Snow and ice in polar and sub-polar seas: numerical modeling and in situ observations

Transcription

1 Snow and ice in polar and sub-polar seas: numerical modeling and in situ observations Bin Cheng, Timo Vihma, Jouko Launiainen, Laura Rontu, Juha Karvonen, Marko Mäkynen, Markku Simila, Jari Haapala, Anna Kontu, Jouni Pulliainen Finnish Meteorological Institute (FMI)

2 0 C T Initial ice formation Freezing season z Thermal equilibrium stage Melting Season

3

4 Objectives to develop snow and ice thermodynamic model HIGHTSI to investigate snow and ice mass balance and temperature regimes. to understand snow and ice physical properties. to improve the snow and ice schemes used as boundary condition for numerical prediction models. to provide physical background information for ice thickness analysis using remote sensing data. to carry out sustainable long term snow and ice observations in Arctic and seasonal ice covered seas. Tasks Snow and ice modeling In situ observations

5 Snow and ice modeling Model validations (Bohai Sea, Baltic Sea, Arctic Ocean) Numerical scheme: spatial resolution on model results External forcing: in situ measurements; NWP model results Effect of snow on ice mass balance: snow ice and superimposed ice formation. Evaluation of albedo schemes applied in ice model. Thermal and optical properties of snow and ice. HIGHTSI model for lake applications. Basin scale ice thermodynamic growth. Field observation Bohai Sea; Baltic Sea CHINARE2003; CHINARE2008 Arctic lakes

6 Snow Ice Air Snow Tsfc Tinx Water Ice Tsnow Tice HIGHTSI: One dimensional snow/ice thermodynamic model considered in a horizontal unit area External forcing: NWP models (HIRLAM/ECMWF) Result: Snow and ice thickness; surface temperature hs Fcs Fci hs Snow/ice pond hi water Snow/Ice Open water/ice concentration inforamtion (SAR, AMSR_E, MODIS)

7 External weather forcing data: - Wind speed (m/s) - Air temperature ( C) - Moisture, in format of relative humidity % - Cloudiness (0-1) - Precipitation, in format of snow liquid water content (mm/t) - Downward shortwave radiative flux (W/m2) - Downward longwave radiative flux (W/m2) Sensible heat flux from water below (W/m2) - Surface albedo (0-1) - Open water/ice concentration inforamtion (SAR, AMSR_E, MODIS)

8

9 The Observed and modeled evolution of (a) snow thickness Hs, (b) ice freeboard, (c) superimposed ice thickness (granular ice) Hsui, and (d) total ice thickness Hi. The observed precipitation (a), total ice thickness (b), snow thickness and freeboard (c), and granular ice growth (d) in the Baltic Sea (Granskog, et al, 2006, J. Glaciol, The observed (symbols) and modelled (lines) snow temperature profiles (a) on day 79 and (b) day 88. The zero depth refers to the snow/ice interface. (Cheng et al, 2006 Ann. Glaciol.) The time series of modelled snow thickness. The white area below the surface indicates the region of active surface and sub-surface melting.

10 Model experiments on snow and ice thermodynamics in the Arctic Ocean with CHINARE 2003 data (Cheng, et al, 2008, JGR)

11 HIGHTSI modeled snow and ice mass balance (Cheng et al, 2008, CJPR) with external forcing data proposed by SIMIP2 (Huwald et al, 2005) - Precipitation x 1.5 Less calculated surface melting against observation The Observed ice thickness and temperature regime during SHEBA annual cycle (Perovich et al, 2003, JGR)

12 Albedo from SIMIP2 (melt pond effect?). Oceanic heat flux was 11W/m2 on the average during the SHEBA year. Overestimated surface melting with coarse spatial resolution. Improved results with superimposed ice formation taken into account, The modeling errors are related to the uncertainties of the snow/ice thermal properties.

13 Tara s drift started in September 2006 in the Laptev Sea north of Siberia. Tara passed near the North Pole to the Fram Strait, where it broke free of the ice on 21st January Tara drift trajectory from NW to SE between 1 April and 30 September. On 18, April, 2007, Tara was located in the center of the large cross

14 HIGHTSI modelled snow and ice thicknesses; in snow and ice temperature field and surface skin temperature

15 Downward shortwave radiative flux Wind speed difference Exp. 4: Hirlam albedo Exp. 5: Tara albedo Difference= Exp. 4 Exp. 5 Albedo: Exp. 4, Exp. 5 Downward longwave radiative flux Temperature difference J day 160 == 9, June Surface temperature difference

16 Calculated surface temperature m 0.1m 0.15m 0.25m 0.3m 0.35m 0.4m 0.5m 1.0m mean values Simulated ice thickness (m) ( oc) (b) Calculated surface temperature ( oc) (a) m 0.2m 0.3m 0.4m 0.5m 1m mean value Simulated ice thickness (m) Surface temperature versus different ice thickness category: (a) a cold period between 3 Jan 0:00-5 Jan 23:00 (b) a warm period between 8 April 0:00 11 April 13:00 (Yang et al, 2012, Tellus) Surface temperature response strongly for thin ice category (<0.5m) in cold condition

17 KaraX sea ice product area Red dots are weather stations. Coverage 1500 by 1350 km. 30 October -2 November, Chinese-Finnish Arctic Seminar

18 Doronin (1971): hs = 0 for hi < 5 cm; hs = 0.05xhi for 5 cm hi 20 cm; hs = 0.1xhi for hi > 20 cm Mäkynen and others (2012): hs = 0 for hi < 5 cm; hs = 0.05xhi for 5 cm hi 20 cm; hs = 0.09xhi for hi > 20 cm Cheng and others (2012): Problems: 1.Snow effect: MODIS surface temperature inverses ice thickness 2.The input of snow thickness for ice modelling

19 A method for sea ice thickness and concentration analysis based on SAR data and a thermodynamic model Karvonen, Cheng, Vihma, Arkett, and Carrieres, 2012, TCD Fig. 10. Ice thickness for the Jan 5, Feb 5, March 5, and Apr (from top to bottom), from HIGHTSI model (middle column), from the CIS ice charts (left column) and based on our SAR algorithm (right column).

20 Ice mass balance buoys invented by SAMS (Scottish Association for Marine Science) Continuous measurements at one location Monitor high resolution temperature profile (sensor interval: 2cm) 81.8N,130.9E Heater element Temperature chain in Hot-Wire mode Data-buoy with Iridium Link Data + power bus Data + Air Power bus Chip Resistor Digital (heater element) temperature sensor Ice-air interface Sea-Ice Digital Thermistor Ice-water Nicolaus, AWI, 05/09/ N,57.4E Ocean Schematic of the temperature chain used to measure the ice air and ice water interface.(by Jeremy Nicolaus, AWI, 22/09/2012

21 19, 12, , 2, , 4, 2012

22 Temperature profiles (air-snow-ice) and temperature field (snow, ice) from IMB. Snow surface, snow/ice interface and ice bottom detected by the IMB data. Snow and ice thicknesses detected from IMB data (lines) and in situ measurement (symbols) in lake Orajärvi. The snow/ice interface is used as reference level; Snow and ice temperature regimes

23 HIGHTSI modelled snow and ice thickness for winter 2011/2012. The external forcing was in situ observations. The green circles and black triangles are in situ observations made in two different locations on Lake Orajärvi

24 11,3, ,4,2012 Ice core samples collected from lake Orajärvi in March and April, winter 2011/ October -2 November, Chinese-Finnish Arctic Seminar

25

26 Conclusions and outlook Model validation is good. Evaluation of external forcing (in situ measurement & NWP results). Improvement of understanding on snow and ice thermodynamics. Multidisciplinary methodology on ice thickness analysis Snow parameterization for Arctic conditions. Sustainable field measurements is important and will continue in the future. Operational services Seasonal forecasts Inter-annual and decadal climate forecasts Close collaborations with Chinese colleagues

27 Acknowledgement to colleagues in China 雷瑞波, 郭井学, 学占海 Dr. Reibo Lei, Dr. Jingxue Guo and Prof. Zhanhai Zhang Polar Research Institute of China (PRIC) 学宇 李志学 学学 Dr candidate:yu Yang, Prof. Zhijun Li and Dr. Peng Lu, Dalian University of Technology (DUT) 学学学 学学碇 Ms. Qinghua Yang, Prof. Huiding Wu National Marine Environmental Forecasting Centre (NMEFC) 石立学 王学茂 Dr. Lijian Shi, Prof. QimaoWang National Satellite Ocean Application Service Centre (NSOAS) 30 October -2 November, Chinese-Finnish Arctic Seminar