Flood Modelling In Pasig-Marikina River Basin

Roy A. Badilla

The Metropolitan Manila has been experiencing recurrent flooding especially in the low lying areas along the Pasig-Marikina River. Because of this problem, the Philippine government has implemented several projects to achieve effective flood control operations in the area. The first project to be completed was the construction of Mangahan floodway. However, after the completion of the project, informal settles have stated building their houses in the side slope of the floodway making it risky to operate the Rosario Weir. Rosario Weir is the structure that controls the inflow to the Mangahan Floodway. This study focuses on the development of a HBV model and a DUFLOW model to study the flood wave behaviour in the study area and to come up with calibrated models which could be used as basis for the operation of Rosario Weir and Napindan Hydraulic Control Structure for effective flood control and early warning in Mangahan Floodway. The HBV-96 model was applied to stimulate the runoff from Pasig-Marikina River Basin using hourly hydrometeorological data. Four rainfall stations and one water level station for a period of three years were used for the calibration and validation of the model. The catchment was extracted from SRTM elevation data and was divided into six sub-basins. Land cover classes applied for this study were field and forest. Other land cover classes that can be specified in the HBV model do not apply to the study area. Two modules of the DUFLOW model were used for his study; the water quantity module and the RAM module. The DUFLOW water quantity module was setup using twelve river cross sections along the upper Marikina River with a river length of 20.10 kilometers. However, the distances between river cross sections along the river length is not uniform. Water level data from Montalban water level station was used as the upstream boundary condition and the data from Mangahan water level station was used as the downstream boundary condition. The calibration of the DUFLOW water quantity module was done at Marikina water level station. Inflow from the intermediate area between the upstream boundary node and the calibration node was handled by RAM module. However, no calibration was done in RAM module because of unavailability of in-situ data although model parameters were specified based on a priori knowledge about the site characteristics. After the DUFLOW calibration, the hydrograph from the HBV model simulation was used as the upstream boundary condition while the downstream boundary condition remains the same. The purpose of integrating the results of the HBV model to the DUFLOW model is to increase the flood lead time which could be beneficial for flooded control and early warning purposes. The results obtained from this research were satisfactory giving a Nash-Sutcliffe efficiency coefficient (R2) of 0.79 and 0.76 for the HBV model calibration and validation data set, respectively. The DUFLOW model was more accurate with R2 of 0.91 for the model that used in the Montalban observed water level data and 0.90 for the model that used the simulated HBV hydrograph. Hence, the calibrated models can be applied for the flood control and early warning system in Mangahan Floodway.

Potential Vorticity Tendency Aspects Of The Motion Of Typhoon Muifa (2004)

Bonifacio Pajuelas

This study attempts to determine the dominant influence on the motion of tropical cyclone (TC) Muifa (2004) using the expanded potential vorticity tendency (PVT) framework to explain TC motion. The framework suggests that a TC is likely to move towards the region of maximum asymmetric PVT which is mainly contributed by the asymmetric components of potential vorticity (PV) advection and diabatic heating (DH). To diagnose the process first, the limited area grid-point Eta model is used to generate the analysis data to obtain air temperature, zonal and meridional components of wind. Cloud top temperature from the Geostationary Observing Environmental Satellite (GOES-9) infrared (IR) images in the wavelength range of 10.2-11.2µm and 11.5-12.5µm were also used. Here, the results are illustrated on a circular grid using radar plot. The circulation of TC Muifa and its environment is illustrated by the typical distribution of sea-level pressure, wind vectors, GOES-9 IR derived temperature, vorticity, and potential temperature. As Muifa intensifies the vertical structure of potential vorticity (PV) becomes more symmetric as this is distributed horizontally and vertically. The role of the individual physical processes of PV advection and diabatic heating (DH) is to generate heat to contribute to the overall process of PVT. Each individual contribution is apparently indicated by the azimuthal asymmetric structure where the maximum magnitude of DH dominates over the PV advection terms. DH maximum occurred at 12 GMT 18 November during which Muifa reaches its peak wind speed at 115 knots. On the average, DH is not maximum during the day which implies that much of the latent heat release occurred during the night. The positive correlation of DH with maximum wind decreases towards the TC center. On the other hand, the negative correlation of PV advection terms with maximum wind increases towards the TC center. The maximum in PV advection terms rotates ahead or is aligned to the direction ahead of the turning motion. But, in instantaneous motion, the maximum in the PV advection terms points to the direction of motion. In one section of the track where the recurving direction is opposite to the direction of turning motion: the magnitude of the SAAPV term to the left of the direction of motion, increases clockwise while the magnitude in the AASPV infront of the direction of motion decreases clockwise. In terms of magnitude, DH is a dominant influence modifying the PVT structure and hence, the motion of TC Muifa while in terms of pattern, the PV advection terms are dominant indicators of the steering flow and the direction of motion.