integration of the major spatial data. Both stand-alone and real-time hydrological
models are supported by geo-information systems such as HEC-1, HEC-RAS
(Knebl et al. 2005), TOPMODEL (Beven et al. 1984), and SWMM (Rossman
2010). Real-time flood monitoring or flood warning system requires continuous
integration and processing of spatial and nonspatial data. Advance GIS brings the
opportunity to collect, integrate, and process real-time data. A system needs to
integrate and process continuous rainfall monitoring, real-time soil moisture
updating system, and surface runoff measuring system in a sophisticated hydrological system for real-time flood monitoring (Al-Sabhan et al. 2003; Hossain 2006).
There are many examples of GIS-based flood monitoring systems such as the
International Flood Network (IFnet), European Flood Alert System (EFAS), and
NRT Global Flood Mapping. Most of these systems use state-of-the-art GIS systems
for flood monitoring integrating meteorological data, rainfall, weather forecast,
observed discharge, and soil condition. Another important aspect of an efficient
flood monitoring system is the user interface. These user interfaces allow users’
visualization as well as analysis of the spatial distribution of model parameters and
simulation through a variety of tools. The advanced GIS and Web system are making
cost-effective real-time flood monitoring a reality.
16.2.4 Event and Duration of the Flood
The magnitude of the flood event mainly depends on the flood extent, flood duration,
and flood frequency. Remote sensing and GIS have also been aiding in the flood
frequency and duration analysis. Dartmouth Flood Observatory (DFO) provides
daily flood data extracted from MODIS for the globe at a 10-degree grid.
RF-CLASS system is providing the duration and frequency of the flood event
extracted from daily flood data from DFO. The daily flood data are processed and
stored for each calendar year, and flood events are extracted based on the certain
consecutive flooding days for a given area. Flood duration is calculated based on
each event.
Flood duration information might contain noises from this automatic processing.
This noise includes many single-day events which might be caused by cloud
contamination of MODIS data or the tidal effect along the shoreline. Noise reduction
techniques can be applied as post-processing techniques. A three-day window filter
might be applied to make a composite flood event to avoid the noise problem
(Fig. 16.6). A rule might be if all days within the window have flood, then this
window is considered as flood otherwise not. This process will remove noise like
one-day or two-day flood. The flood frequency and the number of events in a given
unit of time can also be extracted in this way.
16 Flood Monitoring and Crop Damage Assessment
331
models are supported by geo-information systems such as HEC-1, HEC-RAS
(Knebl et al. 2005), TOPMODEL (Beven et al. 1984), and SWMM (Rossman
2010). Real-time flood monitoring or flood warning system requires continuous
integration and processing of spatial and nonspatial data. Advance GIS brings the
opportunity to collect, integrate, and process real-time data. A system needs to
integrate and process continuous rainfall monitoring, real-time soil moisture
updating system, and surface runoff measuring system in a sophisticated hydrological system for real-time flood monitoring (Al-Sabhan et al. 2003; Hossain 2006).
There are many examples of GIS-based flood monitoring systems such as the
International Flood Network (IFnet), European Flood Alert System (EFAS), and
NRT Global Flood Mapping. Most of these systems use state-of-the-art GIS systems
for flood monitoring integrating meteorological data, rainfall, weather forecast,
observed discharge, and soil condition. Another important aspect of an efficient
flood monitoring system is the user interface. These user interfaces allow users’
visualization as well as analysis of the spatial distribution of model parameters and
simulation through a variety of tools. The advanced GIS and Web system are making
cost-effective real-time flood monitoring a reality.
16.2.4 Event and Duration of the Flood
The magnitude of the flood event mainly depends on the flood extent, flood duration,
and flood frequency. Remote sensing and GIS have also been aiding in the flood
frequency and duration analysis. Dartmouth Flood Observatory (DFO) provides
daily flood data extracted from MODIS for the globe at a 10-degree grid.
RF-CLASS system is providing the duration and frequency of the flood event
extracted from daily flood data from DFO. The daily flood data are processed and
stored for each calendar year, and flood events are extracted based on the certain
consecutive flooding days for a given area. Flood duration is calculated based on
each event.
Flood duration information might contain noises from this automatic processing.
This noise includes many single-day events which might be caused by cloud
contamination of MODIS data or the tidal effect along the shoreline. Noise reduction
techniques can be applied as post-processing techniques. A three-day window filter
might be applied to make a composite flood event to avoid the noise problem
(Fig. 16.6). A rule might be if all days within the window have flood, then this
window is considered as flood otherwise not. This process will remove noise like
one-day or two-day flood. The flood frequency and the number of events in a given
unit of time can also be extracted in this way.
16 Flood Monitoring and Crop Damage Assessment
331