Assessing habitat quality is a major goal of conservationists and restoration practitioners, but to associate habitat quality with biomarkers of vagile animals, the biomarkers must respond rapidly. Here we identified a biomarker capable of rapidly detecting food limitation in the imperiled Delta Smelt (Hypomesus transpacificus), a pelagic fish endemic to the San Francisco Estuary (SFE). We conducted an experiment with fed and unfed treatments of hatchery-raised, sub-adult Delta Smelt that were sampled at 12 time points: 0, 1, 2, 3, 4, 5, 6, 7, 9, 11, 14, and 21 days. We then compared four biomarkers using Day 21 fish: RNA/DNA in liver, triglycerides in liver, glycogen in liver, and glycogen in muscle. Of the three liver endpoints, glycogen had the largest, most significant difference between treatments at Day 21, so we compared it to muscle glycogen across all time points. Liver glycogen declined after just one day of fasting and remained depressed in the fasting treatment across all subsequent time points. Muscle glycogen also responded rapidly, taking only two days to decline in the fasting treatment, but the difference was small and inconsistent across subsequent time points. When applied to hatchery-released Delta Smelt collected from the SFE, we found that liver glycogen concentrations were less than half that of the fed hatchery fish, consistent with the hypothesis of food limitation in the wild, but also several other potential causes. This study highlights the utility of biochemically determining liver glycogen concentrations of wild-caught Delta Smelt to provide insight into local habitat quality.