The real time monitoring of storms is important for the management and prevention of flood risks. However,
in the southeast of Spain, it seems that the density of the rain gauge network may not be sufficient to
adequately characterize the rainfall spatial distribution or the high rainfall intensities that are reached during
storms. Satellite precipitation products such as PERSIANN-CCS (Precipitation Estimation from Remotely
Sensed Information using Artificial Neural Networks - Cloud Classification System) could be used to complement
the automatic rain gauge networks and so help solve this problem. However, the PERSIANN-CCS product has
only recently become available, so its operational validity for areas such as south-eastern Spain is not yet known.
In this work, a methodology for the hourly validation of PERSIANN-CCS is presented. We used the rain gauge
stations of the SIAM (Sistema de Información Agraria de Murcia) network to study three storms with a very
high return period. These storms hit the east and southeast of the Iberian Peninsula and resulted in the loss of
human life, major damage to agricultural crops and a strong impact on many different types of infrastructure.
The study area is the province of Murcia (Region of Murcia), located in the southeast of the Iberian Peninsula,
covering an area of more than 11,000 km2 and with a population of almost 1.5 million. In order to validate
the PERSIANN-CCS product for these three storms, contrasts were made with the hyetographs registered by
the automatic rain gauges, analyzing statistics such as bias, mean square difference and Pearson’s correlation
coefficient. Although in some cases the temporal distribution of rainfall was well captured by PERSIANN-CCS,
in several rain gauges high intensities were not properly represented. The differences were strongly correlated
with the rain gauge precipitation, but not with satellite-obtained rainfall. The main conclusion concerns the
need for specific local calibration for the study area if PERSIANN-CCS is to be used as an operational tool for
the monitoring of extreme meteorological phenomena.
|