H a z a r dM a n a g e m e n ta n dP u b l i cWa r n i n g

weather. Severalsonic snow depthsensors will be installed
to complement the radar.
This fall, constructionon a pair of divertingwalls, above
the village of Flateyri, will be started. It is expected that it
will take two years to complete it.They range from 15 to
20meters in height and have a total length of1.6kmor
about one mile. Whencompleted, an avalancheradar will
be installed ontop ofthe diverting wall tomeasure ava-
lanche velocities.The aimis to investigate the effective-
nessofadivertingwall and obtaindata toverifymodel
calculationsused to obtain the height of the divertingwall.
The actual run-up will bemeasured and mapped by the
local avalanche observer.It is ourhope that this will pro-
vide valuable data that will be of use in future avalanche
defense structures in Iceland and the rest of the world.
In Iceland,the lack of vegetation precludes the use of
vegetationdatatodeterminehistoricalrunoutofava-
lanches. Weare hencelooking for other ways to determine
themaximumrunoutofavalanches.Thus,duringthe
construction ofthe above mentioned diverting dams,on
site inspection will be conducted so as to gatherdata for
thepossibleidentificationofavalanchedebris.The
avalanchefrequencyis very high in this path and we hope
that during the excavation of the fluvial fan, to collect in-
formation that will aid us in identifying past runoutof ava-
lanches. In addition,every effortwill be made tocollect
similardata from other sites.
Aninvestigationintowhat isan acceptable risklevel
whenhazard zoningisconsideredisunderway.Thisis
being done in the preparation for new regulationsand will
provide the governing bodies with a foundation fordeter-
mining what should be adopted as the acceptable level of
risk in Iceland.
Furthermore, an effort is being made to develop a meth-
odology to evaluate the return period of avalanchesin Ice-
land based on existing data.

ACKNOWLEDGMENTS.

The author would like to thank his colleagues at the Ava-
lanche Division of the Icelandic Meteorological Office.I
would alsolike tothank The Nordic CouncilOf Minis-
ters supplied the funding fora part of theworkbehind
this paper.

REFERENCES

A udunsdo ttiret.al.(1 9 96 )Ry m ingarsv aediv egna
snjoflodahaettu.(Evacuationlevels inresponsetoavalanche
danger). VI-G96002-UR02March 1996,Internal Report.

Johannesson, et. al. (1996) An Overview Of The Need For Ava-
lanche Defenses In Iceland. VI-R96003-UR2 September 1996,
Internal Report.

Canadien Avalanche Association (1995).Observation Guide-
linesandRecordingStandardsforWeather,Snowpackand
Avalanches.

Kristersen & Magnusson. (1996). Snow Avalanch Forecasting
In Iceland And Norway. NGI report 581250-2January 1996.

7