 |
|
 |
|
 |
 |
 |
 |
 |
 |
 |
|||||
|
|
|
Polar View snow monitoring services are provided in three geographical regions: 1) the Baltic Sea Area by the Finnish Environment Institute (SYKE), 2) Central Europe by Vista, and 3) Northern Eurasia by the Finnish Meteorological Institute.
The lowlands in Northern Europe and in the Baltic Sea drainage basin are characterized by several accumulation and melting periods until a permanent seasonal snow cover appears. In spring, the melting process may cause rapid changes on water storages which leads to flooding. The boreal zone is characterized by a thick seasonal snow pack. As a temporary water storage, snow is a valuable source of energy when a large volume of water is released during the melting process. On the other hand, sudden runoff raises a risk of flooding in certain areas. Service Description The snow service provided by Finnish Enviroment Institute (SYKE) comprises mapping of regional fraction of snow covered area (SCA) in the Baltic Sea drainage area. This nearrealtime service is based on daily Terra/MODISimagery; an automated processing system is used to produce the SCA for 5km´5km grid cells over the target area. The processing line includes radiometric and geometric corrections, cloud masking and the actual SCAestimation. The following figure shows Snow Maps compiled for 24th, 26th, and 27th March 2007.
Methodology SCA is estimated applying a semiempirical reflectance model developed at SYKE. In this model, the reflectance from a target area is expressed as a function of SCA. Average effective forest canopy transmissivity for each grid cell and generally applicable reflectance values for wet snow, snowfree ground and dense forest canopy are applied as model parameters. Through transmissivity, the forest coverage for each unit area is individually determined. The key idea is that the effective transmissivity is estimated using the Earth observation data, and includes even using data similar to that employed in the actual SCA estimation. Neither auxiliary land cover data nor forest data is needed; only water mask is necessary. This type of approach enables operational snow mapping in an extensive and heterogeneous area such as the boreal forest zone. Service Validation The SCA product is validated against snow course network (including SCA observations) covering the whole Finland. A snow course is a 24 km long trail on the ground representing the different terrain and land cover types typical for the locality. Snow depth, snow density and SCA are observed in several points along the course. MODISderived SCA maps from year 20042005 Have been validated with these data. In addition to this, online validation will be carried out during a monitoring period. This validation procedure used SCA data from dailymade observations at weather station (200 stations in Finland). Area of Interest and End Users In Polar View, almost the whole of the Baltic Sea drainage basins will be covered. Hence also areas outside boreal zone are included. The algorithm is planned to work in areas where seasonal snow cover stays at least several weeks and then gradually melts off. The figure below depicts the target area evolution during Polar View. End users (with SLA) for the first year will be SYKE hydrological services division (responsible for national hydrological forecasting for the entire Finland) and Kemijoki ltd (hydropower company). The target area for SCA monitoring by SYKE in Polar view
Central Europe (VISTA) The lower mountainous regions in Central Europe are characterized by the frequent temperature changes causing the snow pack to accumulate and melt off several times each winter season. Rainfall combined with snowmelt is the most critical situation for the origin of floods. Remote Sensing provides the desired spatial characterizations of snow properties to control and update hydrological models. This service provides frequent updates on snow cover relating to relevant information for flood forecasting and early warning in Central Europe. The snow-monitoring service consists of two different processing chains for optical and microwave data.
Technical Details The service is delivered using two different processing chains for optical and SAR data implemented at VISTA.
Remote Sensing Resources The selection of sensor is based on own experiences and existing developments. For the daily update of the snow covered area (if cloud situation permits) the NOAA AVHRR sensor is used. Several overflights from the sensors provide daily repeatedly datasets. The direct reception of data (using the satellite-receiving station in Munich) allows a very fast generation of products. Spatial resolution of 1km is sufficient for the snow cover products in Central Europe. For wet snow detection, in connection with melting snow, only SAR sensors are appropriate. Covering a swath from about 400km the WideSwath datasets from ENVISAT ASAR are most favourable for larger watersheds. Since data access using the ESA rolling archive is possible, a Near-Real-Time processing is possible. In contrast to the daily optical datasets (NOAA), the availability of ASAR data is limited to precursory data orders. For urgent cases (e.g. when the flood situation is critical) a priority programming of the data acquisition is possible. Service Validation and Quality The service has been validated through continuous application over 18 months for NOAA-AVHRR data. A service based on ENVISAT data was tested in the summer of 2004 and processing provided operationally in winter season 2004/2005. Products are regularly delivered within 1 hour after reception of NOAA-AVHRR data and 3 hours after ASAR data availability. For the winter season 2002/2003 the quality of the snow cover products was 95.6% based on validation with 14,000 measurements. The users will provide an intensive database of snow cover in-situ measurements (approx. 100 stations) to allow a long-term validation. Overview on VISTA snow service for Southern Germany / Central Europe (Source: VISTA)
Size and location of the target areas for this service are determined by the watersheds, for which the end users provide a flood forecast. Presently, the end users are the Flood Forecast centres of Baden-Württemberg (http://www.hvz.baden-wuerttemberg.de) and Rheinland Pfalz, (http://www.hochwasser-rlp.de) responsible for the Upper Rhine, Neckar , and Mosel.
Finnish Meteorological Institute together with Finnish Environment Institute and Helsinki University of Technology have developed an operational Northern Eurasian wide snow monitoring system using satellite observations assimilated with snow depth measurements made on synoptic weather stations across the area. This service provides rough scale map of aerial extent, snow depth and snow water equivalent over the Northern Eurasia. Service description The snow service provided by FMI comprises mapping of snow depth and snow water equivalent over the northern Eurasia (between latitudes 50-85 and longitudes 0-180). This service is based on assimilation of EOS-Aqua/AMSR-E and in-situ observations and it is provided every second week. The map is provided in 0.25 degree grid. The following image shows snow depth in March 8th, 2007.
Methodology The method used in SWE/SD data production is an assimilation approach, in which the SWE/SD level for a given location is obtained by combining space-borne microwave radiometer data with the SWE/SD estimates interpolated from distributed ground data, namely synoptic weather stations. The Bayesian approach developed and applied utilises the forward modelling of radiometer observations as a function of snow pack and forest characteristics. It takes into account the spatial and temporal accuracy characteristics of the employed brightness temperature model, and moreover, the estimated spatial accuracy of discrete in-situ based SD estimates (interpolated values). The applied brightness temperature model is a semi-empirical radiative transfer approach-based HUT Snow Emission Model. The accuracy of SWE retrieval is expected to be better than 20 mm. Figure below shows a typical setup of weather station locations giving snow depth information. The number of in-situ observations vary between 350 and 550, in this case there are 515 stations providing information.
Service validation The independent reference data used for accuracy assessment include ground-based snow course observations around the region of Finland obtained from the Finnish Environment Institute (SYKE) and snow observations from synoptic weather stations in Northern Eurasia. These observations provide regional (corresponding to areas of few km2) estimates on SD, SWE and snow density. The Sodankylä region in northern Finland is used as an intensive ground validation site providing continuous in situ reference data (Arctic Research Centre of the Finnish Meteorological Institute, FMI-ARC). End Users The end users are research users working on the field of climate modelling and hydrology. Currently 3 SLAs have been signed. |
||||||||||||||||||||||||||||||
 |