Influence of Roadside Environment and Road Structures on Blowing-Snow-Induced Visibility Hindrance on Winter Roads

: Analysis using the results of weather observations by a visibility observation vehicle during blowing snow.

2016 International Conference & Workshop on Winter Maintenance and Surface Transportation Weather

Civil Engineering Research Institute for Cold Region Hirotaka Takechi, Masaru Matsuzawa, Yasuhiko Ito and Tetsuya Kokubu

Hokkaido

Sapporo

Tokyo

Outcomes

To clarify the influence of following items on blowing-snow-induced visibility hindrance

Road structures Length of windward flat land Width of roadside tree belt Width of roadside rows of houses Presence/absence of guardrails

Background and objectives Method of moving weather observation Method of multivariate analysis Results of analysis Conclusions

Outline

Background

Hokkaido

Tokyo

Winter Road Problems in Hokkaido

Poor visibility and snowdrift caused by snowstorms Traffic hazards

Snowbreak Woods

Background

It is important to accurately determine which locations are at risk of blowing snow.

Snow fences

Installation of blowing-snow control facilities to mitigate traffic hazards caused by snowstorms

Objective

The Highway Snowstorm Countermeasure Manual The Degree of Snowstorm Danger

It is necessary to develop the technology for risk assessment of snowstorms along entire routes for appropriate installation.

Risk assessment of snowstorms (Image)

Present technology for risk assessment of snowstorms

http://www2.ceri.go.jp/fubuki_manual/index_e.html http://www2.ceri.go.jp/fubuki_manual/pdf/manual.eng.pdf

On The Highway Snowstorm Countermeasure Manual

Snowfall(snow depth)

Road structure

Windward flat land

Roadside trees and houses

Volume of snow settled at

snowdrifts

Frequent occurrence of blowing snow

ランク 総合評点 判定

A 44点以上 吹雪による障害の危険性が極めて高い

B 21-43 吹雪による障害の危険性がある

C 1-20 吹雪による障害に対して注意を要する

D 0以下 吹雪対策の必要なし

Objective

Snowfall(snow depth)

Roadside trees and houses

To survey by moving weather observation on winter road during blowing snow

To clarify the influence of the road environment and structures on blowing-snow-induced visibility hindrance

Items and scores used to assess the degree of snowstorm danger were determined empirically

The degree of Snowstorm Danger

Poor visibility factor

Snowdrift factor

Degree of snowstorm danger

Visibility observation vehicle

Moving weather observations were conducted during blowing snow by a visibility observation vehicles.

Method of observation

Visibility, Wind velocity, Wind direction, Air temperature, Latitude and Longitude data ( Sampling interval was 0.1 sec. )

Sonic wind sensor (CLIMATEC,Inc CYG-85004) Installation height of 2.8m

Forward scattering visibility meter (Meisei. TZF-31A) Installation height of 1.75m

Locations of observation Method of observation

R238 Saroma city, Monbetsu city

R232 Enbetsu city, Haboro city

R231 Iishikari city

R238 Hamatonbetus city, Sarufutsu city

Number of totalobservation times

2012/12/26 10:49 - 16:23 14.7 - 41.4 92013/3/10 15:15 - 18:11 14.2 - 51.7 42014/1/13 12:48 - 16:04 14.3 - 51.7 32014/1/28 18:26 - 22:15 15.0 - 52.0 52014/1/31 14:44 - 17:44 15.0 - 52.0 4

2014/3/6 14:51 - 17:48 17.0 - 26.0 32015/2/2 15:13 - 20:05 15.0 - 42.0 8

2015/2/15 10:22 - 18:19 15.0 - 42.0 72012/12/27 8:52 - 16:05 35.9 - 76.5 6

2013/2/2 19:49 - 20:26 52.8 - 73.0 12013/2/3 8:32 - 9:32 51.8 - 68.5 22013/2/5 14:58 - 15:34 44.3 - 70.4 1

2013/2/20 6:32 - 12:51 41.6 - 71.2 72014/1/11 11:25 - 15:22 36.0 - 71.4 42014/1/28 16:00 - 21:01 35.9 - 71.4 42014/1/29 8:50 - 14:11 35.3 - 57.5 52014/1/31 17:33 - 18:52 36.0 - 71.4 2

2014/2/3 9:54 - 12:43 36.0 - 64.4 42014/3/7 15:44 - 16:43 47.0 - 64.0 22015/1/7 15:40 - 21:58 42.0 - 75.0 72015/1/8 11:35 - 11:50 71.0 - 81.0 1

2015/3/11 17:20 - 19:58 38.0 - 70.0 2

Route Date of observations KP※Maximum section

Number of observations

Route231in Ishikari city(KP14.2-52.0)

43

Route 232from Enbetsu to

Haboro city(KP35.3-81.0)

48

Observation period : Three winters periods of 2012 to 2013, 2013 to 2014, 2014 to 2015

Method of observation

Date of observations

Number of totalobservation times

2013/2/8 9:54 - 16:51 63.7 - 102.3 62013/3/2 15:05 - 17:23 52.1 - 77.3 12014/2/9 17:41 - 23:27 54.6 - 90.2 2

2014/2/17 8:07 - 10:55 74.0 - 102.0 12013/1/26 11:40 - 16:11 229.2 - 235.6 8

2013/2/3 12:40 - 15:52 217.9 - 274.3 32013/12/27 18:17 - 22:20 217.1 - 258.5 2

2014/1/26 16:10 - 16:41 236.9 - 258.5 12014/2/5 5:31 - 5:49 240.0 - 251.9 1

2014/2/10 14:44 - 18:38 219.6 - 257.4 42014/2/18 7:50 - 14:18 216.0 - 262.0 62015/2/27 8:27 - 20:05 225.0 - 257.0 92015/4/15 21:53 - 0:26 222.0 - 258.0 2

Route 238from

Hamatonbetsu toSarufutsu city

(KP216.0-274.3)

36

Route 238from Saroma toMombetsu city(KP52.1-102.3)

10

Route Date of observations KP※Maximum section

Number of observations

Method of observation Observation period : Three winters periods of 2012 to 2013, 2013 to 2014, 2014 to 2015 Date of observations

Method of analysis

Data were gathered for each 50-m section of road.

Average of observed visibility values

Data on road structure, roadside environmental conditions from road inventories, maps, and aerial photographs

Multivariate analysis Mathematical quantification theory class I ( Hayashi's quantification method I )

1. Response variable Visibility hindrance occurrence rate (%)

Method of analysis

Number of occurrences of avg visibility < 200m

Number of total observation times ×100%

0 Cut of higher than 5.0m1 Cut of 0-5.0m2 Fill of 0-1.0m3 Fill of 1.0-3.0m4 Fill of 3.0-5.0m5 Fill of higher than 5.0m0 N/A1 10-100m2 100-300m3 More than 300m0 N/A1 Scarce2 10～30m3 30～50m4 50～100m5 100～200m6 200～300m7 More than 300m0 N/A1 Scarce2 10～30m3 30～50m4 50～100m5 100～200m6 200～300m7 More than 300m0 Absent1 Present

Windward

Road structure(Height of cut and fill )

Independent variableItem

Length offlat land windward

Width of tree belt

Width of rows ofhouses

Presence/absenceguardrails

Category2. Independent variable

Method of analysis

Road structure Length of windward flat land Width of roadside tree belt Width of roadside rows of

houses Presence/absence of

guardrails

A road section without blowing-snow control facilities were studied.

Response variable Categoryscore

Itemrange

Partialcorrelationcoefficient

p-value

0： Cut of higher than 5.0m -0.101 1.950 0.104 6.5×10^-71： Cut of 0-5.0m 0.3932： Fill of 0-1.0m -0.4623： Fill of 1.0-3.0m -0.197

※The rate (%)= 4： Fill of 3.0-5.0m 1.4885： Fill of higher than 5.0m 0.7230： N/A -0.547 1.681 0.114 4.5×10^-81： 10-100m 0.0872： 100-300m 1.3053： More than 300m 1.1340： N/A 0.120 1.591 0.056 7.6×10^-31： Scarce -0.2452： 10～30m -0.1663： 30～50m -0.0524： 50～100m -0.3865： 100～200m -0.7906： 200～300m -1.1897： More than 300m -1.4710： N/A -0.294 6.090 0.156 4.8×10^-141： Scarce 0.9332： 10～30m 1.7763： 30～50m 4.4124： 50～100m 1.7605： 100～200m 0.2066： 200～300m -0.3637： More than 300m -1.6770： Absent -0.375 1.532 0.118 1.3×10^-81： Present 1.157

2.761

Independent variable( Item and category )

Road structure(Height of cut and fill )

Length ofwindward flat land

Visibility hindranceoccurrence rate (%)

Width of windwardtree belt

Width of windwardrows of houses

Presence/absence ofguardrails

Constant

Number of occurrences of avg visibilty < 200m

Number of total observation times

×100%

Results of analysis

-2

-1

0

1

2

3

4

5

Cut

of h

ighe

r th

an 5

.0m

Cut

of 0

-5.0

m

Fill

of 0

-1.0

m

Fill

of 1

.0-3

.0m

Fill

of 3

.0-5

.0m

Fill

of h

ighe

r th

an 5

.0m

N/A

10-1

00m

100-

300m

Mor

e th

an 3

00m

N/A

Scar

ce

10～

30m

30～

50m

50～

100m

100～

200m

200～

300m

Mor

e th

an 3

00m

N/A

Scar

ce

10～

30m

30～

50m

50～

100m

100～

200m

200～

300m

Mor

e th

an 3

00m

Abs

ent

Pres

ent

Road structure(Height of cut and fill )

Length ofwindward flat

land

Width of windward tree belt Width of windward rows ofhouses

G C

Cat

egor

y Sc

ore

Gua

rdra

ils

Cons

tant

Category for each item

Category score of each item Results of analysis

-2

-1

0

1

2

3

4

5

Cut

of h

ighe

r th

an 5

.0m

Cut

of 0

-5.0

m

Fill

of 0

-1.0

m

Fill

of 1

.0-3

.0m

Fill

of 3

.0-5

.0m

Fill

of h

ighe

r th

an 5

.0m

N/A

10-1

00m

100-

300m

Mor

e th

an 3

00m

N/A

Scar

ce

10～

30m

30～

50m

50～

100m

100～

200m

200～

300m

Mor

e th

an 3

00m

N/A

Scar

ce

10～

30m

30～

50m

50～

100m

100～

200m

200～

300m

Mor

e th

an 3

00m

Abs

ent

Pres

ent

Road structure(Height of cut and fill )

Length ofwindward flat

land

Width of windward tree belt Width of windward rows ofhouses

G C

Cat

egor

y Sc

ore

Gua

rdra

ils

Cons

tant

Ｘ

Category score of each item Results of analysis

Category for each item

0.01.02.03.04.05.06.07.0

Hei

ght o

f cut

and

fill

Leng

th o

fw

indw

ard

flat

land

Wid

th o

fw

indw

ard

tree

belt

Wid

th o

fw

indw

ard

row

s of

hous

es

Pres

ence

/abs

ence

of g

uard

rails

Item

rang

e Results of analysis

Item

Range of category score of each item

The visibility hindrance occurrence rate during blowing snow tended to be high for roads with long windward flat land. With flat land of 100 m or longer, the rate tended to be high.

On roads with a windward tree belt of 100 m or wider or with rows of houses of 300 m or wider, the rate tended to be remarkably low. The visibility tended to be good at such locations.

The rate tended to be high at fills of about 3 m or higher and at cuts of lower than 5 m.

At sections with guardrails, the rate tended to be higher than at road sections without guardrails.

Conclusions

Thank you for your kind attention For more information, question snow@ceri.go.jp hiro-takechi@ceri.go.jp Snow and Ice Research Team Civil Eng. Res. Institute for Cold Region