Developed Model
Figure 4 Flow Chart of Runoff Analysis
In this study, a hydraulic model is based on a grid series tank model. Figure 4 shows flow chart of runoff analysis. Firstly, a precipitation is supplied to each DEM grid that is one of the tanks. The supplied precipitation shich means inlet content is calculated by following equation
Q in = K I R L
2
Qin: Inlet Content (m
3)
Ki: Infiltration
R: Precipitation (m)
L: Grid Size (m)
In this equation, infiltration is usually depend upon landcover, Nakano (1) showed relationship between infiltration and landcover. In this study, we adopted his results, and the infiltration was led from NDVI which is calculated from MOS-1 MESSR data. The relationship between infiltration and NDVI is shown as follows'
| | | Bare Soil | | Glassland | | Forest | |
| NDVI : | -1.0 | ¨ | 0.1 | ¨ | 0.25 | ¨ | 1.0 |
| Infiltration (mm/h) | | 30 | | 100 | | 270 | |
Next, the inlet content divides into three discharges which are evapotranspiration, infiltration and surface flow. Moreover, the infiltrated we\ater was divided into three discharges which are sbsurface flow, base flow and dead water.
A decision of runoff percentage of surface flow and subsurface flow is very important. The runoff percentage is changed by season, which is influenced by precipitation intensity. Table 1 shows runoff percentage in each season, which were calculated from the hydrograph (Figure 3) The runoff percentage shows tendency to increase according to rainy wseason. In this study the runoff percentages can be set by proportional equation with calculated runoff percentage in dry season and rainy season. The surface runoff and the subsurface runoff in each tank can be expressed by following equations.
Q
in = Kr Q
in
Qin : Inlet Content (m
3)
Kr: Runoff P rcentage
Q in : Effectiv Inlet (m
3)
After that, the surface flow and the subsuface flow must be discharged to next grid according to slope aspect and velocity That is to say flow tracking. The slope aspect is calculated from DEM the velocity can be estimated from slope gradient which is also calculated from DEM. The surface flow and the subsurface flow in the grid can be experessed by a continuous equation as follows;
Q t+t = q in -q out)
D t
Q : Remaining Countent (m
2)
Qin: Inlet (m
3/s)
Qout: Outlet (m
3/s)
DT: Time (s)
By the way, an estimation of the base runoff is very difficult because the discharge includes ground water which is always supplied in spite of precipitation. In this study, the base runoff discharge was led from the observed hydrograph. A straight broken line in Figure 3 shows estimated base runoff discharge by using visual interpretation.
Table 1 Runoff Percentage in Each Season
| | Dry Season Rainy | ® | Season |
| Base Runoff | 13.84 | 7.53 | 9.36 |
| Subsurface Runoff | 1.95 | 3.01 | 6.92 |
| Surface Runoff | 2.13 | 4.22 | 14.12 |
| Total Runoff | 17.91 | 14.75 | 30.40 |
Results and Conclusions
By using previous model, estimation of total discharge was carried out. Figure 5 shows hydrograph in 1987 which are consist of estimated discharge, actual discharge and pecipitation. The result shows almost high accuracy. However, in the end of June and the beginning of October, etimated discharge was much different from actual one. The climate data and discharge data were observed at only one point. In monsoon area and mountainous area, climate is very changeable. Moreover, the cachment area in this study were too huge. It has been already mentioned in the previous chapter. So, it is impossible to represent climate data by only one observatory point. In future, bydrograph should be made from many climate data though we couldn't have enough observation points.
Figure 5 Estimated Hydrograph in 1987
In this study, estimation of runoff discharge was succeeded with use of remote sensing data, DEM and climate data. It will be used as time interpolation of runoff discharge. After all remote sensing data and DEM showed very helpful data for runoff analysis.
References
- Hideaki NAKANO, "Hydrology in Forest". Kyouritsu Shuppan, pp. 55-72 (in Japanese)
- Shiro OCHI, Shunji MURAI and Suvit VIBULSRESTH, " Flood Disaster Prediction Model using Remote Sensing Data and Geographic Information System", proceedings of the 10th ACRS, PP B-3-1-b-3-6
- Sahido SUSANTO and Yoshihiro KAIDA, "Tropical Hydrology Simulation Model for Watershed Management", Journal of Japan Society of Hydrology Simulation Model for Watershed Management Journal of Japan Society of Hydrology and Water Resource, Vol.4 No. 2, PP. 430-53
