Solar flares are sudden releases of magnetic energy, driving adverse space weather that can disrupt radio communications, degrade GPS accuracy, and pose risks to satellites and astronauts. While multi-day forecasts exist, operational entities require actionable, minute-scale nowcasts for immediate mitigation. This talk presents a new, observationally-driven nowcasting approach to address this critical gap.

We introduce a real-time method using high-cadence observations of the chromosphere in the Ca II K line (393.4 nm) from the Dunn Solar Telescope. An efficient algorithm monitors intensity in active regions to detect the sharp, impulsive rise phase of a flare, a direct thermal response to initial particle beams. We compare the timing of these Ca II K peaks against the standard 1-8 Å soft X-ray (SXR) flux curves from NOAA’s GOES satellites, which define the official flare peak.
Our analysis of C- to M-class flares demonstrates a consistent, operationally significant result: the impulsive Ca II K peak typically precedes the GOES SXR peak by 20 to 30 minutes. This lead time provides a reliable warning signal, offering a crucial window for stakeholders to take protective action before the flare's most intense phase.

Additionally, we present a complementary analysis of an X-class event, revealing a distinct pattern of multiple, sequential secondary brightenings deep into the post-flare decay phase. These "aftershocks" suggest continued, smaller-scale energy release not captured by standard models. We will discuss representative examples and explore working hypotheses for their drivers.

This talk connects the development of a high-impact operational nowcast with new scientific questions about flare energy release. We will conclude by charting next steps for refining this method and its potential for adoption across a global network of ground-based solar observatories.