Evaluation of Rainfall Estimation Derived From Commercial Interactive DVB Receivers Using Disdrometer, Rain Gauge, and Weather Radar

Accurate measurement and monitoring of precipitation is crucial for many applications, such as flood and drought risk assessment and management. Conventionalmeteorological devices for estimating precipitation (i.e., rain gauges, disdrometers, active and passive remote sensors, be they ground-based, spaceborne, or airborne) have their own strengths and weaknesses. The latter are often related to time and space resolution, coverage, and cost. In the last two decades, several studies have been carried out to exploit opportunistic signals of terrestrial microwave communication links to improve precipitation estimation capability. This study describes and evaluates a method to retrieve rainfall rate from the signal-to-noise ratio obtained from commercial interactive digital video broadcasting (DVB) receivers [referred as to smartlow-noise blocks (smartLNBs)]. During a 1-year measurement campaign carried out in Tuscany (Italy) with purposely deployed instruments, the precipitation values estimated from a set of SmartLNBs were compared with measurements from co-located rain gauges and disdrometer. The normalized mean absoluteerror (NMAE) and root-mean-square error (RMSE) obtained comparing the total cumulative precipitation from a SmartLNB and a disdrometer are 48.8% and 7.46 mm, respectively.Encouraging results also come from comparing the totalprecipitation amounts as measured by the SmartLNBs and bythe rain gauges, with values of NMAE (respectively, RMSE)ranging between 44% and 82% (respectively, 5.2 and 11.5 mm).