We have long known the existence of the so-called “problem geomagnetic storms,” the progenitors of which, i.e. solar eruptions, are not readily identified in Earth-based remote-sensing observations.  We propose an ISSI team to investigate the origins of recent problem geomagnetic storms and, more broadly, to advance our understanding of the formation and geoeffectiveness of interplanetary CMEs (ICMEs) without clear solar origins.  Here we call them “orphan ICMEs.”  During solar cycle 23, thanks to the continuous coverage and improved sensitivity of SOHO/LASCO, many major geomagnetic storms were traced back to wide (full or partial halo) CMEs.  For a considerable fraction of these CMEs, however, it was not possible to find the associated low coronal signatures (LCSs). In fact, it has been noted since cycle 23 that many ICMEs cannot even be paired with clear white-light CMEs near the Sun.  Furthermore, during the minimum between cycles 23 and 24, coronagraphs on the STEREO spacecraft in near-quadrature with the Earth observed Earth-directed CMEs, including CMEs with narrow angular widths, which often lacked LCSs as viewed against the solar disk. These CMEs have been called “stealth CMEs.”  The proposed ISSI team will conduct an in-depth study of the origins of problem geomagnetic storms and orphan ICMEs.  In particular, events since the launch of SDO in 2010 will be investigated, since EUV images of the corona from the Atmospheric Imaging Assembly (AIA) on SDO can reveal much weaker eruptive signatures than past instruments. This improvement is aided by STEREO observations from different vantage points that can give 3D trajectories of CMEs over large distances. MHD simulations will also be used as guide to understand how stealth CMEs are triggered and how the characteristics of the ambient solar wind affect the propagation of slow and weak CMEs in the inner heliosphere.  Our main goal is to find out if the mechanisms and processes during the eruptions responsible for problem geomagnetic storms are distinct from those associated with more typical CMEs.  Our team consists of leading experts in analyzing remote-sensing and in situ data, modeling solar and interplanetary magnetic fields, and simulations of eruptions in the corona and the heliosphere.