ICAM/MAP Brig 19-23 May 20031 Ten years of operational numerical simulations of snow and mountain weather conditions and recent developments at Météo-France.

ICAM/MAP Brig 19-23 May 2003 1

Ten years of operational numerical simulations of snow and mountain weather conditions and

recent developments at Météo-France.

 Y. Durand, G. Guyomarc'h, L. Mérindol, G. Giraud, E. Brun, E. Martin.

Météo-France, Centre d'Études de la Neige (CNRM), 1441 rue de la Piscine, F 38406 Saint Martin d'Hères.

1) Introduction

2) The Safran/Crocus/Mepra (SCM) suite

3) Operational use of SCM (analysis and forecast modes)

4) Research use of SCM

5) Current developments

6) Conclusion

1

Main Characteristics/Limitations of the Operational SCM massif-scale

avalanche hazard modeling.

• -homogeneous massifs (~400 km2) with different elevations (~10), aspects (7) and slopes(3)

• -hourly simulation of a lot of complete snow profiles (T, Z, , LWC, stratigraphy, stability) under the assumption that, at that scale, the snowpack evolution is completely controlled by the atmospheric forcing

• -no realistic orography but « idealized » slopes• -use only of « atmospheric » observations (no use of data concerning the

snow cover)• -use of NWP models: ARPEGE / ALADIN• -no local small scale features forcing • -crude simulation of snowdrift effects (in fresh snow crystal modification

only)

2

ICAM/MAP Brig 19-23 May 2003 3

Regional avalanche forecasting tool

Analyse and forecast of the snow pack evolution by massif, elevation, aspect and slope

Meteorological Data : Observations,meteo model…

SAFRAN

Meteorological analysisCROCUS

Snow model

MEPRAExpert system model

2

ICAM/MAP Brig 19-23 May 2003 4

SAFRAN• Meteorological

analysis for mountain regions

• notions of massif, altitude, aspect

• 2 days forecast version by adaptation of NWP models

• OI and variationnal methods used

MeteorologicalNWP model

Snow and weather observations

Classical meteo Observations and atmospheric sounding

Analysis /6H

24H Analysis for précipitation

Hourly Interpolation for all the parameters

Hourly meteorological parameters affecting snowpack evolution

Radar (in dev.)Satellite

•temperature and humidity•wind velocity/ direction•radiative fluxes•snow and rain precipitation•cloudiness

2

ICAM/MAP Brig 19-23 May 2003 5

SAFRAN « PRACTICAL » VERSION

• O/I scheme and Intermittent Analysis (6h) with ARPEGE or ALADIN as guess-field for :

1. Wind (Div. + Rot.) (verticale profile + surface)2. Humidity (verticale profile + surface)3. Cloudiness (3 layers)4. Temperature (verticale profile + surface)

• 24h Rainfall Analysis based on climatology and typical synoptic patterns.

• Variational Analysis at 1 hour step on 6h windows.

• Several algorisms and modelling for :

1. Radiations 2. Diurnal Variations

3. Vertical Snow-Rain Limits

2

ICAM/MAP Brig 19-23 May 2003 6

CROCUS

2

1D numerical snow model

ICAM/MAP Brig 19-23 May 2003 7

MEPRA1D mechanical analysis

– Additional mechanical characteristics , – Ram resistance profile – Shear strength profile (C)

– Estimation of the applied shear stress (n for snow,sfor skier)– Wet snow instability– Slab occurrence in superficial layers, presence of buried weak layer

• Natural stability index

• Accidental stability index

• Natural avalanche risk on a 6 level scale (very low, low, moderate increasing, moderate decreasing, high, very high)

• Accidental avalanche risk on a 4 level scale (very low, low, moderate,high)

• Avalanche types (fresh dry, fresh wet, fresh mixed, surface slab, surface wet, bottom wet)

n

CS

sn

CS '

2

ICAM/MAP Brig 19-23 May 2003 8

3

ICAM/MAP Brig 19-23 May 2003 9

SAFRAN Results (2)

24h analyzed rainfalls

(15/11/02)

3

ICAM/MAP Brig 19-23 May 2003 10

24h forecated rainfall(11/01/99)

SAFRAN Results (3)

3

ICAM/MAP Brig 19-23 May 2003 11

3/5

ICAM/MAP Brig 19-23 May 2003 12

Tests on annual rainfall climatology

(SIM Projet).

Year 2000.

Symposium Zoning

Safran with full observation network.

Safran with real-time observation network.

3/5

ICAM/MAP Brig 19-23 May 2003 13

CROCUS results (1)

3

ICAM/MAP Brig 19-23 May 2003 14

CROCUS results (2)

3

ICAM/MAP Brig 19-23 May 2003 15

Scatter diagrams of measured and simulated means snow depth on 37 test sites during 4 different months over 10 winter seasons

10 years of measured (dotted) and simulated (solid) snow depths at the Tignes ski resort, Vanoise massif, 2080m.

Safran-Crocus Validation (1)

3

ICAM/MAP Brig 19-23 May 2003 16

Safran-Crocus Validation (2)

Comparisons between observed and numerically simulated snowpack structures during the 1996-97 winter at the ski resort of "La Plagne" (Vanoise massif) on the snow pit location of "Montchavin" (2100m, NE). The different panels illustrate the weekly observed snow pits and the corresponding computed profile. The vertical axis (in cm) represents the snow depth and the blue and green curves respectively temperature and density profiles (with two different scales on the horizontal axis in °C and %V). On the right side of each profile , the stratigraphic profile is illustrated by color code while vertical hachures show crusts.

ICAM/MAP Brig 19-23 May 2003 17

MEPRA operational results (1)

MEPRA : detection of an unstable layer

In this case (1st January 1996), the model has detected a snow structure favourable to an avalanche triggering by overloading (e.g. skiers). A weak layer is buried under 40 cm of new snow constituting a slab after some cohesion processes.

3

ICAM/MAP Brig 19-23 May 2003 18

MEPRA operational results (2)

MEPRA : risk of spontaneous (natural) avalanches due to an unstable fresh snow amount

3

ICAM/MAP Brig 19-23 May 2003 19

Mepra results (3)

Symbolic representation (elevation/aspects) of MEPRA natural

avalanche risks in a typical spring situation (9 March 94, 12 UTC)

3

ICAM/MAP Brig 19-23 May 2003 20

Snow and Climate Change

Snow duration (1500 m)

4

ICAM/MAP Brig 19-23 May 2003 21

Avalanches and climate change

• Rainfall +10% Température +1.8°C

a

0

0,2

0,4

0,6

0,8

1

1,2

1,4

Nov. Dec. Jan. Feb. Mar. Apr. May June

ind

ex

Ref.

PT

Tc

0

0,2

0,4

0,6

0,8

1

1,2

1,4

Nov. Dec. Jan. Feb. Mar. Apr. May June

ind

ex

Ref.

PT

T

All avalanches Melting Avalanches

4

ICAM/MAP Brig 19-23 May 2003 22

Local avalanche forecasting tools

CrocusMepra PCCrocusMepra PC

-Local simulation of the internal state and the mechanical stability of the snow pack

- Analyse and forecast mode

4

ICAM/MAP Brig 19-23 May 2003 23

SCALE DEFINITIONS

Downscaling

Massif scale : ~ 500 km2

Time step : 1 hour

Symbolic topography

Local scale : ~ 1 km2

Time step : 30 min

Fine scale orography

Avalanche path scale : ~ 100 m2

Time step : ~ 5 min

100 200 300 400 500 600

100

200

300

400

500

600

700

800

5

ICAM/MAP Brig 19-23 May 2003 24

Snow Drift

• Slab formation, effects of snow accumulation and erosion… • Objectives :

– Better estimation of accidental avalanche risks

– Better localisation of avalanche risks

• Experimentation in a specialized field laboratory• Modelling:

- SAMVER : surface wind field estimation

- SYTRON1 : wind transport at the massif scale (nearly operationnal)

- SYTRON2 : wind transport at local scale

5

ICAM/MAP Brig 19-23 May 2003 25

Snow Drift

5

Snow depth differences due to snow drift effects

ICAM/MAP Brig 19-23 May 2003 26

Safran/Crocus/Mepra

chain• Valuable avalanche forecasting tool for regional forecasters :

meteo, snow, stability and risks• « analyse » mode with all the snow/meteorological data• 24 and 48 H forecast with the runs of the ALADIN and

ARPEGE meteorological models and analyses by a nearest neighbour method

• Validations : – Meteorological (Col de Porte, Lac Blanc)

– Snow depths

– Snow profiles

– By the avalanche forecasters

6

Conclusion (1)

The SCM chain is presently running over:24 alpine massifs23 Pyrenean massifs12 islandic areas

It is operationally used by numerous forecasters.

It is still under development (snow drift, …)

It is also a research tool (climate ….)

It is used in related applications (hydrology, …)

Good skiing and thank you for your attention

6

Conclusion (2)