JP2006201024A - Distribution type of evaluation method for snowice as water resource - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a distribution type of an evaluation method for snowice as a water resource such as predicting future variation by grasping a variation in accurate snowice water resource quantity in objective region and variation in outflow rate of the objective region, for example, by the eventual output, outflow rate for snowfall and snow melting in order to construct a system by verifying and correcting actual outflow and calculation outflow using a distribution type of an outflow model. <P>SOLUTION: The objective region is partitioned in a grid. From snowfall in each grid by meteorological data and each geographical conditions and the outflow of each grid including snow melting by a phenomenological model based on snowfall quantity, snow melting, soil penetration and evaporation, the outflow rate of the whole objective region is calculated from a pseudo river channel network by actual river data and geological data. By verifying and comparing the outflow and the actual outflow and correcting parameters for correcting the snowfall quantity in each grid and calculating the snowfall quantity, snow melting or outflow, accurate snowice water resources in the objective region is grasped with the distribution type of the evaluation method for the snowice water resources. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention applies, for example, a distributed snowmelt runoff model capable of quantitatively calculating the distribution of snow cover in mountainous areas to target basins, and evaluates and predicts fluctuations in snow ice water resources and runoff with high accuracy. is there.

  Dams have the role of water use, water hydrophilicity, and flood control, and their management has a great impact on the basin society and natural environment.

  That is, precipitation as water source falls as snow or rain, and the overall trend such as seasonal variation is generally grasped. However, there are many uncertain aspects in terms of quantitative variation every year. In particular, snowfall in mountainous areas is not well understood and its measurement data is not small, so it is difficult to grasp the quantitative distribution.

  For this reason, it is difficult to make judgments when performing appropriate water resource management.

  Conventional river runoff models target runoff associated with rainfall, and runoff associated with snowfall and melting has not been reflected in the river runoff model.

  Even if snowfall, snowfall, and snowmelt are estimated by independent model calculations, it is difficult to grasp the actual amount, so it is not verified and the reliability is low.

  Since it is difficult to collect actual data, it is difficult to verify the calculated values and improve the accuracy of the calculated snowfall and the calculated snowmelt output from the snowfall model and the snowmelt model.

Appendix: "Distributed Snowmelt Runoff Model Research Achievements"

  Therefore, in the present invention, a distributed snowmelt runoff model capable of quantitatively calculating snowfall distribution in mountain areas is applied to the target basin, and fluctuations in the amount of snow ice water resources and runoff are accurately estimated and predicted. .

  In the present invention, the following output is obtained.

  (1) We propose a method to evaluate the quantity of snow and ice water resources and runoff in the dam catchment area.

(2) Estimated results of snow and ice water resources and runoff in typical meteorological patterns (water-filled years, flat-water years, drought years).

(3) Estimate the amount of snow, ice and water resources in the basin from the meteorological data up to the end of the current fiscal year, and show the runoff prediction during the snowmelt season (March to June) based on each weather pattern.

  (4) Based on long-term meteorological fluctuation forecasts, the amount of snow and ice water resources and their impact on runoff are shown.

  By utilizing the results of the present invention, it becomes possible to grasp in real time the quantitative amount of snow and ice resources in a vast dam catchment area and to predict changes in the amount of water resources and the flow rate during the snowmelt season. As a result, water resource management becomes easy, and it is considered that early measures can be taken from the prediction information.

  The gist of the present invention will be described with reference to the accompanying drawings.

  The target watershed is divided into grids, and the amount of snow on each grid is determined by meteorological data and topographical conditions, and the amount of runoff from each grid including snowmelt is determined by a physical phenomenon model based on snowfall, snowmelt, soil infiltration, and evapotranspiration. In addition, the outflow amount of the entire target basin is obtained based on the pseudo river network obtained from the actual river channel data and the topographic data, and the outflow amount of the entire target basin obtained by this analysis is compared with the actual outflow amount by verifying and comparing The snow ice water resource is characterized by grasping the snow ice water resource amount with high accuracy in this target basin by correcting the snow accumulation amount of each grid or correcting the parameter for calculating the snow water amount, the snow melting amount or the runoff amount. This relates to the distributed evaluation method.

  2. The distribution type of snow and ice water resources according to claim 1, wherein a change in snow and ice water resources with high accuracy in the target basin and a change in runoff of the target basin are grasped and future changes are predicted. It relates to the evaluation method.

  Moreover, the discharge area of this water dam is controlled and managed by grasping the amount of snow, ice, and water resources in the water basin as the target water basin. This relates to the distribution-type evaluation method for snow and ice water resources described in 1.

  Since the present invention is configured as described above, in order to construct a system that verifies and corrects the calculated runoff amount based on the distributed runoff model and the actual runoff amount, and constructs a system, the amount of snowfall and the amount of snowmelt are finally output. It is possible to estimate values that are more reliable than existing models that have not been verified, and to grasp snow and ice water resources with high accuracy in the target basin. This is an extremely excellent distribution-type evaluation method for snow and ice water resources, such as predicting future changes by grasping changes in the amount of snow and ice water resources with high accuracy in the target basin and changes in the amount of runoff from the target basin.

  Embodiments of the present invention that are considered suitable (how to carry out the invention) will be briefly described with reference to the drawings, illustrating the operation of the present invention.

  For example, based on meteorological data such as rainfall and temperature obtained by existing weather stations and other methods, snow accumulation and snow melting in a river basin and the accompanying snow melting are converted into a grid of tens to hundreds of meters in the river basin. Divide and calculate runoff for each grid.

  A numerical water channel (pseudo river network) that connects the grids is created from the elevation data and actual river channel data of each grid point in consideration of the basin boundary and undulations.

  The amount of runoff through the simulated river network is calculated using a distributed runoff model method that introduces a physical model of the snow melting process.

  The amount of snowfall and its change at an arbitrary time in an area where weather data is not available are estimated with high reliability.

Estimating snow and ice water resources
(a) Snow cover analysis in mountainous areas (Analysis of existing water resources due to snow cover)
Use this model to analyze the amount of snow, ice, and water resources in the basin quantitatively for each year of the analysis, and to quantitatively determine the amount of snow, ice, and water resources in the basin as a time series at each grid point .

(b) Water resource fluctuation analysis and environmental impact change analysis due to weather fluctuations Long-term fluctuation trends of weather are input into the runoff and snowmelt analysis model, the amount of water resources in the future is grasped, and snow and ice of long-term weather fluctuations such as global warming phenomenon Analyze the impact on water resources.

(c) Application to the management and control method of water utilization dams that use snow as the main water source Using snowmelt runoff analysis models, the amount of snow and ice water in the catchment area of the water utilization dam is quantitatively determined for each grid point, and future dam It is applied to control and manage discharge for flood and drought.

Snowmelt disaster risk prediction
(a) Snowmelt mud flow analysis The snowmelt runoff model is used to estimate the amount of snowfall in active volcanoes, and to predict the size and risk of snowmelt mudflow due to volcanic activity during the snowfall season.

(b) Snowmelt flood analysis Meteorological forecast data (precipitation, temperature, cloud cover, etc.) provided by the Japan Meteorological Agency is used as input data to predict runoff to any time ahead of snowmelt floods due to rainfall during the snowmelt season or temperature rise. .

  Specific embodiments of the present invention will be described with reference to the drawings.

  This model is a “distributed runoff model” corresponding to the terrain of the basin and the heterogeneity of the weather, unlike the “centralized runoff model” that calculates the basin information by averaging. Since the terrain and weather conditions of the basin are uneven, the basin is divided into fine grids to reproduce a more realistic situation, and each grid is calculated taking into account geography and weather conditions. is there.

  For runoff calculation, runoff is calculated for each precipitation (rain or snow) input to each grid using a physical phenomenon model (precipitation, snowmelt, soil infiltration, evapotranspiration, etc.) Follow-up calculation to the basin exit.

  This model incorporates snowfall and melting processes into each grid, and can reproduce the long-term basin flow including the snowfall and snowmelt periods.

  That is, the present invention comprises a runoff analysis model and a snowmelt runoff model. The entire system uses meteorological data such as precipitation and temperature as input values for a given river basin, divides the basin at grid points at regular intervals, calculates runoff flow and snowfall on the grid points, The water flow is traced from the grid point to the grid point via the pseudo river network, and finally the river flow at the basin exit is obtained.

  Therefore, in this system, watersheds ranging from several tens of square kilometers to several thousand square kilometers are represented by lattice points with an interval of about 50 m, and the pseudo river network connecting the lattice points to the place where the water is from high to low. It requires a technology that relies heavily on computers that are built according to the principle of flow. Next, the precipitation is classified into rain and snow from the precipitation and temperature given at each grid point, and the snow accumulated at each grid point is calculated for each grid point to melt according to the temperature and solar radiation. For the runoff generated at each grid point, the river flow at the basin exit is calculated by a calculation method that efficiently tracks a complex pseudo river network. In this series of methods, it is necessary to identify the parameters that determine the distribution of snowfall (snow accumulation), and the model that achieves this identification can reproduce the flow rate at the basin outlet with high accuracy and at each grid point. The amount of snow runoff and the amount of snow are given with high accuracy.

  Using a model that can output in this way, it can be applied to the evaluation of snow and ice water resources. In other words, snow accumulation in the mountains is considered an important water resource. In this model, the amount of snow can be obtained with high accuracy and distribution. For example, information on how much snow has accumulated in the mountains at the end of March when snow does not fall is important information for spring water resource management. Moreover, the snowmelt flow rate can be obtained with high accuracy by this model. Furthermore, it is possible to determine with high accuracy how the amount of snow, ice and water resources is affected by climate change such as global warming. In this way, this technology is considered to include technology applied to water resources management.

<Differences from conventional models>
Conventional runoff analysis models usually considered the basin as a single vessel and considered that there was an average amount of precipitation there. Although this type of model has recently started to divide the basin for each grid and calculate the runoff amount in a distributed manner, the present invention is the first idea to calculate the snowfall amount and apply it. In other words, in this model, since the amount of runoff, the amount of snowfall, etc. can be obtained with high accuracy at each grid point, there has been no method for quantitatively knowing the amount of snow accumulated in the mountains in the past. In this model, the information on the amount of snow accumulation in the mountainous area can be obtained with high accuracy for the first time from this information, and this is the first application for snow and ice water resource management.

  This will be described in more detail below.

1. Features of the “distributed snowmelt runoff model” When dealing with snow and ice water resources, it is necessary to handle the three processes of “falling snowfall process”, “snowmelt process”, and “runoff process” before obtaining the output runoff amount. There is. In the past, each process was rarely integrated, and it was rare to relate snow ice water resources and snowmelt runoff in association with the entire basin.

  This model expresses a realistic snowmelt runoff process by applying a model based on physical phenomena from past research results in each process and combining them.

(1) Non-uniformity of basin conditions is expressed by `` distribution type '' The `` distribution type '' adopted in this model is more detailed for the `` distribution type '' that considers the basin as a uniform condition. It is a technique that can be divided into a grid and capture the originally uneven weather conditions and topographic conditions.

In this distributed method, it is possible to analyze by reflecting the weather and geographical conditions with spatial bias by calculating each process in each grid.

(2) The channel structure of the entire basin is expressed by the `` pseudo river network '' The pseudo river network is based on the principle that `` water flows from a high place to a low place ''. Is expressed numerically.

The amount of outflow that is the output of each grid is tracked and calculated via this "pseudo river network"
It is output as the outflow of the entire basin at the end of the river.

(3) Accurate and reliable estimation of snowfall and snowmelt volume This model integrates the process of snowfall and snowmelt to reproduce the long-term basin flow including the snowfall and snowmelt periods in the model. As a by-product of the calculation process, it is possible to calculate snow cover and snow melt.

The calculated snow cover and snow melt are integrated with each other in a continuous manner. Therefore, by verifying the calculated runoff, which is the final output, with the actual runoff, it is highly accurate and reliable. High value.

(4) Estimate the amount of snowfall taking into account the altitude and error of observation equipment Snowfall in mountainous areas has little measurement data and the distribution status is unknown, so grasp the quantitative distribution The only way to do this is to estimate it from several observation data provided in the area.

Therefore, when comparing the basin average precipitation calculated simply from the observed precipitation and the integrated basin outflow height (basin outflow / basin area), the former is smaller than the latter. The reason is
A) Precipitation meter during snowfall captures only a certain percentage of true precipitation. B) In mountainous areas, precipitation during snowfall is altitude-dependent.
Therefore, this model does not reflect the increase in precipitation at high altitudes.In this model, the snowfall altitude correction coefficient that corrects the difference in snowfall due to altitude and the lack of snow particle capture by the rain gauge are corrected. A rain gauge correction factor has been introduced.

(5) “Snow melting” is expressed based on solar radiation and topographical conditions. Conventionally, snow melting is based on the temperature only as a parameter.
In this model, the heat balance method is used to calculate the radiation balance and sensible heat latent heat, and the effects of topography such as slope direction and gradient are taken into account.

2. Analysis flow of distributed snowmelt runoff model
(1) Snowfall (precipitation) process
(a) Allocation of precipitation data Generally, precipitation data is known only in several precipitation gauges in the basin, and these values are distributed to the grid around each precipitation gauge.

The distribution of precipitation data is based on the nearest neighbor method (
The same principle as the Thiessen method is used, and the precipitation at the weather station closest to each grid is used.

(b) Discrimination of precipitation type Precipitation given to each grid element as an input value is judged as either rain or snowfall depending on the temperature of the grid corrected for elevation. In the case of snowfall, snow is stored in each grid as snow accumulation.

(c) Correction of snowfall In the calculation of snowfall in this model, the commonly-known `` snowfall increases linearly with altitude '' is modeled, and the difference in snowfall due to altitude is corrected. Snowfall altitude correction factor ”has been introduced.

In addition, since previous studies have reported that rain gauge observation data contain errors, a rain gauge correction factor has been introduced to correct the lack of snow particle capture by rain gauges. ing.

Summarizing the above, precipitation P (h) as snow at the grid point at altitude h can be written as follows using the rain gauge observation value P (h ₀ ) at the observation point at altitude h ₀ .
P (h) = A · P (h ₀ ) · [1 + B · (h−h ₀ )

(2) Snow melting process The stored snow is melted by meteorological factors such as temperature and solar radiation, and snow melting runoff occurs. This model employs a heat balance method that takes into account radiant energy, sensible heat, and latent heat, and is expressed by an advanced and realistic model.

(3) Spill process
(a) Runoff in the grid Precipitation (rainfall, snowmelt) input in the grid becomes the runoff of the grid point through the Shin Yasue model taking into account the water retention capacity and wet state of the soil. Outflow downstream.

The Shin Yasue model takes into account the runoff process of evapotranspiration and evapotranspiration based on the superstored runoff theory, and is known to be widely applicable in wet regions such as Japan.

The theory of superstored runoff is the idea that “a runoff does not occur until the water storage capacity of the soil is exceeded for a certain precipitation”, and the excess precipitation flows out to the “basal runoff” component that penetrates underground and the soil surface. Divided into two components, “Direct Outflow” component, and outflowed.

Evapotranspiration contains a component that evaporates directly from the soil and a component that evaporates from the leaves of the tree, and takes away moisture sequentially from the upper layer divided into upper, lower, and deep layers. The reverse is true for replenishment by precipitation or melting snow.

(b) Runoff between each grid Runoff from each grid element is traced to the basin exit through the path connecting the grid points (pseudo river network).

For tracking calculation of this model,
The kinematic wave method (kinematic wave method, equivalent roughness method) is used to calculate the outflow based on the basic equation according to the physical law.

(c) Runoff path between each grid The numerical water flow path connecting each grid is called `` pseudo river network '', and the basin boundaries and undulations are taken into account from the altitude data and actual river channel data at each grid point. It is created and combined with the actual river channel to numerically represent the channel structure of the entire basin.

The Geographical Survey Information provided by the Geospatial Information Authority of Japan has a grid interval of 50 m, and this model creates a “pseudo river network” by incorporating this information.

3. About accuracy improvement of estimated snow cover Since the distributed snowmelt runoff model is a runoff model that integrates each process (falling snow, snowmelt, runoff), it verifies the runoff that is the final output of the interaction of each process. By doing so, the upper snowfall process and the snow melting process are verified.

  Therefore, it is possible to estimate the amount of snowfall and the amount of snowmelt with higher accuracy than the existing model.

  The differences from the existing model are shown below.

(1) Process flow based on existing model Since it is difficult to collect actual data on the amount of snowfall and the amount of snowmelt that are output from the snowfall and snowmelt processes, it is difficult to collect actual data. It is difficult.

(2) Process flow based on runoff model The calculated runoff output from each process is based on the runoff at the end of the basin where actual data is collected, compared to long-term values including the snowfall and snowmelt periods. Verified.

This verification will improve the accuracy of the “fall snow process”, which is a higher-level process, and will make it possible to estimate the amount of snow, ice and water resources in mountainous areas where actual observation is difficult.

4). How to proceed
(1) Collecting and organizing data
(a) Collecting and organizing existing documents and related materials in the target basin The following materials are collected and organized as related documents and materials in the target basin.

・ Documents related to water resource use in the basin ・ Documents related to snowfall in the basin and mountainous areas

(b) Collect and organize meteorological and hydrological data in the target basin Collect and organize various meteorological and hydrological data in the target basin and the surrounding area.

[Meteorological and hydrological data]
・ Temperature, precipitation, snow depth, solar radiation, wind direction and wind speed ・ Dam management data (inflow, discharge, storage)
・ Amount of water taken from the target river (power generation, tap water, industrial water, agricultural water, etc.)

[Candidate information source]
・ Ministry of Land, Infrastructure, Transport and Tourism hydrological observation data ・ Meteorological Agency AMeDAS data ・ Meteorological observation data of each dam and surrounding dam management station

If it is judged that verification of accuracy is necessary, such as when stations are not distributed in the basin on average, observation intervals, accuracy is not uniform, or there are many missing data, on-site observations may be conducted.

Verification is performed by observing the amount of snow water in a flat, open area that is less affected by wind blowoffs and puddles, and comparing it with model calculations.

(c) Examination of data extraction conditions for each meteorological pattern Extract the meteorological patterns for the wet, normal, and dry years from the collected meteorological and flow data, and apply them to the prediction and estimation of snow and ice water resources and runoff.

Here, precipitation data for the last 30 years (according to the general definition of meteorology) is collected for the water-filling year, the normal water year, and the drought year, and the frequency of occurrence is extracted by analysis. The data is 3
When it is not available in 0 years, it shall be advanced within the available range.

A) The annual amount of annual precipitation is ¼ from the top (according to the definition of annual water flow)
. (Ministry of Construction River Sabo Technical Standards (draft), explanation, survey, p.57)).

        B) A plain water year is the year closest to the average precipitation.

C) The drought year is 1 / 10th from the bottom (according to the definition of annual drought flow.
Year corresponding to the River Sabo Technical Standard (draft), the same commentary, survey, p.57)).

(2) Creating a distributed runoff model in the target basin
(a) Data set creation Create input data sets (meteorological data) and verification data sets (runoff data) for parameter identification.

Basically, hourly observation data is used as input data. If uniform data cannot be obtained due to missing measurements, etc., the analysis proceeds by techniques such as interpolation and analysis period limitations.

The types of input data include the following.
[Input weather data]: Temperature, precipitation, solar radiation, wind direction and wind speed

(b) Creation of runoff network The runoff network is based on topographic data and actual river channel data at intervals of 10 to 100 meters.
It is created taking into account the basin boundaries and undulations, and combined with the actual river channel, the channel structure of the entire basin is numerically expressed.

(3) Verification of runoff model by parameter identification Since runoff conditions vary depending on the catchment area, the “distributed snowmelt runoff model” expresses the catchment characteristics by providing several parameters. Therefore, it is necessary to set unique parameters in each basin (hereinafter this operation is called “identification”).

The main method for parameter identification is to calculate multi-year meteorological data as input data, calculate the runoff amount while changing the parameters, and obtain the value with the smallest error rate from the actual flow rate.

      The main parameters are the following three.

A) Parameters indicating the amount of water retained and wetness of the soil B) Parameters for correcting the amount of snowfall by the rain gauge C) Parameters for correcting the increase in snowfall due to altitude

Of these, parameters similar to A) can be determined using data during low or low water. The parameters of B) and C) correct the total amount of precipitation,
It is the most important parameter for the snowmelt runoff model.

(4) Snow / ice water resource analysis by runoff model
(a) Analysis of the amount of snow and ice water resources using a snowmelt runoff model
Using Dell, it can be obtained quantitatively and as distribution information.

・ In other words, the time series change at each grid point of snow / ice / water resources for each year
The amount of snow, ice and water resources in the entire basin can be obtained.

・ This allows quantitative trends in snow and ice water resources in normal and dry years, as well as
The degree of fluctuation can be obtained.

(b) Real-time water resource prediction incorporating predicted weather data ・ At the beginning of each year's snowmelt, the current basin flow rate and basin snow ice water resource amount are calculated according to this model.
Can be obtained. Using this as the initial condition, weather prediction during the snowmelt period is performed and
And predict the flow during the snowmelt period.

-Since the monthly weather forecast cannot be expected to have high accuracy,
Analyze results and evaluate the range of fluctuation.

        ・ That is, work follows the following procedures.

          A) Obtain the basin snow / ice water resources and flow rate from the runoff model.

B) Estimate the weather pattern during the snowmelt period from monthly weather forecast information. simultaneous
Taking into account the results of (a) above, the fluctuation range of the predicted weather pattern is determined.

C) By inputting the predicted weather pattern into the runoff model, the catchment runoff during the snowmelt period
Make a prediction.

        ・ The above results can be used for dam management.

(5) Understanding the impact of long-term weather fluctuations on snow, ice and water resources • By using this runoff model, the impact of long-term weather fluctuations such as global warming on snow and ice water resources can be evaluated.

・ By entering the long-term fluctuation trend of weather, the amount of water resources in the future can be easily grasped.

(6) Schematic study of snow / ice water resources and runoff analysis system Using the “distributed snowmelt runoff analysis model” for the basin once completed, a “snow melt forecasting system” useful for daily dam management can be developed.

In other words, an analysis system incorporating this distributed snowmelt runoff analysis model is incorporated into the dam management system, and analysis is performed based on various meteorological and hydrological data that are constantly input, and the current amount of snow, ice, and water resources is calculated.

Next, the weather pattern expected from the present time is obtained from these quantities and the analysis result of the weather pattern incorporated into the system. As a result, the fluctuations in snow and ice resources and runoff predicted from the present time are analyzed and output.

This analysis system can provide useful information for difficult judgments in dam management because it can predict the amount of snow and ice water resources and runoff by using daily updated weather data as input data.

      -Judgment on local water supply requests can be considered based on quantitative data.

      ・ Because the amount of water resources due to snow can be grasped quantitatively, early measures can be taken.

It is a flowchart which shows the business flow of a present Example. It is 1 in the “Distributed snowmelt runoff model research work” table listing the prior literature. It is 2 of the "distributed snowmelt runoff model research work" table listing the prior literature. It is 3 in the “Distributed snowmelt runoff model research work” table listing the prior literature.

Claims (3)

  The target watershed is divided into grids, and the amount of snow on each grid is determined by meteorological data and topographical conditions, and the amount of runoff from each grid including snowmelt is determined by a physical phenomenon model based on snowfall, snowmelt, soil infiltration, and evapotranspiration. In addition, the outflow amount of the entire target basin is obtained based on the pseudo river network obtained from the actual river channel data and the topographic data, and the outflow amount of the entire target basin obtained by this analysis is compared with the actual outflow amount by verifying and comparing The snow ice water resource is characterized by grasping the snow ice water resource amount with high accuracy in this target basin by correcting the snow accumulation amount of each grid or correcting the parameter for calculating the snow water amount, the snow melting amount or the runoff amount. Distributed evaluation method.   The distribution-type evaluation method for snow and ice water resources according to claim 1, wherein a change in snow and ice water resources with high accuracy in the target basin and a change in runoff from the target basin are grasped and future changes are predicted. . The drainage of this water dam is controlled and managed by determining the watershed of the water dam as the target basin and grasping the amount of snow and ice water in the target basin. Distribution type evaluation method of snow and ice water resources in Japan.

Images