Event-based rainfall-runoff modeling for a data-scarce semi-urban
catchment using PC-SWMM: A case study for Vizianagaram town, Andhra
Pradesh, India.
Abstract
The rainfall-runoff model is essential to derive a relationship between
Rainfall and Runoff, in which the hydrological response of the catchment
can be derived. A real case study is chosen to simulate the
rainfall-runoff modeling, in which small urban catchments are selected
and are located in the center part of Vizianagaram Town. The present
case study aims to develop an event-based rainfall-runoff model for
upstream and downstream catchments of peddacheruvu catchment (PC). In
this study, Cartosat-10mDEM, Hourly rainfall data are taken from
(weather station), i.e., 1st July to 30th September, the Maximum and
Minimum Infiltration rates, Evaporation data, soil data, Groundwater
parameters, and dry weather flow patterns are used as in input for model
simulation to know the wet weather flow and dry weather flow quantity
contributing from the catchment. The model simulation is carried out by
using the stormwater management model i.e., PC-SWMM version 5.7.1868.
The model simulation is performed at two outlet points in the catchment.
The upstream and downstream catchments are selected for computing the
total runoff hydrographs. The model calibration is done for nine
selected streamflow events from 1st July to 31st August 2019, and the
remaining three streamflow events are chosen from 1st to 30th September
2019 are set for model validation. The model performance was checked by
computing nine goodness of fit measures. The results of this study
suggest that simulated runoff values have satisfactory results with the
observed streamflow. In recent years, understanding the hydrological
modeling and process has become more important in water resource
management, especially in analyzing extreme hydrological events like
floods or droughts. The availability of metrological and hydrological
data is often scarce in a semi-urban catchment. Some of the significant
issues are associated with obtaining reliable long-term hydrological
data in the semi-urban region. This study investigates the performance
of event-based modeling for data-scarce semi-urban catchments using
PC-SWMM in computing the total runoff hydrograph. A real case study,
i.e., Peddacheruvu (PC) Upstream and Downstream catchment, were
selected, and model performance was examined using 12 streamflow events
from 1 July 2019 to 31 September 2019. The model performances are
evaluated using five goodness of fit measures like root mean square
error (RMSE), Nash Sutcliffe efficiency (NSE), coefficient of
determination (R2), RSR, and Kling Gupta efficiency (KGE). The model
performance is acceptable throughout model calibration (1 July to 31
August 2019) as the NSE and R2 varies between 0.75 to 0.77 and 0.76 to
0.78, respectively. Similarly, the model validation performances (1
September to 31 September 2019) revealed best fitted with the observed
hydrograph for NSE and R2 were 0.62 to 0.64 and 0.62 to 0.85. KGE for
model calibration and validation model varies between 0.65 to 0.75 and
0.62 to 0.75.