5:15 PM - 6:45 PM
[HGM03-P04] Changes in River Morphology over 130 Years in the Tottabetsu-River in Hokkaido
Keywords:river morphology, sediment control works
The headwaters of the Tokachi River basin in Hokkaido, characterized by steep gradients and fragile geology, tend to experience frequent landslides due to heavy rain. Since the Meiji era, massive amounts of sediment have flowed downstream to the floodplain, causing frequent floods. In 1961, an area spanning 15 km downstream from the headwaters to the upper edge of the floodplain (Takusei Bridge) was designated as a directly managed sediment control zone. Sand control facilities have been continuously introduced in this area. This study examined the relationship between the introduction of these sediment control facilities and changes in river morphology downstream to the floodplain.
The study area includes three sections: Section I (9 km), Section II (3.5 km), and Section III (2.5 km) downstream from Takusei Bridge. In Section I, impermeable (introduced in 1987 at the 1.7 km point) and permeable (introduced in 2014 at the 2.5 km point)sediment control weirs, and 15 bed stabilizers (introduced between 1987 and 2010 from the 2.7 km to 9 km points) have been installed. Bed stabilizers are installed to control rapid erosion of the riverbed and the sudden movement of unstable sediment within the river channel while allowing sediment to flow downstream. In Section II, soft rocks belonging to the floodplain long-flow branch inner layer are partially exposed, leading to significant deepening of the riverbed and meandering. In Section III, bank protection has been in place since around 1968, with little change in the channel course, but the riverbed has been continuously lowering.
Historical maps, aerial photographs, and survey data related to the introduction of sediment control facilities have been available for this area since 1896. Since the 2000s, regular aerial laser surveys and photography have been conducted. Using these data, the river morphology was classified every 200 meters based on Rosgen's (1996) criteria. Here, the target area was classified into 8 types: 1point bars, 2 point bars with few mid-channel bars, 3 numerous mid-channel bars, 4 side bars, 5 diagonal bars, 6 main channel branching with numerous mid-bars and islands, 7 side bars and mid-channel bars with length exceeding 2 to 3 times channel width, and 8 delta bars. Areas without dense vegetation on aerial photographs were identified and subtracted at 200-meter intervals to estimate the pseudo-river width for each section. The riverbed elevation for each year was determined every 200 meters based on the above data and the identification results of the sedimentary terraces from field surveys. These results were summarized for three periods: Period A, before the introduction of sediment control facilities to the basin (before ~1968); Period B, until the bed stabilizers were installed (~2010); and Period C, up to the present. During Period A, landslides were frequent in the headwaters in the 1950s. In Period C, a flood occurred in August 2016, estimated to have a recurrence interval of 100 to 200 years.
From the results, it was found that in all sections, 1963, before the introduction of sediment control facilities, had the widest river width and highest riverbed elevation. This is thought to be due to the large amount of sediment that flowed and accumulated in the river during the 1950s. During this period, intrusion of riparian forests was also observed, and river morphologies 3 and 7 were generally predominant. During the introduction of sand control facilities, the riverbed elevation tended to rise until 1982, but the river width decreased. Regarding morphology, types 4,7 were mainly observed in Section I, types 3,4 in Section II, and type 4 in Section III. After the introduction of sediment control facilities, the river width decreased throughout all sections, and in Sections I and III, the channel was fixed, and type 4 became predominant. However, in Section II, type 4 also predominated, but this was because the rate of erosion into the soft rock exceeded the rate of intrusion of riparian forests, causing the channel to constantly change. In Section II, this process has been rapidly progressing since the 2016 flood, likely influenced by the sediment transport control function of the sediment control facilities in Section I during and after flooding.
The study area includes three sections: Section I (9 km), Section II (3.5 km), and Section III (2.5 km) downstream from Takusei Bridge. In Section I, impermeable (introduced in 1987 at the 1.7 km point) and permeable (introduced in 2014 at the 2.5 km point)sediment control weirs, and 15 bed stabilizers (introduced between 1987 and 2010 from the 2.7 km to 9 km points) have been installed. Bed stabilizers are installed to control rapid erosion of the riverbed and the sudden movement of unstable sediment within the river channel while allowing sediment to flow downstream. In Section II, soft rocks belonging to the floodplain long-flow branch inner layer are partially exposed, leading to significant deepening of the riverbed and meandering. In Section III, bank protection has been in place since around 1968, with little change in the channel course, but the riverbed has been continuously lowering.
Historical maps, aerial photographs, and survey data related to the introduction of sediment control facilities have been available for this area since 1896. Since the 2000s, regular aerial laser surveys and photography have been conducted. Using these data, the river morphology was classified every 200 meters based on Rosgen's (1996) criteria. Here, the target area was classified into 8 types: 1point bars, 2 point bars with few mid-channel bars, 3 numerous mid-channel bars, 4 side bars, 5 diagonal bars, 6 main channel branching with numerous mid-bars and islands, 7 side bars and mid-channel bars with length exceeding 2 to 3 times channel width, and 8 delta bars. Areas without dense vegetation on aerial photographs were identified and subtracted at 200-meter intervals to estimate the pseudo-river width for each section. The riverbed elevation for each year was determined every 200 meters based on the above data and the identification results of the sedimentary terraces from field surveys. These results were summarized for three periods: Period A, before the introduction of sediment control facilities to the basin (before ~1968); Period B, until the bed stabilizers were installed (~2010); and Period C, up to the present. During Period A, landslides were frequent in the headwaters in the 1950s. In Period C, a flood occurred in August 2016, estimated to have a recurrence interval of 100 to 200 years.
From the results, it was found that in all sections, 1963, before the introduction of sediment control facilities, had the widest river width and highest riverbed elevation. This is thought to be due to the large amount of sediment that flowed and accumulated in the river during the 1950s. During this period, intrusion of riparian forests was also observed, and river morphologies 3 and 7 were generally predominant. During the introduction of sand control facilities, the riverbed elevation tended to rise until 1982, but the river width decreased. Regarding morphology, types 4,7 were mainly observed in Section I, types 3,4 in Section II, and type 4 in Section III. After the introduction of sediment control facilities, the river width decreased throughout all sections, and in Sections I and III, the channel was fixed, and type 4 became predominant. However, in Section II, type 4 also predominated, but this was because the rate of erosion into the soft rock exceeded the rate of intrusion of riparian forests, causing the channel to constantly change. In Section II, this process has been rapidly progressing since the 2016 flood, likely influenced by the sediment transport control function of the sediment control facilities in Section I during and after flooding.