Xinjiang region snow depth process simulation method
By correcting and adjusting meteorological and geographical data for Xinjiang, the simulation bias problem of existing snow cover models in Xinjiang has been solved, and higher accuracy in simulating snow depth has been achieved.
Patent Information
- Application Number
- CN202411003364.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-08
AI Technical Summary
Existing snow cover models exhibit significant simulation biases in Xinjiang, particularly in areas with complex terrain, making it impossible to accurately estimate snow depth. This is primarily due to the fact that model parameters are global averages or constants, and there is a lack of snow depth/snow water equivalent observational data available for calibration and validation.
By acquiring meteorological and geographical data from Xinjiang, bias corrections were performed on the atmospheric reanalysis dataset, and the new snow density parameter in the model was adjusted. The model was then validated using actual observation data to ensure the input data closely matches the Xinjiang region and the parameters are compatible.
The accuracy of snow depth simulation has been improved, reducing the underestimation of snow depth in low-altitude simulations and the overestimation of snow depth in high-altitude simulations, thus ensuring the accuracy of the model's simulation of snow accumulation processes in Xinjiang.
Smart Images

Figure CN118917085A_ABST
Abstract
Description
Technical Field
[0001] The invention mainly relates to the intersecting field of geography and meteorology, and specifically is a method for simulating snow depth process in Xinjiang. Background Art
[0002] Snow accumulation is an important water resource in the arid areas of Xinjiang and has a significant impact on the ecology. An appropriate amount of snow accumulation is an important guarantee for developing winter tourism and sports. However, snow accumulation can also bring corresponding disasters. Therefore, it is necessary to simulate the physical process of snow accumulation in order to predict the snow depth and its changes over time.
[0003] Snow models range from one layer (Noah model) to two layers (DHSVM, VIC and other models) to multiple layers (Noah-MP, CLM, CROCUS, SNTHERM and other models). These snow models can be directly coupled to the land surface model (LSM) and the global climate model (GCM) or regional climate model (RCM). One of the main reasons for model errors is that the development and testing of the schemes were completed in North America or Russia, where the snow is relatively thick and covers a large area, and some key snow parameters of land surface processes are global averages or constants in the models. When these schemes and parameters are directly applied to Xinjiang, large simulation deviations will occur (for example, the simulated snow depth is underestimated at low altitudes and overestimated at high altitudes), resulting in an inability to accurately estimate the snow depth in Xinjiang. In addition, due to the lack of snow depth / snow water equivalent observation data sets that can be used for calibration and verification, especially in areas with complex terrain (Xinjiang has mountain basins and complex terrain, with only 105 stations in the whole region. The two highest meteorological stations in China's Tianshan Mountains are Daxigou Station and Turugart Station, with altitudes of 3539m and 3504.4m respectively. The stations are scarce and mainly located in plains, with poor station representativeness and no data above high altitudes), the model has great challenges in simulating snow cover in Xinjiang. Summary of the invention
[0004] The technical solution of the present invention aims at the technical problem that the existing technical solutions are too single, and provides a solution that is significantly different from the existing technologies. It mainly provides a method for simulating snow depth process in Xinjiang, so as to solve the technical problem that the existing snow accumulation model mentioned in the above background technology has a large simulation deviation.
[0005] The technical solution adopted by the present invention to solve the above technical problems is:
[0006] A method for simulating snow depth process in Xinjiang region comprises the following steps:
[0007] S1. Obtain relevant variables in the atmospheric reanalysis data set of Xinjiang;
[0008] S2, preprocessing the atmospheric reanalysis data set in step S1, and using the relevant variable data provided by the Xinjiang Meteorological Bureau to perform bias correction on the precipitation data in the atmospheric reanalysis data set;
[0009] S3, defining the area simulated by the Noah-MP model according to the latitude and longitude of Xinjiang, obtaining static geographic data containing information such as topography, land use, and vegetation cover, and then compiling and running the atmospheric reanalysis data set prepared in step S2 and the static geographic data to generate the driving field data required for model calculation;
[0010] S4. Observe the new snow density after each snowfall event during the three different periods of snow accumulation, stability and melting to determine the real new snow density in Xinjiang and adjust the new snow density parameters in the model accordingly.
[0011] S5, substituting the driving field data in step S3 into the Noah-MP model with the new snow density parameter modified in step S4, and specifying a time step that is consistent with the driving data, and performing an integral calculation to obtain the snow depth in Xinjiang;
[0012] S6, evaluating the snow depth obtained in step S5, using the snow depth data provided by the Xinjiang Meteorological Bureau as the true value to verify the model accuracy;
[0013] S7. Plot the snow depth obtained in step S5 to obtain spatial distribution data of the snow depth.
[0014] As an implementation manner, in step S1, the atmospheric reanalysis data dataset is the ERA5-land dataset, with a spatial resolution of 0.1°×0.1° and a temporal resolution of 1 hour.
[0015] Specifically, the relevant variables include temperature, pressure, wind, humidity, radiation, and precipitation.
[0016] Specifically, step S2 includes:
[0017] S2-1. Resample the precipitation data in the atmospheric reanalysis data set to the longitude and latitude of the meteorological station, and set a precipitation frequency threshold P for the daily precipitation data in the atmospheric reanalysis data set based on the monthly scale. ERA5_WDT , so that the precipitation frequency of the atmospheric reanalysis data set exceeding the threshold is equal to the precipitation frequency of meteorological observations;
[0018] S2-2. Select a research reference period, calculate the monthly proportional coefficient S of the reference period, and perform bias correction on the daily precipitation data of the atmospheric reanalysis data set:
[0019]
[0020] Among them, μ m is the average monthly precipitation, P obs and P ERA5 are the daily precipitation data of meteorological observation and atmospheric reanalysis data set, respectively. obs_WDT is the precipitation frequency threshold when the daily precipitation is ≥1 mm, P ERA5_WDT It is the precipitation frequency threshold of the atmospheric reanalysis data set;
[0021]
[0022] Among them, P ERA5 (d) and P ERA5_correct (d) are the daily precipitation data of the atmospheric reanalysis data set before and after correction. The precipitation of the atmospheric reanalysis data set after correction has the same precipitation frequency and intensity as the meteorological observation precipitation data.
[0023] Specifically, in step S3, the area simulated by the Noah-MP model is projected as Lambert, with a spatial resolution of 9×9 km and a temporal resolution of 1 hour.
[0024] Specifically, in step S4, the density of new snow is determined using a Snowfork snow characteristics analyzer and manual sampling and weighing observations, and the two observation results assist each other in recording, calibration and verification.
[0025] Specifically, in step S4, the density of new snow in the arid region of Xinjiang is 130-160 kg / m 3 .
[0026] Specifically, in step S6, correlation coefficient and root mean square error are used as evaluation methods.
[0027] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned snow depth process simulation method is implemented.
[0028] The present invention also provides a computer device, comprising: a memory for storing instructions; a processor for executing the instructions, so that the device performs the operations of the above-mentioned snow depth process simulation method.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] (1) The present invention uses the relevant variable data provided by the Xinjiang Meteorological Bureau to perform bias correction on the precipitation data in the atmospheric reanalysis data set, so that the corrected ERA5 precipitation has the same precipitation frequency and intensity as the meteorological observation precipitation data. This reduces the large simulation bias caused by directly applying the global average or constant to Xinjiang, which is conducive to accurately estimating the snow depth in Xinjiang.
[0031] (2) The present invention obtains static geographic data containing information such as topography, land use, and vegetation coverage according to the longitude and latitude of Xinjiang, and compiles and runs the static geographic data with the pre-processed atmospheric reanalysis data set to generate the driving field data required for model calculation, so that the data input to the model is more in line with the actual situation in Xinjiang, reducing the large simulation deviation caused by directly using the original data of the model, which is conducive to accurately estimating the snow depth in Xinjiang.
[0032] (3) The present invention conducts actual observations on the new snow density in Xinjiang and adjusts the new snow density parameters in the model, thereby improving the matching degree between the model parameters and the Xinjiang region. This can further reduce the large simulation deviation caused by directly applying the global average or constant to the Xinjiang region, which is conducive to accurately estimating the snow depth in Xinjiang.
[0033] (4) After obtaining the snow depth in Xinjiang through the model, the present invention uses the snow depth data provided by the meteorological stations in Xinjiang as the true value to verify the model accuracy, which can improve the reliability of the model calculation results, thereby facilitating the accuracy of the snow depth process simulation. In addition, in areas with complex terrain such as Xinjiang, although there is a lack of snow depth / snow water equivalent observation data sets that can be used for calibration and verification, the accuracy of the model simulation is ensured by adjusting the input data and model parameters as described above, so that the snow accumulation conditions simulated by the model are very accurate. Based on this, the current low-altitude snow depth / snow water equivalent observation data sets can be used to calibrate and verify the model, and finally the snow accumulation processes at low and high altitudes simulated by the model have sufficient accuracy.
[0034] In summary, the present invention is based on the general Noah-MP model. On the one hand, it uses the relevant variable data provided by the Xinjiang Meteorological Bureau to perform bias correction on the precipitation data in the atmospheric reanalysis data set through the monthly proportional coefficient of the reference period, and compiles it with the static geographic data to generate the driving field data required for the model calculation, thereby improving the fit between the data of the input model and the Xinjiang region; on the other hand, it adjusts the new snow density parameters in the model through actual observation, thereby improving the matching degree between the model parameters and the Xinjiang region, and uses the snow depth data provided by the Xinjiang Meteorological Bureau to verify the model accuracy, thereby improving the shortcomings of the snow depth error caused by inaccurate parameters, avoiding the underestimated snow depth in the low altitude simulation and the overestimated snow depth in the high altitude simulation, improving the snow depth simulation effect, and improving the spatiotemporal resolution of the snow depth simulation.
[0035] The present invention will be explained in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 It is a flow chart of the steps of the present invention. DETAILED DESCRIPTION
[0037] To facilitate understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings, but the present invention can be implemented in different forms and is not limited to the embodiments described in the text. On the contrary, these embodiments are provided to make the content disclosed in the present invention more thorough and comprehensive.
[0038] Example 1: Please refer to the attached Figure 1 Taking Xinjiang as an example, a snow depth process simulation method in Xinjiang includes the following steps:
[0039] S1. Obtain the temperature, pressure, wind, humidity, radiation, precipitation and other variables in the ERA5 atmospheric reanalysis data set in the Xinjiang latitude and longitude range. The ERA5 data is an atmospheric reanalysis data set provided by the European Centre for Medium-Range Weather Forecasts (ECMWF), which is publicly available at the Copernicus Climate Center (https: / / cds.climate.copernicus.eu / cdsapp#! / home) and can be downloaded for free by registered users. This embodiment uses the ERA5-land data set, which provides a time coverage range from January 1950 to the present, a spatial resolution of 0.1°×0.1°, and a time resolution of 1 hour.
[0040] S2. Preprocess the ERA5 data. Compared with other variables, precipitation is the variable with the largest error. Therefore, it is necessary to use the precipitation data from 1980 to 2020 provided by the Xinjiang Meteorological Bureau to correct the ERA5 precipitation data. First, resample the ERA5 precipitation data to the longitude and latitude of the meteorological station, and set a precipitation frequency threshold P for the ERA5 daily precipitation data based on the monthly scale. ERA5_WDT (i.e., if it is greater than the threshold, it is judged as a rainy day), so that the ERA5 precipitation frequency exceeding the threshold is equal to the meteorological observation precipitation frequency. The second step is to consider 1980-2020 as the research reference period, calculate the monthly proportional coefficient S of the reference period, and perform bias correction on the ERA5 daily precipitation data:
[0041]
[0042] Among them, μ m is the average monthly precipitation, P obs and P ERA5 are meteorological observations and ERA 5-day precipitation data, respectively. obs_WDT is the precipitation frequency threshold when the daily precipitation is ≥1 mm, P ERA5_WDT is the ERA5 precipitation frequency threshold.
[0043]
[0044] Among them, P ERA5 (d) and P ERA5_correct (d) are the ERA5 daily precipitation data before and after correction. The corrected ERA5 precipitation has the same precipitation frequency and intensity as the meteorological observation precipitation data.
[0045] S3. Define the area simulated by the Noah-MP model. The central latitude and longitude of the simulated area is 41.919N, 84.919E. The number of grids in the east-west direction is 222, the number of grids in the north-south direction is 189, the projection is Lambert, the spatial resolution is 9×9km, and the temporal resolution is 1 hour. Download and generate static geographic data containing information such as topography, land use, and vegetation cover (https: / / www2.mmm.ucar.edu / wrf / users / download / get_sources_wps_geog.html), and then compile and run the ERA5 meteorological reanalysis data and static geographic data prepared in step S2 to generate the driving field data required for model calculation.
[0046] S4. Adjust the new snow density parameters, observe the new snow density after each snowfall event in three different periods: snow accumulation period (November-December), stable period (January-February), and ablation period (March-May), and dig longitudinal profiles of the snow on the ground. The snow profile should not be located in the direct sunlight direction. According to the international snow layer classification standards, identify different types of snow according to different snow particle sizes and snow layers, and observe the new snow layer at the top of the snow profile.
[0047] The density of new snow is determined using a Snowfork snow characteristics analyzer and manual sampling and weighing observations. The two observation results assist in recording, calibrating and verifying each other.
[0048] Snowfork snow characteristics analyzer can be used to measure physical quantities such as snow density, water content, and dielectric constant. During observation, observations are made and recorded at intervals of 5 cm from the top to the bottom of the snow profile surface.
[0049] Manual sampling observation uses a square snow sampler (100cm 3 ), when in use, insert it horizontally into the snow profile perpendicular to the snow profile, weigh it together with the sampler when taking it out, and finally subtract the weight of the sampler. When observing, sample and weigh it at intervals of 5 cm from the top to the bottom of the surface of the snow profile.
[0050] Through the snow survey in the previous study area, it was found that the new snow density of the "dry cold type" in the arid area of Xinjiang is 130-160 kg / m 3 In this embodiment, the average value is 145kg / m 3 Improve the parameters in the Noah-MP model. This step is to modify the model code directly. The Noah-MP model originally used the global average value of 100kg / m 3 , which is inconsistent with the actual situation in the study area and will cause deviations in snow simulation.
[0051] S5, substituting the driving field data in step S3 into the Noah-MP multi-parameterized land surface process model with the new snow density parameter modified in step S4, specifying a time step that is consistent with the driving data, and performing integral calculation to obtain the snow depth in Xinjiang;
[0052] S6. Evaluate the above snow depth, use the snow depth data provided by Xinjiang Meteorological Bureau as the true value to verify the accuracy of the model, and use the correlation coefficient (R2) and root mean square error (RMSE) as the evaluation method;
[0053] S7. Plot the above snow depth to obtain spatial distribution data of snow depth.
[0054] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned snow depth simulation method.
[0055] A computer device includes: a memory for storing instructions; a processor for executing the instructions, so that the device performs operations to implement the above-mentioned snow depth simulation method.
[0056] Embodiment 2: The difference between this embodiment and the embodiment is that: in step S1, the ERA5-land dataset is replaced by other datasets, such as the China Regional Ground Meteorological Element Driven Dataset (1979-2018) (https: / / data.tpdc.ac.cn / zh-hans / data / 8028b944-daaa-4511-8769-965612652c49) and the like.
[0057] The rest is the same as the first embodiment.
[0058] The above is an exemplary description of the present invention in combination with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as such non-substantial improvements are made by adopting the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.
Claims
1. A method for simulating snow depth in Xinjiang, characterized by: The steps include: S1. Obtain relevant variables in the atmospheric reanalysis data set of Xinjiang; S2, preprocessing the atmospheric reanalysis data set in step S1, and using the relevant variable data provided by the Xinjiang Meteorological Bureau to perform bias correction on the precipitation data in the atmospheric reanalysis data set; S3, defining the area simulated by the Noah-MP model according to the latitude and longitude of Xinjiang, obtaining static geographic data containing information such as topography, land use, and vegetation cover, and then compiling and running the atmospheric reanalysis data set prepared in step S2 and the static geographic data to generate the driving field data required for model calculation; S4. Observe the new snow density after each snowfall event during the three different periods of snow accumulation, stability and melting to determine the real new snow density in Xinjiang and adjust the new snow density parameters in the model accordingly. S5, substituting the driving field data in step S3 into the Noah-MP model with the new snow density parameter modified in step S4, and specifying a time step that is consistent with the driving data, and performing an integral calculation to obtain the snow depth in Xinjiang; S6, evaluating the snow depth obtained in step S5, using the snow depth data provided by the Xinjiang Meteorological Bureau as the true value to verify the model accuracy; S7. Plot the snow depth obtained in step S5 to obtain spatial distribution data of the snow depth.
2. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: In step S1, the atmospheric reanalysis data dataset is the ERA5-land dataset, with a spatial resolution of 0.1°×0.1° and a temporal resolution of 1 hour.
3. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: In step S1, the relevant variables include temperature, pressure, wind, humidity, radiation, and precipitation.
4. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: Step S2 includes: S2-1. Resample the precipitation data in the atmospheric reanalysis data set to the longitude and latitude of the meteorological station, and set a precipitation frequency threshold P for the daily precipitation data in the atmospheric reanalysis data set based on the monthly scale. ERA5_WDT , so that the precipitation frequency of the atmospheric reanalysis data set exceeding the threshold is equal to the precipitation frequency of meteorological observations; S2-2. Select a research reference period, calculate the monthly proportional coefficient S of the reference period, and perform bias correction on the daily precipitation data of the atmospheric reanalysis data set: Among them, μ m is the average monthly precipitation, P obs and P ERA5 are the daily precipitation data of meteorological observation and atmospheric reanalysis data set, respectively. obs_WDT is the precipitation frequency threshold when the daily precipitation is ≥1 mm, P ERA5_WDT It is the precipitation frequency threshold of the atmospheric reanalysis data set; Among them, P ERA5 (d) and P ERA5_correct (d) are the daily precipitation data of the atmospheric reanalysis data set before and after correction. The precipitation of the atmospheric reanalysis data set after correction has the same precipitation frequency and intensity as the meteorological observation precipitation data.
5. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: In step S3, the area simulated by the Noah-MP model is projected as Lambert, with a spatial resolution of 9×9 km and a temporal resolution of 1 hour.
6. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: In step S4, the density of new snow is determined using a Snowfork snow characteristics analyzer and manual sampling and weighing observations. The two observation results assist each other in recording, calibration and verification.
7. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: In step S4, the density of new snow in the arid region of Xinjiang is 130-160 kg / m 3 .
8. The method for simulating snow depth in Xinjiang according to claim 1, characterized in that: In step S6, correlation coefficient and root mean square error are used as evaluation methods.
9. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the snow depth process simulation method described in any one of claims 1 to 8.
10. A computer device comprising: A memory for storing instructions; A processor is used to execute the instructions so that the device performs operations to implement the snow depth process simulation method described in any one of claims 1-8.
Citation Information
Cited By
Poisson process-based day-by-day ground snow load sequence stochastic simulation method and system
CN119647065A
