| Summary: | 碩士 === 國立臺東大學 === 進修部環境經濟資管碩專(假日) === 105 === “Disasters are recurrent and selective.” Humans exist between heaven and earth. Heavy rain pouring from the sky brings disasters from “Heaven”, and trembling of ground and mountains causes changes on the “Earth.” At the thought of these “heavenly disasters and earthly movements,” fear inevitably arises in our mind. The purpose of engineering is to improve the aftermaths of recent disasters or reduce the possibility of future recurrence. Well-designed construction can protect the environment by reducing the likelihood of recurrence; it can also help prevent disasters altogether by improving the conditions of environment previously prone to disasters. Although men still can’t really control when and how disasters happen, we can plan and implement construction through engineering. Thus, there should be an absolute correlation between occurrence of disasters and constructions.
This paper will focus on Taitung County (Do not include Green Island Township and Lan Yu Township), analyzing records of natural disasters and engineering blueprints from 2001 to 2016, with a total of 1,055 historical cases, including the 539 cases related to hill-like terrains. Engineering blueprints are collected from records of mudslide and flood control engineering for slope disasters between 2001 and 2015, taitung to 1,513 samples. The data are sorted according to the cause, time, and location of occurrence; they are then compiled into a database accordingly. Circumstances surrounding the slope disasters and the subsequent construction are discussed respectively according to the year of the record and spatial distribution. After disaster and engineering clusters are generated by spatial clustering analysis through kernel density estimation (KDE), chronological and spatial clustering characteristics are examined to establish the correlation between disaster occurrence and disaster prevention construction from temporal and spatial clustering characteristics.
An analysis of slope disasters in relation to geographical axis shows the Coast Line with the lowest number of disasters and the South-Link Line with the highest number of disasters. Statistical analysis according to types of slope disaster return landslide as the main type, followed by road-construction-related accidents. Spatial clustering analysis of disasters shows that the number of slope disasters does not increase in proportion to the area of the slope but demonstrate a positive correlation to the number of potential debris flow torrents. Those with a higher number of potential debris flow torrents also tend to sustain a higher number of disasters.
According to statistics, the main engineering type is treatment engineering, followed by environmental conservation. According to clustering analysis of disaster-related engineering, projects implemented before 2003 were mainly distributed along the Coast Line and the Valley Line, and there was no obvious cluster along the South-Link Line. Since 2004, large-scale construction sites started to form in the Taimali Township. Although there is no significant diminution in the overall area of previous clusters in the Coast Line and the Valley Line, there is a decline in its central area. The kernel density of the clusters along the Valley Line and the Coast Line dropped significantly until 2010, indicating a redirection of the overall engineering focus toward the South-Link Line.
For the correlational analysis of disaster and construction, this study defines the ratio of quantified disasters to the number of constructions as Ri , the disaster management rate representing spatial clustering characteristics as Pi and engineering utility efficiency as Qi . Based on the results of analyses, a quantified Ri of 30% is taken as a reference point for evaluating the efficiency of engineering constructions. Analyses show that due to the extensive area of disasters, it is difficult to implement restorative constructions at all the locations right away, resulting in a poor “disaster management rate.” With regard to “engineering utility efficiency”, the regression analysis returns a highly positive correlation, indicating high density of constructions following occurrence of disasters and their efficiency in reducing the likelihood of future disasters.
|
|---|
