1. © ECMWF GCW contribution from ECMWF Use of snow observations in NRT Analysis, Reanalysis and Verification prepared by Patricia de Rosnay, Gianpaolo Balsamo, Paul Poli

2. © ECMWF Snow observations used for operational data assimilation at ECMWF Status in January 2015

3. © ECMWF Snow components at ECMWF Snow Model: Component of H-TESSEL Single layer snowpack - Snow water equivalent SWE (m), ie snow mass - Snow Density ρ s, between 100 and 400 kg/m 3 - Snow Albedo between 0.5 and 0.85 Observations: - Conventional snow depth data: SYNOP and National networks - Snow cover extent: NOAA NESDIS/IMS daily product (24km & 4km) Data Assimilation: -Optimal Interpolation (OI) in oper IFS - Analysed variable: SWE, ρ s de Rosnay et al., Survey of Geophysics 2014 Prognostic variables Drusch et al., JAM, 2004 ; de Rosnay et al., SG 2014 de Rosnay et al., ECMWF Res. Mem. R48.3/PdR/1028 2010, and ECMWF Res. Mem. R48.3/PdR/1139 2011 Balsamo et al., JHM, 2009 and Dutra et al., JHM 2010

4. © ECMWF 2015 01 01 at 06UTC Additional Snow data Additional data from national networks (7 countries): Sweden (>300), Romania(78), The Netherlands (33), Denmark (43), Hungary (61), Norway (183), Switzerland (332).  Dedicated BUFR (2011) (de Rosnay et al. ECMWF Res. Memo, R48.3/PdR/1139, 2011) Snow Depth (cm) Available on the GTS (Global Telecommunication System) Snow SYNOP and National Network data Snow Observations SYNOP

5. © ECMWF GTS SYNOP Snow depth availability Operational snow observations monitoring (TAC only at the moment): http://old.ecmwf.int/products/forecasts/d/charts/ monitoring/conventional/snow/ Status in January 2014 Gap in USA, China and southern hemisphere NRT data exist and is available (more than 20000 station in the US) But it is not on the GTS for NWP applications. 2015 01 01 at 12UTC  WMO GCW Snow Watch initiative to improve in situ snow depth data access (NRT and rescue), Brun et al 2013  BUFR template (WMO approved in April 2014) or SYNOP report  WMO Members States encouraged to put their snow depth data on the GTS Snow Observations

6. © ECMWF GTS SYNOP Snow depth availability 10 January 2015: 14h availability on the GTS of all in situ observations types: Coverage from all obs type: TAC SYNOP, BUFR SYNOP and national data Snow Observations BUFR SYNOP  observations from China now available on the GTS. (de Rosnay et al, « BUFR SYNOP snow depth observations », ECMWF Research Memorandum RD14-440, Nov. 2014) (Snow depth in m)

7. © ECMWF Snow observations used for reanalysis, verification, model development at ECMWF Status in January 2015

8. © ECMWF ERA-Interim/Land snow verification over European domain is performed thanks to the SYNOP-GTS observations that enable to demonstrate the added value of the land reanalysis ERA-Interim/Land: snow performance Evolution of snow depth errors (mean RMSE/BIAS in Europe) of ERA-Interim and ERA-Interim/Land compared to USSR in-situ observations over a recent year (2010). More than 600 observations each day Balsamo et al. (2015 HESS)

9. © ECMWF ERA-Interim/Land snow density verification on Russia domain is performed thanks to the non-GTS observations gathered by ROSHYDROMET ERA-Interim/Land: snow performance Evolution of snow density evolution of ERA-Interim (left) and ERA-Interim/Land (right) compared to USSR in-situ observation averaged between 1979 and 1993. Balsamo et al. (2015 HESS)

10. © ECMWF Snow at ECMWF: Summary 2009 20102011 20122013 2014 Snow Model. Liq. Water. Density. Albedo. Fraction Dutra et al., JHM 2010. OI. 4km IMS. Obs preproc/QC. IMS latency/acquisition. Additional in situ obs. New BUFR template. WMO/SnowWatch action. IMS data assimilation. obs error revision de Rosnay et al., Res Memo 2010, 2011 Brun et al., Snow Watch 2013 de Rosnay et al., Surv. Geophys 2014 Balsamo et al HESS 2015 Snow Obs and DA Snow Model & DA. Multi-layer model. Snow cover Fract. BUFR SYNOP. RT modelling. Snow COST action (ES1404) ECMWF Land Data Assimilation System: https://software.ecmwf.int/wiki/display/LDAS/LDAS+Home

11. © ECMWF A suggested action list (an ECMWF perspectives) 1)NRT snow observations are a top priority for initialization of operational NWP models and the inclusion in the GTS network made them accessible to operational NWP Centres 2)Error in the format and/or missing information (e.g. failure to report zero snow) prevent the use of the in-situ observation (e.g. TAC-BUFR conversion has some cases of erroneous conversion) even if observations are on the GTS 3)Retrospective observations are extremely important for reanalysis, model verification, and model development and there is not a well defined system in place to receive observations that do not succeed entering the GTS. 4)Survey actions in other projects such as in the reanalysis context can be linked to GCW to optimally share information (e.g. CORE-CLIMAX coordination meeting on snow, see executive summary on GCW site and the presentations at: http://www.coreclimax.eu/?q=Snowhttp://www.coreclimax.eu/?q=Snow