11:15 AM - 11:30 AM
[ACG43-08] Comparison of terrestrial analysis results using the Japan Ocean-River-RunOff model (JORRO) with existing hydrologic models and observed flow rates.
Keywords:hydrological model, ocean model
To evaluate the ecosystems of brackish and coastal areas where freshwater, seawater, and nutrients from the ground surface mix, it is necessary to analyze the behavior of water in rivers and oceans. In the past, calculations have been performed by combining different river and ocean models, but the water flow at the boundary between them has yet to be fully understood. JORRO, the Japan Ocean-River-RunOff model used in this presentation, can analyze water flow in terrestrial and oceanic areas. However, compared to existing hydrological models, the accuracy of the analysis over land has not been sufficiently verified, and the effects of land use and groundwater have yet to be introduced. In this presentation, we compare the results of JORRO's overland flow analysis with the hydrologic model CDRMV3.3.1 and observed flows.
Based on the ocean layer thickness model, JORRO considers freshwater and seawater to be different layers due to the difference in density, resulting in a one-layer model for land areas and a two-layer model for ocean areas. The model's simplicity makes it easy to apply it to other areas and evaluate the calculation results using input data logically. On the other hand, CDRMV3.3.1, an existing hydrologic model, models the soil in three layers, allowing calculations that consider not only surface runoff but also intermediate flows. However, parameters such as thickness and porosity of each layer are identified by calibration with observed flow rates, which makes it challenging to transfer the values to different calculations, although the reproducibility is high.
The analysis was performed for the Shirakawa River in Kumamoto, using rainfall data from radar AMeDAS during a typhoon. The Shirakawa River flows through the gentle terrain inside the caldera, the steep slope in Tateno, and the gentle terrain in the urban area of Kumamoto. By analyzing this terrain, JORRO proved to capture the river flow as it changes from subcritical, supercritical, to subcritical flow. The midstream flow stations reproduced two peaks similar to the observed flow, and compared to the results of CDRMV 3.3.1 before optimization, the response from rainfall is closer to the observed flow. However, there are deviations in their absolute values and attenuation. This presentation will detail the causes and solutions to this problem.
Based on the ocean layer thickness model, JORRO considers freshwater and seawater to be different layers due to the difference in density, resulting in a one-layer model for land areas and a two-layer model for ocean areas. The model's simplicity makes it easy to apply it to other areas and evaluate the calculation results using input data logically. On the other hand, CDRMV3.3.1, an existing hydrologic model, models the soil in three layers, allowing calculations that consider not only surface runoff but also intermediate flows. However, parameters such as thickness and porosity of each layer are identified by calibration with observed flow rates, which makes it challenging to transfer the values to different calculations, although the reproducibility is high.
The analysis was performed for the Shirakawa River in Kumamoto, using rainfall data from radar AMeDAS during a typhoon. The Shirakawa River flows through the gentle terrain inside the caldera, the steep slope in Tateno, and the gentle terrain in the urban area of Kumamoto. By analyzing this terrain, JORRO proved to capture the river flow as it changes from subcritical, supercritical, to subcritical flow. The midstream flow stations reproduced two peaks similar to the observed flow, and compared to the results of CDRMV 3.3.1 before optimization, the response from rainfall is closer to the observed flow. However, there are deviations in their absolute values and attenuation. This presentation will detail the causes and solutions to this problem.