1. Introduction

Taiwan is located in the subtropical region and the western North Pacific typhoon belt, in which the rivers are torrential because of rugged terrain and short streams. The rainfall seasons in Taiwan concentrates on July to August, which often brings heavy rain to cause serious flooding disaster in Taiwan coastal and low-lying areas. According to statistic data of Taiwan Fire department, a total of 350 typhoons and at least 1000 torrential rains attacked Taiwan in the past 100 years. It can be seen that flooding is the most serious natural disaster in Taiwan. To sum up disaster damage in the years of 1958-2000, there are 108 deaths, 250 people injured, 3458 buildings destroyed and 5575 buildings half-destroyed. To convert disaster damage into currency, the average damage of flooding disaster per a year before 1990 years approached 5.5 billion NT dollars and that after 1990 years exceeded 30 billion NT dollars. It is obvious that the disaster damage of flooding increases gradually year by year. The flooding disaster is induced by multi-factors including torrential rain, abundant sediment, poor drainage, improper land development and so on. Based on historical records, several typhoons, the 2009 Morakot typhoon and the 2013 Kong-rey typhoon for example, hit Taiwan, and brought abundant rainfall to cause serious damage, especially in Tainan, Taiwan.

There are a lot of studies in disaster risk and disaster preparedness plan for disaster mitigation. The disaster perception was also addressed because the disaster perception contributes to mitigation works. Wong and Zhao (2001) pointed out that people live in the flood-potential area because of high population density and limited dwelling and residents do not longer care about flooding control engineering. Instead, residents mind emergency relief and the post-disaster recovery action. The similar phenomenon occurs in Bengal and the residents in Bengal are ready to live with flooding (Rahman, 1996). In addition, 69%-86% residents in the Freiston coast of England thought the coastal protection measures is necessary for prevention from flooding and only 36%-50% residents believed the decision made by government departments (Myatt et al., 2003). Chen (2005) investigated the disaster risk perception of residents in disaster areas to supply a basic data of disaster risk management. This paper aims to investigate the disaster perception of residents in the disaster potential area. The Tainan located in south Taiwan is selected in this paper. The analysis findings will supply a helpful reference for mitigation works.

2. Method

2.1. Study Design

The close-ended questionnaire survey is adopted to explore residents' perception and attitude toward flooding disaster in this paper. The familiar Liker five-point scale from "strongly disagreeable" to "strongly agreeable" is measured. Except for demographics, the questionnaire consisted of eight sections comprising a total of 52 items (see Table 1).

Table 1. Summary of means and standard deviations (SD) of sections and items

The first section of the questionnaire includes possible factors lead to flooding occurrence. The second section of the questionnaire investigates possible action reducing flooding damage. The third section explores response after issuing a heavy rain warning. The fourth section surveys the possible worry about flooding damage. The fifth section examines information sources to make a decision of .The sixth section investigates the disaster prevention strategy of government for residents. The seventh section is to understand residents’ worry when leaving home to take refuge. The eighth section investigates residents’ satisfaction degree for residence. The demographic items included sex, age, and highest education.

In this paper, a reliability scale test is carried out for all sections in order to assessing the internal consistency of variables. According to document of Babbie (1992), the value of Cronbach’s alpha was classified based on a reliability index in which 0.90 –1.00 is very high, 0.70 - 0.89 is high, 0.30 – 0.69 is moderate, and 0.00 – 0.29 is low. As shown in Table 1, the resulting alpha values ranged from 0.65 to 0.89, which fall into the classification of moderate and high and very high. That is to say, the alpha values in this paper are within the acceptable range to assure reliability.

2.2. Sample Selection and Data Collection

The sample is randomly selected from four villages of Rende district, Tainan city, Taiwan including jhong-sheng village, bao-an village, er-sing village, and tai-jia village. These villages spread along Erren river bank and experienced several flooding induced by typhoon. In this paper, the door-to-door questionnaire is conducted. According to household data, there are 881 households in the study area and a sample of at least 268 households is targeted with a significant level of 95%. In this paper, of 270 households sampled, 16 questionnaires are unusable because the answer is not well completed. Thus, complete questionnaires are obtained from 254 households for an effective response rate of 94.07% and a significant level of 94.8%.

Of the respondents, majority of respondents are men with a rate of 66.5%. 42.2% are aged between 31 to 49 years old and 31.5% are aged between 21 to 30 years old. 49.6% had a bachelor diploma and 36.2% had a degree of high school.

3. Findings

Table 1 provides a summary of the means and standard deviations (SD) of sections and items. The items within each of sections are summed to obtain a mean score. The findings are listed as follows.

§ The major factors leads to flooding are “houses below the pavement”, “water does not discharge due to rising tide” and “water from outside overflow embankments”.

§ Implementing the necessary evacuation, joining disaster exercises held by community and requesting government to arrange evacuation lines and shelters can reduce the flooding damage.

§ Checking gas and wires, paying attention on any holiday and purchasing consumer goods are the popular responding behaviors when a typhoon warning is issued.

§ Losses of housing price, crop and production equipment will bring residents a serious harassment. It should be noted that losses of life, health and psychology is minor for residents.

§ Information sources to make a decision of evacuation come from the village chief or fire Department, neighbors, relatives and friends, and the elder or adult children.

§ For residents, holding a disaster prevention lecture and drills, purchasing product soaked in water, and providing timely disaster prevention information will reduce the flooding disaster. However, issuing damage compensation or grants, providing the punching bags and waterproof sheet, and rebuilding drainage cannot decrease the flooding disaster.

§ House, car and furniture, and job are major harassments when people decide to leave home to take refuge.

§ The average satisfaction of dwelling is high for residents living along Erren river bank, Taiwan.

4. Conclusions

The disaster occurrence is closely related to residents’ reactions. Thus it is necessary to explore risk perception, attitude and behaviour of disaster for implementing the disaster management strategy. These residents along Erren river bank, Taiwan are selected and the questionnaire survey method is adopted to understand their flooding perception. The study results show that residents tend to be at peace with flooding, which is consistent with the study of Rahman (1996). Moreover, compared to disaster prevention by government, the autonomous disaster prevention is possible to reduce the flooding damage for residents.

ACKNOWLEDGEMENTS

The authors wish to acknowledge financial support of the National Science Council (NSC), Taipei, Taiwan through Grant No. NSC 103-2119-M-464 -001. For this assistance, the authors are very grateful.