The Tower Flight Data Manager (TFDM) is a new terminal automation platform that will provide an integrated tower-user display suite including an extended electronic flight strip or "flight data management" (FDM) display. The integrated information exchange and processing environment established by TFDM will support a suite of automation-assisted user support tools collectively designated as the Arrival/Departure Management Tool or A/DMT. A/DMT will develop and manage an integrated plan for arrival, scheduled (and to the extent possible) non-scheduled departure operations at the airport, based on 4D-trajectory assignments. A primary concern of A/DMT is the efficient use of the runway complex to meet service demand from both arrivals and departures. In addition, A/DMT seeks to reduce fuel usage and engine emissions on the airport surface, to permit more efficient use of gates and holding areas, and to enhance the safety of surface operations. We first put forth a strategy for developing a scalable TFDM-A/DMT Information Management Architecture (TIMA) employing standard information exchange models, services and data formats. This architecture will be consistent with evolving System Wide Information Management (SWIM) technologies and data standards, and will support efficient insertion of processing algorithms (e.g. surface trajectory management algorithms) developed by the research community and/or industry. Next, we describe TIMA . While TIMA makes use of Service-Oriented Architecture (SOA) principles, it is primarily an information-oriented architecture; we discuss why this architectural style is necessary for TFDM, and how it is also beneficial for SWIM. We conclude with a description of a general model for managing temporal aspects of information within TFDM. TIMA needs to support not only real-time operations, but post-facto analysis as well. A major difficulty in conducting analyses involving different data sources is time synchronization of data. We describe a method for associating temporal information with data sources in a data-agnostic manner, so that data can be retrieved from a variety of sources in a uniform manner.