TY - JOUR
T1 - Rigid migration of Unzen lava rather than flow
AU - Goto, Akio
AU - Fukui, Keiichi
AU - Hiraga, Takehiko
AU - Nishida, Yasunori
AU - Ishibashi, Hidemi
AU - Matsushima, Takeshi
AU - Miyamoto, Tsuyoshi
AU - Sasaki, Osamu
N1 - Funding Information:
We are grateful to T. Nakakoji, K. Sueyoshi and K. Yabe for their supports on viscosity measurements, and late N. Matsuwo who recorded and left the video images distributed from NHK during the activity. Discussions with S. Nakada, and S. Okumura were helpful. Critical reviews by Nick Varley and anonymous reviewers improved our manuscript substantially. Editorial handling by Kelly Russell is gratefully acknowledged. This study was supported by grant in aid for cooperative work from Earthquake Research Institute, University of Tokyo , Japan.
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/12/1
Y1 - 2020/12/1
N2 - We have reassessed the mode of advance of lava erupted in the 1990s from Unzen Fugen-dake Volcano in northwestern Kyushu, Japan. The apparent viscosity of Unzen lava had been estimated to be 0.9–4.2 × 1010 Pa s from the observed lava surface velocity in 1991, using Jeffreys model, based on the assumption that the lava was a Newtonian or Bingham fluid. In contrast, the viscosity of synthetic groundmass glass derived from Unzen lava had been measured in the laboratory at expected lava temperatures (780 °C–880 °C) to be 1011–1013 Pa s, considerably higher than the viscosity based on field observation. If the suspension effect of crystals is taken into consideration for groundmass glass, several orders of magnitude higher viscosity than observed should be expected for Unzen lava. To seek the source of the disparity in viscosity estimates, we first performed uniaxial-compression viscometry on natural Unzen lava samples collected in 1991 and thereby confirmed that the lava viscosity was actually higher than determined by field observations and equal to or higher than the fluid–solid threshold viscosity (1012 Pa s). This suggests that the observational viscosity had been underestimated. From reanalysis of the lava migration using video and photographs, contemporaneous with the field-based viscosity estimation, we found that the lava behavior was far from that of a fluid. Apparently, the lava lobe lost its fluidity at the flow front and instead advanced rigidly, thereby indicating that the application of Jeffreys model had incorrectly led us to underestimate viscosity. Although we used only the 1991 lava samples and image data, we infer that rigid migration was common for all exogenous lava lobes during the 1991–1995 activity. Our comparative study demonstrates that cooled and/or degassed lavas could show higher mobility than expected from experiment- or model-based viscosity. Apparently this occurs when the frictional force is smaller between lobe base and underlying slope than the viscous resistance inherent in lava.
AB - We have reassessed the mode of advance of lava erupted in the 1990s from Unzen Fugen-dake Volcano in northwestern Kyushu, Japan. The apparent viscosity of Unzen lava had been estimated to be 0.9–4.2 × 1010 Pa s from the observed lava surface velocity in 1991, using Jeffreys model, based on the assumption that the lava was a Newtonian or Bingham fluid. In contrast, the viscosity of synthetic groundmass glass derived from Unzen lava had been measured in the laboratory at expected lava temperatures (780 °C–880 °C) to be 1011–1013 Pa s, considerably higher than the viscosity based on field observation. If the suspension effect of crystals is taken into consideration for groundmass glass, several orders of magnitude higher viscosity than observed should be expected for Unzen lava. To seek the source of the disparity in viscosity estimates, we first performed uniaxial-compression viscometry on natural Unzen lava samples collected in 1991 and thereby confirmed that the lava viscosity was actually higher than determined by field observations and equal to or higher than the fluid–solid threshold viscosity (1012 Pa s). This suggests that the observational viscosity had been underestimated. From reanalysis of the lava migration using video and photographs, contemporaneous with the field-based viscosity estimation, we found that the lava behavior was far from that of a fluid. Apparently, the lava lobe lost its fluidity at the flow front and instead advanced rigidly, thereby indicating that the application of Jeffreys model had incorrectly led us to underestimate viscosity. Although we used only the 1991 lava samples and image data, we infer that rigid migration was common for all exogenous lava lobes during the 1991–1995 activity. Our comparative study demonstrates that cooled and/or degassed lavas could show higher mobility than expected from experiment- or model-based viscosity. Apparently this occurs when the frictional force is smaller between lobe base and underlying slope than the viscous resistance inherent in lava.
KW - Bingham fluid
KW - Lava viscosity
KW - Newtonian fluid
KW - Rigid body
KW - Unzen Fugen-dake
UR - http://www.scopus.com/inward/record.url?scp=85096229439&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85096229439&partnerID=8YFLogxK
U2 - 10.1016/j.jvolgeores.2020.107073
DO - 10.1016/j.jvolgeores.2020.107073
M3 - Article
AN - SCOPUS:85096229439
VL - 407
JO - Journal of Volcanology and Geothermal Research
JF - Journal of Volcanology and Geothermal Research
SN - 0377-0273
M1 - 107073
ER -