Czech Republic has experienced different types of floods in recent years, such as summer floods due to the long time precipitation (July 1997, August 2002), flash floods (July 1998, June 2009) but also significant floods from snow melt (March 2000, March and April 2006). Wide range of possible types of floods brings also very wide and complex demand on the input data and hydrological methods and models that are used.
Czech Hydrometeorological Institute (CHMI) regularly evaluates quantity of snow in the Czech Republic. Unfortunately, there is always some degree of uncertainty in the ground observations and in the modelling of the the situation outside the area covered by the meteorological stations. Especially when there is not snow cover over the whole territory of the country, uncertainty of the snow amount estimate may reach even several tens of percents.
The particular problem is the amount of snow in forested areas under the tree cover. Snow accumulates slowly at the beginning of the winter season and its melt is also slower due to the vegetation cover. Snow remains in the forest areas for a longer time comparing to the neighbouring open areas where most of meteorological stations are located. This effect significantly contributed to the extent of flood in the spring of 2006 for instance. Observations from meteorological stations didn’t report any snow, but there was still snow hidden in forest which continually melted and caused floods.
Information obtained from the ground stations is naturally not sufficient. In this case Earth Observation satellites may help significantly as they bring the spatial information. Satellite imagery can also provide additional data useful for predicting of floods, such as information about the properties of snow melting in surface layer of the snow cover. In the summer time different type of information can be derived, for instance the soil water content and soil saturation help to predict responses to coming rainfalls.
These facts, among others, were stimulation to come up with FLOREO project (“Demonstration of ESA Environments in support to FLOod Risk Earth Observation monitoring,”) having the aim to bring more reliable data about the current status of snow cover and additional important ground parameters to improve flood prediction services.
FLOREO turns into the first operational phase now. Its main focus is to build complex system of landscape monitoring that provides important data for flood prediction services. CHMI currently integrates the project results into their processes and believes that the entire system will be fully in operation during this year.
Nowadays the snow cover and its changes are monitored using several satellite systems, such as Terra MODIS imagery with spatial resolution of 250 m or Envisat ASAR radar imagery with spatial resolution of 150 m. The advantage of MODIS data is daily acquisition with full coverage of the area of interest and more straightforward data processing but clouds can hide the land surface. In that case radar data can be processed independently of the cloud cover but only with the period of acquisitions about 10 to 14 days. FLOREO combines both optical and radar imagery.
Inevitable part of the project is the information infrastructure providing user-friendly access to selected datasets and analytical tools via internet mapping services. Sprinx Systems, the project leader, is responsible for the design and operation of the whole infrastructure system. Gisat delivers the concept and processing algorithms that are then programmed by Sprinx Systems to build a complex system including control procedures, data visualization and provision of the results to the end user (CHMI).
See the FLOREO web site to find more information about the project activities or visit the public version of the map portal.
Watch the video demonstrating the main features of the FLOREO map portal: