| dc.description.abstract | Hydrological modeling of large river basins is an infrequent topic in Ecuador, since a
large part of modeling has been carried out for small basins using deterministic hydrological
models. This is due to the large amount of information required and the demanding computational
processes that is involved.
In the present study, the semi-distributed hydrological model MGB-IPH was successfully
applied in the high Andean basin of the Cuenca river in Ucubamba (approximately 1618 km2
area), using precipitation and flow data measured in the field and meteorological extra data from
a reanalysis base. The basin was discretized into five main drainage sub-basins (Machángara,
Tomebamba, Yanuncay, Tarqui and Urbana). Six scenarios were considered for this study: three
hydrological modeling cases using drainage networks of different densities, with minimum
drainage area thresholds of 1.6 km2
(Case 1), 4 km2
(Case 2) and 16 km2
(Case 3), and in addition
in each case, two different flow routing methods were implemented: Muskingum-Cunge and
Inertial Hydrodynamic. The model parameters were calibrated and validated for each defined subbasin and for each case raised.
The best performance of the model is found in the scenario with the densest drainage
network (Case 1), for which efficiency values were obtained in the Urban sub-basin (closing subbasin) of 0.750, 0.821 and -3.678% for Nash-Sutcliffe, modified Nash-Sutcliffe and BIAS
respectively in the validation of the model, and with the Muskingum-Cunge method.
Then, a hydrological simulation for an extended period of ten years (2003-2012) was
carried out. This was done, in order to show the different products offered by the MGB-IPH
model, especially when performing the simulation with the Inertial Hydrodynamic method, such
as: daily series of mean flow and average water level in any micro-basin, a series of the total
flooded area of the basin and a map of flooded areas of the basin. For this simulation, the following
criteria was used: the spatial discretization with the drainage network of case 1, the parameters of
the model of each sub-basin that were calibrated and validated for said case, the precipitation data
observed in the period of ten years and the climatological data of reanalysis.
The efficiency and performance of the MGB-IPH model in the Cuenca river basin in
Ucubamba have been successful in the calibration and parameter validation processes, and with
few input data required. In addition, it is suitable for simulating average flow events, low-water
flows and water balance in a high Andean mountain basin, and underestimating peak flows. | |
