การถดถอยและการป้องกันการกัดเซาะตลิ่งของคลองอู่ตะเภา จังหวัดสงขลา

Abstract

Riverbank retreat is a global problem. It causes the loss of land alongside the river. Moreover, excessive sediment from riverbank erosion yields the reduction of river drainage capacity. The U-Tapao River has experienced riverbank retreat for sometimes. Due to the lack of riverbank retreat study in Thailand, this study was conducted in order to constructing knowledges regarding the riverbank retreat of the U-Tapao River. The objectives of this study were 1) to study the characteristics of the river, 2) to evaluate riverbank retreat and develop the riverbank retreat model, and 3) to study riverbank erosion protection that can be applied to the U-Tapao riverbanks. Riverbank characterization study consisted of determinations of physical, geotechnical and erosional properties. Riverbanks from upstream to downstream areas were surveyed and soil samples were collected for laboratory testing. A submerged jet device was built and used in determining soil erodibility parameters in-situ. Study results show that the U-Tapao riverbanks can be classified into cohesive, non-cohesive, and composite banks. Soil erodibility parameters of the riverbank soils with high retreat rate found in the midstream areas ranged from 1.03-20.93 Pa and 2.23-89.07 cm/N.s, for critical shear stress and erodibility coefficient, respectively. These soils were classified as erodible to very erodible soils. Evaluations of retreated area and length were conducted using aerial imagery analysis. Aerial images representing the years 2002, 2010, and 2016 were obtained for the analysis. Evaluation results indicate that from 2002 to 2010, the eroded and reclaimed areas were 247.82 and 52.13 acre which corresponded to a net land loss of 195.70 acre. The riverbanks were highly eroded in Hat Yai district area with the eroded area of 5.86 acre per one kilometer of riverbank length which was equivalent to average erosion length of 4.35 to 13.87 m. For the midstream area, six high bank retreat locations were specifically studied. It was found that from 2002 to 2016 the retreat length for these six locations ranged from 7.52 to 31.83 m with the average of 17.21 m. A bank retreat model was developed using Bank Stability and Toe Erosion Model (BSTEM) and the model was calibrated using the lumped parameter (a). The developed bank retreat model was found to be able to compute accurate bank retreat lengths comparing to those obtained from areal imagery interpretation. The developed model can be used to predict future bank retreat at these locations. In riverbank retreat protection study, three pilot sites were designed to have para rubber sheet and planting for erosion protection purposes. Pilot site 1 (para rubber sheet with vetiver grass), pilot site 2 (para rubber sheet with Sennasiamea), and pilot site 3 (para rubber sheet only) were constructed. Surveying and visual inspection were conducted to monitorthe erosion of the pilot sites for 10 months. Monitoring results indicated that there was no erosion for all sites. Thus, all three protection methods were proved to be efficient.