Understanding public media channel preferences can inform preparedness plans, response strategies and long-term recovery. However, questions remain about how media consumption changes across pre-crisis, crisis, and post-crisis phases. Past theories argue that media use may change for several reasons, including during times of societal conflict and challenge. These theories point to the belief that, during a crisis, we expect media channel use to change because media preferences during a crisis will be fundamentally different compared with everyday routines. This paper takes advantage of a survey fielded to Texas residents soon after Hurricane Harvey. Here, we ask the following: (a) "what media channels are most prominent in each crisis phase?" and (b) "do media channel preferences change across crisis phases?". We use simple descriptive statistics and chi-square tests to describe media channel preferences across the three crisis phases by demographics. Additionally, we use alluvial diagrams to visualize media channel preferences over time. In total, 62% (n = 174) of respondents reported no changes in channel preferences. However, chi-square tests identified significant differences in media use changes related to a handful of demographic characteristics. These findings are explored alongside theories that would hypothesize likely media use changes across pre-crisis, crisis and post-crisis phases.