Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
years considering uncertainty of sea level changeTakai, Shizuka; Uchikoshi, Emiko*; Shimada, Taro; Takeda, Seiji
MRS Advances (Internet), 6 Pages, 2026/00
In Japan, regulations require the disposal of intermediate-level radioactive waste at depths of at least 70 m below ground surface for 10 years. Therefore, it is essential to assess the long-term safety of landscape evolution by uplift, denudation, and sea-level change, which will decrease the depth of disposal and change groundwater flow. Numerical simulations are invaluable for analyzing landscape evolution while taking into account future uncertainties. However, the projection over the next 10 years has not been conducted for coastal areas, which are preferable to disposal in terms of waste transportation. In this study, we developed a framework for projection of coastal landscape evolution on a 10 years based on simulations over the past 125,000 years. Using the LEMs developed by JAEA, the constructed framework following two steps was verified in the Kamikita coastal plain. Step 1) Simulation from past to present: Based on simulations of the past 125,000 years (the last glacial-interglacial cycle), we first identified the best parameter set that can reproduce the actual topography. The deviation between reproduced and recent topography is within 
15 m for 80% of the area. Step 2) Projection over the next 10 years: Using the best parameters, we make projections over the next 125,000 years. On the basis of sea level change in the late Quaternary and previous research on the glacial isostatic adjustment modelling, three scenarios of future global sea level were considered; base case (the next glacial period starts 5,000 years after present), glacial case (the next glacial period starts from the present), and global warming case (the current interglacial lasts over the next 125,000 years). In the case of a hypothetical disposal depth of 100 m, erosion depth should be less than 30 m to maintain the waste at a depth of 70 m below ground. For the next 125,000 years, erosion exceeding 30 m is projected to occur in the river (mid and downstream) and coastal areas (~below highest projected sea level of 15 m). Vertical erosion by fluvial incision is the most affected by sea level decline. In the base and global warming cases, sea level rises up to 15 m; erosion occurs several hundred meters landward from the current 15-m contour. We also confirmed that the average basin relief increases significantly in accordance with sea level fall.
Takai, Shizuka; Sanga, Tomoji*; Shimada, Taro; Takeda, Seiji
no journal, ,
Prediction of long-term future landscape evolution is indispensable for safety assessment for intermediate radioactive waste disposal, whose safety assessment period is ~10 years. To assess the coastal landscape evolution considering the uncertainty of future sea-level change, the numerical simulation based on the landscape evolution models (LEMs) will be valuable. However, the applicability of LEMs over 10 years has not been verified in coastal areas. JAEA has developed a LEMs (JAEAsmtp) coupling hillslope and fluvial transport, tectonics, marine sedimentation, sea-level and climate change, and lithology. In this study, we demonstrate the capabilities of JAEAsmtp in assessing the coastal landscape evolution over the last glacial-interglacial cycle. Our target area (250 km
) is located on the Kamikita coastal plain (sedimentary rock), where the marine terraces (MIS5e, 7, 9) are widely distributed. First, based on the marine terraces, borehole, and sonic prospecting data, the spatial distributions of the present and paleo-elevation, uplift rate, and alluvial deposits were estimated. Second, the LEMs parameters for fluvial incision and erodibilities were obtained from the slope-area analysis and soil-test data, respectively. Finally, the landscape evolution from MIS5e (125 ka) to present was simulated. By incorporating the drift sand into JAEAsmtp, the applicability was confirmed via reproducibility of the present coastal line and the distribution of alluvial deposits.
Takai, Shizuka; Sanga, Tomoji*; Shimada, Taro; Takeda, Seiji
no journal, ,
In the safety assessment of intermediate-level radioactive waste disposal (the safety assessment period of 
10 years), numerical landscape evolution models (LEMs) are invaluable to assess the future considering climate change. However, the most parameters of LEMs cannot be determined directly from the measurement. In this study, we constructed a framework of calibration and validation of LEMs parameters based on the simulation over the last 125 ka. Using the LEMs developed by JAEA, the constructed framework was verified in the Kamikita coastal plain. (i) calibration (the Tokusari River watershed): we found the set of LEMs parameters that satisfy the constraint conditions of the past landscape evolution (terrace erosion rates, depth of the buried valley, longitudinal river profile, width of erosional valley-floor). For efficient calibration, we firstly selected parameters for calibration based on the Morris sensitivity analysis. Then, the best parameter set was estimated by approximating the constraint conditions by second-order polynomial surrogate models of selected parameters. Through the calibration, five parameters are identified as important: the coefficients of hillslope transport, fluvial incision, coastal erosion, and climate change. The developed surrogate models are approximated the simulation results with the 0.7 or more coefficient determination. (ii) validation (the Futamata River watershed): we finally validate the result by applying the calibrated parameters to an independent watershed. As the result of simulation from 125 ka to present, the residual of reproduced and recent topography is less than 15 m for 80% in the area both of the calibration and validation domain. The validity of calibration was further verified through reproducibility of buried valley, terrace cliff, and terrace erosion rates.
Takai, Shizuka; Shimada, Taro; Uchikoshi, Emiko*; Takeda, Seiji
no journal, ,
In geological disposal, landscape evolution by uplift, denudation, and sea-level change will change geological environment and decrease the depth of disposal. This may lower safety functions of disposal system: therefore, the effect needs to be evaluated properly. Landscape evolution can be evaluated quantitatively using landscape evolution models. In general, the evaluation is based on extrapolation of the past. However, the future sea-level change may differ from the past because of greenhouse gases. In this study, we constructed the evaluation method for future long-term landscape evolution based on the past landscape evolution. We confirm the applicability in the typical basin consisting of mountain, river, plain, and sea. The effect to future landscape evolution by uncertainties of sea-level change were evaluated.
Takai, Shizuka; Sanga, Tomoji*; Shimada, Taro; Takeda, Seiji
no journal, ,
In the safety assessment of intermediate-level radioactive waste disposal (the safety assessment period of 10 years), numerical landscape evolution models (LEMs) are invaluable to assess the future considering future climate change. However, a limited number of studies have reported on the development of LEMs in the coastal area (Salles et al., 2016; Inoue and Tanaka, 2013). Moreover, the applicability of LEMs has not been verified for both erosion and deposition over a span of 10 years. In this study, we verified the capability of the JAEAsmtp (the LEMs developed by the Japan Atomic Energy Agency) in assessing coastal landscape evolution. First, we introduced the marine deposition model to the JAEAsmtp. From the feature of Holocene sedimentary structure at six coastal areas in Japan, the deposition is modelled by the planar distribution centering at the river mouth (gaussian: silt sedimentation around offshore, jet: sand sedimentation around the river mouth). In the virtual simulation of longitudinal river profile in the postglacial period, the typical delta response to sea-level change was formed. Second, the landscape evolution over the last glacial-interglacial cycle was simulated on the Kamikita coastal plain, where marine terraces (MIS5e, 7, 9) are widely distributed. The model parameters were calibrated through reconstruction of past landscape evolution (e.g., the erosion rate of marine terraces, the depth of buried valley). The validity of the LEMs was verified through the reconstruction of recent shoreline, the position of MIS5e marine terraces, and the formation of varied valley.
