The length change of a rod of an extensometer, which is caused by a temperature change and force acting on an extensometer, produces an error in the measurement by an extensometer. The length change of a rod depends on its coefficient of linear expansion, its coefficient of elasticity and so on. Observations by the super-invar extensometer and the fused quartz extensometer installed on the common base have been carried out at Osakayama Crustal Deformation Observatory. The measurement by the super-invar extensometer (E4B) and one by the fused quartz extensometer (E5B) are in good agreement except the case of remarkable temperature changes. It is concluded that measurements by extensometers used in the present study are the sum of the ground strain and the length change of rods of extensometers, i.e. super-invar and fused quartz, caused by a temperature change. Coefficients of linear expansion of super-invar and fused quartz on the observational condition have been estimated by making use of the phenomenon that an adiabatic temperature change caused by a pressure change whose period is a few hours produces only the length change of the extensometer. The coefficient of linear expansion of super-invar is negative and its absolute value is about 1 × 10
-7/°C. That of fused quartz is positive and its value is about 1 × 10
-7/°C. These values are effective in case that periods of temperature changes are less than 1 day. In an analysis of Earth tidal strains, the combination of the measurement by the super-invar extensometer and one by the fused quartz extensomemeter, i.e. (E4B+E5B)/2 is free from temperature changes and indicates the ground strain. As a result of analyzing (E4B+E5B)/2, amplitude ratios to Body tidal strains (1=0.0832, h=0.6114) are constant of 0.5-0.6 for M
2, S
2, and O
1: and the amplitude ratio for K
1 varies during the period from October 1983 to March 1984.
View full abstract