Abstract:
High spatio-temporal variability is a characteristic of extreme rainfall. In
mountainous regions like the Tropical Andes, where intricate orography and
mesoscale atmospheric dynamics greatly impact rainfall systems, this particularly
holds for mountain areas like the Tropical Andes. Thus, the absence of
operational rainfall monitoring networks with high spatio-temporal resolution
has imposed difficulties for a proper analysis of extreme rainfall events in the
Ecuadorian Andes. Nowhere, we present our improved knowledge on rainfall
extremes based on newly available rainfall radar data of this region. In our
study we employ a clustering approach to identify types of extreme rainfall
events and analyze their spatio-temporal characteristics. Based on 3 years of
data obtained from an X-band scanning weather radar data, the study was conducted
in the southern Ecuadorian Tropical Andes at 4450 m a.s.l. By applying
a rainfall threshold, 67 extreme rainfall events were selected. The rainfall characteristics
of each extreme rainfall event, such as the amount of rain, its duration,
its hour, and month of occurrence were determined and used as input
variables of a k-means clustering analysis to group the events into different
classes. The result revealed three main classes of extreme rainfall events. The
first class is characterized by highest rain intensity and lowest duration. The
second class is characterized by its month of occurrence, during the first
5 months of the year. The third class showed lowest rain intensity and highest
duration mainly occurred at higher elevations. The typology of events
advances our understanding of the spatio-temporal characteristics of extreme
rainfall in the Tropical Andes.