The Earth's electron outer radiation belt is a highly variable region in which the populations can vary by several orders of magnitude in a minute to hours. Such extreme dynamics depends on a complex and delicate balance between source and loss processes that are ultimately driven by the interactions between the interplanetary medium and the Earth’s magnetosphere. In recent years, several efforts have been carried out to improve our understanding of the controlling processes driving radiation belt dynamics and to improve the predictability of the relativistic electrons that populate it. In the first part of this dissertation, we study the solar wind parameters that are relevant for the understanding and prediction of relativistic electron enhancement events and relativistic electron persistent depletion events at geostationary orbit and what have the largest potential for prediction. We then use these results to explore the extent of the effects of the solar wind through the outer belt by comparing relativistic electron enhancement events at geostationary orbit with the response at lower radial distances and so try to understand how deep within the inner magnetosphere the solar wind influence can reach. In the second half of this dissertation we explore the recently discovered phenomena of ultrarelativistic remnant belts that can lead to a triple belt configuration of the Earth's radiation belts. Such events were first reported shortly after the launch of the Van Allen Probes mission but are poorly understood. We identify three-belt events to characterize their occurrence rate, and the geomagnetic conditions under which they occur. We investigate their location, characteristic energy and general properties of the magnetosphere and solar wind that can favor their formations. We finally look at remnant belt persistence by statistically calculating their lifetime and decay rates. Using these results, we study the physical mechanism of ultrarelativistic remnant belts decay by comparing with previously reported analytical estimates.