Development of Threshold Criteria for Rainfall-Triggered Lahar

ANI HAIRANI, Adam Pamudji Rahardjo

2020 | Disertasi | DOKTOR TEKNIK SIPIL

Lahar is a sediment disaster that is usually preceded by torrent rainfall in the upper stream of volcanic area. It can damage destroy everything along its path, such as buildings, bridges, roads, and also trap the people in prone areas. To reduce the risk of lahar disaster, development of a lahar warning system is important. Rainfall threshold or commonly known as critical line is widely used to predict lahar. However, it highly relies on the number of data. A simulation-based critical line is deemed as one of the solutions to improve the reliability of lahar warning system. This research was aimed to develop rainfall threshold for lahar warning coupling with rainfall-runoff model. The initiation of lahar was identified by modifying Takahashi formula. The modification was done by taking into account the effect of hydrodynamics force. It was strengthened through the flume test. A distributed rainfall-runoff model were applied to simulate response of a basin for several cases of rainfall triggering lahar in Merapi. Area in which lahar initiates were identified based on overland flow depth resulted from the simulation and the developed theory. Simulation results and the historical lahar were added to determine the best critical line for lahar warning. The critical line was derived from MLIT Japan guideline and RBFN (Radial Basis Function Network) analysis. The experiment of initiation of sediment showed that modification of Takahashi corresponding to the bed shear stress caused by hydrodynamic force of surface runoff can be applied to identify initiation process of lahar. Effect of bed shear stress due to hydrodynamic force apparently weakens with the increase of slope. Fluctuation of rainfall intensity reflects changes of lahar initiation area. The more severe rainfall intensity, the larger volume of overland flow, and thus the greater lahar initiation takes place. By using statistical parameter that was recommended by Sabo Research Center and lahar simulation results, the output value of RBFN that was equal to 1 can capture the most lahar occurrences among all candidates.

Lahar is a sediment disaster that is usually preceded by torrent rainfall in the upper stream of volcanic area. It can damage destroy everything along its path, such as buildings, bridges, roads, and also trap the people in prone areas. To reduce the risk of lahar disaster, development of a lahar warning system is important. Rainfall threshold or commonly known as critical line is widely used to predict lahar. However, it highly relies on the number of data. A simulation-based critical line is deemed as one of the solutions to improve the reliability of lahar warning system. This research was aimed to develop rainfall threshold for lahar warning coupling with rainfall-runoff model. The initiation of lahar was identified by modifying Takahashi formula. The modification was done by taking into account the effect of hydrodynamics force. It was strengthened through the flume test. A distributed rainfall-runoff model were applied to simulate response of a basin for several cases of rainfall triggering lahar in Merapi. Area in which lahar initiates were identified based on overland flow depth resulted from the simulation and the developed theory. Simulation results and the historical lahar were added to determine the best critical line for lahar warning. The critical line was derived from MLIT Japan guideline and RBFN (Radial Basis Function Network) analysis. The experiment of initiation of sediment showed that modification of Takahashi corresponding to the bed shear stress caused by hydrodynamic force of surface runoff can be applied to identify initiation process of lahar. Effect of bed shear stress due to hydrodynamic force apparently weakens with the increase of slope. Fluctuation of rainfall intensity reflects changes of lahar initiation area. The more severe rainfall intensity, the larger volume of overland flow, and thus the greater lahar initiation takes place. By using statistical parameter that was recommended by Sabo Research Center and lahar simulation results, the output value of RBFN that was equal to 1 can capture the most lahar occurrences among all candidates.

Kata Kunci : initiation of lahar, distributed rainfall-runoff model, critical line, lahar warning system