Snow melting runoff mixed forecasting method and system

Abstract

The invention discloses a mixed forecasting method and system for snow-melting runoff, relates to the technical field of hydrological forecasting, and solves the technical problem of insufficient reliability of snow-melting runoff forecasting under complex terrain and meteorological conditions in the prior art. The method comprises the steps that a long-short-term memory network model, namely a DS-LSTM model, based on dynamic-static hybrid driving is constructed, training verification is conducted on the DS-LSTM model through a data set generated by research area data, and the DS-LSTM model captures the influence of early-stage meteorological elements on accumulated snow characteristics through long-period memory characteristics; describing the effect of a local static underlying surface on accumulated snow features through a multi-dimensional learning ability, replacing a snow accumulation and melting calculation module in a snow melting runoff model with a verified DS-LSTM model to obtain a snow melting runoff mixed forecasting model, and adopting the snow melting runoff mixed forecasting model to realize snow melting runoff mixed forecasting; according to the method, snow accumulation and melting calculation data of multi-factor dynamic correction can be provided for the snow melting runoff model.

Description

Technical Field

[0001] The invention relates to the technical field of hydrological forecasting, and in particular to a snowmelt runoff hybrid forecasting method and system. Background Art

[0002] In recent years, with climate change, the melting of snow and glaciers has intensified, and the temporal and spatial distribution of snowmelt runoff has changed, further increasing the uncertainty of snowmelt runoff simulation forecast. Improving the accuracy of snowmelt runoff simulation forecast is of great significance to reservoir flood control, power generation and comprehensive water resources management. Most of the current snowmelt runoff models are physical-based models, which are mainly divided into two categories. One is the conceptual model based on degree-day factors, and the other is the model based on energy balance. The model based on energy balance comprehensively considers the input and output of various types of energy in the snowmelt process and can describe the snowmelt process more accurately. However, because it requires high-quality observation data and the parameterization process is complex, especially for long-term simulation of large-scale areas, the computational cost is high and it is difficult to run in real time; the conceptual model based on degree-day factors assumes that the snowmelt rate is linearly related to the temperature, and uses empirical formulas to determine the relationship between the average daily temperature and the amount of snowmelt. This type of model is simple and easy to use, with high computational efficiency, and is widely used in snowmelt runoff forecasting in alpine areas. However, the model performs poorly under complex terrain and meteorological conditions. It only uses temperature as the main driving factor and is unable to accurately capture the nonlinear and dynamic changes of snow accumulation, resulting in insufficient reliability in snowmelt runoff prediction. Summary of the invention

[0003] In order to solve the above-mentioned problems existing in the prior art, the present invention provides a hybrid forecasting method and system for snowmelt runoff, which solves the technical problem that the prior art has insufficient reliability in predicting snowmelt runoff under complex terrain and meteorological conditions.

[0004] A hybrid forecasting method for snowmelt runoff includes constructing a long short-term memory network model based on a "dynamic-static" hybrid drive, namely a DS-LSTM model, training and verifying the DS-LSTM model using a data set generated by study area data, wherein the DS-LSTM model uses long-term memory characteristics to capture the influence of previous meteorological factors on snow accumulation characteristics, and depicts the effect of local static underlying surfaces on snow accumulation characteristics through multi-dimensional learning capabilities, and replaces the snowmelt calculation module in the snowmelt runoff model with the verified DS-LSTM model to obtain a hybrid forecasting model for snowmelt runoff, and uses the hybrid forecasting model for snowmelt runoff to realize hybrid forecasting of snowmelt runoff.

[0005] Furthermore, the study area data include dynamic meteorological data, static underlying surface characteristic data, snow characteristic site measured data and snow remote sensing data.

[0006] Furthermore, the data acquisition process of the study area includes:

[0007] Dynamic meteorological data: collect and integrate precipitation, temperature, radiation, etc. in the basin. These data are refined in space and time through interpolation and reanalysis techniques to ensure data integrity and consistency.

[0008] Static underlying surface characteristic data: obtain DEM, aspect, slope, land use type, soil, etc., perform spatial rasterization on the collected data (for example, unify to 0.25° resolution), and use interpolation methods to fill in missing values ​​and process abnormal data to ensure data integrity and consistency.

[0009] Snow characteristic site measured data: collect existing snow water equivalent measured site data, and select sites with long-term series observation data, so that the selected sites have less missing data and higher observation frequency;

[0010] Snow remote sensing data: Select snow remote sensing products with a long series and stable update frequency, and perform preliminary data processing after completing data download.

[0011] Furthermore, the generated data set includes snow observation sites selected based on measured data of snow characteristic sites, determining the grid position of each site after selection, and extracting dynamic meteorological data, static underlying surface characteristic data and snow remote sensing data from the study area data based on the grid position.

[0012] Furthermore, the DS-LSTM model includes a dynamic input module, a static feature integration module and a feature fusion module. The dynamic input module uses the memory unit of LSTM to capture the dynamic change characteristics of meteorological elements in time series as dynamic features, so as to accurately predict future snow characteristics; the static feature integration module combines the multi-dimensional data mining capability of the model to capture the influence of local underground surface characteristics on snow distribution and characteristics as static features; the feature fusion module introduces an embedding layer to map the static features to the same feature space of the dynamic features, splices the dynamic features output by the dynamic input module with the static features output by the static feature integration module to form a mixed feature vector, and then introduces an attention mechanism to weight the mixed feature vector to enhance the contribution of key features to the prediction of snow characteristics.

[0013] Furthermore, the training verification includes setting the loss function, the hyperparameter optimization algorithm, the early stopping strategy, and the Patience parameter. If the verification performance does not improve within the multiple continuous training cycles specified by the Patience parameter, the training will be terminated. After the training is terminated, any snow observation site is extracted as a verification site, and the snow water equivalent remote sensing data, dynamic meteorological data, and static underlying surface data at the site location are used as input data and brought into the trained DS-LSTM model for verification. The RMSE is calculated based on the verification data to perform a business evaluation on the corrected snow water equivalent data. If the evaluation is qualified, the final DS-LSTM model is obtained. Furthermore, the snowmelt runoff model adopts the HBV model, which includes a snowmelt calculation module, a soil module, and a confluence module. The snowmelt calculation module in the HBV model is replaced with the trained and verified DS-LSTM to obtain the snowmelt runoff hybrid forecast model DS-LSTM-HBV, which realizes the dynamic correction of snow accumulation.

[0014] Furthermore, after the construction is completed, the DS-LSTM-HBV is calibrated and verified to improve the accuracy of snowmelt runoff mixing prediction.

[0015] A snowmelt runoff hybrid forecasting system includes a data input module, a prediction module and an output module, the prediction module includes a snowmelt runoff hybrid forecasting model constructed using a snowmelt runoff hybrid forecasting method, the data input module is used to input measured data of snow observation sites, the prediction module is used to predict the snow water equivalent of the snow observation sites based on the measured data input by the data input module, and the output module is used to output the snow water equivalent predicted by the prediction module.

[0016] Furthermore, it also includes a dynamic database of meteorological elements, a static database of underlying surface characteristics, a snow characteristic site measurement database and a snow remote sensing database, which are used to store various measured data of snow observation sites.

[0017] The beneficial effects of the present invention include:

[0018] 1. It can provide snowmelt calculation data with multi-factor dynamic correction for snowmelt runoff model.

[0019] 2. Dynamic snow data correction based on snow observation station data can achieve accurate correction and spatial interpolation of snow remote sensing data under different terrain and climatic conditions, providing more reliable input data for watershed management and hydrological forecasting. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a snowmelt runoff hybrid forecasting method involved in an embodiment of the present application.

[0021] Figure 2This is a diagram showing the verification of the snow water equivalent correction results at the Zhuoma site involved in the embodiment of the present application.

[0022] Figure 3 This is a verification effect diagram of the simulation results of the snowmelt runoff model at the Xinlong station involved in the embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0024] A hybrid snowmelt runoff forecasting method, such as Figure 1 As shown, it includes constructing a long short-term memory network model based on a "dynamic-static" hybrid drive, namely a DS-LSTM model, and training and verifying the DS-LSTM model using a data set generated by the study area data. The DS-LSTM model uses the long-term memory characteristics to capture the influence of previous meteorological factors on snow accumulation characteristics, and depicts the effect of local static underlying surfaces on snow accumulation characteristics through multi-dimensional learning capabilities. The snowmelt calculation module in the snowmelt runoff model is replaced with the verified DS-LSTM model to obtain a snowmelt runoff hybrid forecast model, and the snowmelt runoff hybrid forecast model is used to realize snowmelt runoff hybrid forecast.

[0025] In another embodiment, the study area data includes dynamic meteorological data, static underlying surface characteristic data, snow characteristic site measured data and snow remote sensing data. Specifically, the study area data acquisition process includes:

[0026] Dynamic meteorological data: collect and integrate precipitation, temperature, radiation, etc. in the basin. These data are refined in space and time through interpolation and reanalysis techniques to ensure data integrity and consistency.

[0027] Static underlying surface characteristic data: obtain DEM, aspect, slope, land use type, soil, etc., perform spatial rasterization on the collected data (for example, unify to 0.25° resolution), and use interpolation methods to fill in missing values ​​and process abnormal data to ensure data integrity and consistency.

[0028] Measured data of snow accumulation characteristics stations: collect existing measured station data on snow water equivalent, and screen out stations with long-term observation data, so that the screened out stations have less missing data and higher observation frequency; among them, for stations with long-term observation data, considering that the measured station data are mostly short and intermittent, it is sufficient to screen out relatively long series.

[0029] Snow remote sensing data: Select snow remote sensing products with a long series and stable update frequency, and perform preliminary data processing after completing data download.

[0030] In this embodiment,

[0031] Dynamic meteorological data: The snow depth dataset (V1.0.1961-2013) of Qinghai-Tibet Plateau stations was downloaded from the spatiotemporal environment big data platform (https: / / poles.tpdc.ac.cn / zh-hans / ); precipitation and temperature data of 2012-2019 cn05 (spatial resolution of 0.25°×0.25, daily) were downloaded from China Meteorological Network; radiation data (1km resolution) were downloaded from the Earth Resources Data Cloud Platform (www.gis5g.com). After downloading, the data were resampled and rasterized to a resolution of 0.25°.

[0032] Static underlying surface characteristic data: Download the national DEM data of China's digital elevation map (1KM) from the spatiotemporal environmental big data platform (https: / / poles.tpdc.ac.cn / zh-hans / ). The elevation data map is a 1km data formed based on China's 1:250,000 contour lines and elevation points, including DEM, hill shade, slope, and aspect maps; download the 30-meter annual land cover dataset and its dynamic changes (1985-2022) from the National Center for Glacier, Frozen Soil and Desert Science Data (http: / / www.ncdc.ac.cn); download China's 30-meter resolution land use data from 1990 to 2022 from the Earth Resources Cloud (www.gis5g.com). Download the basic attribute dataset of China's high-resolution national soil information grid (2010-2018) from the spatiotemporal environmental big data platform (https: / / poles.tpdc.ac.cn / zh-hans / ). After downloading, the data were resampled and rasterized to a uniform resolution of 0.25°.

[0033] Snow characteristics site measured data: from the National Glacier, Frozen and Desert Data Center

[0034] The snow depth dataset of ordinary stations in typical snow-covered areas in China (2017-2019) was downloaded from http: / / www.ncdc.ac.cn / portal / . The dataset contains the manual observation and instrument automatic observation data of 41 ordinary meteorological stations in typical snow-covered areas in my country from 2017 to 2019. 25 stations with continuous long series of data were selected. The interpolation method was used to fill the missing values ​​and process the abnormal data to ensure the integrity and consistency of the data.

[0035] Snow remote sensing data: Log in to Earthdata official website (https: / / urs.earthdata.nasa.gov) to query AMSR data products (spatial resolution: 0.25°, time: July 2, 2012 to present, daily snow water equivalent data), write Python code to automatically download daily snow water equivalent data, and convert the initial AMSR data in he5 format to nc format. AMSR snow remote sensing data information is converted from EASE-GRID projection to WGS84 projection.

[0036] In another embodiment, the generated data set includes snow observation sites screened according to measured data of snow characteristic sites, determining the grid position of each site after screening, and extracting dynamic meteorological data, static underlying surface characteristic data and snow remote sensing data from the study area data based on the grid position.

[0037] In this embodiment, the grid positions of 25 meteorological observation stations in typical snow-covered areas were screened out from the measured data of snow characteristic stations, and the daily-scale dynamic meteorological data (precipitation, temperature and radiation), static underlying surface data (elevation, slope aspect, land use and soil type), and AMSR snow remote sensing data from 2017 to 2019 were extracted to ensure that the model input covers the key dynamic and static factors affecting snow changes, providing a solid data foundation for the accuracy and applicability of the model.

[0038] In another embodiment, the DS-LSTM model includes a dynamic input module, a static feature integration module and a feature fusion module. The dynamic input module uses the memory unit of LSTM to capture the dynamic change characteristics of meteorological elements in time series as dynamic features, so as to accurately predict future snow characteristics; the static feature integration module combines the multi-dimensional data mining capability of the model to capture the influence of local underground surface characteristics on snow distribution and characteristics as static features; the feature fusion module introduces an embedding layer to map the static features to the same feature space of the dynamic features, that is, to map the static features to the same grid position of the dynamic features, and splices the dynamic features output by the dynamic input module with the static features output by the static feature integration module to form a mixed feature vector, and then introduces an attention mechanism to weight the mixed feature vector to enhance the contribution of key features to snow feature prediction.

[0039] In another embodiment, the training verification includes setting a loss function, a hyperparameter optimization algorithm, an early stopping strategy, and a Patience parameter. If the verification performance has not improved within a plurality of consecutive training cycles specified by the Patience parameter, the training will be terminated. After the training is terminated, any snow observation site is extracted as a verification site, and the snow water equivalent remote sensing data, dynamic meteorological data, and static underlying surface data at the site location are used as input data and brought into the trained DS-LSTM model for verification. The RMSE is calculated based on the verification data to perform business evaluation on the corrected snow water equivalent data. If the evaluation is qualified, the final DS-LSTM model is obtained.

[0040] In this embodiment, the observation data of 25 ordinary meteorological stations in typical snow-covered areas are used as the true value, of which 2017 and 2018 are the training period and 2019 is the verification period. Set the parameter range, the number of hidden layer neurons (50-100), the time step (10-100), and the learning rate (0.001-0.01), and input the extracted daily dynamic meteorological data and static geographic feature data from 2017 to 2019 into the LSTM model for training. By continuously adjusting the model parameters, the model performance is optimized until RMSE ≤ 0.85, indicating that the model has a high correction accuracy and the model completion rate is determined.

[0041] Model effect evaluation. To verify the universality of the model, representative stations were selected, and the AMSR remote sensing data, dynamic and static data of the stations were used as input data of the DS-LSTM model to obtain the corrected snow water equivalent data. Figure 2 The correction results of AMSR snow data at Zhuoma station from October 1 to December 31, 2017 are shown. The corrected RMSE value is 0.85, which meets the validation requirements of the model.

[0042] In another embodiment, the snowmelt runoff model adopts the HBV model, which includes a snowmelt accumulation calculation module, a soil module and a confluence module. The snowmelt accumulation calculation module in the HBV model is replaced with a trained and verified DS-LSTM to obtain a snowmelt runoff hybrid forecasting model DS-LSTM-HBV, thereby realizing dynamic correction of snow accumulation.

[0043] Specifically, the calculation of snow accumulation and melting is the key. This module uses the degree-day factor method to calculate snow accumulation and snowmelt, and its size is controlled by the basin temperature and the previous snow accumulation. The DS-LSTM remote sensing snow correction model obtained in the above steps replaces the snow accumulation calculation module of the HBV model, and constructs the DS-LSTM-HBV snowmelt runoff hybrid forecast model based on the intelligent correction of remote sensing snow products to realize the dynamic correction of the model's snow accumulation.

[0044] In another embodiment, after the construction is completed, the DS-LSTM-HBV is calibrated and verified to improve the accuracy of snowmelt runoff mixing prediction.

[0045] In this embodiment, the newly constructed DS-LSTM-HBV model is calibrated and verified to obtain a snowmelt runoff model with high accuracy. Figure 3 The model simulation results of the Xinlong Station snowmelt runoff model before and after replacing the snowmelt calculation module from March 1 to May 31, 2021 are shown. The R2 value of the snowmelt runoff simulation before replacement was 0.76, and the R2 value of the snowmelt runoff simulation after replacement was 0.87. The model accuracy was improved after replacement, indicating that the remote sensing snow intelligent correction method that replaces the snowmelt calculation module can improve the simulation and forecast accuracy of the snowmelt runoff model.

[0046] A snowmelt runoff hybrid forecasting system includes a data input module, a prediction module and an output module, the prediction module includes a snowmelt runoff hybrid forecasting model constructed using a snowmelt runoff hybrid forecasting method, the data input module is used to input measured data of snow observation sites, the prediction module is used to predict the snow water equivalent of the snow observation sites based on the measured data input by the data input module, and the output module is used to output the snow water equivalent predicted by the prediction module.

[0047] In another embodiment, it also includes a dynamic database of meteorological elements, a static database of underlying surface characteristics, a snow characteristic site measurement database and a snow remote sensing database, which are used to store various types of measured data of snow observation sites.

[0048] The above-mentioned embodiments only express the specific implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A hybrid snowmelt runoff forecasting method, characterized in that: It includes constructing a long short-term memory network model based on "dynamic-static" hybrid drive, namely the DS-LSTM model, and training and verifying the DS-LSTM model using a data set generated by data from the study area. The DS-LSTM model uses long-term memory characteristics to capture the impact of previous meteorological factors on snow accumulation characteristics, and depicts the effect of local static underlying surfaces on snow accumulation characteristics through multi-dimensional learning capabilities. The snowmelt calculation module in the snowmelt runoff model is replaced with the verified DS-LSTM model to obtain a snowmelt runoff hybrid forecast model, and the snowmelt runoff hybrid forecast model is used to realize snowmelt runoff hybrid forecast.

2. A snowmelt runoff mixed forecasting method according to claim 1, characterized in that: The data of the study area include dynamic meteorological data, static underlying surface characteristic data, snow characteristic site measured data and snow remote sensing data.

3. A snowmelt runoff mixed forecasting method according to claim 2, characterized in that: The data acquisition process of the study area includes: Dynamic meteorological data: collect and integrate precipitation, temperature, and radiation in the basin, and refine them in space and time through interpolation and reanalysis techniques; Static underlying surface characteristic data: obtain DEM, aspect, slope, land use type, soil, perform spatial rasterization on the collected data, and use interpolation methods to fill in missing values ​​and process abnormal data; Snow characteristic site measured data: collect existing snow water equivalent measured site data and select sites with long time series observation data; Snow remote sensing data: Select snow remote sensing products, complete data download, and perform preliminary data processing.

4. A snowmelt runoff mixed forecasting method according to claim 2, characterized in that: The generated data set includes snow observation sites selected based on measured data of snow characteristic sites, determining the grid position of each site after screening, and extracting dynamic meteorological data, static underlying surface characteristic data and snow remote sensing data from the study area data based on the grid position.

5. A snowmelt runoff mixed forecasting method according to claim 2, characterized in that: The DS-LSTM model includes a dynamic input module, a static feature integration module and a feature fusion module. The dynamic input module uses the memory unit of LSTM to capture the dynamic change characteristics of meteorological elements in time series as dynamic features; The static feature integration module combines the multi-dimensional data mining capabilities of the model to capture the impact of local underground surface characteristics on snow distribution and characteristics as static features; The feature fusion module introduces an embedding layer, maps static features to the same feature space of dynamic features, concatenates the dynamic features output by the dynamic input module with the static features output by the static feature integration module to form a mixed feature vector, and then introduces an attention mechanism to perform weighted processing on the mixed feature vector.

6. A snowmelt runoff mixed forecasting method according to claim 1, characterized in that: The training verification includes setting a loss function, a hyperparameter optimization algorithm, an early stopping strategy, and a Patience parameter. If the verification performance does not improve within a plurality of consecutive training cycles specified by the Patience parameter, the training will be terminated. After the training is terminated, any snow observation site is extracted as a verification site, and the snow water equivalent remote sensing data, dynamic meteorological data, and static underlying surface data at the site location are used as input data and brought into the trained DS-LSTM model for verification. The RMSE is calculated based on the verification data to perform business evaluation on the corrected snow water equivalent data. If the evaluation is qualified, the final DS-LSTM model is obtained.

7. A snowmelt runoff mixed forecasting method according to claim 1, characterized in that: The snowmelt runoff model adopts the HBV model, which includes a snowmelt calculation module, a soil module and a confluence module. The snowmelt calculation module in the HBV model is replaced with a trained and verified DS-LSTM to obtain a snowmelt runoff hybrid forecasting model DS-LSTM-HBV.

8. A snowmelt runoff mixed forecasting method according to claim 7, characterized in that: After the construction is completed, DS-LSTM-HBV is calibrated and verified.

9. A snowmelt runoff hybrid forecasting system, characterized in that: It includes a data input module, a prediction module and an output module, the prediction module includes a snowmelt runoff hybrid prediction model constructed by a snowmelt runoff hybrid prediction method according to any one of claims 1 to 8, the data input module is used to input measured data of snow observation sites, the prediction module is used to predict the snow water equivalent of the snow observation sites according to the measured data input by the data input module, and the output module is used to output the snow water equivalent predicted by the prediction module.

10. A snowmelt runoff mixed forecasting system according to claim 9, characterized in that: It also includes a dynamic database of meteorological elements, a static database of underlying surface characteristics, a measured database of snow characteristics stations, and a snow remote sensing database, which are used to store various measured data from snow observation stations.

Images