Japan Geoscience Union Meeting 2021

Presentation information

[E] Oral

A (Atmospheric and Hydrospheric Sciences ) » A-GE Geological & Soil Environment

[A-GE27] Subsurface Mass Transport and Environmental Assessment

Thu. Jun 3, 2021 10:45 AM - 12:15 PM Ch.12 (Zoom Room 12)

convener:Hirotaka Saito(Department of Ecoregion Science, Tokyo University of Agriculture and Technology), Chihiro Kato(Faculty of Agriculture and Life Science, Hirosaki University), Yuki Kojima(Department of Civil Engineering, Gifu University), Shoichiro Hamamoto(Department of Biological and Environmental Engineering, The University of Tokyo), Chairperson:Hirotaka Saito(Department of Ecoregion Science, Tokyo University of Agriculture and Technology), Zi feng Wu (Guangzhou University), Shoichiro Hamamoto(Department of Biological and Environmental Engineering, The University of Tokyo),Yuki Kojima(Department of Civil Engineering, Gifu University)

11:20 AM - 11:35 AM

[AGE27-03] [Comparison of experiments and model calculations on the capillary rise of water and the air entrapment in rock pores]

*Ryoto Nomura1, Tadashi Yokoyama2, Naoki Nishiyama3 (1.Graduate School of Advanced Science and Engineering, Hiroshima University, 2.Graduate School of Integrated Arts and Sciences, Hiroshima University, 3.University of Tsukuba)


A rock usually has pores of various sizes. Pore diameter distribution is widely measured as a fundamental information on pore structure. When a rock contacts water, water infiltrates into pores by capillary rise. In many cases, a certain amount of air is trapped in the pore and the pores are not completely filled with water. As a conventional concept that can be used to predict the extent to which the pore of each diameter is filled with water, one assumption is that the pores are filled with water in order from narrow to wide pores (van Genuchten, 1980), and the other assumption is that each pore is filled with water at the same ratio (the same ratio as whole rock water saturation) (Parker and Lenhard, 1987). Below, the former is referred to as VG model and the latter as PL model. In both the models, the amount of water in each pore diameter can be predicted based on the pore diameter distribution using a known water saturation. In this study, we are working on creating a new model that can predict the amount of water in each pore diameter in the presence of entrapped air even if the water saturation is unknown.



Our model built upon the models of Tsunazawa et al. (2016) and Yokoyama et al. (2020). The model considers a situation in which water rises from the bottom of the Y-shaped tube (radius rsupply), the tube is separated to narrow tube (radius rnarrow) and wide tube (radius rwide) at the branching point, and then merges again in an inverted-Y-shape tube at the top. When water moves in such a tube, after the branching point, water first advances only inside the narrow tube because the capillary pressure in the narrow tube is greater than that in the wide tube (Mehrabian et al. 2011; Sadjadi et al. 2015). However, when water in the narrow tube advances by a certain distance of L, the advantage of pressure in the narrow tube is lost due to the viscous resistance in the tube, and water starts to advance in the wide tube. This L value is a criterion for evaluating in which of wide tube and narrow tube the air entrapment occurs, and can be calculated theoretically by setting the values of rsupply, rnarrow, rwide, and grain size. By performing such calculations for various combinations of pore radii in the rock, the amount of water in each pore diameter in the presence of entrapped air can be predicted.



The amount of water in each pore diameter inside a water-bearing rock can be measured directly by experiments (for example, water expulsion method; Nishiyama et al., 2012; Yokoyama et al., 2020). The data of water volume in each pore diameter determined after the capillary rise experiments using various rock samples were compared with the values predicted by each of the VG model, PL model, and our model. The experimental results showed that in many cases not only narrow pores but also wide pores contained a certain amount of water. This feature can be better reproduced by the PL model and our model, compared with the VG model which predicts that narrower pores are fully filled with water whereas wider pores are kept vacant.