Journal of the Japanese Society of Snow and Ice

A study on surface layer avalanche run-out limit by probable linear regression analysis

Hideki TERADA, Jiro OURA, Shigeru SUHARA

1994 Volume 56 Issue 4 Pages 315-323
Published: December 15, 1994
Released on J-STAGE: February 05, 2010

DOIhttps://doi.org/10.5331/seppyo.56.315

JOURNAL FREE ACCESS

Show abstractHide abstract

Probable linear regression analysis, based on the fuzzy logic, was applied to evaluating the run-out danger degree of surface layer avalanches. Probable linear regression analysis is basically the same as the multiple linear regression analysis, except that the regression coefficient acquired for the model are fuzzy numbers. In other words, the answers it produces have a degree of tolerance that corresponds to the impreciseness of the input data. This method is particularly effective when making a model involving a subjective judgment. A feature of this analysis is that objective models can be created using small amount of data, and the models are easy to handle.
To investigate the effect of various factors on run-out danger degree, two models (one combining nine factors and the other based on the optimization of factors) were created and tested. The results show that the main factors determining the run-out danger degree of the surface layer avalanche are the size of the avalanche and the condition of the slope (especially, the length of the slope) on which the avalanche occurs.

View full abstract

Download PDF (1963K)
Experimental study on frost heaving of water-saturated soil under tri-axial stress

Hideo YAMAMOTO, Yasushi UEDA, Hisao IZUTA

1994 Volume 56 Issue 4 Pages 325-333
Published: December 15, 1994
Released on J-STAGE: August 07, 2009

DOIhttps://doi.org/10.5331/seppyo.56.325

JOURNAL FREE ACCESS

Show abstractHide abstract

Heaving tests were conducted to study the characteristics of frost heaving in soils under tri-axial stress in an open system apparatus, in which the stress σ₁ in the direction of heat flow and stress σ₃ in the rectangular direction of heat flow were arbitrarily assigned to the specimen. Frost expansion occurred not only in the direction of heat flow but also in the rectangular direction and depended on σ₁ and σ₃. A σ₁ increased, the frost heave ratio ξ₁ in the direction of heat flow decreased and frost heave ratio ξ₃ in the rectangular direction increased. A σ₃ increased, ξ₁ increased and ξ₃ decreased. Volume frost heave ratio ξ_v decreased with the increase in σ₁ or σ₃, but the decrease tended to stop when these stresses increased above a certain value. The experimental results of ξ₁ and ξ₃ could be fitted to the experimental formulas expressed as a function of σ₁ and σ₃.The results from a different heaving test, performed by confining the specimen in a cylinder, were in satisfactory agreement with the values calculated by formulas based on ξ₃=0.

View full abstract

Download PDF (2657K)
Development of machines for hydraulic transportation of snow

Part 3: Ditch end snow mixing machine

Teruyoshi UMEMURA, Kazunori NAKAMURA, Nobuhiko AZUMA, Yoshihiro OGAWA, ...

1994 Volume 56 Issue 4 Pages 335-340
Published: December 15, 1994
Released on J-STAGE: February 05, 2010

DOIhttps://doi.org/10.5331/seppyo.56.335

JOURNAL FREE ACCESS

Show abstractHide abstract

In order to develop a combined system of open channels and pipes for transportation of snow, the optimum shape, dimension and operating condition of blades, which are to be installed at the end of an open channel to crush the snow lumps to mix with water, have been experimentally investigated. A modified grass cutter type of blades has been developed and its performance expressed in terms of the flow ratio of snow to water as R^s = 1.74×10^-4·N/ F^r for F^r of 0.2 to 0.4. Where N is the number of revolution per minute of the blades, and Fr the Froude number of the open channel. Use of this equation is shown to give the actual dimension and the number of revolution for a snow dumping open channel under practical use.

View full abstract

Download PDF (2847K)
Annual evaporation from continental type glaciers

Hiroyuki OHNO, Tetsuo OHATA

1994 Volume 56 Issue 4 Pages 341-351
Published: December 15, 1994
Released on J-STAGE: August 07, 2009

DOIhttps://doi.org/10.5331/seppyo.56.341

JOURNAL FREE ACCESS

Show abstractHide abstract

Annual evaporation, melting and internal accumulation were computed for Xiao Dongkemadi Glacier in the Tanggla Mountains, using a heat transfer model with meteorological elements measured at 5600 m a.s.l. for a period of over a year (May 14, 1992 to May 13, 1993). Estimated total evaporations in melting and non-melting season are 74±10 mm and 46±14 mm, respectively. These amount considered to occupy approximately 10% of annual ablation of the sate. This ratio for Chongce Ice Cap, West Kunlun Mountains where climate is drier was estimated to comprise more than 30%. Our analysis quantitatively confirms that the ablation reduces under drier conditions in spite of increase in the heat supply to the glacier. On the condition of ice exposure, that are often observed in some continental type glaciers in winter, evaporation increeases by 3-5 times depending on optical properties of the ice. Different meteorological elements influence evaporation for different seasons. The water vapor pressure is dominant in the melting season whereas the short-wave radiation is in the non-melting season. Daily evaporation from glacier is governed by two conditions; one is the heat supply and the other is the one for diffusion. The daily evaporation from Xiao Dongkemadi Glacier peaked twice in a year; one in spring and the other in autumn, at an amount of 1 to 1.5 mm. Glaciers in drier or in higher altitude places would show different annual pattern of the evaporation such as those without the autumn peak or with the later spring peak.

View full abstract

Download PDF (5398K)
Prediction methods of winter road conditions in mountainous area

Takashi NAKATSUJI, Toru HAGIWARA, Takashi FUJIWARA, Terutoshi KAKU

1994 Volume 56 Issue 4 Pages 353-361
Published: December 15, 1994
Released on J-STAGE: August 07, 2009

DOIhttps://doi.org/10.5331/seppyo.56.353

JOURNAL FREE ACCESS

Show abstractHide abstract

Road Conditions in winter vary a great deal in time and space, particularly so in mountainous areas. Therefore, it is very difficult to predict them by any physical methods. Based on field survey data, we constructed two statistical models that have promising self-organizing characteristics; a GMDH (Group Method of Date Handling) model and a multilayer neural network model. We modified a GMDH model so that we can treat not only quantitative data such as temperature, but also qualitative data such as the state of snow removal conducted or not. Moreover, we proposed a multilayer neural network model with multiple input sources, which makes it possible to treat data at each locality, such as sunshine condition, separatoly from the area-wide data, or not. Predicting the friction coefficient 2 hours ahead every 1km road section in a mountainous area, we obtained the prediction precision of more than 70% with the both models.

View full abstract

Download PDF (4721K)
Comments on “Experiments on formation of water chamels in glacier” and editorial review

Katutoshi TUSIMA

1994 Volume 56 Issue 4 Pages 363-364
Published: December 15, 1994
Released on J-STAGE: August 07, 2009

DOIhttps://doi.org/10.5331/seppyo.56.363

JOURNAL FREE ACCESS

Download PDF (1155K)
Reply to the comments by K Tusima

Ken KOIZUMI, Renji NARUSE

1994 Volume 56 Issue 4 Pages 364
Published: December 15, 1994
Released on J-STAGE: August 07, 2009

DOIhttps://doi.org/10.5331/seppyo.56.364

JOURNAL FREE ACCESS

Download PDF (133K)

Register with J-STAGE for free!