Mass measurements of gravitational microlenses require one to determine the microlens parallax \pi _ { E } , but precise \pi _ { E } measurement , in many cases , is hampered due to the subtlety of the microlens-parallax signal combined with the difficulty of distinguishing the signal from those induced by other higher-order effects .
In this work , we present the analysis of the binary-lens event OGLE-2017-BLG-0329 , for which \pi _ { E } is measured with a dramatically improved precision using additional data from space-based Spitzer observations .
We find that while the parallax model based on the ground-based data can not be distinguished from a zero- \pi _ { E } model at 2 \sigma level , the addition of the Spitzer data enables us to identify 2 classes of solutions , each composed of a pair of solutions according to the well-known ecliptic degeneracy .
It is found that the space-based data reduce the measurement uncertainties of the north and east components of the microlens-parallax vector \mbox { \boldmath$ \pi$ } _ { E } by factors \sim 18 and \sim 4 , respectively .
With the measured microlens parallax combined with the angular Einstein radius measured from the resolved caustic crossings , we find that the lens is composed of a binary with components masses of either ( M _ { 1 } ,M _ { 2 } ) \sim ( 1.1 , 0.8 ) M _ { \odot } or \sim ( 0.4 , 0.3 ) M _ { \odot } according to the two solution classes .
The first solution is significantly favored but the second can not be securely ruled out based on the microlensing data alone .
However , the degeneracy can be resolved from adaptive optics observations taken \sim 10 years after the event .