The increasing rainfall intensity and cumulative rainfall induced by climate change magnify the flow rate of a river and significantly erode a dyke. Securing the integrity of a dyke to protect the land is an essential topic in disaster prevention and water resource management. A concrete-faced river dyke increases erosion resistance and is usually used along the main river in south Taiwan. However, eroded caves behind the thick concrete face are difficult to detect. This study attempts to develop a new visual-based statistical model to estimate the degree of cavity erosion behind the concrete-faced river dyke. Because removing the in-situ concrete face of the dyke is usually forbidden, a non-destructive ground-penetrating radar (GPR) image is used to confirm the location and the size of the cavity. Therefore, an indoor test is conducted to identify the GPR signal of the cavity behind a concrete plate. The detection of in-situ caves associated with the removal of the concrete face during dyke repair is used to validate the statistical model. The degree of cavity erosion is classified based on the in-situ GPR detection results. The outlook factors of the concrete faces are collected by a visual survey to correlate the outlook factors of the concrete dyke to the internal cavity erosion degree by multiple linear regression analysis. The accuracy of the statistical model is verified by comparing the cavity erosion degree predicted by the statistical model and that defined by GPR.
|Number of pages||12|
|Publication status||Published - 2013 Nov 1|
All Science Journal Classification (ASJC) codes
- Geography, Planning and Development
- Safety, Risk, Reliability and Quality
- Environmental Science (miscellaneous)
- Earth and Planetary Sciences (miscellaneous)