Forecasting Rain: Radars For Estimating Rainfall Rates

ScienceDaily (Nov. 17, 2008) — To be effective, flood warning systems use rainfall data available in real time. These data come from the ground observation network and estimations made based on the national network of climate radars operated by Météo France.Today, mountain zones are only partially covered by this rain detection technology. Within the INTERREG project, a new generation of radars is being tested by Cemagref in the Var department, a mountainous region with a high flood risk. The radar is currently located in the countryside immediately inland from Nice.

Measuring the intensity of rain as it is falling is indispensable to anticipating rapidly rising waters and reacting to the associated flood risk. To the classical rain gauges installed throughout the area, new radar technologies were added in the 1990s to detect rain and measure the accumulation of precipitations in real time.

Today, Météo France has a national network of 24 weather radars available within approximately 100 km. However, all regions in France are not covered by this mesh, in particular the mountain zones where the relief masks downstream rain zones by creating an obstacle to wave displacement. Within the FRAMEA project, a new radar technology developed by Novimet is being tested at the Aix-en-Provence Cemagref. The experiments conducted in the Maures massive have proven to be highly promising.

More compact and less expensive radars

The radars used at the beginning to monitor planes flying overhead were extended to the detection and quantification of precipitations. Large-scale radars, 6–8 m in antenna diameter, are used today in weather stations located in the plains. In mountain zones, the number of radars must be multiplied in relation to the relief, which requires smaller and less expensive models. The new Hydrix radar responds to these demands. However, by reducing the diameter of the parabolic antenna to 1.5 m, the wave frequency must be boosted, which increases the attenuation of waves during their displacement.

To compensate this signal attenuation effect, a profiling algorithm (ZPHI) is used. Finally, the radar operates in double polarization, which provides information on the size of the rain drops and estimates precipitations without resetting ground network observations. Today, in a doctoral dissertation supervised by both Cemagref and the firm Novimet, this new radar technology is being tested in the Var department, a mountainous region that experiences very intense flash floods.

Results that are coherent with ground readings

The Hydrix is installed near Réal Collobrier, Cemagre’s instrumented research catchment, located in the Maures massif. The total rainfall in autumn 2006 calculated by the radar was compared to the rain gauge readings on the ground and to the accumulation estimated by one of the nearby radars belonging to the Météo France network. Within a 60- to 80-km radius, the data supplied by the radar were in coherence with the quantities of rain collected on the ground. In addition, the algorithmic signal processing retransmitted rain gauge data in real time that were as good quality as the data sent by the classical radar managed by Météo France. Today, the research is continuing so as to integrate the rain gauge data supplied by the radar into existing rainfall-runoff models.

By converting rainfall into runoff, these mathematical tools can calculate the runoff of rivers at the outlet of a catchment. These rainfall and runoff data then feed the flood warning systems, such as the Aiga system developed by Cemagref and Météo France in 2005. By completing the existing radar network, the Hydrix technology will contribute to the extension of the flood warning system over the entire area, including mountainous zones.


Adapted from materials provided by Cemagref.
Email or share this story:

Reblog this post [with Zemanta]
E-mail me when people leave their comments –

You need to be a member of SWMM 5 or SWMM or EPASWMM and SWMM5 in ICM_SWMM to add comments!

Join SWMM 5 or SWMM or EPASWMM and SWMM5 in ICM_SWMM