The Scientific Context of WFIRST Microlensing in the 2020s

As discussed in Exoplanet Science Strategy (National Academies of Sciences Engineering and Medicine 2018), WFIRST (Akeson et al. 2019) is uniquely capable of finding planets with masses as small as Mars at separations comparable to Jupiter, i.e., beyond the current ice lines of main sequence stars. In semimajor axis, these planets fall between the close-in planets found by Kepler (Coughlin et al. 2016) and the wide separation gas giants seen by direct imaging (e.g. Lagrange et al. 2009) and ice giants inferred from ALMA observations (Zhang et al. 2018). Furthermore, the smallest planets WFIRST can detect are smaller than the planets probed by radial velocity (Mayor et al. 2011; Bonfils et al. 2013) and Gaia (Perryman et al. 2014) at comparable separations. Interpreting planet populations to infer the underlying formation and evolutionary processes requires combining results from multiple detection methods to measure the full variation of planets as a function of planet size, orbital separation, and host star mass. Microlensing is the only way to find planets from 0.5 to 5M⊕ at separations of 1 to 5 au. Fundamentally, the case for a microlensing survey from space has not changed in the past 20 years: going to space allows wide-field diffraction-limited observations that can resolve main-sequence stars in the bulge, which in turn allows the detection and characterization of the smallest microlensing signals including those from planets with masses at least as small as Mars (Bennett & Rhie 2002). What has changed is that ground-based microlensing is reaching its limits, which underscores the scientific necessity for a space-based microlensing survey to measure the population of the smallest planets. Ground-based microlensing has found a break in the mass-ratio distribution at about a Neptune mass-ratio (Suzuki et al. 2016; Jung et al. 2018), implying that Neptunes are the most common microlensing planet and that planets smaller than this are rare. However, ground-based