Rain and snow temperature threshold parameterization method and device and medium

Abstract

The invention discloses a rain and snow temperature threshold parameterization method and device and a medium, and relates to the technical field of meteorological recognition. The method comprises the following steps: acquiring daily average temperature and rainfall data of a target area; constructing a plurality of models, wherein the plurality of models comprise a first model and a second model; respectively inputting daily average temperature and rainfall data of a target area into the first model and the second model to obtain a first identification result and a second identification result; wherein the first recognition result and the second recognition result both comprise the rain and snow critical temperature of the target area and different-phase-state precipitation time characteristics of the target area; and based on the real meteorological data, respectively calculating errors between the first identification result and the real meteorological data and between the second identification result and the real meteorological data, and selecting a model for identifying the rain and snow critical temperature of the target area and / or identifying the precipitation time characteristics of different phase states of the target area according to the calculated errors. The method provided by the invention solves the problem of large precipitation phase state identification error in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of meteorological identification technology, and more specifically, to a parameterization method, device and medium for rain and snow temperature thresholds. Background Art

[0002] Precipitation changes are the result of the combined effects of atmospheric dynamics and thermodynamics on the water cycle, energy cycle, ecological environment, and human activities within the climate system, with significant repercussions on the Earth's climate system. Precipitation in different regions exhibits distinct phases due to differences in underlying surface conditions, surface air temperature, and regional atmospheric circulation. Identifying precipitation phases has been a long-standing challenge for meteorological researchers.

[0003] The rain-to-snow ratio is closely related to regional temperature and reflects the ratio of rainfall to snowfall. First, under the backdrop of global warming, changes in snowfall amount will directly affect the rain-to-snow ratio. Rising temperatures will also affect the pattern of precipitation events, with some snowmelt turning into rain, further affecting the ratio. Furthermore, changes in the snowfall ratio will affect Earth's surface reflectivity. When precipitation is solid snow, Earth's surface reflectivity increases rapidly, reducing solar radiation absorbed by the surface and producing a cooling effect. When precipitation is liquid rain, Earth's surface reflectivity decreases, leading to an increase in solar radiation absorbed by the surface. In regions with glaciers, changes in the snowfall ratio can influence glacier melt to a certain extent. An increase in the snowfall ratio will increase the glacier's surface reflectivity, reducing the amount of radiant heat absorbed by the glacier, ultimately leading to a stagnant or slowing trend in glacier melt. Conversely, a decrease in snowfall will reduce glacier surface reflectivity, increasing radiant heat absorbed by the glacier and accelerating glacier melt. Therefore, changes in the snowfall ratio can serve as an indicator of climate change.

[0004] Using precipitation phases to reflect changes in snowfall proportions is a common approach. The single temperature threshold method can identify precipitation phases by finding a critical temperature value to distinguish between rain and snow, but this approach also presents a new problem. This method ignores the fact that precipitation phases can coexist in multiple forms within a certain temperature range, resulting in a certain degree of error. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a parameterization method, device and medium for rain and snow temperature thresholds to solve the problem of large errors in existing precipitation phase identification.

[0006] In a first aspect, the present invention provides a parameterization method for rain and snow temperature thresholds, the method comprising:

[0007] Obtain the average daily temperature and precipitation data for the target area;

[0008] constructing a plurality of models, the plurality of models including a first model and a second model;

[0009] Inputting the daily average temperature and precipitation data of the target area into the first model and the second model respectively to obtain a first recognition result and a second recognition result; wherein the first recognition result and the second recognition result both include the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area;

[0010] Based on the real meteorological data, the errors between the first recognition result and the second recognition result and the real meteorological data are calculated respectively, and a model for identifying the critical temperature of rain and snow in the target area and / or identifying the time characteristics of precipitation in different phases in the target area is selected according to the calculated errors.

[0011] Furthermore, the first model includes:

[0012] The first probability calculation unit is used to determine the proportion of snowfall to total precipitation according to the following formula (1):

[0013] sp=1 / (1+exp(-1.48+1.25x)) (1)

[0014] Where sp is the proportion of snowfall to total precipitation; exp is an exponential function with a natural constant as the base;

[0015] a first critical temperature identification unit, configured to identify a critical temperature of rain and snow in a target area according to a ratio of the snowfall to the total precipitation;

[0016] The first phase identification unit is used to identify the precipitation time characteristics of different phases in the target area according to the critical temperature of rain and snow in the target area.

[0017] Furthermore, identifying the critical temperature of rain and snow in the target area according to the ratio of the snowfall to the total precipitation includes:

[0018] The maximum daily average temperature corresponding to the proportion of snowfall to total precipitation is 100% is taken as the maximum critical temperature of solid snowfall T 降雪 ;

[0019] The lowest daily average temperature corresponding to the snowfall ratio of 0% in the total precipitation is taken as the lowest critical temperature of liquid snowfall T. 降雨 ;

[0020] According to the maximum temperature T of the solid snowfall 降雪 and the lowest temperature of liquid snowfall T 降雨Determine the critical temperature of rain and snow in the target area; wherein, the critical temperature of rain and snow in the target area includes the critical low temperature T 最小 and critical high temperature T 最大 .

[0021] Furthermore, according to the maximum temperature T of the solid snowfall 降雪 and the lowest temperature of liquid snowfall T 降雨 , the critical temperature of rain and snow in the target area is determined by the following formula:

[0022] T 最大 =T 降雨 -0.1℃

[0023] T 最小 =T 降雪 +0.1℃

[0024] Where, T 最大 and T 最小 represent the critical high temperature and critical low temperature respectively.

[0025] Furthermore, identifying the temporal characteristics of precipitation in different phases in the target area according to the critical temperature of rain and snow in the target area includes:

[0026] The average daily temperature is greater than or equal to T 最大 The time period is identified as the rainfall phase;

[0027] The average daily temperature is less than or equal to T 最小 The time period is identified as the snowfall phase.

[0028] Furthermore, the second model includes:

[0029] a second probability calculation unit, configured to determine an intersecting and overlapping portion of a snowfall temperature and a rainfall temperature based on the input daily average temperature and precipitation data, divide the intersecting and overlapping portion into a plurality of temperature sub-ranges at preset temperature intervals, and calculate the snowfall probability and rainfall probability in each temperature sub-range;

[0030] a second critical temperature identification unit, the second critical temperature identification unit being configured to construct a snowfall variation curve and a rainfall variation curve in a coordinate system using temperature values in a small temperature range and their corresponding snowfall probabilities and rainfall probabilities as abscissas and ordinates, and using the intersection of the snowfall variation curve and the rainfall variation curve as the critical temperature for rain and snow in the target area;

[0031] The second phase identification unit is used to identify the precipitation time characteristics of different phases in the target area according to the critical temperature of rain and snow in the target area.

[0032] Furthermore, in each temperature range, the snowfall probability and rainfall probability are calculated using the following formula (2):

[0033]

[0034] Where p1 and p2 are the probability of rainfall and snowfall, respectively; exp is an exponential function with a natural constant as the base.

[0035] Furthermore, after selecting a model for identifying the critical temperature of rain and snow in the target area and / or identifying the temporal characteristics of precipitation in different phases in the target area based on the calculated error, the method further includes:

[0036] When different models are selected to identify the critical temperature of rain and snow in the target area and to identify the time characteristics of precipitation in different phases in the target area, the unit modules in the model for identifying the critical temperature of rain and snow in the target area and the unit modules for identifying the time characteristics of precipitation in different phases in the target area are combined to construct a parameterized model, which is used to identify the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area.

[0037] In a second aspect, the present invention provides a parameterization device for rain and snow temperature thresholds, the device comprising:

[0038] a data acquisition unit configured to acquire daily average temperature and precipitation data of a target area;

[0039] a model building unit configured to build a plurality of models, wherein the plurality of models include a first model and a second model;

[0040] a parameter identification unit configured to input the daily average temperature and precipitation data of the target area into the first model and the second model, respectively, to obtain a first identification result and a second identification result; wherein the first identification result and the second identification result both include the critical temperature for rain and snow in the target area and the time characteristics of precipitation in different phases in the target area;

[0041] The model screening unit is configured to calculate the errors between the first recognition result and the second recognition result and the real meteorological data respectively, and select a model for identifying the critical temperature of rain and snow in the target area and / or identifying the time characteristics of precipitation in different phases in the target area based on the calculated errors.

[0042] In a third aspect, the present invention provides a readable storage medium, wherein the readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the method as described above.

[0043] The present invention has at least the following beneficial effects:

[0044] The present invention constructs a first model and a second model, and identifies the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area based on the first model and the second model respectively, and determines the model used to identify the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area according to the error between the recognition result and the real data. In addition, when different models are selected to identify the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area, the two models are combined or the first model is used to correct the second model, so as to form a new recognition model. The recognition model can accurately identify the critical temperature of rain and snow and the time characteristics of precipitation in different phases, and ensure the accuracy of the parameterization of the rain and snow temperature threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a parameterization method for rain and snow temperature thresholds according to an embodiment of the present invention is shown.

[0046] Figure 2 The structure of the first model and its data processing schematic diagram according to an embodiment of the present invention are shown.

[0047] Figure 3 The structure of the second model and its data processing schematic diagram according to an embodiment of the present invention are shown.

[0048] Figure 4 A target area overview diagram according to an embodiment of the present invention is shown.

[0049] Figure 5 The rain and snow threshold temperatures for the target area from 2019 to 2021 according to an embodiment of the present invention are shown (a. exponential equation, b. rain and snow temperature distribution, c. changes in rain and snow frequency, d. rain and snow probability distribution).

[0050] Figure 6 A graph showing changes in precipitation and air temperature in different phases in the target area from August 2019 to August 2020 according to an embodiment of the present invention is shown.

[0051] Figure 7 A graph showing changes in precipitation and temperature in different phases in the target area from October 2020 to September 2021 according to an embodiment of the present invention is shown.

[0052] Figure 8 A seasonal variation diagram of precipitation and temperature in different phases in a target area according to an embodiment of the present invention is shown.

[0053] Figure 9 A diagram showing the interannual variation of precipitation and temperature in different phases in a target area according to an embodiment of the present invention is shown.

[0054] Figure 10A structural diagram of a parameterization device for rain and snow temperature thresholds according to an embodiment of the present invention is shown.

[0055] Figure 11 A diagram for judging different phases of a target area according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0056] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments. The embodiments of the present invention are further described in detail below with reference to the accompanying drawings and specific embodiments, but are not intended to limit the present invention. For the various steps described herein, if there is no necessity for a contextual relationship between each other, the order in which they are described as examples herein should not be regarded as limiting, and those skilled in the art should know that they can be adjusted in order as long as the logic between them is not destroyed, resulting in the inability to implement the entire process.

[0057] The embodiment of the present invention provides a parameterization method for rain and snow temperature thresholds, such as Figure 1 FIG. 1 is a flow chart of a parameterization method for rain and snow temperature thresholds. The parameterization method for rain and snow temperature thresholds includes steps S10 to S40, which are described in detail below.

[0058] S10: Obtain the daily average temperature and precipitation data of the target area.

[0059] In this embodiment, the daily average temperature and precipitation data of the target area are obtained through historical measured data of the automated weather station in the target area, wherein the daily average temperature is aligned with the precipitation data in the time dimension.

[0060] S20: Construct multiple models, where the multiple models include a first model and a second model.

[0061] S30: Inputting the daily average temperature and precipitation data of the target area into the first model and the second model respectively to obtain a first recognition result and a second recognition result; wherein the first recognition result and the second recognition result both include the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area.

[0062] In some embodiments, as Figure 2 FIG2 is a schematic diagram of the structure of the first model and its data processing. The first model 200 includes a first probability calculation unit 201 , a first critical temperature identification unit 202 and a first phase identification unit 203 .

[0063] The first probability calculation unit 201 responds to the input daily average temperature and precipitation data and determines the proportion of snowfall to total precipitation according to formula (1):

[0064] sp=1 / (1+exp(-1.48+1.25x)) (1)

[0065] Where sp is the ratio of snowfall to total precipitation; exp is an exponential function with a natural constant as the base.

[0066] It should be noted that the ratio of snowfall to total precipitation is the probability of snowfall when a precipitation event occurs. Precipitation events are determined based on precipitation data. Generally speaking, if there is precipitation of a set amount of precipitation within a certain time period, it is considered that a precipitation event has occurred. The average daily temperature in formula (1) is the average daily temperature when the precipitation event occurs.

[0067] In this embodiment, the empirical parameters a and b can be obtained by fitting based on historical measured data. For example, based on the known historical measured data, the historical measured data are processed using a visual interpretation method to determine the critical temperature for rain and snow in the target area, that is, to determine the critical low temperature and critical high temperature, and at the same time determine the proportion of snowfall to total precipitation. That is, when the SP and the corresponding average daily temperature are known, the empirical parameters a and b in formula (1) are fitted and solved using a large amount of data, and the values of a and b are finally determined. Then, for the newly input average daily temperature, the SP can be calculated.

[0068] The first critical temperature identification unit 202 responds to the input of the first probability calculation unit 201 and identifies the critical temperature of rain and snow in the target area according to the ratio of snowfall to total precipitation.

[0069] Specifically, the critical temperature of rain and snow in the target area is identified based on the proportion of snowfall to total precipitation, including: the maximum daily average temperature corresponding to the proportion of snowfall to total precipitation is 100% as the maximum critical temperature of solid snowfall T 降雪 The lowest daily average temperature corresponding to the snowfall ratio of 0% to the total precipitation is taken as the lowest critical temperature of liquid snowfall T 降雨 ; According to the maximum temperature of solid snowfall T 降雪 and the lowest temperature of liquid snowfall T 降雨 Determine the critical temperature of rain and snow in the target area; the critical temperature of rain and snow in the target area includes the critical low temperature T 最小 and critical high temperature T 最大 .

[0070] In some embodiments, the maximum temperature T of solid snowfall is determined based on the 降雪 and the lowest temperature of liquid snowfall T 降雨 , the critical temperature of rain and snow in the target area is determined by the following formula:

[0071] T 最大 =T 降雨 -0.1℃

[0072] T 最小 =T 降雪 +0.1℃

[0073] Where, T 最大 and T 最小 represent the critical high temperature and critical low temperature respectively.

[0074] It should be noted that the purpose of 0.1°C in the calculation formula of the critical temperature for rain and snow in the target area above is to reduce the boundary value error.

[0075] The first phase identification unit 203 responds to the input of the first critical temperature identification unit 202 and identifies the time characteristics of precipitation in different phases in the target area according to the critical temperature of rain and snow in the target area.

[0076] In some embodiments, identifying the time characteristics of precipitation in different phases in the target area according to the critical temperature of rain and snow in the target area includes: 最大 But not less than T 最小 The time period with the average daily temperature less than T 最小 The time period is identified as the snowfall phase.

[0077] In some embodiments, as Figure 3 The second model 300 includes a second probability calculation unit 301 , a second critical temperature identification unit 302 and a second phase identification unit 303 .

[0078] The second probability calculation unit 301 determines the intersecting and overlapping parts of the snowfall temperature and the rainfall temperature based on the input average daily temperature and precipitation data, and divides the intersecting and overlapping parts into multiple temperature small ranges at preset temperature intervals, and calculates the snowfall probability and rainfall probability in each temperature small range respectively.

[0079] In some embodiments, in each temperature range, the snowfall probability and the rainfall probability are calculated respectively by the following formula (2):

[0080]

[0081] Where p1 and p2 are the probability of rainfall and snowfall, respectively; exp is an exponential function with a natural constant as the base.

[0082] The second critical temperature identification unit 302 responds to the input of the second probability calculation unit 301, takes the temperature values in a small temperature range and their corresponding snowfall probability and rainfall probability as the horizontal and vertical coordinates, constructs a snowfall change curve and a rainfall change curve in a coordinate system, and takes the intersection of the snowfall change curve and the rainfall change curve as the critical temperature of rain and snow in the target area.

[0083] The second phase identification unit 303 responds to the input of the second critical temperature identification unit 302 and identifies the time characteristics of precipitation in different phases in the target area according to the critical temperature of rain and snow in the target area.

[0084] In this embodiment, the second model 300 identifies only one critical temperature for rain and snow in the target area, which corresponds to the critical low temperature identified by the first model 200. That is, when the average daily temperature is lower than the critical temperature for rain and snow in the target area identified by the second model 300, the time period corresponding to the average daily temperature is identified as a snowfall phase, otherwise it is identified as a rainfall phase.

[0085] S40: Based on the real meteorological data, respectively calculate the errors between the first recognition result and the second recognition result and the real meteorological data, and select a model for identifying the critical temperature of rain and snow in the target area and / or identifying the time characteristics of precipitation in different phases in the target area according to the calculated errors.

[0086] In this embodiment, the real meteorological data refers to the critical temperature of rain and snow and the time characteristics of precipitation in different phases in the target area determined by visual discrimination or field investigation methods. These methods are manual. Generally speaking, the accuracy of real meteorological data is higher than that of machine models. The performance of the first model and the second model is evaluated, and the best model suitable for the current target area is selected to respectively identify the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area.

[0087] In some embodiments, after selecting a model for identifying the critical temperature for rain and snow in the target area and / or identifying the temporal characteristics of precipitation in different phases in the target area based on the calculated error, the method further includes:

[0088] When different models are selected to identify the critical temperature of rain and snow in the target area and to identify the temporal characteristics of precipitation in different phases in the target area, the unit modules in the model used to identify the critical temperature of rain and snow in the target area and the unit modules used to identify the temporal characteristics of precipitation in different phases in the target area are combined to construct a parameterized model. The parameterized model is used to identify the critical temperature of rain and snow in the target area and the temporal characteristics of precipitation in different phases in the target area.

[0089] For example, when using a first model to identify the temporal characteristics of precipitation in different phases in a target area and a second model to identify the critical temperature for rain and snow in the target area, the first and second models are combined, with the second phase identification unit removed from the second model to obtain a parameterized model. In this parameterized model, the acquired data is input into the first and second models respectively. The first model outputs the temporal characteristics of precipitation in different phases in the target area, i.e., identifies snowfall, and the second model outputs the critical temperature for rain and snow in the target area. The output critical temperature for rain and snow in the target area is used to simulate the rain-snow ratio. It should be noted that the method for simulating the rain-snow ratio using the critical temperature for rain and snow in the target area here uses existing methods and is not described in detail in this embodiment.

[0090] Of course, if the errors of the critical temperature of rain and snow in the target area and the temporal characteristics of precipitation in different phases in the target area identified by the first model or the second model are the smallest, the first model or the second model is used as the parameterized model.

[0091] The following embodiments of the present invention will further illustrate the feasibility and progress of the present invention in conjunction with specific implementation cases.

[0092] like Figure 4 The target area selected in this implementation case is within the range of 26°47′N-27°28′N, 99°51′N-100°27′E, with an area of 4.6×10 3 km 2 The terrain is highly undulating, with significant elevation differences. Its primary atmospheric systems are the high-altitude westerly circulation and the southwest monsoon, resulting in a low-latitude plateau monsoon climate. Since the 1960s, temperatures in the target area have risen significantly, while precipitation has remained stable or slightly decreased.

[0093] The data used in this implementation case study are field data from automated weather stations within the target area, including temperature and precipitation data and daily ground-level photographs taken at 12:00 PM and 6:00 PM from the monitoring platform. The timeframe covers August 1, 2019, and September 31, 2021. Due to poor environmental conditions, some monitoring data was missing and discontinuous. Therefore, the target area's meteorological data was sorted and screened chronologically. A total of 655 days of baseline data were available. Due to missing temperature data, 516 valid data sets were obtained, resulting in 185 days of precipitation. The estimated probability of precipitation within the target area was 35.85%. Visual inspection of ground-level photographs taken at 12:00 PM and 6:00 PM at the corresponding stations was used to determine if snowfall had occurred. Due to the high altitude and high humidity in the target area, heavy fog often preceded snowfall events, with a probability of 60%. Because fog makes it difficult to determine ground conditions, hourly precipitation data was used to identify specific precipitation periods, allowing for accurate determination of snowfall occurrences. The data results show that there were 48 days of snowfall during the study period, and the estimated snowfall probability was 9.30%. See Table 1 for details.

[0094] Table 1 Characteristics of precipitation and temperature in different phases

[0095]

[0096] For the first model, based on 516 visually discriminated daily precipitation phase meteorological data, two critical temperatures T are found in the rain and snow phases. 最小 and T 最大 As can be seen from Table 1, T 降雨 -3.67℃, T 降雪 is 3.65℃, then the two critical temperatures are between -3.77℃ and 3.75℃. Use formula (1) to fit the relationship between SP and daily average temperature in the region (such as Figure 5 a) in the middle, Figure 5 The 13 points in the figure represent the actual SP under the corresponding daily average temperature. 最小 (-0.265℃) or below corresponds to SP of 100%, T 最大 (4.994℃) and above correspond to a snowfall ratio of 0%; Figure 5 The curve a represents the SP index equation for the target area, with a = -148 and b = 1.24. Within the two critical temperature ranges, the horizontal axis is divided into intervals of 0.5°C, corresponding to the SP of 13 points in the figure, excluding the endpoints. A correlation test was conducted between the simulated SP corresponding to the fitted equation and the actual snowfall conditions, and the correlation coefficient was 0.96, indicating a strong correlation. Figure 5 As shown in Figure a, the SP changes rapidly between -0.5℃ and 3.05℃, and the precipitation phase in the target area is extremely unstable. The SP is higher in the temperature range below -0.5℃ and lower in the temperature range above 3.05℃.

[0097] For the second model, the station temperature and precipitation data of the target area from 2019 to 2021 were used. After preliminary visual identification, it was found that since the measuring instruments were greatly affected by the weather, it was easy for snowfall to occur but the precipitation data was 0. Only the snowfall and rainfall with precipitation data and the daily average temperature were counted to obtain the temperature distribution of rainfall and snowfall at the target area stations ( Figure 5 b). Affected by the accuracy of the measuring instrument data, this embodiment only considers the data based on visual identification, marking the data visually identified as solid snowfall as 1 and the data identified as liquid rainfall as 0, and sorting them in positive order according to the daily average temperature to obtain the temperature interval of the precipitation period. The number of rainfall and snowfall in each temperature interval of the target area station is counted with a step size of 0.5℃ ( Figure 5 From the statistical results, we can see that the ranges of rain and snow are not independent, and there is an overlapping part between -4℃ and 3.5℃ ( Figure 5In (c), the temperature overlap is divided into 16 temperature sub-intervals with a step size of 0.5°C. The probability of snowfall and rainfall is calculated in each sub-interval. Formula (2) is used to fit the probability of rain and snow. The red and blue lines are respectively the rain and snow probabilities fitted by the S-type exponential equation. The critical temperature of rain and snow is 0.15°C ( Figure 5 (d)

[0098] A comparison of the first and second models revealed their respective strengths and weaknesses. The first model achieved a root mean square error of 0.11 with the measured data. While the model was unable to determine the exact amount of snowfall, it provided a good simulation of both rainfall and snowfall. The second model achieved a more accurate rain-to-snow ratio simulation at the critical temperature, achieving an accuracy of 0.94.

[0099] Based on visual identification, we identified 34 snowfall events where the measuring instrument failed to record precipitation data (data indicated in brackets in Table 2). Due to the limited snowfall data, interpreting the snowfall period is difficult. This example analyzes only the number of days of snowfall events. During the study period, there were 17 snowfall days in spring, 16 in winter, 4 in summer, and 11 in autumn. This suggests that snowfall is concentrated in winter and spring.

[0100] Table 2 shows that precipitation in the target region showed no significant change between the two years (p > 0.05), while snowfall showed a significant downward trend (p < 0.05). Precipitation in period a decreased from 636.56 mm to 330.08 mm compared to period b, a decrease of 48.15%. Average precipitation increased from 6.366 mm to 6.472 mm, and the number of days with precipitation was halved from 100 to 51. While precipitation and the number of days with precipitation decreased significantly, the average annual precipitation increased slightly, indicating a more concentrated precipitation trend in the target region. Generally speaking, the SPR showed an overall downward trend, decreasing significantly at an average rate of 1.11% / (10a). Throughout the entire period, precipitation increased, snowfall decreased, and the SPR decreased. However, the rate and significance of change were higher in warm periods than in cold periods: warm-season precipitation accounted for approximately 90% of the annual total on the plateau. The proportion of warm-season snowfall to annual snowfall in most areas of the plateau showed a downward trend (-0.29% / (10a)). The results obtained in this embodiment are consistent with the above cognition.

[0101] from Figure 8 From the data, the temperature in the target area showed an insignificant upward trend (p>0.05); there was more precipitation in spring and summer, with the most concentrated precipitation in summer and the increase in extreme rainstorms. On August 31, 2021, the single-day precipitation reached 76.56mm, and the precipitation roughly changed with the temperature. Figure 6 、 Figure 7 、 Figure 9As shown in Table 2, the peak periods of precipitation and snowfall occurred in 2020. During period a, summer rainfall was abundant, accounting for 71.94% of the total rainfall, while spring snowfall accounted for a significant portion, with more snowy days in winter and spring. During period b, precipitation decreased overall, maintaining a trend of concentrated summer rainfall, while the number of snowy days decreased in winter and spring. This indicates that precipitation and snowfall are concentrated in the target area, with significant interannual snowfall variability. The concentration of precipitation, especially rainfall, in the summer is strongly related to the monsoon climate and geographical location of the low-latitude plateau. In summer, a thermal low pressure system forms on the Qinghai-Tibet Plateau, drawing air from all directions toward the plateau. The target area is located on the southwestern edge of the plateau, making orographic precipitation highly likely to form as air ascends and reaches the plateau. In winter, a cold plateau forms on the plateau, and an anticyclonic system forms, diverging air from the plateau. Airflow descends, resulting in lower temperatures and unfavorable precipitation formation. Daily visual interpretation of photographs reveals that snow melts more rapidly in the summer, which is related to the average summer temperature of 5.074°C.

[0102] Table 2 Seasonal characteristics of precipitation in the target area

[0103]

[0104] Based on the method proposed in the present invention, the snowfall amount can also be determined by combining snowfall events and photos.

[0105] Specifically, precipitation on the glacier surface is very important for glacier melting and accumulation, especially snowfall, which directly inputs and increases glacier mass. Snowfall is extracted from the specific precipitation total and is interpreted manually by visual inspection. Determining daily precipitation as daily snowfall requires two steps. First, find the statistical day with precipitation (i.e., daily precipitation greater than zero) from the start and end days. Second, determine the type of daily precipitation using the corresponding daily photo or multiple daily photos and other data. There are three main criteria for determining precipitation and snowfall: (I) Compare photos of a certain day with photos of the previous or next day to determine whether the daily precipitation type is rain or snow. (II) Combine snow depth data with photos of precipitation events on consecutive days to determine the daily precipitation type. (III) Combine hourly precipitation, snow depth, and photos of recent days to determine the daily precipitation type. This is mainly due to the fact that there may be continuous foggy precipitation weather, resulting in no favorable photos for one to several days.

[0106] Identifying precipitation phase from photos is an important step in this method. The identified precipitation phase data will be used to build a model for determining precipitation phase. This example uses camera photos from August 1, 2019 to July 11, 2021 to identify snowfall events through manual visual inspection. Some field photos can be found in Figure 11 , the following Figure 11 a and b are a group, Figure 11 c and d are a group, Figure 11In the figure, e and f are taken as a group for example to further illustrate the method proposed in the present invention.

[0107] One judgment situation is that there is only one day with precipitation records among several days. In this case, the precipitation stage pattern can be determined based on the daily precipitation records and photos. Take the snowfall event on September 21, 2019 as an example. According to the meteorological data records, the precipitation on September 20, 2019 was 0 mm, and the precipitation on the 21st was 11.15 mm, so it can be determined that precipitation occurred on the 21st. Then use photos to determine whether a snowfall event occurred. In the photos on the 20th ( Figure 11 In the photo (a), no snowfall was observed on the glacier surface or the mountain behind it, but in the photo taken on the 21st ( Figure 11 In b), snowfall was observed on the glacier surface. Based on the above, snowfall was determined to have occurred on the 21st, and the daily precipitation was defined as snowfall.

[0108] Another judgment situation is that there are precipitation records for several consecutive days, and the corresponding photos are complete for judgment. In this case, the daily precipitation and snow depth records can also be combined with photos to judge the precipitation stage. Take the snowfall event on October 5, 2019 as an example. The meteorological data records that the precipitation on October 4 was 8.34 mm and the precipitation on October 5 was 7.75 mm. Therefore, it can be determined that precipitation events occurred on October 4 and October 5. Then use photos to determine whether a snowfall event occurred. In the photo on October 4 ( Figure 11 In the image (c), no snowfall was observed on the glacier surface or on the mountain behind it, but in the image (c) on October 5, Figure 11 In (b), snowfall was observed on the glacier surface. Combined with the snow depth data, the snowfall on October 4th decreased by 1.09 cm, while the snowfall on October 5th increased by 5.62 cm. Based on this, it is determined that rain occurred on October 4th, and the corresponding daily precipitation is defined as rain. Snow occurred on October 5th, and the corresponding daily precipitation is defined as snow.

[0109] In addition to the above two cases, some snowfall events also need to be judged in combination with photos of the previous and next few days, daily precipitation, hourly precipitation and snow depth data. Take the snowfall event on December 28, 2019 as an example ( Figure 11 On December 26, there was little snow in the back mountains, with daily precipitation of 0 mm and no change in snow depth data ( Figure 11 On the 27th, the ground was obscured by heavy fog, and the snowfall could not be determined from the photos. However, the daily and hourly precipitation were both 0 mm, and the snow depth data did not change. On the 28th, the surface of the back mountain was clearly covered with snow, with daily precipitation of 9.4 mm and snow depth data increasing by 8.22 cm ( Figure 11 Based on the above, we can conclude that there was no snowfall on the 27th, but there was snowfall on the 28th, and the corresponding precipitation amount was also determined to be snowfall.

[0110] exist Figure 11 In the figure, the judgment of (o)-(p) and (i)-(j) is determined by standard I; the judgment of (k)-(l) and (m)-(n) is determined by standard II; the judgment of (g)-(h) is determined by standard III. There is no precipitation on January 29 and February 1, and the precipitation on January 30 and January 31 is judged as snowfall.

[0111] The embodiment of the present invention also provides a parameterization device for rain and snow temperature thresholds, such as Figure 10 As shown, the device includes:

[0112] The data acquisition unit 1001 is configured to acquire the daily average temperature and precipitation data of the target area;

[0113] A model building unit 1002 is configured to build a plurality of models, wherein the plurality of models include a first model and a second model;

[0114] The parameter identification unit 1003 is configured to input the daily average temperature and precipitation data of the target area into the first model and the second model, respectively, to obtain a first identification result and a second identification result; wherein the first identification result and the second identification result both include the critical temperature for rain and snow in the target area and the time characteristics of precipitation in different phases in the target area;

[0115] The model screening unit 1004 is configured to calculate the errors between the first recognition result and the second recognition result and the real meteorological data respectively, and select a model for identifying the critical temperature of rain and snow in the target area and / or identifying the time characteristics of precipitation in different phases in the target area based on the calculated errors.

[0116] In some embodiments, the first model comprises:

[0117] The first probability calculation unit is used to determine the proportion of snowfall to total precipitation according to the following formula (1):

[0118] sp=1 / (1+exp(-1.48+1.25x)) (1)

[0119] Where sp is the proportion of snowfall to total precipitation; exp is an exponential function with a natural constant as the base;

[0120] a first critical temperature identification unit, configured to identify a critical temperature of rain and snow in a target area according to a ratio of the snowfall to the total precipitation;

[0121] The first phase identification unit is used to identify the precipitation time characteristics of different phases in the target area according to the critical temperature of rain and snow in the target area.

[0122] In some embodiments, identifying the critical rain and snow temperature in the target area based on the ratio of the snowfall to the total precipitation includes:

[0123] The maximum daily average temperature corresponding to the proportion of snowfall to total precipitation is 100% is taken as the maximum critical temperature of solid snowfall T 降雪 ;

[0124] The lowest daily average temperature corresponding to the snowfall ratio of 0% in the total precipitation is taken as the lowest critical temperature of liquid snowfall T. 降雨 ;

[0125] According to the maximum temperature T of the solid snowfall 降雪 and the lowest temperature of liquid snowfall T 降雨 Determine the critical temperature of rain and snow in the target area; wherein, the critical temperature of rain and snow in the target area includes the critical low temperature T 最小 and critical high temperature T 最大 .

[0126] In some embodiments, the maximum temperature T of the solid snowfall 降雪 and the lowest temperature of liquid snowfall T 降雨 , the critical temperature of rain and snow in the target area is determined by the following formula:

[0127] T 最大 =T 降雨 -0.1℃

[0128] T 最小 =T 降雪 +0.1℃

[0129] Where, T 最大 and T 最小 represent the critical high temperature and critical low temperature respectively.

[0130] In some embodiments, identifying the temporal characteristics of precipitation in different phases in the target area based on the critical temperature of rain and snow in the target area includes:

[0131] The average daily temperature is less than or equal to T 最大 But not less than T 最小 The time period is identified as the rainfall phase;

[0132] The average daily temperature is less than or equal to T 最小 The time period is identified as the snowfall phase.

[0133] In some embodiments, the second model comprises:

[0134] a second probability calculation unit, configured to determine an intersecting and overlapping portion of a snowfall temperature and a rainfall temperature based on the input daily average temperature and precipitation data, divide the intersecting and overlapping portion into a plurality of temperature sub-ranges at preset temperature intervals, and calculate the snowfall probability and rainfall probability in each temperature sub-range;

[0135] a second critical temperature identification unit, the second critical temperature identification unit being configured to construct a snowfall variation curve and a rainfall variation curve in a coordinate system using temperature values in a small temperature range and their corresponding snowfall probabilities and rainfall probabilities as abscissas and ordinates, and using the intersection of the snowfall variation curve and the rainfall variation curve as the critical temperature for rain and snow in the target area;

[0136] The second phase identification unit is used to identify the precipitation time characteristics of different phases in the target area according to the critical temperature of rain and snow in the target area.

[0137] In some embodiments, in each temperature range, the snowfall probability and the rainfall probability are calculated respectively by the following formula (2):

[0138]

[0139] Where p1 and p2 are the probability of rainfall and snowfall, respectively; exp is an exponential function with a natural constant as the base.

[0140] In some embodiments, the model screening unit is further configured to: when selecting different models to identify the critical temperature of rain and snow in the target area and to identify the time characteristics of precipitation in different phases in the target area, combine the unit modules in the model for identifying the critical temperature of rain and snow in the target area and the unit modules for identifying the time characteristics of precipitation in different phases in the target area to construct a parameterized model, wherein the parameterized model is used to identify the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area.

[0141] It should be noted that the structure of the parameterization device for each rain and snow temperature threshold described in this embodiment belongs to the same technical concept as the parameterization method for rain and snow temperature threshold described previously, and achieves the same beneficial effects through the same principle, which will not be repeated here.

[0142] An embodiment of the present invention further provides a readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the method described in any of the above embodiments.

[0143] Furthermore, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present invention having equivalent elements, modifications, omissions, combinations (e.g., schemes where various embodiments intersect), adaptations, or changes. The elements in the claims are to be interpreted broadly based on the language employed in the claims and are not limited to the examples described in this specification or during the prosecution of this application, which examples are to be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered as examples only, with the true scope and spirit being indicated by the following claims and the full scope of their equivalents.

[0144] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of their solutions) can be used in combination with each other. For example, those of ordinary skill in the art may use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features can be grouped together to simplify the present invention. This should not be interpreted as an intention that a feature of an invention that is not claimed for protection is necessary for any claim. On the contrary, the subject matter of the present invention may be less than all the features of the embodiments of a particular invention. Thus, the following claims are incorporated into the specific embodiments as examples or embodiments, wherein each claim is independently a separate embodiment, and it is considered that these embodiments can be combined with each other in various combinations or arrangements. The scope of the present invention should be determined with reference to the appended claims and the full scope of equivalents to which these claims are entitled.

Claims

1. A parameterization method for rain and snow temperature thresholds, characterized in that: The method comprises: Obtain the average daily temperature and precipitation data for the target area; constructing a plurality of models, the plurality of models including a first model and a second model; Inputting the daily average temperature and precipitation data of the target area into the first model and the second model respectively to obtain a first recognition result and a second recognition result; wherein the first recognition result and the second recognition result both include the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area; Based on the real meteorological data, the errors between the first recognition result and the second recognition result and the real meteorological data are calculated respectively, and a model for identifying the critical temperature of rain and snow in the target area and / or identifying the time characteristics of precipitation in different phases in the target area is selected according to the calculated errors.

2. The parameterization method of rain and snow temperature threshold according to claim 1 is characterized in that: The first model includes: The first probability calculation unit is used to determine the proportion of snowfall to total precipitation according to the following formula (1): sp=1 / (1+exp(-1.48+1.25x)) (1) Where sp is the proportion of snowfall to total precipitation; exp is an exponential function with a natural constant as the base; a first critical temperature identification unit, configured to identify a critical temperature of rain and snow in a target area according to a ratio of the snowfall to the total precipitation; The first phase identification unit is used to identify the precipitation time characteristics of different phases in the target area according to the critical temperature of rain and snow in the target area.

3. The parameterization method of rain and snow temperature threshold according to claim 2 is characterized in that: Identify the critical temperature for rain and snow in the target area based on the ratio of the snowfall to the total precipitation, including: The maximum daily average temperature corresponding to the proportion of snowfall to total precipitation is 100% is taken as the maximum critical temperature of solid snowfall T 降雪 ; The lowest daily average temperature corresponding to the snowfall ratio of 0% in the total precipitation is taken as the lowest critical temperature of liquid snowfall T. 降雨 ; According to the maximum temperature T of the solid snowfall 降雪 and the lowest temperature of liquid snowfall T 降雨 Determine the critical temperature of rain and snow in the target area; wherein, the critical temperature of rain and snow in the target area includes the critical low temperature T 最小 and critical high temperature T 最大 .

4. The parameterization method of rain and snow temperature threshold according to claim 3 is characterized in that: According to the maximum temperature T of the solid snowfall 降雪 and the lowest temperature of liquid snowfall T 降雨 , the critical temperature of rain and snow in the target area is determined by the following formula: T 最大 =T 降雨 -0.1℃ T 最小 =T 降雪 +0.1℃ Where, T 最大 and T 最小 represent the critical high temperature and critical low temperature respectively.

5. The parameterization method of rain and snow temperature threshold according to claim 3 is characterized in that: Identifying the temporal characteristics of precipitation in different phases in the target area based on the critical temperature of rain and snow in the target area includes: The average daily temperature is greater than or equal to T 最大 The time period is identified as the rainfall phase; The average daily temperature is less than or equal to T 最小 The time period is identified as the snowfall phase.

6. The parameterization method of rain and snow temperature threshold according to claim 1, characterized in that: The second model includes: a second probability calculation unit, configured to determine an intersecting and overlapping portion of a snowfall temperature and a rainfall temperature based on the input daily average temperature and precipitation data, divide the intersecting and overlapping portion into a plurality of temperature sub-ranges at preset temperature intervals, and calculate the snowfall probability and rainfall probability in each temperature sub-range; a second critical temperature identification unit, the second critical temperature identification unit being configured to construct a snowfall variation curve and a rainfall variation curve in a coordinate system using temperature values in a small temperature range and their corresponding snowfall probabilities and rainfall probabilities as abscissas and ordinates, and using the intersection of the snowfall variation curve and the rainfall variation curve as the critical temperature for rain and snow in the target area; The second phase identification unit is used to identify the precipitation time characteristics of different phases in the target area according to the critical temperature of rain and snow in the target area.

7. The parameterization method of rain and snow temperature threshold according to claim 6, characterized in that: In each temperature range, the snowfall probability and rainfall probability are fitted by the following formula (2): Where p1 and p2 are the probability of rainfall and snowfall, respectively; exp is an exponential function with a natural constant as the base.

8. The parameterization method of rain and snow temperature threshold according to claims 1 to 7, characterized in that: After selecting a model for identifying the critical temperature of rain and snow in the target area and / or identifying the temporal characteristics of precipitation in different phases in the target area based on the calculated error, the method further includes: When different models are selected to identify the critical temperature of rain and snow in the target area and to identify the time characteristics of precipitation in different phases in the target area, the unit modules in the model for identifying the critical temperature of rain and snow in the target area and the unit modules for identifying the time characteristics of precipitation in different phases in the target area are combined to construct a parameterized model, which is used to identify the critical temperature of rain and snow in the target area and the time characteristics of precipitation in different phases in the target area.

9. A parameterization device for rain and snow temperature thresholds, characterized in that: The device comprises: a data acquisition unit configured to acquire daily average temperature and precipitation data of a target area; a model building unit configured to build a plurality of models, wherein the plurality of models include a first model and a second model; a parameter identification unit configured to input the daily average temperature and precipitation data of the target area into the first model and the second model, respectively, to obtain a first identification result and a second identification result; wherein the first identification result and the second identification result both include the critical temperature for rain and snow in the target area and the time characteristics of precipitation in different phases in the target area; The model screening unit is configured to calculate the errors between the first recognition result and the second recognition result and the real meteorological data respectively, and select a model for identifying the critical temperature of rain and snow in the target area and / or identifying the time characteristics of precipitation in different phases in the target area based on the calculated errors. 10 . A non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, perform the method according to claim 1 .

Images