RD 6

RCAO, a coupled atmosphere-ice-ocean-land model of the Arctic: 

RCAO, a coupled atmosphere-ice-ocean-land model of the Arctic R. Döscher*, K. Wyser*, M. Meier**, G. Broström*, P. Samuelsson*, P. Graham*, M. Qian*** *SMHI/Rossby Centre/Norrköping/Sweden **SMHI/Ocean research/Norrköping/Sweden ***UQAM/Montreal/Canada

Sea ice extent: 

Sea ice extent Spinup phase

Summer sea ice extent in four ensemble runs: 

Summer sea ice extent in four ensemble runs Spinup phase

Summer sea ice extent anomaly in four ensemble runs: 

Summer sea ice extent anomaly in four ensemble runs --- coupled runs --- ERA40 or sat observation

Sea ice extent: 

Sea ice extent summer winter Ensemble mean ERA

Interannual variability in sea ice extent: 

Interannual variability in sea ice extent summer winter ERA Ensemble mean

Trends in sea ice extent: 

Trends in sea ice extent summer winter ERA Ensemble mean

Slide8: 

summer winter Trends in sea ice thickness Ensemble mean compare with Rothrock et al. 2003

Mean T2m anomaly: 

Mean T2m anomaly --- red RCAO ensemble --- black ERA-40

Surface pressurein standalone atmosphereandcoupled model: 

Surface pressure in standalone atmosphere and coupled model

Predictability: 

Predictability Predictability is defined as the extent to which variability of Arctic variables can potentially be controlled by external forcing. The potential predictability is low when the internal (=internally generated) variability high. Predictability studies are useful to assess the feasibility of prediction systems. Internal variability = internally generated variability = mean variability among the ensemble members not predictable by boundary conditions common to all ensemble members External variance = externally driven variability = variance of the ensemble-averaged anomalies SNR = external variance/internal variability = indicator for predictability

Predictability of summer ice thickness: 

Predictability of summer ice thickness internal variability external variability total variability external/internal

Slide13: 

Red areas in the signal-to-noise ratio indicate strong control by external driving forces rather than internal non-linear processes. Predictability of winter ice thickness internal variability external variability total variability external/internal

Seasonal cycle in total variability for t2m: 

Seasonal cycle in total variability for t2m ERA-40 Model ensemble weak signal during summer, due to almost zero ice temperature

Predictability t2m winter: 

Predictability t2m winter Dominating internal variability in Fram Straits internal variability external variability total variability external/internal

Slide16: 

Climatology of freshwater Beaufort gyre Total Arctic freshwater volume: ocean: 74345 km3 Sea ice: 10450 km3 Freshwater originates mostly form Siberian rivers and is transported into the Beaufort Sea. (Proshutinsky et al., 2002)‏ Data from: Polar hydrographic climatology (Steele et al.). 1x1 degree data set. Freshwater height Sea surface salinity

Freshwater height: 

Freshwater height

Freshwater export Fram Strait: 

GSA 1990s event Freshwater export Fram Strait

Freshwater export Framstraits: 

Figure from Lemke et al. : Simulated and observed sea ice transport through Fram Strait 1990-1996 (Hilmer, 1999, private communication; Vinje et al., 1998)‏ Freshwater export Framstraits

Freshwater export Greenland-Island: 

Freshwater export Greenland-Island Solid part of freshwater export reduced in Danmark/Greenlan Straits Freshwater transport is similar in all ensemble members

Freshwater fluxescombined results from observations and modelling: 

Freshwater fluxes combined results from observations and modelling ??? Dickson et al. 2007

Slide22: 

Foreseen needed development of RCA for Arctic applications Now: diagnostic soil frost Future: prognostic soil ice and its impact on soil hydrology Now: soil depth is 3m Future: deeper soil and more layers to include perma frost processes Now: Baltic Sea river routing Future: flexible river routing applicable anywhere Now: forest and open land (grass) Future: include wet land processes Now: physical lake (FLake) Future: biochemical processes Now: only snow Future: glacier included

Slide23: 

RCA3 coupled to the dynamic vegetation model GUESS (Ben Smith et al., Lund University) Air temperature Soil temperatures Soil water SW net CO2 LAI Fraction forest Fraction deciduous LAI Fraction of vegetation Future: utilize the potential in GUESS to include carbon and nitrogen soil processes

Runoff Routing in RCA2 ...: 

Runoff Routing in RCA2 ... In: mm/km2 Out: m3/s focus on the Baltic Basin

Runoff Routing in RCA2 ...: 

Runoff Routing in RCA2 ... a more universal approach with added detail is desired In: mm/km2 Out: m3/s focus on the Baltic Basin

Runoff Routing in RCA3 ...: 

Runoff Routing in RCA3 ... example: 0.2 deg

Runoff Routing in RCA3 ...: 

Runoff Routing in RCA3 ... Torne-Kalix Basin example: 0.2 deg River outflow

Runoff Routing in RCA3 ...: 

Runoff Routing in RCA3 ... example: 0.2 deg choose a basin!

New datasets coming in Dec, 2007Hydro1k/Hydrosheds: 

New datasets coming in Dec, 2007 Hydro1k/Hydrosheds

Results so far: 

Results so far The regional coupled model gives realistic sea ice extent trends during the 1980s and 1990s. Externally forced variability in the Arctic is generally stronger or of similar amplitude than internally generated variability. Internally generated variability (due to non-linear ocean-ice-atmosphere interaction within the Arctic) is of similar importance as external forcing for future decadal prediction efforts Freshwater export shows some basically realistic features. the regional coupled model potentially provides the tool for an integrated freshwater analysis

Future plans: 

Future plans Further model improvements Land surface scheme River routing Ice classes Further predictability studies under different climate Recent climate simulations forced with with improved reanalysis products Ocean standalone simulations of the complete 19th century within the AOMIP project. Within the EU-IPY-project DAMOCLES, Regional Arctic climate scenarios based on Bergen Climate model and other GCMs will be carried out.