A method for dynamically predicting snow cover rate

Abstract

The present invention discloses a method for dynamically predicting snow cover rate, including: S1, selecting representative historical data groups where the snow cover rate decreases as the temperature increases; determining the snow critical temperature as the reference temperature for cumulative temperature; S2, taking the snow critical temperature as the starting standard, calculating the cumulative temperature, and based on the cumulative temperature, plotting multiple groups of snow recession curves with different snow cover rates as the starting values; S3, determining the snow recession curve to be used during the prediction period according to the snow cover rate on the day before the prediction period, and predicting the snow cover rate for each day during the prediction period; S4, selecting a new snow recession curve according to the measured snow cover rate value on the first day of the prediction period, the snow critical temperature, and the cumulative temperature on the first day of the prediction period. The advantages are: it can flexibly achieve snow prediction for each region at the grid scale, can reflect the spatial distribution of snow from the space, more accurately predict the snow recession process, and provide support for accurate snowmelt runoff forecasting and simulation.

Description

Technical Field

[0001] The present invention relates to the technical field of snow remote sensing product processing and prediction, and particularly relates to a method for dynamically predicting snow cover rate. Background Art

[0002] The change of snow cover has an important impact on the simulation and prediction of snowmelt runoff in areas replenished by snowmelt runoff. However, there are few current snow prediction methods, and the prediction methods have great uncertainties.

[0003] Currently, the main methods for snow prediction are as follows. One is to analyze the self-characteristics of snow cover data, including the snow cover range, the relationship between snowfall and months and seasons, and the relationship between snowfall and precipitation and temperature within a year. Using historical snow cover data, the future snowfall situation is theoretically predicted, but the prediction error is relatively large. The second is to rely on climate model models that understand the snow climate effect of numerical models to predict future snow cover. Currently, due to the relatively imperfect meteorological measured data in high-altitude areas such as high mountains and plateaus, the prediction of snow cover mainly stays in a theoretical analysis area. Theoretical prediction can only analyze the snow cover change trend within a large range, and the specific snow cover rate in a small area cannot be obtained, and detailed hydrological prediction work cannot be carried out. The snow cover rate prediction of climate model models is relatively complex, requires a large amount of data preparation, and there will be problems with poor correlation between prediction skills and prediction starting time. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for dynamically predicting snow cover rate, so as to solve the foregoing problems existing in the prior art.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A method for dynamically predicting snow cover rate includes the following steps:

[0007] S1. Screen snow and temperature data with good reliability and consistency, and select representative historical data groups in which the snow cover rate decreases as the temperature increases; according to the actual situation of the research area, determine the snow critical temperature as the reference temperature for cumulative temperature;

[0008] S2. Taking the snow critical temperature as the starting standard, calculate the cumulative temperature, and based on the selected representative historical data groups and the cumulative temperature, draw multiple groups of snow recession curves with different snow cover rates as the starting values;

[0009] S3. Extract the snow cover rate on the day before the prediction period, determine the snow recession curve to be used in the prediction period based on this snow cover rate, and locate the corresponding snow cover rate on this snow recession curve for each day in the prediction period proportionally based on the cumulative temperature relationship between the predicted temperature values for each day in the prediction period and the snow critical temperature.

[0010] S4. Based on the initial prediction in step S3, when the measured snow cover rate on the first day of the prediction period is obtained, select a new snow recession curve according to the measured snow cover rate value on the first day of the prediction period, the snow critical temperature, and the cumulative temperature on the first day of the prediction period to achieve dynamic adjustment of the prediction.

[0011] Preferably, there is a step S0 before step S1, and step S0 includes

[0012] S01. Preliminary construction of the meteorological land surface database: Collect the DEM basic data of the target basin and the air temperature and snow data, and interpolate these data to the same spatial resolution to preliminarily construct the meteorological land surface database.

[0013] S02. Obtaining the snow cover rate of the study area: Extract the snow cover data and the temperature spatial distribution data within the target basin range, perform format conversion, clipping, fusion, and snow value extraction on the snow cover data to obtain the snow cover rate of the study area.

[0014] Preferably, in step S2, draw 10 groups of snow recession curves starting from 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1 respectively.

[0015] Preferably, in step S3, select the snow recession curve with the smallest difference between the starting value and the snow cover rate on the day before the prediction period as the snow recession curve to be used in the prediction period.

[0016] Preferably, the calculation formula for the cumulative temperature in step S3 is

[0017]

[0018] where T i is the predicted temperature value for the i-th day, and T crit is the snow critical temperature; E i is the cumulative temperature for the i-th day, i = 1, 2, …, n, and n is the total number of days.

[0019] Preferably, there is also a step S5 after step S4. Specifically, when the measured values of the snow cover rate for each day in the prediction period are known, compare the magnitude relationship between the difference between the measured value of the snow cover rate for each day in the prediction period and the predicted value of the snow cover rate for the corresponding day and the preset error. If the difference is less than the preset error, it means that the prediction accuracy meets the requirements, and the snow cover rate is output; otherwise, adjust the snow recession curve until the snow prediction effect within each DEM range meets the accuracy requirements, and terminate the adjustment of the snow recession curve.

[0020] The beneficial effects of the present invention are as follows: 1. The method of the present invention predicts the future snow cover rate of the study area by constructing snow recession curves under different snow cover degrees in the study area and then according to the current snow cover scenario and future temperature changes. The prediction method is relatively simple and has good correlation. 2. The method of the present invention overcomes the problems that the existing theoretical analysis prediction cannot accurately reflect the snow cover rate, and the climate model prediction is relatively complex and the correlation is relatively unclear. It can provide accurate data information for the snowmelt runoff forecast and simulation in basins lacking data, and provide support for related application research. 3. The method of the present invention can flexibly realize snow prediction in each region at the grid scale, can reflect the spatial distribution of snow from space, and more accurately predict the snow recession process, providing support for accurate snowmelt runoff forecast and simulation. Description of the Drawings

[0021] Figure 1 is a schematic flowchart of the method in the embodiment of the present invention;

[0022] Figure 2 is a schematic diagram of the snow recession curve in the upper reaches of the Lancang River in the embodiment of the present invention;

[0023] Figure 3 is a schematic diagram of snow prediction by the snow recession curve in the upper reaches of the Lancang River in the embodiment of the present invention. Detailed Embodiments

[0024] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] Embodiment 1

[0026] As Figure 1As shown in the figure, in this embodiment, a method for dynamically predicting snow cover rate is provided. Starting from the perspective of establishing the relationship between snow cover rate and accumulated temperature, the present invention collects and processes the MODIS snow cover rate gridded public data product, establishes snow recession curves with different starting recession values in the study area, and then dynamically predicts the snow value in the study area for the next 7 days according to the future temperature change, improving the simulation effect of snow prediction. The method of the present invention specifically includes seven parts:

[0027] I. Preliminary construction of meteorological land surface database

[0028] This part corresponds to the preliminary construction of the meteorological land surface database in step S0: collecting the DEM basic data, temperature and snow data of the target basin, interpolating these data to the same spatial resolution, and preliminarily constructing the meteorological land surface database.

[0029] II. Data processing and selection

[0030] This part corresponds to the acquisition of the snow cover rate in the study area in step S0: extracting the snow cover data and temperature spatial distribution data within the range of the target basin, performing format conversion, clipping, fusion and snow value extraction on the snow cover data, and obtaining the snow cover rate in the study area.

[0031] III. Data screening

[0032] This part corresponds to step S1. From the meteorological land surface database in step S0 and the snow cover rate in the study area, snow and temperature data with good reliability and consistency are screened, and representative historical data groups with decreasing snow cover rate as the temperature increases are selected; according to the actual situation of the study area, the snow critical temperature T crit is determined as the reference temperature for cumulative temperature.

[0033] IV. Drawing of snow recession curves

[0034] This part corresponds to step S2. Taking the snow critical temperature T crit as the starting standard, the cumulative temperature is calculated, and the selected representative historical data groups and the cumulative temperature are formed into a recession curve array. Based on the recession curve array, multiple groups of snow recession curves are drawn with different snow cover rates as the starting values; due to different starting snow cover rates, 10 groups of snow recession curves are drawn with 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1 as the starting values respectively.

[0035] The number of recession curves is determined according to the actual situation of the study area, and may be less than 10 groups or more than 10 groups (interpolating snow recession curves to improve accuracy). The snow recession curves should be representative of the local snow recession process in the study area.

[0036] V. Initial snow cover rate calculation

[0037] This part corresponds to step S3. Extract the snow cover rate of the day before the prediction period, determine the snow recession curve to be used in the prediction period according to the difference between the snow cover rate and the starting value of each snow recession curve, and based on the cumulative temperature relationship between the predicted temperature values of each day in the prediction period and the snow critical temperature T crit , locate the value of the predicted snow cover rate on this snow recession curve proportionally, and then predict the snow cover rate of each day in the prediction period.

[0038] Specifically, select the snow recession curve with the smallest difference between the starting value and the snow cover rate of the day before the prediction period as the snow recession curve to be used in the prediction period.

[0039] The formula for calculating the cumulative temperature is

[0040]

[0041] where T i is the predicted temperature value of the i-th day, and T crit is the snow critical temperature; E i is the cumulative temperature of the i-th day, i = 1, 2,..., n, and n is the total number of days.

[0042] VI. Dynamic prediction of snow cover rate

[0043] This part corresponds to step S4. Based on the initial prediction, when the measured snow cover rate of the first day in the prediction period is obtained, select a new snow recession curve according to the measured snow cover rate value of the first day in the prediction period, the snow critical temperature T crit and the cumulative temperature of the first day in the prediction period to achieve dynamic adjustment of the prediction.

[0044] VII. Correction of snow recession curve

[0045] This part corresponds to step S5. When the measured values of the snow cover rate of each day in the prediction period are known, compare the magnitude relationship between the difference between the measured value and the predicted value of the snow cover rate of each day in the prediction period and the preset error. If the difference is less than the preset error, it means that the prediction accuracy meets the requirements, and the snow cover rate is output; otherwise, adjust the snow recession curve until the snow prediction effect within each DEM range meets the accuracy requirements, and terminate the adjustment of the snow recession curve.

[0046] Embodiment 2

[0047] In this embodiment, taking the snow data in the Lancang River Basin as an example, the execution process of the method of the present invention is described in detail. The specific process is as follows:

[0048] 1. Obtain MODIS data from the Earthdata official website. Download the snow cover data of four modules, namely h25V05, h26v05, h26v06, and h27v06 in MOD10A from 2007 to 2018 according to the longitude and latitude of the Lancang River elevation zone. Obtain and download the national DEM data from the official website of the Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences, and cut out the data within the scope of the Lancang River Basin. Download the China Regional High Spatiotemporal Resolution Ground Meteorological Element Driving Dataset (CMFD) with a spatial resolution of 0.1° from the National Tibetan Plateau Data Center. Interpolate the dataset to a unified grid with a resolution of 500m.

[0049] 2. Cut out the upstream basin range of the Lancang River from the MOD10A1 500m grid snow dataset, splice the modules, extract the grid snow values, and calculate the snow cover rate. Cut out the upstream area range of the Lancang River from the temperature data.

[0050] 3. Select representative snow cover and temperature datasets. According to the actual situation in the upstream of the Lancang River, T crit is selected as 0°C.

[0051] 4. Use the selected datasets to draw the snow recession curve of the upstream of the Lancang River. Due to the actual snow situation on the spot, the snow cover rate range is mainly concentrated in the interval of 0 - 0.3. Therefore, draw the snow recession curves starting from 0.1, 0.2, and 0.3, as Figure 2 shown.

[0052] 5. Assume to predict the snow cover rate for the next 7 days starting from April 18, 2023. First, extract the snow cover rate of 0.32 on April 17. Since 0.32 only differs from 0.3 by 0.02, select recession curve 3 as the snow prediction recession curve (if the snow cover rate on the day before the prediction period differs from the initial value of the recession curve by > 0.05, an interpolation can be made on the existing snow recession curves to obtain a snow recession curve that meets the value). Select recession curve 3 and obtain the predicted temperature data for the next 7 days, which are 0, 0.66, 1.47, 2.09, 4.16, 1.48, and 2.07 degrees Celsius respectively. The cumulative temperature is calculated by the following formula:

[0053]

[0054] where, T i is the predicted temperature value on the i-th day, and T crit is the snow critical temperature; E i$T_i$ is the cumulative temperature on the $i$-th day, where $i = 1, 2, \ldots, n$ and $n$ is the total number of days. The 7-day cumulative temperatures are 0, 0.66, 2.13, 4.22, 8.38, 9.86, and 11.93 degrees Celsius respectively, and the corresponding snow cover rates obtained on the snow recession curve are 30.31, 25.10, 16.33, 9.67, 7.77, 7.97, and 6.60 respectively. See Figure 3 .

[0055] 6. After obtaining the measured snow cover rate value on April 18, a new snow recession curve is selected according to the measured snow cover rate to dynamically and slidingly predict the snow cover.

[0056] 7. After all the measured snow cover values for 7 days are known, the snow recession curve is further modified and improved according to the prediction accuracy, so as to ensure that the snow recession curve is representative and the prediction is accurate.

[0057] By adopting the above technical solutions disclosed in the present invention, the following beneficial effects are obtained:

[0058] The present invention provides a method for dynamically predicting the snow cover rate. The method of the present invention predicts the future snow cover rate in the study area by constructing the snow recession curves under different snow cover degrees in the study area, and then according to the current snow cover scenario and future temperature changes. The prediction method is relatively simple and has a good correlation. The method of the present invention overcomes the problems that the existing theoretical analysis prediction cannot accurately reflect the snow cover rate, and the climate model prediction is relatively complex and the correlation is not clear. It can provide accurate data information for the snowmelt runoff forecast and simulation in the basins lacking data, and provide support for relevant applied research. The method of the present invention can flexibly realize the snow cover prediction in each region at the grid scale, can reflect the spatial distribution of the snow cover from the space, and more accurately predict the snow recession process, providing support for accurate snowmelt runoff forecast and simulation.

[0059] The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for dynamically predicting snow cover rate, characterized in that: It includes the following steps: S1. Screen the snow cover and air temperature data with good reliability and consistency, and select representative historical data groups in which the snow cover rate decreases as the temperature increases; according to the actual situation of the research area, determine the snow cover critical temperature as the reference temperature for the cumulative temperature; S2. Taking the snow cover critical temperature as the starting standard, calculate the cumulative temperature, and based on the selected representative historical data groups and the cumulative temperature, draw multiple groups of snow cover decline curves with different snow cover rates as the starting values; S3. Extract the snow cover rate on the day before the prediction period, determine the snow cover decline curve to be used in the prediction period according to this snow cover rate, and based on the cumulative temperature relationship between the predicted temperature values of each day in the prediction period and the snow cover critical temperature, locate and predict the snow cover rate corresponding to each day in the prediction period on this snow cover decline curve proportionally; In step S3, select the snow cover decline curve with the smallest difference between the starting value and the snow cover rate on the day before the prediction period as the snow cover decline curve to be used in the prediction period; In step S3, the calculation formula for the cumulative temperature is Among them, T i is the temperature prediction value for the i-th day, and T crit is the snow cover critical temperature; E i is the cumulative temperature for the i-th day, where i = 1, 2, …, n and n is the total number of days; S4. Based on the initial prediction in step S3, when the measured snow cover rate on the first day of the prediction period is obtained, select a new snow cover decline curve according to the measured snow cover rate value on the first day of the prediction period, the snow cover critical temperature, and the cumulative temperature on the first day of the prediction period to achieve dynamic adjustment of the prediction; There is also step S5 after step S4. Specifically, in step S5, when the measured snow cover rate values of each day in the prediction period are known, compare the magnitude relationship between the difference between the measured snow cover rate value of each day in the prediction period and the predicted snow cover rate value of the corresponding day and the preset error. If the difference is less than the preset error, it means that the prediction accuracy meets the requirements, and output the snow cover rate; otherwise, adjust the snow cover decline curve until the snow cover prediction effect within each DEM range meets the accuracy requirements, and terminate the adjustment of the snow cover decline curve.

2. The method for dynamically predicting snow cover rate according to claim 1, wherein: There is step S0 before step S1. Step S0 includes S01. Preliminary construction of the meteorological land surface database: Collect the DEM basic data of the target basin and the air temperature and snow cover data, and interpolate these data to the same spatial resolution to preliminarily construct the meteorological land surface database; S02. Obtaining the snow cover rate of the research area: Extract the snow cover data and the temperature spatial distribution data within the target basin, perform format conversion, clipping, fusion, and snow value extraction on the snow cover data to obtain the snow cover rate of the research area.

3. The method for dynamically predicting snow cover rate according to claim 1, wherein: In step S2, draw 10 groups of snow cover decline curves with 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1 as the starting values respectively.

Images