Effective decision-making in disaster contexts is critically dependent on access to relevant information about a situation. This dissertation proposes a refined decision-making model that explicitly integrates information sources within Endsley's (1995) Situational Awareness framework and also builds upon Reason's (1990) Human Error Model. This refined model emphasizes the crucial role of information sources in improving perception of a situation and thus enhancing situational awareness, particularly in the perception stage. Through five diverse case studies employing computational approaches, this work explores the generation and provision of information to enhance situational understanding in various disaster scenarios, including public health emergencies, urban search and rescue operations, crisis information management, and environmental hazard mitigation. The first case study quantifies the impact of a satisficing algorithm on route efficiency for medical countermeasure dispensing during a public health emergency. The second case study evaluates the feasibility of employing small Uncrewed Aircraft Systems for wide-area search in urban search and rescue operations through simulation. The third and fourth case studies leverage large language models (LLMs): mining unstructured storm narratives for impacts and benchmarking zero-shot classification of crisis-related social media content. Finally, the last case study develops prioritization strategies for infrastructure improvements based on environmental and population risk, utilizing historical sewage spill data and a geospatial over-land spill model. These computational methodologies illustrate empirical information sources within a larger decision-making context for a variety of disaster scenarios, ultimately enhancing decision-makers' perception of a situation thus improving decision-making.