CN116224472A - Tropical cyclone rainfall forecast and inspection method based on regional tree - Google Patents
Tropical cyclone rainfall forecast and inspection method based on regional tree Download PDFInfo
- Publication number
- CN116224472A CN116224472A CN202310144318.6A CN202310144318A CN116224472A CN 116224472 A CN116224472 A CN 116224472A CN 202310144318 A CN202310144318 A CN 202310144318A CN 116224472 A CN116224472 A CN 116224472A
- Authority
- CN
- China
- Prior art keywords
- precipitation
- tropical cyclone
- forecast
- checking
- forecasting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000001556 precipitation Methods 0.000 claims abstract description 156
- 239000008186 active pharmaceutical agent Substances 0.000 claims abstract description 21
- 230000001186 cumulative effect Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 4
- 239000006185 dispersion Substances 0.000 description 4
- JTJMJGYZQZDUJJ-UHFFFAOYSA-N phencyclidine Chemical class C1CCCCN1C1(C=2C=CC=CC=2)CCCCC1 JTJMJGYZQZDUJJ-UHFFFAOYSA-N 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Environmental & Geological Engineering (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a tropical cyclone rainfall forecast inspection method based on a regional tree, which relates to the field of meteorological research and comprises the following steps: step one, acquiring tropical cyclone data, and calculating average observation and forecast tropical cyclone positions; step two, defining an inspection area, and extracting and observing and forecasting precipitation of a precipitation field in the inspection area; step three, constructing a regional tree structure for observing and forecasting precipitation; step four, calculating three precipitation form distribution indexes by utilizing information in the regional tree structure; step five, calculating forecast errors of precipitation distribution morphology indexes WDL, WF and DS; and step six, calculating error indexes of three precipitation distribution morphology indexes and integral error indexes of the tropical cyclone precipitation distribution morphology. According to the method, the inspection index is defined according to the tropical cyclone precipitation distribution characteristics, and the difference of the tropical cyclone precipitation distribution characteristic forecasting capability of different modes, different forecasting timeliness and the like can be reflected through the comparison of the error indexes, so that support is provided for improving the tropical cyclone precipitation forecasting.
Description
Technical Field
The invention relates to the field of meteorological research, in particular to a tropical cyclone rainfall forecast and inspection method based on regional trees.
Background
Tropical cyclones are low-pressure vortices occurring on tropical or subtropical ocean surfaces, one of the most damaging natural disasters in the world, and precipitation is one of the most dominant disaster-causing factors of tropical cyclones. The quantitative rainfall forecast of the tropical cyclone is very important for guaranteeing the life and property safety, the quantitative rainfall forecast of the tropical cyclone mainly depends on the mode forecast at present, whether the mode forecast can better reflect the observed rainfall distribution characteristics and which defects exist are clearly known, and the method is a precondition for using and improving the mode forecast. In the past, the traditional point-to-point inspection method is mostly adopted for inspecting the tropical cyclone rainfall forecast, and the rainfall forecast error is not fully analyzed.
The tropical cyclone rainfall forecasting is very challenging, namely, the tropical cyclone path forecasting is accurate, the rainfall characteristics along the path can be accurately forecasted by the mode forecasting, the tropical cyclone path forecasting and the mode forecasting are indispensable, and if the mode forecasting cannot accurately describe the distribution of the rainfall around the center of the tropical cyclone, the rainfall forecasting cannot be good even if the path forecasting is accurate again. Whether the mode forecast can better forecast the rainfall around the center of the tropical cyclone is an important standard for measuring the level of the mode forecast.
In the past, the inspection of the rainfall forecast around the center of the tropical cyclone lacks the inspection of the rainfall distribution form, and the distribution of the tropical cyclone rainfall is different from the characteristics of other weather system rainfall, such as a plurality of rain belts distributed around the center of the tropical cyclone, and the mature tropical cyclone has the spiral rain belt which is spirally converged towards the eye wall and the like. In order to evaluate the predictive ability of pattern predictions on these tropical cyclone precipitation profiles, it is necessary to develop new methods that add this section of inspection to more fully evaluate tropical cyclone pattern precipitation predictions.
Disclosure of Invention
The invention aims at: the invention provides a tropical cyclone precipitation prediction inspection method, which is based on a region tree method to establish inspection indexes, and can be used for inspecting various distribution forms of precipitation in tropical cyclone precipitation prediction, including the degree of the whole precipitation rain belt far away from the tropical cyclone center, the dispersion degree of the precipitation form and the spiral degree of the precipitation form.
The technical content is as follows: a tropical cyclone rainfall forecast inspection method based on a regional tree comprises the following steps:
step one, acquiring tropical cyclone data in a certain period, and calculating average observation and forecast tropical cyclone positions in the period;
selecting a checking radius, defining a checking area, and extracting and observing and forecasting precipitation of a precipitation field in the checking area;
step three, constructing a regional tree structure for the observed and forecast precipitation extracted in the step two respectively;
step four, the three precipitation form distribution indexes are calculated respectively by utilizing the information in the regional tree structure for the observation and forecast regional tree structure constructed in the step three; the three precipitation morphology distribution indexes are blade weighted average distance WDL, weighted crushing degree WF and solidity difference DS;
step five, calculating prediction errors of precipitation distribution morphology indexes WDL, WF and DS, namely the difference between prediction and observation;
and step six, calculating error indexes of three precipitation distribution morphology indexes and integral error indexes of the tropical cyclone precipitation distribution morphology.
Further, in the first step, the tropical cyclone data includes the tropical cyclone positions observed and forecasted at the beginning and ending time of the period and the grid point accumulated precipitation data observed and forecasted in the period.
Further, in the first step, when a thermal zone cyclone accumulated precipitation forecast in a certain period is selected for inspection, the applicable accumulated precipitation duration is 3 hours;
the observed tropical cyclone position can be obtained from tropical cyclone optimal path data ibtrucs with a time resolution of 3h; the predicted tropical cyclone position may be obtained using a vortex tracking procedure GFDL Vortex Tracker;
the precipitation data of the observation grid point can be obtained from a global precipitation observation plan GPM, and the time resolution is 30min; the forecast grid point rainfall data is 3h cumulative rainfall forecast in the mode forecast of the required inspection.
In the second step, the radius of inspection is selected to be 500km or the precipitation in the tropical cyclone circulation is separated, and the radius of precipitation in the circulation is taken as the radius of inspection.
In the second step, the specific method of defining the inspection area is to make circles with the average tropical cyclone position observed and forecasted as the center and the selected inspection radius as the radius, and the inspection area of tropical cyclone precipitation is defined in the circles; the specific operation method is that if the distance between the grid point and the center of the tropical cyclone does not exceed the checking radius, the precipitation on the grid point is reserved, otherwise, the precipitation is set as a lack-measured value.
Further, in the third step, the method for constructing the region tree structure includes:
1) Selecting a set of precipitation thresholds for dividing the precipitation field;
2) Converting the precipitation field into a binary field by using the set of thresholds, identifying connected areas in the binary field, constructing a node for each area and storing the grid point coordinates and the grid point values of the nodes;
3) Calculating the attribute of the region and storing the attribute in the node;
4) Identifying overlapping relationships between the regions;
5) A tree structure is used to store information for all nodes.
Further, in the fourth step:
1) The weighted average distance WDL of the blades is the weighted average distance from the geometric center of the leaf nodes in the regional tree to the center of the tropical cyclone, and the weight is the depth of the leaf nodes; the WDL calculation formula is as follows:
wherein N is the number of leaf nodes in the regional tree, and X i For the ith leaf in the region treeGeometric center of node, X TC Is the center coordinate of the tropical cyclone, dist is the large circle distance between two points, dp i Depth for leaf node i;
2) The weighted breaking degree WF is the percentage of the weighted number of broken areas in the area tree to the total area number, wherein the weight is the reciprocal of the depth of the area, the broken areas are the number of areas separated from each other on a certain layer, if only one area exists on the certain layer, the area is complete, and the broken area number is 0; the WF calculation formula is as follows:
wherein D is the depth of the region tree, l is the first layer of the region tree, m is the number of broken regions in the first layer, and n is the number of regions in the first layer;
3) The solidity difference DS is the difference degree of solidity of all nodes in the regional tree, and the solidity is the ratio of the area of the region to the area of the minimum convex polygon containing the region; DS is calculated as follows:
wherein N is the number of nodes in the region tree, S i Is the solidity of the node i and,
the average solidity of all nodes in the regional tree is represented by the absolute value symbol.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention defines the check index according to the tropical cyclone precipitation distribution characteristics, and has more physical significance.
2. The three precipitation morphology distribution indexes in the invention are not only used for inspection, but also can be used for researching the change rule of tropical cyclone precipitation morphology, influence factors thereof and the like.
1) The WDL reflects the degree of the whole rainfall rain belt far away from the tropical cyclone center, wherein the maximum value with large rainfall is given more weight, the subjective judgment of human eyes is more met, and the larger the WDL is, the more the whole rainfall is far away from the tropical cyclone center;
2) WF reflects the dispersion degree of the precipitation form, the precipitation areas are more dispersed when the precipitation intensities are lower, namely, the precipitation areas are separated from each other, and the larger WF indicates the more dispersed precipitation;
3) DS reflects the degree of helicity of the precipitation morphology, with a larger DS indicating a more pronounced heliciform profile of precipitation.
3. The method can find out the prediction deviation of the mode prediction on different distribution forms of the tropical cyclone precipitation, reflect the difference of prediction capacities of different modes, different prediction timeliness and the like through the comparison of error indexes, and provide powerful support for improving the prediction of the tropical cyclone precipitation.
1) Prediction errors of the rainfall distribution morphology indexes WDL, WF and DS reflect the degree that the whole rainfall forecasting rain belt is closer to or far from the tropical cyclone center than the observation, the degree that the rainfall forecasting is more scattered or concentrated than the rainfall distribution, and the difference degree of the spiral degree of the rainfall morphology forecasting and the rainfall observation respectively;
2) The error indexes of the three form indexes of the precipitation distribution are respectively normalized, the form indexes of the precipitation distribution are different at different moments, and the error level of the forecast index relative to the observed index can be better reflected by adopting the error indexes.
3) The three error indexes are averaged by the integral error index of the tropical cyclone precipitation distribution form, and the integral forecasting capability of the tropical cyclone precipitation distribution form is reflected by one index, so that the integral forecasting capability of different forecasting in different modes, different forecasting timeliness and the like can be evaluated and compared conveniently.
Drawings
FIG. 1 is a schematic flow chart of a tropical cyclone precipitation prediction inspection method based on a regional tree;
FIG. 2 is a schematic diagram of a tree structure and common terminology;
FIG. 3 is a schematic diagram of a tree structure for constructing tropical cyclone precipitation areas;
FIG. 4 illustrates a vane weighted average distance (WDL) for heat zone cyclonic precipitation in various embodiments;
FIG. 5 illustrates Weighted Fragility (WF) for heat zone cyclonic precipitation in various embodiments;
FIG. 6 illustrates the corresponding solidity Differential (DS) for the heat zone cyclonic precipitation in various embodiments.
Detailed Description
The following detailed description of the technical solution of the present invention will be given with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, the method for forecasting and checking tropical cyclone precipitation based on regional trees provided by the invention comprises the following steps:
step one, acquiring tropical cyclone information in a certain period, wherein the information comprises observation and forecast of tropical cyclone positions at the starting and ending moments of the period and grid point accumulated precipitation information in the period, and calculating average observation and forecast of tropical cyclone positions in the period;
when the thermal zone cyclone accumulated precipitation forecast in a certain period is selected for inspection, the applicable accumulated precipitation time length is the best 3 h.
Observing tropical cyclone position can be obtained from tropical cyclone optimal path data IBTrACS (International Best Track Archive for Climate Stewardship); the time resolution is 3h, and the website is https:// claatedatagide.
The observation grid point precipitation data may be obtained from a global precipitation observation plan (Global Precipitation Measurement, GPM); the time resolution is 30min, the website is https:// giovanani. Gsfc. Nasa. Gov/giovanani/, and the accumulated precipitation data in a given time period can be directly output from the website.
The predicted tropical cyclone position can be obtained using a vortex tracking program GFDL Vortex Tracker with a website https:// dtcaster.
The forecast grid point rainfall data is 3h cumulative rainfall forecast in the mode forecast of the required inspection.
Selecting a checking radius, defining a checking area, and extracting and observing and forecasting precipitation of a precipitation field in the checking area;
selecting a certain checking radius, and using 500km; preferably, the precipitation in the tropical cyclone circulation can be separated, and the radius of the precipitation in the circulation is taken as the checking radius.
The specific method for defining the inspection area is to make circles by taking the average tropical cyclone position observed and forecasted as the center of a circle and the selected inspection radius as the radius, and the inspection area of the tropical cyclone precipitation is defined in the circles. The specific operation method is that if the distance between the grid point and the center of the tropical cyclone does not exceed the checking radius, the precipitation on the grid point is reserved, otherwise, the precipitation is set as a lack-measured value.
Step three, constructing a regional tree structure for the observed and forecast precipitation extracted in the step two respectively;
fig. 2 is a schematic diagram of a tree structure and common terminology. FIG. 3 is a schematic diagram of a tree structure for constructing tropical cyclone precipitation zones. The precipitation fields and the segmentation threshold values in the figure are only shown schematically, for the extracted tropical cyclone precipitation (fig. 3 a), the precipitation fields are firstly converted into binary fields by the threshold values (fig. 3 b), then the connected areas in the binary fields are identified as precipitation areas (fig. 3 c), and various information contained in each precipitation area is stored in an area tree structure as shown in fig. 3 d. Specific construction steps are given below:
1) A set of precipitation thresholds { ti } i=1, 2, … n for dividing the precipitation field is selected, t1 being set to 0mm for describing the precipitation field using a single tree structure, and t2 being set to 1mm if precipitation above 1mm is analyzed. A uniformly increasing precipitation threshold may be used, ti=t (i-1) +d, i=3, 4 … n, with ti increasing until ti is greater than the maximum value of the precipitation field, d being the interval of the precipitation threshold, typically a desirable d of 1mm. The precipitation threshold value which is not increased uniformly can also be used, a plurality of precipitation threshold values are selected according to the requirements, but the precipitation threshold values are not too sparse;
2) And converting the precipitation field P into a binary field { Pi }, wherein when the binary field { Pi } is converted, the grid point with the precipitation value larger than or equal to ti in the precipitation field P is assigned to be 1, and otherwise, the grid point is assigned to be 0. In particular, P1 has a grid point value of 1 and other grid points of 0 in the range of the tropical cyclone center q=2500 km, which is also for describing the precipitation field by using a single tree structure, wherein the value of q can be other values without affecting the final result, but needs to be larger than the inspection radius of all samples;
3) Identifying connected areas (interconnected lattice points with a value of 1) in Pi, constructing a node n (r) for each area r, storing lattice point coordinates and lattice point values in r, calculating attributes of the areas r such as area, geometric center coordinates, solidity and the like, and storing the attributes in the node n (r);
4) Identifying the overlapping relation between the areas obtained in the third step, assuming that rm and rn are areas identified in Pi and P (i+1), if the coordinates of the grid points in rn are a subset of the coordinates of the rm grid points, rn is a sub-area of rm, and node n (rn) is a sub-node of n (rm);
5) And storing information of all nodes by using a tree structure, wherein each node comprises grid point information and attribute information of a certain area, and parent node and child node information.
Step four, the three precipitation form distribution indexes are calculated respectively by utilizing the information in the regional tree structure for the observation and forecast regional tree structure constructed in the step three;
(1) The first index is the leaf weighted average distance (weighted distance of leaves, WDL), defined as the weighted average distance from the geometric center of the leaf node in the region tree to the center of the tropical cyclone, the weight being the depth of the leaf node; WDL is calculated as follows:
wherein N is the number of leaf nodes in the regional tree, and X i Is the geometric center of the ith leaf node in the region tree, X TC Is the center coordinate of the tropical cyclone, dist is the large circle distance between two points, dp i Is the depth of leaf node i.
The WDL reflects the degree to which the precipitation rain belt is entirely away from the tropical cyclone center, wherein a large maximum value of precipitation is given more weight, the subjective judgment of human eyes is more met, and the greater the WDL is, the further the precipitation is from the tropical cyclone center.
Fig. 4 shows WDLs corresponding to heat-induced cyclonic precipitation in different embodiments, with fig. 4a-d being respectively observed 3h cumulative precipitation for the tropical cyclonic GPM at 2019, 11, 15, 6, 9, 14, 6, 9, 15, 18, 21, and 6, 13, 2017. The precipitation of fig. 4a closely surrounds the tropical cyclone center, the WDL is small, and the precipitation of fig. 4b-d is progressively larger as a whole further from the tropical cyclone center.
(2) The second index is a weighted crush degree (weighted fragmentation, WF), defined as the percentage of the weighted number of crushed areas in the area tree to the total area number, wherein the weight is the reciprocal of the depth of the area, the crushed areas are the number of areas separated from each other on a certain layer, if there is only one area on a certain layer, the areas are complete, and the crushed area number is 0; WF is calculated as follows:
wherein D is the depth of the region tree, l is the first layer of the region tree, m is the number of broken regions in the first layer, and n is the number of regions in the first layer.
WF reflects the degree of dispersion of the precipitation morphology, with precipitation areas appearing more dispersed at lower precipitation intensities, i.e., separated from each other, with larger WF indicating more dispersed precipitation.
FIG. 5 shows WF for heat-induced cyclonic precipitation in various embodiments, and FIGS. 5a-d are respectively taken as heat-induced cyclonic GPM observations for 3h accumulated precipitation at 2019, 11, 13, 18, 21, 8, 5, 21, 8, 6, 0, 30, 0, 3, and 15, 12, 2015. In fig. 5a, precipitation has only one precipitation area at each intensity, no scattered precipitation area exists, precipitation is concentrated most, and WF is 0. The precipitation of fig. 5b appears to be more concentrated overall, but the stronger precipitation distribution is more sporadic, with slightly larger WF. Fig. 5c and d are scattered with several large value centers in addition to the vicinity of the tropical cyclone center, wherein fig. 5d is separated from each other in the lower intensity precipitation area, the precipitation distribution appears very scattered, and WF is maximum.
(3) The third index is the solidity difference (dispersion of solidity, DS), defined as the difference in solidity of all nodes in the region tree, the solidity being the ratio of the region area to the minimum convex polygon area containing the region; DS is calculated as follows:
wherein N is the number of nodes in the region tree, S i Is the solidity of the node i and,
the average solidity of all nodes in the regional tree is represented by the absolute value symbol.
DS reflects the degree of helicity of the precipitation morphology, with a larger DS indicating a more pronounced heliciform profile of precipitation.
Fig. 6 shows DS corresponding to heat zone cyclonic precipitation in different embodiments, and fig. 6a-d are respectively observed for 3h cumulative precipitation for a tropical cyclonic GPM at 2015, 8, 11, 6, 9, 10, 13, 0, 3,4, 3, 6, and 0, 3. In fig. 6a, the precipitation areas of various intensities are substantially ellipsoidal, with very high solidity and very small DS. In FIGS. 6b-d, the spiral distribution of precipitation is more and more pronounced and DS is more and more significant.
Step five, calculating prediction errors of precipitation distribution morphology indexes WDL, WF and DS, namely the difference between prediction and observation;
the three prediction errors respectively reflect the degree that the whole rainfall forecasting rain belt is closer to or far from the center of the tropical cyclone than the observation, the degree that the rainfall forecasting is more scattered or concentrated than the distribution of the observed rainfall, and the difference degree of the spiral degree of the morphology of the forecasting and the observed rainfall.
Step six, calculating error indexes of three precipitation distribution morphology indexes and integral error indexes of the tropical cyclone precipitation distribution morphology;
the error index of the three precipitation distribution morphology indexes, namely the absolute value of the difference between the forecast index and the observation index is divided by the sum of the forecast index and the observation index; the integral error index of the tropical cyclone precipitation distribution form is a numerical value obtained by averaging the error indexes of the three indexes.
Claims (7)
1. The tropical cyclone rainfall forecast inspection method based on the regional tree is characterized by comprising the following steps of:
step one, acquiring tropical cyclone data in a certain period, and calculating average observation and forecast tropical cyclone positions in the period;
selecting a checking radius, defining a checking area, and extracting and observing and forecasting precipitation of a precipitation field in the checking area;
step three, constructing a regional tree structure for the observed and forecast precipitation extracted in the step two respectively;
step four, the three precipitation form distribution indexes are calculated respectively by utilizing the information in the regional tree structure for the observation and forecast regional tree structure constructed in the step three; the three precipitation morphology distribution indexes are blade weighted average distance WDL, weighted crushing degree WF and solidity difference DS;
step five, calculating prediction errors of precipitation distribution morphology indexes WDL, WF and DS, namely the difference between prediction and observation;
and step six, calculating error indexes of three precipitation distribution morphology indexes and integral error indexes of the tropical cyclone precipitation distribution morphology.
2. The method according to claim 1, wherein in the first step, the tropical cyclone information includes tropical cyclone positions observed and forecasted at the beginning and ending time of the period and grid point accumulated precipitation information observed and forecasted during the period.
3. The method for forecasting tropical cyclone precipitation inspection based on regional tree according to claim 2, wherein in the first step, when the thermal cyclone accumulated precipitation forecast in a certain period is selected for inspection, the applicable accumulated precipitation duration is 3h;
the observed tropical cyclone position can be obtained from tropical cyclone optimal path data ibtrucs with a time resolution of 3h; the predicted tropical cyclone position may be obtained using a vortex tracking procedure GFDL Vortex Tracker;
the precipitation data of the observation grid point can be obtained from a global precipitation observation plan GPM, and the time resolution is 30min; the forecast grid point rainfall data is 3h cumulative rainfall forecast in the mode forecast of the required inspection.
4. The method for forecasting and checking tropical cyclone precipitation based on regional trees according to claim 1, wherein in the second step, the checking radius is selected to be 500km or the precipitation in the tropical cyclone circulation is separated, and the precipitation radius in the circulation is taken as the checking radius.
5. The method for forecasting and checking tropical cyclone precipitation based on regional tree according to claim 1, wherein in the second step, the specific method for defining the checking region is to make circles with the average tropical cyclone position observed and forecasted as the center and the selected checking radius as the radius, and the circle is the checking range of tropical cyclone precipitation; the specific operation method is that if the distance between the grid point and the center of the tropical cyclone does not exceed the checking radius, the precipitation on the grid point is reserved, otherwise, the precipitation is set as a lack-measured value.
6. The method for forecasting and checking tropical cyclone precipitation based on regional tree according to claim 1, wherein in the third step, the method for constructing the regional tree structure is as follows:
1) Selecting a set of precipitation thresholds for dividing the precipitation field;
2) Converting the precipitation field into a binary field by using the set of thresholds, identifying connected areas in the binary field, constructing a node for each area and storing the grid point coordinates and the grid point values of the nodes;
3) Calculating the attribute of the region and storing the attribute in the node;
4) Identifying overlapping relationships between the regions;
5) A tree structure is used to store information for all nodes.
7. The method for forecasting and checking tropical cyclone precipitation based on regional trees according to claim 1, wherein in the fourth step:
1) The weighted average distance WDL of the blades is the weighted average distance from the geometric center of the leaf nodes in the regional tree to the center of the tropical cyclone, and the weight is the depth of the leaf nodes; the WDL calculation formula is as follows:
wherein N is the number of leaf nodes in the regional tree, and X i Is the geometric center of the ith leaf node in the region tree, X TC Is the center coordinate of the tropical cyclone, dist is the large circle distance between two points, dp i Depth for leaf node i;
2) The weighted breaking degree WF is the percentage of the weighted number of broken areas in the area tree to the total area number, wherein the weight is the reciprocal of the depth of the area, the broken areas are the number of areas separated from each other on a certain layer, if only one area exists on the certain layer, the area is complete, and the broken area number is 0; the WF calculation formula is as follows:
wherein D is the depth of the region tree, l is the first layer of the region tree, m is the number of broken regions in the first layer, and n is the number of regions in the first layer;
3) The solidity difference DS is the difference degree of solidity of all nodes in the regional tree, and the solidity is the ratio of the area of the region to the area of the minimum convex polygon containing the region; DS is calculated as follows:
wherein N is the number of nodes in the region tree, S i Is the solidity of the node i and,
the average solidity of all nodes in the regional tree is represented by the absolute value symbol. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310144318.6A CN116224472B (en) | 2023-02-21 | 2023-02-21 | Tropical cyclone rainfall forecast and inspection method based on regional tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310144318.6A CN116224472B (en) | 2023-02-21 | 2023-02-21 | Tropical cyclone rainfall forecast and inspection method based on regional tree |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116224472A true CN116224472A (en) | 2023-06-06 |
| CN116224472B CN116224472B (en) | 2024-01-19 |
Family
ID=86569066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310144318.6A Active CN116224472B (en) | 2023-02-21 | 2023-02-21 | Tropical cyclone rainfall forecast and inspection method based on regional tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116224472B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101092380B1 (en) * | 2011-04-29 | 2011-12-13 | 공주대학교 산학협력단 | Estimating method using ensemble average based on the regional climate model performance |
| CN113515871A (en) * | 2021-09-08 | 2021-10-19 | 江苏铨铨信息科技有限公司 | Object-oriented artificial intelligence inspection method for rainfall forecast |
| CN114637806A (en) * | 2022-04-11 | 2022-06-17 | 中国气象局上海台风研究所(上海市气象科学研究所) | Visual analysis method for tropical cyclone forecast inspection index data |
-
2023
- 2023-02-21 CN CN202310144318.6A patent/CN116224472B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101092380B1 (en) * | 2011-04-29 | 2011-12-13 | 공주대학교 산학협력단 | Estimating method using ensemble average based on the regional climate model performance |
| CN113515871A (en) * | 2021-09-08 | 2021-10-19 | 江苏铨铨信息科技有限公司 | Object-oriented artificial intelligence inspection method for rainfall forecast |
| CN114637806A (en) * | 2022-04-11 | 2022-06-17 | 中国气象局上海台风研究所(上海市气象科学研究所) | Visual analysis method for tropical cyclone forecast inspection index data |
Non-Patent Citations (1)
| Title |
|---|
| 黄伟;余晖;梁旭东;: "GRAPES-TCM对登陆热带气旋降水的预报及其性能评估", 气象学报, no. 05 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116224472B (en) | 2024-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111815037B (en) | Interpretable short-critical extreme rainfall prediction method based on attention mechanism | |
| CN111639237B (en) | Electric power communication network risk assessment system based on clustering and association rule mining | |
| Marzban et al. | Ceiling and visibility forecasts via neural networks | |
| Cannon et al. | Synoptic map-pattern classification using recursive partitioning and principal component analysis | |
| CN111352113A (en) | Strong convection weather short-term forecasting method and system, storage medium and terminal | |
| Mahmoodi et al. | Outlier detection in ocean wave measurements by using unsupervised data mining methods | |
| CN114611834A (en) | Electric power station site selection evaluation planning method based on multi-dimensional feature analysis | |
| Nita et al. | Changes in cyclone intensity over Romania according to 12 tracking methods | |
| CN116224472B (en) | Tropical cyclone rainfall forecast and inspection method based on regional tree | |
| CN116911806A (en) | Internet + based power enterprise energy information management system | |
| Mantri et al. | Weather prediction and classification using neural networks and k-nearest neighbors | |
| Cui et al. | Data-based windstorm type identification algorithm and extreme wind speed prediction | |
| Miller et al. | A preliminary assessment of using spatiotemporal lightning patterns for a binary classification of thunderstorm mode | |
| Otero et al. | Zonda wind classification using machine learning algorithms | |
| Refonaa et al. | Cognitive computing techniques based rainfall prediction—a study | |
| Guzman et al. | Heavier inner-core rainfall of major hurricanes in the North Atlantic Basin than in other global basins | |
| Englehart et al. | Diagnosing warm‐season rainfall variability in Mexico: a classification tree approach | |
| Çınaroğlu et al. | A data mining application of local weather forecast for Kayseri Erkilet Airport | |
| Cui et al. | Data-based wind disaster climate identification algorithm and extreme wind speed prediction | |
| CN107977727A (en) | A kind of method that probability is blocked based on social development and climatic factor prediction cable network | |
| Yang et al. | Radar‐Derived Quantitative Precipitation Estimation Based on Precipitation Classification | |
| CN113486514A (en) | Building heat consumption prediction modeling method based on automatic calibration | |
| Duong et al. | An adaptive neuro-fuzzy inference system for seasonal forecasting of tropical cyclones making landfall along the Vietnam coast | |
| Michaelides et al. | Artificial neural networks for the diagnosis and prediction of desert dust transport episodes | |
| Zhao et al. | Characteristics of large-scale atmospheric circulation patterns conducive to severe spring and winter wind events over beijing in china based on a machine learning categorizing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |