Abstract

     Culverts are used to allow runoff to pass through roads, railways, and embankments. Accumulation of debris during flood events reduces the culvert flow capacity and hence flow overtopping results in culvert failure both hydraulically and structurally. This paper presents the results of an experimental study of temporal variations of blockage upstream of culverts due to woody debris under unsteady flow conditions. To simulate flood conditions, a synthetic flow hydrograph was produced in the laboratory. Cylindrical wooden dowels with two different diameters were used to simulate the woody debris carrying during flood events. Two culvert shapes including box and circular pipe culverts are examined here. The results showed that the maximum percentage of blockage occurs during the falling limb of the hydrograph. Although the feeding rate of smaller diameter woody debris into the flow is of considerable importance in the culvert blockage, the blockage percentage is not influenced by the feeding rate of large woody debris. It was also found that the pipe culvert is more susceptible to blockage than the box-shaped culvert. Using regression analysis, predictive equations are suggested to estimate the percentage of culvert blockage during flood events.

Introduction

     While the structural design of the culvert is simple, evaluating its hydraulic performance is still a challenging task due to the blockage problem with the presence of various types of floating debris (Ahmed et al., 2021; Iqbal et al., 2021b; Osman & Taha, 2022). A culvert is a short-length underground drainage structure to convey flood flows from one side of infrastructures such as embankments, levees, roadways, and railroads in urban areas or in the countryside to the other side. Culverts are also designed in different shapes and for a particular purpose for example to allow the passage of fish or animals besides their main purpose as a drainage structure(Chanson et al., 2021; Li et al., 2022). While the structural design of the culvert is simple, evaluating its hydraulic performance is still a challenging task due to the blockage problem with the presence of various types of floating debris (Balkham et al., 2011; Blanc et al., 2014; Iqbal et al., 2021a; Paik & Park, 2011; Zhong et al., 2021). Three different types of debris may be considered in hydraulic engineering including floating debris (generally different types of vegetation), non-floating debris (e.g. sediments), and urban debris such as structural materials and all types of fly-tipping (Balkham et al., 2011; Weeks et al., 2009). Some studies show that there is a range of locations and conditions where the blockage may be a concern in hydraulic designs (Diehl, 1997; Friedrich et al., 2022; Furlan et al., 2018; Spreitzer et al., 2021; Streftaris et al., 2013).

Conclusion

     In this study, experimental tests were carried out under unsteady flow conditions using a synthetic hydrograph during flood events to simulate blockage at the inlet of box and pipe culverts. The results show that for both culvert types, the maximum percentage of blockage (B) at the culvert entrance occurred at the final time step of the falling limb of the hydrograph. It was found that the maximum blockages for wood with 12 and 23 mm diameters were 65% and 70%, respectively. For a given wood diameter and a specific feeding rate, the percentage of blockage in the pipe culvert was higher than that of the box culvert. As a result, the percentage of blockage in the P4 and P3 tests with a 23 mm diameter wood debris were 45% and 37% higher than that in the B4 and B3 tests with the same wood debris diameter, respectively. Furthermore, it was found that the shape of the culvert influences the height of the accumulated woods. The height of the blockage in the P4 and P3 tests were 10% and 23% less than that in the B4 and B3 tests, respectively. Moreover, the maximum increase in the upstream flow depth (H*) for both types of culverts was observed at the falling limb of the hydrograph. Also, the values of H* in the P4, P3, P2, and P1 tests were18.3%, 15%, 13.5%, and 13% higher than corresponding values in the B4, B3, B2, and B1 tests, respectively. Finally, some empirical equations have been presented based on regression analysis for the calculation of blockage percentage and its height which can be used by hydraulic designers to increase the reliability of their designing against blockage. Also, the results may be interesting for the stakeholders in the operational phase to increase the reliability of the culvert performance.