Use of cell phone networks for rainfall estimation in the city of Luxembourg


CALL: 2008

DOMAIN: SR - Water Resources under Change


LAST NAME: Pfister




KEYWORDS: Enhancement of rainfall measur

START: 2009-04-01

END: 2011-03-31


Submitted Abstract

Continuously increasing computational resources have led to substantial progress in the identificationand quantification of uncertainties in hydrological modelling. Moreover, new investigation techniques ofhydro-climatological processes provide innovative and complementary data that allow both new modelconcepts and a substantial improvement of the predictive power of hydrological models throughreductions of their equifinality potential. In the chain of environmental process monitoring,conceptualisation and modelling, rainfall remains one of the main sources of uncertainty. Rainfallcharacteristics, such as intensity and spatial distribution, are major input variables in hydrologicalmodels. However, frequent approximations in rainfall estimations are inherent to natural and technicalconstraints:(i) The high spatial variability of rainfall (especially during convective events) that cannotbe fully captured by traditional raingauges.(ii) The influence of external forcings on standard raingauges (e.g. wind blasts, evaporationof collected rainfall).(iii) The difficulties in measuring solid precipitation (snow, hail). These shortcomings areamplified in densely built areas (high site-specificity of monitoring sites, due to urban structures).Over the past two decades, weather radars have helped to partially overcome these uncertainties inprecipitation measurements. While weather radars provide quite accurate spatial representations ofrainfall, they still experience difficulties in providing accurate rainfall intensities and amounts. Due tothese limitations, weather radar rainfall data quality can be improved to a certain extent by ground-truthvalidations based on rainfall measurements provided by point raingauge networks. Site-specific effects,such as wind turbulences induced by buildings, increase the uncertainties in urban rainfallmeasurements. Microwave signals used in modern telecommunication networks have been recentlyidentified as a tool offering the possibility to improve the quantitative accuracy of point raingaugenetworks and weather radars, especially with respect to urban hydrology applications.This project thus aims to test the potential offered by the cell phone communication technology toincrease the precision of the rainfall input signal used in urban hydrology modelling applications. Basedon a technological collaboration with the P&T (L) and the IRM (B), the project is structured in twocomplementary topics:(i) The first topic consists of a fundamental research activity that will investigate thepotential of cell phone communication networks for an improvement of quantitative rainfallestimations over the city of Luxembourg.(ii) The second topic investigates the qualitative increase of urban hydrological modeloutputs due to the integration of the microwave signal technique for rainfall measurements.Of clearly applied character, the second topic of the project will also include information exchangeactivities in the framework of a project observer group, consisting of TR Engineering and the City ofLuxembourg, as well as the Ministry of Interior (Administration des Services de Secours &Administration de la Gestion de l’Eau). The high percentage of sealed surfaces generates both very highsurface runoff coefficients (risk of flash floods), and a possible water contamination through first-flushprocesses. Future quantitative and qualitative management techniques of urban waters need to rely onboth accurate rainfall measurements and modelling performances; needs that are to be investigated andaddressed via the RAINCELL project.

This site uses cookies. By continuing to use this site, you agree to the use of cookies for analytics purposes. Find out more in our Privacy Statement