Slide 1

NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria CL4: EMS2015-144 Simulated versus Satellite Retrieval Distribution Patterns of the Snow Water Equivalent over Southeast Europe Hristo Chervenkov National Institute of Meteorology and Hydrology Bulgarian Academy of Sciences 1784 Sofia Tsarigradsko Shose 66 Slide 2 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria Outline of the presentation 1.Preface Motivation and Difficulties in the Objective Snow Cover Assessment Studies 2.Sensing Principle and Snow Water Equivalent by Microwave Radiometry. Kinds of Satellite-derived Snow Water Equivalent Product 3.Methodology, Performed Calculations and Comparisons 4.Results and Comments 5.Summary and Conclusions 6.Acknowledgement Slide 3 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria 1. Preface Motivation and Difficulties in the Objective Snow Cover Assessment Studies The snow is a very important component of the climate system which controls surface energy and water balances and is the largest transient feature of the land surface. It has an effect on atmospheric circulation through changes to the surface albedo, thermal conductivity, heat capacity and aerodynamic roughness. The snow properties of surface water storage control the availability of water in many ecosystems and to a sixth of the worlds population. The various properties characterizing snow are highly variable and so have to be determined as dynamically active components of climate. To fully understand global snow water trends, the most fundamental metric to assess is the snow water equivalent (SWE), with snow depth (SD) a close second. However, on large spatial scales the properties of snow are not easily quantified either from modelling or observations. For example, station based snow measurements (usual scarce and infrequent) often lack spatial representativeness, especially in regions where the topography, vegetation and overlaying atmosphere produce considerable heterogeneity of the snow-pack distribution. Since neither observations (ground measurements or satellite retrievals) nor models alone are capable of providing enough adequate information about the time space variability of snow properties, it becomes necessary to combine their information. Main aim of the presented study is to compare the simulated for relatively long (more than 10 years) by regional climate model (RCM) SWE distribution patterns with the satellite retrievals aiming to reveal systematic disagreements for the selected geographical domain of interest. Slide 4 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria 2. Sensing Principle and Snow Water Equivalent by Microwave Radiometry. Kinds of Satellite-derived Snow Water Equivalent Product (1 from 3) The SWE is measured by passive MW-radiometers carried by polar orbiting platforms since 1988. The conical scanners provide images with constant zenith angle, that implies constant optical path in the atmosphere and homogeneous impact of the polarisation effects (see fig. 1). Also, conical scanning provides constant resolution across the image, though changing with frequency. The soil under the snow-pack emits stronger MW-signal and these microwaves are scattered and attenuated by the snow as they travel upwards. The attenuation depend on the wavelength and the snow-pack properties, in particular the amount of snow (SWE) grain size, snow density, presence of ice lenses and the amount of liquid water. The shorter the wavelength the greater the scattering for a given set of snow-pack properties, thus the difference between the signal at two wavelengths should be related to the amount of snow. This has been theoretically derived and tested over certain areas to give the following equation: Fig. 1 Geometry of conical scanning for AMSR-E (EUMETSAT HSAF courtesy) where Tb 18H and Tb 36H are the brightness temperatures at 18 and 36GHz horizontal polarisation correspondingly, and a is constant. This can be converted to SWE by multiplying by snow density (assumed to be 300 kg.m -3 ). This is known as the Chang algorithm and forms the basis of the AMSR-E and SSM/I inversions. Although it has been validated to some extend, this approach is known to suffer from a number of issues: dependence from the grain size (The larger the snow grains the greater the scattering for a given SWE) dependence of liquid water amount (it affects the snow-packs dielectric constant, increasing attenuation) Slide 5 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria dependence from the vegetation cover (A forest correction factor, c, where f is the fractional forest cover within the pixel, according the equitation above) saturation in deeper snow covers (SWE>120 mm). Therefore accuracies are reduced in deep snow. Shallow snow-packs cause little scattering and can be missed by microwave sensors; a minimum detectable SWE of 15 mm has therefore been suggested. All four products cover at least the northern hemisphere at 25 km spatial resolution although the standard Globsnow product masks the mountainous areas. 2. Sensing Principle and Snow Water Equivalent by Microwave Radiometry. Kinds of Satellite-derived Snow Water Equivalent Product (2 from 3) Tabl. 1 Main information about the most popular four SWE satellite products Slide 6 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria 2. Sensing Principle and Snow Water Equivalent by Microwave Radiometry. Kinds of Satellite-derived Snow Water Equivalent Product (3 from 3) The last two products are actually the result of an assimilation process (see fig. 2). The basic (very sparse) ground network of stations performing snow depth observation provides a first guess field that is converted into MW brightness temperatures by an HUT emission model that also accounts for forests. The assimilation process forces the first guess field to optimally match the retrieved by the sensor brightness temperatures. The unrestricted and unlimited access to long- term (>35 years) validated SWE data-sets (daily, weekly and monthly aggregated) makes the Globsnow product very suitable for intercomparisson studies as the presented here. The most modern passive MW radiometer available is used (SMMR 1978- 1988, SSM/I 1988-2002, AMSR-E 2002-2011 and SSM/I since October 2011). Ground measurements of snow depth are taken from the ECMWF WMO weather stations. Fig. 2 Flowchart of the processing chain of the Globsnow product (EUMETSAT HSAF courtesy) Slide 7 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria Fig. 3 Visualization of the Globsnow product for January 2008 as example for month with significant snow-pack over the domain of interest (see further description) Slide 8 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria Fig. 4 Visualization of the Globsnow product for January 2011 as example for month with shallow snow-pack over the domain of interest (see further description) Slide 9 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria 2. Methodology, Performed Calculations and Comparisons (1 from 3) The regional climate model RegCM in nutshell The RCM system RegCM is maintained in the International Centre of Theoretical Physics (ICTP) in Triest, Italy (http://gforge.ictp.it/gf/project/regcm/). It can be applied to any region of the World, with grid spacing of up to about 10 km (hydrostatic limit), and for a wide range of studies, from process studies to paleoclimate and future climate simulation. RegCM is a 3-dimensional, primitive equation RCM with dynamical core based (version 2 and later) on the hydrostatic version of the NCAR-PSU Mesoscale Model 5 (MM5). B-Arakawa staggered horizontal grid, (terrain following) vertical coordinate system, FDM for the PDEs Radiative transfer package: from the Community Climate Model v. 3 (CCM3). The large-scale cloud and precipitation computations are performed by Subgrid Explicit Moisture Scheme (SUBEX). The model is flexible (in particular by selection of different ICBC- datasets, parameterizations/closure assumptions and etc.), portable and easy to use. Fig. 5 Simplified flow-chart of RegCM and data flow in the system Slide 10 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria 3. Methodology, Performed Calculations and Comparissons (2 from 3) Computational setup The model domain is centered over Bulgaria, includes the whole Balkan peninsula, as shown in Fig. 6 and consists of 7277 20 km20 km gridcells. The simulation period is from 1 th November till 31 th March (hereafter: winter) for 14 consecutive years between 2000 and 2014. Four different RegCM-configurations are used: - ICBC ERAInterim (hor. res.: 1,51,5), land surface model BATS (EIN15/BATS run) - ICBC NCEP/NCAR RII (hor. res.: 2,52,5), land surface model BATS (NNRP2/BATS run) - ICBC ERAInterim (hor. res.: 1,51,5), land surface model CLM (EIN15/CLM run) - ICBC NCEP/NCAR RII (hor. res.: 2,52,5), land surface model CLM (NNRP2/CLM run) Model output is the gridded distribution of the SWE on 6-hourly basis (i.e. at 00, 06, 12 and 18 UTC) Model version: RegCM4.4-rc8 (Parallelized (MPI) version) Convective parameterization: Grell (with closure scheme of Arakawa & Schubert) PBL parameterization: Holtslag Nonconvective moisture and cloud processes (incl. precipitation) parameterization: SUBEX S: 64bit Linux (Slackware distribution), graphical environment KDE. netCDF-file manipulation: ncdump, cdo Automatization: bash-shell scripts Post processing: FORTRAN90 (gfortran compiler) Visualization&graphics: Ncview, GrADS, VERDI, GS SURFER&GRAPHER Slide 11 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria Fig. 6 Example of (processed) RegCM output (Average SWE (unit: kg.m -2 ) for February 2012). Due to its physical and hydrological consistency, the average SWE can be accepted as most informative snow climate index. Slide 12 NATIONAL INSTITUTE OF METEOROLOGY AND HYDROLOGY BULGARIAN ACADEMY OF SCIENCES 15 th EMS Annual Meeting & 12 th European Conference on Applications of Meteorology (ECAM) | 0711 September 2015 | Sofia, Bulgaria 4. Results and comments (1 from 2) The model output (only results for the EIN15/BATS run are presented) are compared with the monthly aggregated Globsnow SWE products for the month of January, traditionally treated as climate representative for the whole winter. The comparison is not possible if: The SWE value in given Globsnow pixel are