees/decade (p<0.05), and the generating latitude of -0.14degrees/decade The rate of decade shows a downward trend. In this embodiment, the landfall position of the tropical cyclone is more easterly and northerly, and the generation position is also more easterly.

Subsequently, the tropical cyclone landing location is divided into two regions, East China and South China, and the generation and development of these two regions and the large-scale environmental differences are analyzed. The research variables include the generation location (longitude and latitude) of the landing tropical cyclone, and the duration at sea. , landing intensity, moving speed within 24 hours after landing and specific humidity of 500hPa at the landing point. Figure 5 shows that, judging from the location of tropical cyclones that landed in East China and South China and their duration at sea (Fig. 5a-c), the longitude, latitude, and duration of tropical cyclones that landed in East China were significantly higher than those in South China, indicating that the tropical cyclones that landed in East China were generated more eastward and northerly than those that landed in South China. This makes the track of tropical cyclones more tortuous and leads to an increase in the duration of tropical cyclones at sea, implying that the duration of tropical cyclones landed in East China is longer than that in South China. Tropical cyclones last longer at sea and may carry more moisture inside the storm. Judging from the difference in the average specific humidity of the middle layer between East China and South China within 6 hours before and after the tropical cyclone landfall point (Fig. 5f), the specific humidity content at 500hPa when the tropical cyclone landed in East China was significantly higher than that in South China, indicating that the duration of the tropical cyclone at sea is limited. An increase potentially increases the water vapor content carried by landfalling tropical cyclones. The amount of water vapor inside the tropical cyclone is an important factor affecting the change of τ. The more water vapor carried by the tropical cyclone when it lands, the slower the intensity of the landfalling tropical cyclone will decay.

Then, combined with the spatial difference distribution of large-scale environmental variables, the physical mechanism of the attenuation change of the intensity of landfalling tropical cyclones is clarified, and the differences in environmental variables between the 10 years with the largest longitude of the landfall center and the 10 years with the smallest longitude of the landfall center are synthesized and analyzed. It is obtained that in the 10 years when the longitude of the landfall center is the largest, the positively anomalous low-level relative vorticity in East China and the anomalous counterclockwise airflow above it are conducive to maintaining a strong anticlockwise cyclone after the TC landfall (Fig. 6a, d); The negative anomalous mid-level vertical velocity (increased ascending motion) is conducive to the maintenance of the central updraft after the tropical cyclone lands (Fig. 6b); the relatively high relative humidity in the middle and low layers and the significantly increased water vapor in the whole layer can promote the condensation of water vapor and release latent heat. Thus maintaining the warm core structure of the tropical cyclone (Figure 6c, f); the abnormally weak vertical wind shear is also conducive to maintaining the warm core structure of the tropical cyclone (Figure 6d); the abnormally high soil moisture in East China is conducive to increasing the thermal The conductivity in turn produces a more persistent latent heat flux, which is conducive to the maintenance of tropical cyclones on land after landfall (Fig. 6h). However, South China does not have the favorable environmental conditions mentioned above.

(8) Identification of landfall tropical cyclone rainfall and its quantitative indicators;

In the present embodiment, based on the landing tropical cyclone obtained in embodiment (2), based on the objective weather map analysis method to identify the landing tropical cyclone rainfall field, and then identify the landing tropical cyclone rainfall index, in order to study the impact of the landing tropical cyclone intensity attenuation on the tropical cyclone For the impact of rainfall, six evaluation indicators are adopted: ① Grid point rainfall mean (Pmean), that is, the average value of rainfall at each grid point within the rainfall range caused by tropical cyclones every three hours; ② Grid point rainfall maximum value (Pmax), That is, the maximum value of the grid point rainfall within the rainfall range caused by the tropical cyclone every three hours; ③ The average value of the total rainfall at the grid point within 24 (48) hours after landfall PTmean24 (PTmean48), that is, the TC within 24 (48) hours after landfall. The average value of the total rainfall at each grid point within the unrestricted rainfall range (the range composed of all grid points that have experienced rainfall in the corresponding research period after the tropical cyclone lands); ④The maximum total rainfall at the grid point within 24 (48) hours after landfall The value PTmax24 (PTmax48), that is, the maximum value of the total rainfall of each grid point within the unrestricted rainfall range within 24 (48) hours after the tropical cyclone landed; -PTmean24, that is, the total amount of rainfall at each grid point within the limited rainfall range within 24 hours after the tropical cyclone landed (the range composed of all grid points that have rainfall at each unit time in the research period after the tropical cyclone landed) The mean value; ⑥The maximum value of the total rainfall at the grid point within 24 hours after landfall re-PTmax24, that is, the maximum value of the total rainfall at each grid point within the limited rainfall range within 24 hours after the tropical cyclone landed.

Taking the tropical cyclone Ken that landed in China in 1989 and its induced rainfall as an example of the above six indicators, the spatial distribution of rainfall caused by tropical cyclone Ken at each unit time within 48 hours and its time-varying curve are shown in Figure 7. The average value and maximum value of rainfall of all grid points in the spatial distribution map of each unit time in Figure 7a-q are Pmean and Pmax, so the change curves of Pmean and Pmax at each unit time within 48 hours caused by Ken are as follows Shown in Figure 7r. The re-PTmean24 and re-PTmax24 calculated by Ken within the limited range are the mean and maximum values of the total rainfall from 0 to 24h for all grid points in the black dot area in Figure 7a-i. The PTmean24 (PTmean48) and PTmax24 (PTmax48) in the unlimited range caused by Ken are the mean and maximum values of the total rainfall of each grid point within 0-24 (0-48) hours.

(9) Analysis of the impact of landfall tropical cyclone intensity attenuation changes on tropical cyclone rainfall;

In the present embodiment, based on the τ value of the landfalling tropical cyclone obtained in embodiment (3) and the six landfalling tropical cyclone rainfall indicators obtained in embodiment (8), first understand the change of the tropical cyclone rainfall rate of change with more detailed τ intervals τ is divided into 10 quantile intervals (0-10th, 10-20th, ..., 90-100th) from small to large, and the tropical cyclone rainfall change rate in these 10 τ quantile intervals is counted , draw a box plot and construct its linear regression model with τ. Judging from the linear correlation between the τ quantile interval and the tropical cyclone rainfall change rate sequence (Figure 8), the change rates of Pmean24, Pmax24, Pmean48, and Pmax48 all increase with the increase of the τ quantile interval. Among them, the rate of change of Pmean48 and Pmax48 had a significant (p<0.01) positive correlation with the τ quantile interval (r was 0.30 and 0.36, respectively; Figure 8c-d), which indicated that the intensity decay of landing TC slowed down to 48 The increase in the rate of change of rainfall rate within hours is more obvious, which means that the slowing down of landfall TC intensity attenuation will enhance TC rainfall within 48 hours after landfall.

Subsequently, the long-term trend of each tropical cyclone rainfall index was calculated by the least square method, and the results showed that PTmean24, PTmax24, PTmean48, PTmax48, re-PTmean24 and re-PTmax24 all showed a significant increase trend during 1979-2018 (p< 0.01 or p<0.05), the increasing trends were 2.0 mm/decade, 23.0mm/decade, 1.9mm/decade, 24.2mm/decade, 13.1mm/decade and 16.7mm/decade (see Figure 9a, c, e, g and Fig. 10a,c). Further, the correlation analysis of the time series of tropical cyclone rainfall indicators and landfall tropical cyclone τ values was carried out, and the results showed that PTmean24, PTmax24, PTmean48, PTmax48, re-PTmean24 and re-PTmax24 were all significantly related to τ (p<0.01) Highly positive correlation (r is 0.64, 0.76, 0.67 and 0.70, respectively; Fig. 9b, d, f, h and Fig. 10b, d). This shows that the time scale τ of the intensity decay of the landfalling tropical cyclone is the main influencing factor of PTmean24, PTmax24, PTmean48, and PTmax48. The slower the intensity of the landfalling tropical cyclone is, the slower the intensity of the landfalling tropical cyclone is. The mean and maximum value of the total amount are correspondingly higher.

The beneficial effects of the present invention are:

(1) The present invention proposes a kind of identification method of landfall tropical cyclone intensity decay speed situation, has identified the long-term change trend of landfall tropical cyclone intensity decay speed, has analyzed the physical mechanism of landfall tropical cyclone intensity decay speed change, which is important for landfall tropical cyclone intensity decay speed. Estimating the destructive potential of a cyclone is of great importance.

(2) The present invention has analyzed the influence of the attenuation change of the intensity of the landfalling tropical cyclone on the temporal and spatial process of the rainfall by the difference of the tropical cyclone rainfall under different landfalling tropical cyclone intensity attenuation intervals, and has clarified that the attenuation of the intensity of the landing tropical cyclone slows down the amplification of the tropical cyclone rainfall effect.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (8)

1. A method for identifying the strength attenuation of a tropical cyclone during landing and evaluating the influence of the tropical cyclone on rainfall is characterized by comprising the following steps: the method comprises the following steps:

step S1: collecting data; collecting actually measured tropical cyclone optimal path data, global precipitation data and meteorological reanalysis data;

step S2: screening the tropical cyclone; screening all login tropical cyclones according to the optimal tropical cyclone path data obtained in the step S1, and selecting login tropical cyclones meeting the principle according to the login tropical cyclone screening principle;

and step S3: identifying the intensity attenuation time scale of the tropical cyclone when landing; combining the login tropical cyclones meeting the principle obtained in the step S2, carrying out Kaplan-DeMaria exponential decay fitting on the intensity of each tropical cyclone within 24 hours after login to identify the corresponding login tropical cyclone intensity decay time scale tau, and excluding login tropical cyclone events with the standard deviation being 2 times larger than the mean value of tau;

and step S4: identifying the characteristic of attenuation change of the intensity of the tropical cyclone during landing; combining the values tau of the login tropical cyclones obtained in the step S3, counting time sequences of the values tau of the login tropical cyclones, analyzing probability density curves of the values tau of the login tropical cyclones in two stages before and after a research period, and identifying time variation characteristics of the values tau of the login tropical cyclones;

step S5: identifying influence factors of the attenuation change of the strength of the tropical cyclone during landing; performing correlation analysis between factors possibly influencing the attenuation of the landing tropical cyclone and the time sequence by combining the time sequence of the tau value of the landing tropical cyclone obtained in the step S4, and identifying all factors influencing the attenuation change of the intensity of the landing tropical cyclone;

step S6: quantifying the relative contribution of the influence factors of the attenuation change of the cyclone strength of the landing tropical; combining all the factors influencing the attenuation change of the intensity of the login tropical cyclone obtained in the step S5, and quantifying the relative contribution of each factor to the attenuation change of the intensity of the login tropical cyclone according to a relative contribution calculation method;

step S7: analyzing a physical mechanism of the attenuation change of the intensity of the landed tropical cyclone; combining the characteristic of the attenuation change of the intensity of the login tropical cyclone obtained in the step S4 with all factors influencing the attenuation change of the intensity of the login tropical cyclone obtained in the step S5, and clarifying a physical mechanism of the attenuation change of the intensity of the login tropical cyclone from two aspects of tropical cyclone path transfer and large-scale environmental characteristics;

step S8: logging in tropical cyclone rainfall and identification of quantitative indexes of the tropical cyclone rainfall; and (3) identifying the login tropical cyclone rainfall field based on an objective weather map analysis method by combining the login tropical cyclone obtained in the step (S2), and further identifying the login tropical cyclone rainfall index.

Step S9: analyzing the influence of the attenuation change of the intensity of the landing tropical cyclone on the rainfall of the tropical cyclone; and (4) combining the tau value of the login tropical cyclone obtained in the step (S3) with the rainfall index of the login tropical cyclone obtained in the step (S8), counting the change condition of the rainfall index of the tropical cyclone, analyzing the correlation between the rainfall index of the tropical cyclone and the tau value of the login tropical cyclone, and further analyzing to obtain the influence of the attenuation change of the strength of the login tropical cyclone on the rainfall of the tropical cyclone.

2. The method for identifying the attenuation of the intensity of landing tropical cyclone and evaluating the influence of landing tropical cyclone on rainfall as claimed in claim 1, wherein: in step S1, the optimal path data of the tropical cyclone includes a cyclone center position and a near-center maximum wind speed of the tropical cyclone every 6 hours;

the global precipitation data are lattice precipitation data with high space-time resolution, the time resolution is 3 hours, and the space resolution is 0.1 degree multiplied by 0.1 degree;

the meteorological reanalysis data comprises warp wind, weft wind, relative vorticity, vertical speed, relative humidity, specific humidity, whole layer water vapor, warp water vapor flux, weft water vapor flux, soil humidity and sea surface temperature.

3. The method for identifying the attenuation of the intensity of landing tropical cyclone and evaluating the influence of landing tropical cyclone on rainfall as claimed in claim 1, wherein: in step S2, the logging-in tropical cyclone screening principle includes four steps:

step 1: the intensity of the hot belt cyclone at the position before landing reaches the storm level of the strong hot belt and above;

step 2: the tropical cyclone stays on land for at least N hours; n is a preset value;

and step 3: the strength of the tropical cyclone after landing does not increase continuously;

and 4, step 4: the tropical cyclone does not have the conditions of temperature zone degeneration and temperature zone transition at one position before landing and four positions after landing.

4. A method of assessing the attenuation of tropical cyclone intensity landing and its impact on rainfall as claimed in claim 3, wherein: in step S3, the formula that the intensity within 24 hours after the tropical cyclone landing is in Kaplan-DeMaria exponential attenuation is as follows:

wherein V (t) is the intensity of the tropical cyclone landing, t is the time within 24 hours after the tropical cyclone landing, the time resolution of the tropical cyclone is 6 hours, and t = t ₁ ，t ₂ ，t ₃ ，t ₄ ，V(t ₁ ) τ is the intensity decay time scale for the intensity of the first location of the landing tropical cyclone.

5. The method for identifying the attenuation of the intensity of landing tropical cyclone and evaluating the influence of landing tropical cyclone on rainfall as claimed in claim 1, wherein: in the step S4, the time change characteristic of the attenuation time scale of the tropical cyclone intensity is logged in to obtain a change trend calculated by a least square method, the least square method is used for fitting and researching a time sequence of variables, so that the obtained slope is the change trend, and the least square fitting formula of the slope is as follows:

wherein b is the slope, n is the total years of the study period, x _i For the ith year of the study period,

mean value of years of study period, y _i To investigate the year i value of a variable over the study period,

mean values of study variables over the study period.

6. A method of identifying and assessing the effect of landing tropical cyclone intensity decay on rainfall as claimed in claim 1 in which: in the step S5, factors which may influence the attenuation of the tropical cyclone landing include the landing center position of the tropical cyclone, the landing intensity, the moving speed within 24 hours after landing and the sea surface temperature;

analyzing the correlation between factors capable of influencing the attenuation of the landing tropical cyclone and a time sequence of a tau value of the landing tropical cyclone by adopting a spearman correlation coefficient;

and (4) checking the significance of the obtained correlation by adopting a t test, and screening all factors influencing the attenuation of the landing tropical cyclone according to the magnitude of the correlation and the significance thereof.

7. A method of identifying and assessing the effect of landing tropical cyclone intensity decay on rainfall as claimed in claim 1 in which: in step S6, the relative contribution calculating method includes three steps:

step 1: dividing the research period into a front period and a rear period, counting the mean value of the landing tropical cyclone tau in the two periods, and calculating the increment N of the mean value of the landing tropical cyclone strength tau in the rear period relative to the previous period _a ；

And 2, step: when calculating the relative contribution of any factor, only considering the influence of the factor on the attenuation of the landing tropical cyclone in the later period, calculating the average value of the landing tropical cyclone tau in the later period, and obtaining the average value of the previous period in the step 1Making a difference to obtain the increment N of the mean value of the logging tropical cyclone tau under the single influence of the factor ₁ ；

And step 3: dividing the increment of the mean value of the logging tropical cyclone tau under the influence of a certain factor by the increment under the influence of all factors to obtain the percentage N ₁ /N _a 100%, where the percentage is the relative contribution of this factor to the impact of landing tropical cyclone strength attenuation.

8. A method of identifying and assessing the effect of landing tropical cyclone intensity decay on rainfall as claimed in claim 1 in which: the total number of the tropical cyclone rainfall indexes logged in the step S8 is six: the average value of the lattice point rainfall, the maximum value of the lattice point rainfall, the average value of the lattice point rainfall total amount in the rainfall unlimited range, the maximum value of the lattice point rainfall total amount in the rainfall unlimited range, the average value of the lattice point rainfall total amount in the rainfall limited range and the maximum value of the lattice point rainfall total amount in the rainfall limited range.

Images