JP3219383B2 - Automatic snow depth measurement system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic snowfall measurement system used for snow removal work on roads and the like and operation control of a snow removal device and a snow melting device.

[0002]

2. Description of the Related Art The vertical depth and snow depth (also referred to as new snow depth) of snow that has accumulated in a certain period of time can be used for snow removal work on roads, operation control of snow removal and snow melting equipment, and disaster prevention measures such as avalanches. Is essential information, and the development of a system to measure and collect these in real time is desired.

[0003]

However, an apparatus for automatically measuring the snowfall depth has not been put into practical use at present, and the snowfall depth is usually set by placing a white board called a snowboard on the snow surface and stacking on it. Measured by measuring snow depth. Accordingly, it is an object of the present invention to provide a system for automatically measuring the snowfall depth.

[0004] The automatic snow depth measurement system of the present invention that solves the above-mentioned problems of the prior art assumes that snow is a compressed viscous fluid, and its viscosity coefficient η is a power function of its dry density ρ, η =
Cρ ^a , C, and a are expressed as constants, and the amount of compression (−Δ) generated in each snow layer having a thickness (h) formed within a unit time Δt.
h) and the subsidence amount of the uppermost snow layer, which is the sum of the compression amounts, is calculated from the load of the upper snow layer by the equation σ = η (−Δh / (h ·
Δt)); and an automatic snowfall measurement system comprising: means for measuring a height change (snow depth difference) of the uppermost layer of the snow layer per unit time and a snowfall water amount per unit time. A value obtained by subtracting the calculated settling amount from the actually measured snow depth difference is set as an index D. When the index D is negative, the value is multiplied by the density estimated from the compression amount (-Δh) to melt snow. It is provided with a calculation / display means for calculating the water content, calculating the water content of each snow layer by distributing the calculated amount of snowmelt to each snow layer, and displaying the calculated water content of each layer in a graph.

[0005]

According to the embodiment of the present invention, by repeating the above calculation while changing the above constants C and a, the optimum values of the constants C and a suitable for the actual measurement results can be obtained. It is configured.

[0006]

First, the principle of the present invention will be described. Since the snow on the ground is compressed by its own weight and the load of the snow on it, each snow particle constituting the snow keeps displacing downward (settling down) at all times. For this reason, the depth of the snowfall accumulated from a certain time t _n-1 to the current time t _n after an appropriate unit time Δt has elapsed, and the snow depth during that time (the vertical depth of the snow accumulated on the ground) The increments do not match. Snowfall depth therebetween, the snow depth of the current t _n, but is given as the difference between the ground clearance of the current t _n snow surface at time t _n-1, the ground current t _n snow surface at time t _n-1 The height is not the same as the snow depth at time t _n-1 , but is lower by the amount of snowfall. Accordingly, the snowfall depth is given as the _sum of the difference in the snow depth between the snowfall and the snowfall at time tn _-1 .

Kojima (1957, 1958, 1967) examined in detail the time-dependent changes in the density of each snow layer constituting snow, and found that an increase in the density and a decrease in thickness of each layer occurred due to the viscous compression of the snow. And the compressive viscosity coefficient η of dry snow with a density of 0.07 to 0.50 g · cm ^-3 is η = exponential function of density ρ
It was shown that it can be expressed by ηo · exp (kρ). Then, it was shown that the vertical distribution of the density and the temporal change of the snow depth can be estimated by the temporal change of the amount of snowfall water using this equation. Motoyama and Kojima (1985) also estimate changes in snow depth using the same method. On the other hand, Endo et al. (1990) reported that
For fresh snow and dry snow of 0.30 g · cm ^-3 , it was shown that the compression viscosity coefficient η can be expressed as ρ = C · ρ ^a as a power function of the density ρ. It is shown that the use of this formula makes it easy to calculate the snow depth and density, and it is also used to predict the occurrence of surface avalanches (Endo, 1993).

As described above, studies on the viscous compression of snow cover have been mainly made on dry snow. The viscosity coefficient of wet snow is generally smaller than that of dry snow when the density of snow containing water is the same (Kinoshita, 1963). However, for wet snow with a water content of 5% or less, the viscosity coefficient and dry density (water relationship between the density of snow parts of ice except) is approximately the same as the relationship to dry interference snow 0 ^o C (Kojima, 1967) have been obtained like that. However, it is difficult to measure the viscosity of wet snow due to snow melting and rapid changes in snow quality, and there are no systematic studies and the details are not clear.

Therefore, here, for convenience of calculation, it is assumed that the viscosity coefficient η (gf · hr / cm ² ) of wet snow is proportional to the power of the dry density ρdry (g · cm ⁻³ ) of the snow. , Η = C (ρdry) ^a (1), and consider the change in thickness of each snow layer due to viscous compression. Here, C and a are constants.

Here, as shown in FIG. 3 and FIG. 4, a layer of snow t _i accumulated from time t _n-1 to t _n inside the snow cover.
Considering the layer (i = 1, 2,... N), the time t (>
In t _i ), the thickness of the t _i layer is h _i (t) (cm), and the weight is w
_{Assuming that i} (t) (g / cm ² ) and the weight (water content) of water contained therein is q _i (t) (g / cm ² ), the weight water content α _i (t) of the t _i layer is represented by _{α i (t) = q i} (t) / w i (t) (2), dry density [ρ _i (t)] dry is [ρ _i (t)] dry = [1- given by α _i (t)] _{w i (t) / h i} (t) (3). The wetting density of the ti layer is ρi (t) = wi (t)
/ hi (t).

The snow pressure at time t acting on the ti layer at time t is defined as σi (t) (gf / cm ² ). With this pressure, ti is defined as a small time from time t to time t + dt. Layer thickness hi
It is assumed that (t) is reduced by dhi (t). In the case of viscous compression, the following relationship holds between them. σi (t) = η (t) [-dhi (t) / [hi (t) · dt]] (4) By substituting equations (1) and (3) into equation (4), the following equation is obtained. σi (t) · dt = [-c [｛1-αi (t)｝ wi (t)] ^a / ｛hi
(t)｝ ^{a + 1} ] ・ dhi (t)

In the above equation, if the weight ｛1-αi (t)｝ wi (t) of the ice portion of the ti layer of interest does not change with time, integration is possible, and both sides can be measured from time tn-1 to time tn. When integrated over the interval, the thickness h'i (tn) (cm) at time tn of the ti layer by viscous compression
Is given by the following equation. h'i (tn) = hi (tn-1) * [1+ (a / c) hi (tn-1) ^a Qi (tn-1, tn) / [｛1-αi (tn-1)｝ wi (tn-1)] ^a ] ^{-1 / a} (5)

Here, the thickness of the ti layer at the time tn due to viscous compression is defined as h'i (tn) in order to distinguish it from the change in thickness due to melting. Hereafter, the thickness in the ti layer by viscous compression is denoted by h'i (tn), and ti as the result of viscous compression and melting
Let the thickness of the layer be represented by hi (tn). Also, hi (tn-1),
wi (tn-1) and αi (tn-1) are the thickness of the ti layer at time tn-1;
Weight and moisture content. Qi (tn-1, tn) (gfhr / cm ² )
Is the integrated value of the pressure applied to the ti layer from time tn-1 to the current time tn, and is given by the following equation when the time interval Δt (= tn-tn-1) is short. Qi (tn-1, tn) = _tn-1 ∫ ^tn σi (s) ・ ds = (1/2) ｛σi (tn-1) + σi (tn)｝ Δ
t However, the symbol _tn-1 ∫ ^tn means integration in the section from tn-1 to tn.

Here, as shown in FIG. 3, σi (tn-1) and σi (tn) are time tn-1 and time tn acting on the ti layer, respectively.
Pressure. As can be seen from FIG.
If the weight wi (tn-1) of each layer on the ti layer at the time tn-1 does not change until the time tn, the following equation is obtained. Qi (tn-1, tn) = ｛wi (tn-1) / 2 + _{i +} 1Σn ^-1 wi (tn-1) + p (tn) / 2｝ Δt (6) where wi (tn-1 ) / 2 is added in consideration of the fact that the contribution of the weight of the ti layer to the compression of the ti layer is 1/2. p (tn) (g / cm ² ) is the precipitation from time tn-1 to tn, and is equal to the weight wn (tn) of the newly accumulated tn layer.

Equation (6) does not hold when the weight of each snow layer changes due to snowmelt or snowmelt water / rainwater permeation, but can be used as an approximate value if the time interval Δt is appropriately shortened. . Therefore, here, between time tn-1 and tn, no snowmelt or snowmelt water / rainwater infiltration occurs, and (5), (6)
After the viscous compression given by the equation has occurred, it is assumed that snowmelt and snowmelt water / rainwater infiltration will occur in the following manner. Then, if the thickness, weight, moisture content, and precipitation p (tn) of each snow layer in the previous time step are known, the snow layer of each snow layer by viscous compression can be obtained from Eqs. (5) and (6). thickness
h'i (tn) will be given.

Index D (tn) and handling of snowmelt water Here, FIG.
As shown in the figure, the snow depth at the current time tn is H (tn), the thickness of each snow layer at time tn by viscous compression is h'i (tn), and the snow bottom is between time tn-1 and tn. G (tn)
And an index D (tn) (cm) in the following equation is considered. D (tn) = H (tn) + g (tn) − _{i = 1} Σ ^n-1 h'i (tn) g (tn) is usually 0.1 cm / day or less. (tn) ≒ H (tn) − _{i = 1} Σ ^n-1 h'i (tn) (7) = ｛H (tn) -H (tn-1)｝ + ｛H (tn-1) − _{i = 1} Σ ^n-1 h'i (tn)｝ (7 ').

As can be seen from equations (7) and (7 '), the index D (t
n) is the time tn predicted from viscous compression from the next snow depth
-1 minus the height of the snow surface at time tn (Equation (7))
Or the sum of the snow depth difference between time tn-1 and time tn plus the amount of sedimentation predicted at time tn of the snow surface at time tn-1 (formula (7 ')).

Here, considering that there is only snowfall without snow melting on the snow surface from time tn-1 to time tn, the index D (tn) is positive, and the value of the index D (tn) is newly accumulated. Layer thickness (
The snowfall during that time) is hn (tn). Conversely, when only snowmelt occurs without snowfall, the index D (tn) is negative, and its value is the depth of snow melted from the snow surface (hereinafter referred to as snowmelt depth for simplicity).
Will be shown. If the time interval Δt is large, both snow melting and snowfall may occur during that time.
Is small, the probability is small, and the index D (tn) is an index indicating the exact snowfall or snowmelt depth during this time.

Therefore, in this specification, Δt is set as short as one hour, and the thickness of each snow layer at time tn is calculated by the following method.
The weight and water content were determined. 1) The thickness h'i (tn) of each snow layer by viscous compression is calculated from the equations (5) and (6), and the index D (tn) is obtained. 2) When the index D (tn) is positive, the value is calculated as the new snow layer (tn).
When D (tn) is negative in addition to the thickness hn (tn) of the layer on the snow surface, the snow layer corresponding to that value is removed from the snow surface. 3) When the index D (tn) is positive, it is considered that snowfall has occurred as the snowfall p (tn), and p (tn) is set as the weight wn (tn) of the tn layer. Index D (t
If n) is negative, treat it as rain and infiltrate the lower layer with snowmelt water melted on the snow surface. 4) It is considered that snowmelt due to infiltration of rainwater and snowmelt does not occur. Infiltration into the lower layer of seepage water adapts a simple tank model, given the maximum moisture content αma with the moisture content of the snow layer
If it moves only when x is exceeded and all layers are saturated at the maximum water content αmax, supersaturated water flows from the bottom layer into the ground. When handled in this manner, the thickness, weight, and moisture content of each snow layer are given as follows.

When D (tn) is positive, the thickness, weight, and moisture content of each snow layer When D (tn) ≧ 0, the snow layers are piled up from time tn-1 to time tn.
The thickness hn (tn) of the tn layer is given by D (tn), and the precipitation p
(tn) is the weight wn (tn) of the tn layer. Since the water content of snowfall is unknown, assuming this to be 0, the thickness, weight, and water content of the newly deposited tn layer are given by the following equations, respectively. hn (tn) = D (tn) wn (tn) = p (tn) αn (tn) = 0

When the index D (tn) is positive, it is considered that only snowfall does not cause snow melting, so the thickness of each layer other than the tn layer is equal to the thickness h '(tn) by viscous compression. The weight and moisture content are equal to the value at time tn-1. Therefore, the values of the other layers are as follows. hn (tn) = h'n (tn) (i = n-1, n-2,…, 1) wi (tn) = wi (tn-1) (i = n-1, n-2,…, 1) αi (tn) = αi (tn-1) (i = n-1, n-2,…, 1)

When D (tn) is negative, the thickness, weight, and moisture content of each snow layer When index D (tn) <0, there is no snowfall and (hn (tn) = 0).
It is determined that the snow layer from the snow surface to the depth -D (tn) has melted. The layer lost by snowmelt is _i ( _k Σ ^n-1 h'i (tn) ≤ D (tn) < _{i = k-1} Σ ^n-1 h'i (tn) (8) ), From the tn-1 layer to the tk layer are completely melted, and tk-1 is partially melted.

Therefore, the thickness of each snow layer is given by the following equation. hi (tn) = 0 (i = n-1, n-2, ..., k) hk-1 (tn) = _{i = k-1} Σ ^n-1 h'i (tn) + Dn-1 (tn) hi (tn) = h'i (tn) (i = k-2,…, 1) Weight of snow (including water) melting on the snow surface m (tn) (g / cm ² )
Is m (tn) = _{i = k} Σ ^n-1 wi (tn-1) + [1- hk-1 (tn) / h'k-1 (tn)] wk-1 (tn-1) (9) And water of m (tn) and precipitation p (tn) penetrates into the lower layer according to the tank model described above.

As a result, assuming that the water content from the tk-1 layer to the tj layer (j <k) reaches the maximum water content αmax, the weight of each snow layer becomes wi (tn) = 0 (i = n− 1, n-2,…, k) wk-1 (tn) = [hk-1 (tn) / h'k-1 (tn)] [[1-αk-1 (tn-1)] / [1 -αmax]] * wk-1 (tn-1) wi (tn) = [[1-αi (tn-1)] / [1-αmax]] wi (tn-1) (i = k-2…, j) wj-1 (tn) = wj-1 (tn-1) + p (tn) + m (tn)-[hk-1 (tn) / h'k-1 (tn)] [(αmax-αk -1 (tn-1)] / [1-αmax]] wk-1 (tn-1) _{-i = j} Σ ^K-2 [[αmax-αi (tn-1)] / [1-αmax]] wi (tn-1) wi (tn) = wi (tn-1) (i = j-2,…, 1)

The moisture content of each snow layer is αi (tn) = 0 (i = n, n−1..., K) αi (tn) = αmax (i = k−1..., J) αj−1 (tn ) = 1- ｛wj-1 (tn-1) -qj-1 (tn-1) / wj-1 (tn)｝ αi (tn) = αi (tn-1) (ij-2,…, 1) Becomes

Here, the tj layer having the water content αmax is given by j satisfying the following equation. [hk-1 (tn) / h'k-1 (tn)] [[αmax-αk-1 (tn-1)] / [1-maxt]] wk-1 (tn-1) + _{i = j} Σ ^k-2 [αmax-αi (tn-1)] / [1-αmax] wi (tn-1) <m (tn) + p (tn) <[hk-1 (tn) / h'k-1 ( tn)] [[αmax-αk-1 (tn-1)] / [1-αmax] * wk-1 (tn-1) + _{i = j-1} Σ ^k-2 [(αmax-αi (tn-1 )] / [1-αmax]] wi (tn-1) (11)

As can be seen from the above, the thickness, weight, and water content of each layer at time tn are determined by the immediately preceding time step tn-1.
If there are these values of and the measured values of snow depth and precipitation at time tn, they can be obtained. Therefore, if the calculation is performed sequentially from time t1 and the result is used for the calculation of the next time step, the thickness, weight, and water content of each layer of snow can be estimated from the snow depth and precipitation at each time. Since in this embodiment taking Δt to 1 hour, the time of snowfall New time tn is given by hn (tn), day snowfall depth up to time tn, day snowfall depth ^{_{= i = n-23 Σ n}} hi (tn) Can be FIG. 5 shows a flowchart of the arithmetic processing.

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In order to calculate values such as daily snow depth by the above method, an automatic snow depth measuring system having the structure shown in FIG. 1 was prepared. 1 is an ultrasonic transducer that constitutes a snow depth gauge, 2 is a rain gauge sensing unit that constitutes a snow depth gauge, 3 is a snow depth gauge conversion unit,
Reference numeral 4 denotes a rain gauge converter, 5 and 6 denote arithmetic units and modems installed in the observation station building, and 7 denotes a telephone line.

The ultrasonic transducer 1 transmits an ultrasonic wave to a snow surface below, receives a reflected wave from the snow surface, measures the distance from the elapsed time of transmission / reception to the snow surface, and transmits the measurement result to a cable. And to the arithmetic unit 5 via the snow depth gauge conversion unit 3. The rain / snow gauge unit 2 measures the weight of rain or snow melted by the heater to measure the amount of precipitation per unit time and notifies the arithmetic unit 5 through the cable and the converter 4. The arithmetic unit 5 installed in the observation station fetches the automatically measured values of snow depth and precipitation every hour, and calculates the thickness, weight, water content, etc. of each snow layer divided in hourly units. Calculate and measure the daily snow depth from 9:00 each day at 9:00 in the morning in real time, and transmit these measured values to a remote recording device through a telephone line.

As a snow depth gauge, Kaijo Co., Ltd.'s ultrasonic snow depth meter SL-143 (1 cm sensitivity) or SL-340 (1 m sensitivity)
m), and a flood-type rain gauge (sensitivity 0.5m)
m) was used. As described in the next section, the parameters used in the calculation were determined so that the error from the actual measured snow depth was minimized. Until then, provisional values were used.

FIG. 2 is an example of a display screen showing each snowfall layer, its thickness, and the magnitude of its water content, by each snowfall layer, its thickness, and the concentration attached thereto. According to this display screen, the state of the distribution of the moisture content in the snow structure becomes clear at a glance, and by combining this with observation results such as temperature and wind speed,
It is possible to determine the possibility of an avalanche. Also, instead of the water content, the weight of each snow layer can be displayed on the screen by a concentration corresponding to the weight.

The experiment for measuring the snow depth by the automatic snow depth measurement system was performed in a snow observation observatory at the Tokamachi test site of the Forestry and Forest Products Research Institute (Tokamachi City, Niigata Prefecture). Experiment period is 1992-9
In three winters, from winter 3 to winter 94-95, operation was almost normal. In addition, to evaluate the estimated value of the snowfall obtained in this way, a snowboard was prepared on the snow surface of the dew field of the test site, and the depth of the snowfall piled on it from 9 am to 9 am the following morning Was measured every morning, and this was used as an actually measured value.

Regarding the weather conditions at the Tokamachi test site during the three winter periods when the observations were made, the average monthly temperature was 3.2 ° C in December and 0.5 in January.
℃, February 0.9 ℃, March 2.8 ℃, the monthly average temperature was positive even in severe winter. The average wind speed is moderate, about 1.1m. The maximum snow depth in each winter is 172 cm in winter 1992-93, 93-94
Winter 163 cm, winter 94-95 226 cm.

The values of C and a in equation (1) used in the present method and the value of the maximum water content αmax are not clear. Therefore, C = 1.11 * 10 ⁶ gf ・ hr ・ cm ^-2・ (g ・ cm ^-3 ) ^-a αmax = 0.15, and the daily snow depth when the value of a was changed variously was calculated and measured. Value.

Table 1 shows the standard errors of the respective estimated values in the three winter seasons.

[Table 1] δ = [｛Σ (estimated value−implemented value) ² ｝ / number of measurements] ^1/2 . As can be seen from Table 1, when a = 3.6, the standard error is the smallest, and its value is δ = 1.71 cm. Therefore, FIG. 6 plots the estimated values of daily snowfall for three winter seasons when a = 3.6 with respect to the actually measured values.

As can be seen from FIG. 6, the estimated value and the measured value are in good agreement, and the maximum error is about ± 5 cm.
Therefore, in this method, it was considered that the use of a = 3.6 was the most accurate and appropriate. As an example, FIG. 7 shows the actual values of the daily snow depth for each day of January in the three winter seasons and the estimated values when a = 3.6.

[0037]

As a guide to know the daily snowfall depth, a daily snow depth and a snow depth difference daily meter are known. The daily snow depth difference is the snow depth at a certain time minus the snow depth at the same time one day before, and the snow depth difference daily meter is the positive value of the hourly difference in snow depth measured at every hour. Is the sum of Therefore, in order to verify the effect of the present invention, when the daily snow depth difference is set as the daily snow depth (this is called the daily snow depth difference method), when the daily snow depth difference is set as the daily snow depth ( The error of the depth difference day meter method was examined and compared with the error of the viscous compression model method of the present invention. The snow depth data necessary for calculating the daily snow depth difference and the snow depth difference daily meter, and the daily snow depth data (conducted values) that are compared with these data are the same as those used in the viscous compression model method. Fig. 8 shows the relationship between daily snow depth and daily snowfall, Fig. 9
Shows the relationship between the daily snow depth difference and the daily snowfall.

According to FIG. 8, when the daily snow depth difference is the daily snow depth, the error is generally negative, and the maximum value is -23 cm.
The standard error δ was 9.72 cm. Such a large negative error is caused by the settling of snow and the melting of snow, and it is considered that the cause of this error is corrected by the viscous compression model method shown here.

In order to correct a large negative error caused by the above-mentioned daily snow depth difference method, the snow depth difference daily measurement method regards a decrease in the snow depth as being caused by the subsidence of snow and the melting of snow, and performs the snow accumulation every hour. This is interpreted as removing the negative value of the depth difference day and adding only the positive value. For this correction, as shown in FIG. 9, the relationship between the snow depth difference daily meter and the daily snowfall is distributed almost evenly around the solid line indicating the one-to-one correspondence, but the maximum value of the error is ± Fifteen
cm and the standard error was 4.52 cm.

According to the automatic snowfall measurement system of the present invention,
The maximum error of the viscous compression model method is ± 5 cm, and the standard error is 1.
This is 71 cm, which indicates that the accuracy of the viscous compression model method employed by the automatic snowfall measurement system of the present invention is significantly higher than the daily snowfall depth method and the snow depth difference daily measurement method.

In order to improve the accuracy of estimating the daily snow depth, several estimation methods based on multiple regression analysis have been proposed in addition to the daily snow depth method and the snow depth difference daily measurement method. Yoshida et al. (1991)
Is a multiple regression formula that divides the average temperature of the day and the previous day into four categories depending on whether it is plus or minus, and calculates the daily snow depth as the objective variable, the snow depth difference between the day and the previous day, and the precipitation on the day as the explanatory variables. Is proposed as an equation for estimating daily snowfall. Then, from this estimation formula, the winter of 1983/84 to the winter of 87/88 at Tokamachi
5 The standard error obtained by substituting weather data for winter is 4.29.
In cm, the standard error calculated by using the same data for the snow depth difference daily method was 7.29 cm.

On the other hand, as described above, 1992-1993
The standard error obtained by the viscous compression model method obtained from the data for the three winter seasons from the winter of 1994 to the winter of 1994-95 was 1.71 cm, and the standard error by the snow depth difference date method using the same data was 4.52 cm. Accurate comparison is not possible due to the different data used, but based on the error of the snow depth difference date meter, the error of Yoshida's estimation formula is 59% by the snow depth difference date measurement method, The error of the viscous compression model method is 38% of the error of the snow depth difference date measurement method, and it is considered that the viscous compression model method has higher accuracy.

The relational expression between the compression viscosity coefficient and the dry density
The coefficient C and the value of a in (1) are, as described earlier, C = 1.11 * 1
0 ⁶ gf · hr · cm− ² · (g · cm ⁻³ ) ^-a , a = 3.6 was considered to be the most appropriate. This result was obtained for snow covered in Tokamachi, which is mostly composed of wet snow even in the middle of winter. Equation (1) is assumed to be applicable to both wet and dry snow. When C and a are set to 0
It is more preferable that the index D be determined to be 0 when the temperature is 0 ° C. or less.

Therefore, for comparison, the relational expression between the compression viscosity coefficient and the dry density used in the present invention and the conventional measurement results obtained for dry fresh snow and frozen snow below the freezing point (Kojima, 19
67; Shinojima, 1967; Nakamura, 1988; Endo et al., 1990;
Kajikawa and Ono, 1990) are plotted on the log-log graph in Fig. 10. The solid line in FIG. 10 is the result of the former, and the broken line is the result of the latter. Figure
As can be seen from FIG. 10, as a whole, the solid line, which is the theoretical value, and the broken line, which is the experimental data, show relatively good agreement. However, when viewed in detail, the viscosity coefficient of the theoretical value, [rho <a larger than the actual viscosity in the region of 0.3 g / cm ^3, also, ρ> 0.3g / cm ³ are in is evaluated to be smaller.
From this, it can be seen that by changing the values of the coefficients C and a according to the density range of the snow layer, a more accurate snowfall depth can be calculated.

Further, according to the present invention, it is possible to estimate the amount of snowmelt on the snow surface from the data of the snow depth and the amount of snowfall, and it is possible not only to estimate and display the settling curve of each snow layer, but also to estimate the density inside the snow layer. Also, the vertical distribution of the water content and the like can be displayed on the screen by shading.

References Y.ichi Endo, Y.Ozeki, S.Niwano, 1990: Relationship between compression viscosity coefficient and density of low-density snow. Snow and ice, 52, 267-274. Y.ichi Endo, 1993: Dry snow surface due to snowfall intensity. Prediction of avalanche occurrence. Snow and ice, 55, 113-120. Masahiro Kajikawa and Noboru Ono, 1990: Relationship between compression viscosity coefficient of new snow cover and crystal form of snowfall snow and ice, 52, 283-287. Kojima, K., 1957: Viscous compression of snow layer III. Low temperature science, physics, 16, 167-196. Kojima, 1958: Viscous compression of snow layer IV. Low temperature science, physics, 17, 53-64. Kojima, k. , 1967: Densification of seasonal snow cov
er.physics of snow andice, ed.H.Oura, The Institute
of Low Temperature Science, Hokkaido Univ., 929-95
2. Motoyama, H. and Kojima, K., 1985: Estimation of snow depth change (in case of dry snow). Low temperature science, Physics. 44, 15-25. Nakamura H. 1988: Studies on the settlement force of
snow as a generation mechanism.Report of the Nati
onal Research Center for Disaster Prevention, 41,36
1-385. Forestry Agency, 1988: Investigation report on comprehensive countermeasures for large-scale surface avalanches. Report on Adjustment Costs for the Comprehensive Land Development Project in 1988, 55-60. Shinojima, k.
mation of depositedsnow.physics of snow and ice, e
dHOura, The Institute of Low Temterature Science,
Hokkaido Univ., 875-907. Yamada, J., 1993: Estimation of New Snow Depth Based on Temporal Meteorological Data. I Rainfall and Snowfall Meter Method and Multiple Regression Analysis. Y. Yoshida, S. Takami and Y. Endo, 1991: A method for estimating snowfall depth from AMeDAS observations. Proceedings of the 19th National Convention of the Ice Society of Japan, 135.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing a configuration of an automatic snowfall measurement system according to an embodiment of the present invention.

FIG. 2 is an example of a snow layer internal monitor display screen showing the thickness of each snow layer and the magnitude of the water content by the thickness of each snow layer and the concentration applied thereto.

FIG. 3 is a conceptual diagram for explaining the principle of the present invention.

FIG. 4 is a conceptual diagram for explaining the principle of the present invention.

FIG. 5 is a flowchart illustrating a procedure of a calculation process according to the principle of the present invention.

FIG. 6 is experimental data showing a relationship between a calculated value of snow depth and an observed value.

FIG. 7 is experimental data showing a relationship between a daily change in a calculated value of snow depth and an observed value.

FIG. 8 is experimental data showing the relationship between the daily snow depth difference and the daily snowfall depth.

FIG. 9 is experimental data showing a relationship between a daily snow depth difference daily meter and a daily snowfall depth.

FIG. 10 is experimental data showing a relationship between a compression viscosity coefficient and a dry density.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Ultrasonic transducer which comprises a snow depth gauge 2 Sensing part which constitutes a snowfall gauge 3 Snow depth gauge conversion part 4 Rainfall gauge conversion part 5 Operation part 6 Modem 7 Telephone line

──────────────────────────────────────────────────続 き Continuing on the front page (72) Shoji Niwano 1 Matsunosato, Kusazaki-cho, Inashiki-gun, Ibaraki Pref. Forestry Agency, Forestry Research Institute (72) Inventor Shuichi Shioda Kaiyo in 5-5, Sakaemachi, Hamura-shi, Tokyo (72) Inventor Tomoji Ozawa 3-1-1 Kaijo, Sakaemachi, Hamura-shi, Tokyo Tokyo, Japan (72) Inventor Shozo Sakai Kaijo, 5-share Company 5-1-1, Sakaemachi, Hamura-shi, Tokyo (56) Reference Document JP-A-5-2084 (JP, A) JP-A-7-113877 (JP, A) JP-A-9-143947 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01W 1/14 G01B 21/18 G01F 23/28

Claims (7)

(57) [Claims] Assuming that snow is a compressed viscous fluid, its viscosity coefficient η is a power function of its dry density ρ, η = Cρ ^a , C, a
Is expressed as a constant, and the amount of compression (−Δh) generated in each snow layer having the thickness (h) formed within the unit time Δt, and the amount of settling of the uppermost snow layer which is the sum of the amounts of compression. From the load of the upper snow layer according to the formula σ = η (−Δh / (h · Δt)); the height change (snow depth difference) and the unit of the uppermost layer of the snow layer in a unit time. And a means for measuring the amount of snowfall in time.
A value obtained by subtracting the calculated settling amount from the actually measured snow depth difference is set as an index D. When the index D is negative, the value is multiplied by the density estimated from the compression amount (−Δh) to obtain a snowmelt amount. Calculation and display means for calculating the water content of each snow layer by distributing the calculated amount of snowmelt to each snow layer, and displaying the calculated water content of each snow layer on a screen. An automatic snow depth measurement system. 2. The method according to claim 1, wherein the screen display of the water content by the calculating / displaying means is performed based on the density of the color applied to each snow layer stacked with the calculated thickness. The feature is an automatic snow depth measurement system. 3. The automatic snowfall depth measuring system according to claim 1, wherein said calculating / displaying means further comprises means for displaying the weight and dry density of each snow layer on a screen. 4. The method according to claim 1, wherein the calculation is repeated while changing the constants C and a to obtain the optimum values of the constants C and a that are suitable for the actual measurement result. Automatic snow depth measurement system. 5. The method according to claim 4, wherein the optimum values of the constants C and a suitable for the actual measurement result are such that the index D becomes 0 when precipitation is 0 and temperature does not fall below 0 ° C. An automatic snowfall depth measurement system characterized by being obtained as follows. 6. The method according to claim 1, wherein the constant C is 1.
11 * 10 ⁶ gf · hr · cm ⁻² · (g · cm ⁻³ ) ^-a and the constant a is set to 3.6. 7. The method according to claim 1, wherein the constants C and a
Is an automatic snow depth measurement system, wherein different values are set according to the dry density and water content of the snow layer.