Uncertainty Analysis of Shihmen Reservoir Capacity Using Monte Carlo Method

Abstract

In this study we create a new statistical model for calculating the lifespan of a reservoir that is able to produce not just one averaged result but can provide a range of results representing uncertainty The model Monte Carlo Reservoir Sedimentation Model (MS Model) was created using the structure of the 3W Model created by Midear and Kondolf (2009) and the statistical theory of the Monte Carlo method The uncertainty considered in this model only includes uncertainty around possible hydraulic events To test the model we conducted reservoir life simulations for Shihmen reservoir located in Northern Taiwan using both the 3W Model and MS Model The models were all run from 1960 until the reservoir capacity reaches 0 and compare the results to the observed data available from 1964-2015 as well as to each other The MS Model is distinct from other models in that it does not only consider a single sedimentation rate but considers sedimentation rates of both watershed soil loss and landslides Different probability distributions of sedimentation rates due to both soil loss and landslides were calculated and the most accurate pair for the study area was found to be included as input in the model The simulation result from the MS Model show the life of Shihmen Reservoir is about 127 years with an error of about ±24 years We compare the result of the MS Model with the previous reports and existing reservoir capacity data which shows that the average value of the MS Model simulation is quite in line with the current situation and also is able to consider the hydrological uncertainty Therefore MS Model can accurately simulate the capacity of Shihmen Reservoir Because of the uncertainty of hydrological events this study explores a large number of simulations and predicts that the predicted 127 year lifespan could be extended or cut short by 24 years based on hydrological uncertainties from unpredictable hydrological events Under the current reality of climate change it is ever more important to pay attention to the impact of hydrological uncertainty

Date of Award	2020
Original language	English
Supervisor	Hsiao-Wen Wang (Supervisor)