Modeled response of snow cover- atmosphere-ocean interactions in the Northern Hemisphere. Gina Henderson, Daniel J. Leathers and Brian Hanson Department.

Modeled response of snow cover-Modeled response of snow cover-atmosphere-ocean interactions in the atmosphere-ocean interactions in the

Northern Hemisphere.Northern Hemisphere.

Gina Henderson, Daniel J. Leathers and Brian Hanson

Department of Geography, University of Delaware

Outline:Outline:

1. Introduction & Background: climate modeling and snow cover

2. The Model: atmospheric, land and ocean components

3. Experiment Design: observed snow datasets

4. Preliminary Results: high vs. low snow experiments for North America and Eurasia

5. Future work

Introduction: Significance of SnowIntroduction: Significance of Snow

Snow cover identified as the most variable land surface condition in both time and space (Cohen, 1994; Gutzler and Rosen, 1992)

Snow cover is a major factor in modulating climate variability and change

http://climate.rutgers.edu/snowcover/index.php

January

1966-99

~ 46.8 x 106 km2

August

1966-99

~ 3.4 x 106 km2

Snow cover forcing climate:Snow cover forcing climate:

Cold dense air above snow surfaces affect local climate and can propagate anomalies to neighboring regions.

Modeled snow forcing:

Snow cover over western Siberia found to be negatively correlated with the leading EOF of sea level pressure in the North Atlantic (Gong, G., D. Entekhabi, and J. Cohen, (2002): A large-ensemble model study of the wintertime AO-NAO and the role of interannual snow perturbations. Journal of Climate, 15).

Snow cover in the northern Great Plains in the U.S. was found to be linked to downstream tropospheric circulation (Klingaman, N. P., B. Hanson, D. J. Leathers, (2008): A teleconnection between forced Great Plains snowcover and European winter climate. Journal of Climate, 21).

This study:

How does an interactive slab ocean affect the land-atmosphere forcing under a forced snow scenario?

The Model:The Model:

Employs 3 modules of the NCAR Community Climate System Model, version 3.1 (CCSM3) The Community Atmospheric Model (CAM3.1) ✔ The Community Land Model (CLM) ✔ The Data Ocean Model (DOM) ✗ The Slab Ocean Model (SOM) ✔

The GCM has 26 vertical levels and a standard baseline spherical truncation at wavenumber 42 (T42).

Grid cells are approximately 2.8º latitude by 2.8º longitude (~ 200 km by 300 km in middle latitudes)

Slab Ocean Model (SOM):Slab Ocean Model (SOM):

SOM is a mixed-layer slab ocean model where mixed layer temperature is the prognostic output variable.

Seasonal deep water exchange is simulated by an internal heat source, Q, set from a control run.

Mixed layer depths vary seasonally and geographically.

F = net atmosphere to ocean heat flux

FS = net downward solar flux absorbed

FL = net upward longwave flux

SH = upward sensible heat flux

LH = upward latent heat flux

T0 = ocean mixed layer temperature

0 & C0 = density and heat capacity of ocean water

h0 = annual mean ocean mixed layer depth

A = fraction of the ocean covered by sea ice

F = net atmosphere to ocean heat flux

Q = internal ocean mixed layer heat flux

Foi & Ffrz = sea ice energy flux terms

0 C0 h0 ( T0/ t) = (1-A)F + Q + AFoi + (1-A)Ffrz

Control Run

Max North American Snow

Max Eurasian Snow

Min North American Snow

Min Eurasian Snow

• Model prescribes snow

• 58 year run, equilibrium after yr 20

• Last 38 years used for analysis

• Branch model runs

• Snow is prescribed not predicted

• Snow depth based on observations

• Ensemble size of 20

Experiment Design:Experiment Design:

Snow Data:Snow Data: 1° X 1° interpolated snow depth data (Dyer and Mote, 2006) from U.S. NWS cooperative stations an the Canadian daily surface observations. Period of record 1900-2000 with daily resolution

Gridded 2.5° X 2.5° snow depth data. NSIDC’s Historical Soviet Daily Snow Depth (HSDSD) Period of record is 1881-1995, we are using 1967-1995 Gridded by Hengchun Ye, CSU.

Experiment Name

Description Length Fixed/Free sea surface

Snow condition

Ctl_free Control run 38 years Slab Ocean (SOM)

Free to vary

Ctl_dom Control run, prescribed SSTs

200 years

Data Ocean

(DOM)

Free to vary

Ctl_clim Control run, prescribed snow to model climatology

20 years Slab Ocean (SOM)

Prescribed

Max_eur Snow prescribed everywhere, max over Eurasia

20 years SOM and DOM

Prescribed

Min_eur Snow prescribed everywhere, min over Eurasia

20 years SOM and DOM

Prescribed

Max_na Snow prescribed everywhere, max over North America

20 years SOM and DOM

Prescribed

Min_na Snow prescribed everywhere, min over North America

20 years SOM and DOM

Prescribed

Snow Data: prescribed North American experimentSnow Data: prescribed North American experiment

Snow Data: prescribed Eurasian experimentSnow Data: prescribed Eurasian experiment

Results: Results: Max – Min Eurasian ExperimentMax – Min Eurasian Experiment

Student’s t-test performed to test for significance.

Dark gray = 95% confidenceLighter gray = 90% confidence

Cooler sea surface temperatures associated with maximum snow.

Anomalies of -1.5 to -2 K in the North Pacific and Atlantic.

Results: Results: Max – Min Eurasian ExperimentMax – Min Eurasian ExperimentLatitudinally averaged SST differenceLatitudinally averaged SST difference

Results: Results: Max – Min Eurasian ExperimentMax – Min Eurasian Experiment

Cooler 2 m temperature with maximum snow prescription

Temperature depressions up to -5 K over Northern Eurasia

Diabatic cooling over North Pacific and Atlantic

Results: Results: Max – Min Eurasian ExperimentMax – Min Eurasian Experiment

Results: Results: Max – Min Eurasian ExperimentMax – Min Eurasian Experiment

Sea level pressure shows an organized pattern but is not significant.

Negative NAO pattern over North Atlantic during maximum snow conditions.

Lack of significance may be an ensemble size problem.

The North Atlantic Oscillation (NAO):The North Atlantic Oscillation (NAO):

Positive NAO: Steep pressure gradient Strong westerlies Warm & wet N Europe Cold & dry Mediterranean

Negative NAO: Weakened pressure gradient Cold & dry N Europe Warm & wet Mediterranean Cold air outbreaks in E U.S.

http://airmap.unh.edu/graphics/

Results: Results: Max – Min North American ExperimentMax – Min North American Experiment

Results: Results: Max – Min North American ExperimentMax – Min North American Experiment

Once again, pattern is organized but not significant.

Maximum snow conditions associated with weakened pressure gradient over the North Atlantic.

Lower pressures over Aleutian Low area under maximum snow conditions.

Summary of Findings:Summary of Findings:

Both experiments show response to prescribed snow forcing, Eurasian response larger.

Significant negative sea surface temperature response is evident in both the Eurasian and North American maximum snow forcing experiments.

Surface and lower atmosphere temperature, and 500 hPa heights show negative response to maximum snow forcing.

Although sea level pressure response is organized and suggests the excitement of modes of Northern Hemisphere atmospheric circulation, results are not statistically significant.

Experiment Name

Description Length Fixed/Free sea surface

Snow condition

Ctl_free Control run 38 years Slab Ocean (SOM)

Free to vary

Ctl_dom Control run, prescribed SSTs

200 years

Data Ocean

(DOM)

Free to vary

Ctl_clim Control run, prescribed snow to model climatology

20 years Slab Ocean (SOM)

Prescribed

Max_eur Snow prescribed everywhere, max over Eurasia

20 years SOM and DOM

Prescribed

Min_eur Snow prescribed everywhere, min over Eurasia

20 years SOM and DOM

Prescribed

Max_na Snow prescribed everywhere, max over North America

20 years SOM and DOM

Prescribed

Min_na Snow prescribed everywhere, min over North America

20 years SOM and DOM

Prescribed

Future work:Future work:

Questions?Questions?

ExtraExtra

Background:Background:

Characteristics of snow: High reflectivity Thermal insulator Sink for latent heat Frozen storage term in hydrologic cycle

Snow cover is a major factor in modulating climate variability and change

The Community Land Model (CLM):The Community Land Model (CLM):

CLM subgrid hierarchy, land biogeophysical and hydrologic processes.

Results: Results: Max – Min North American ExperimentMax – Min North American Experiment

Similar response to Eurasian experiment, weaker values.

Anomalies of -0.5 to -1 K in the North Pacific and Atlantic.

Results: Results: Max – Min North American ExperimentMax – Min North American Experiment

Cooler surface temperatures over North Atlantic and Pacific of -1 K.

Cooler surface temperatures over southern North America of -1 to -5 K.

Max - Min Eurasian

Max - Min North American