Developing Pre-Supernova Neutrino Support for sntools

This paper presents the development and validation of pre-supernova neutrino event generation capabilities within the sntools package, aiming to enable early warning detection and unified analysis of pre-collapse stellar neutrinos for current and next-generation detectors like Hyper-Kamiokande.

The Gist

Imagine a massive star is like a giant, aging furnace. For millions of years, it burns different types of fuel (hydrogen, helium, carbon, etc.) to stay hot and stable. But as it runs out of fuel, it starts to sputter and shake. Eventually, it's going to explode in a spectacular event called a Supernova.

For decades, we've been able to "hear" the explosion itself—the moment the star collapses and sends out a massive burst of energy. But until now, we've been blind to the whispers the star makes before it explodes. These whispers are pre-supernova neutrinos, tiny ghost particles that leak out during the star's final, frantic days of burning fuel.

Here is what this paper is about, broken down simply:

1. The Problem: We Need a Better "Weather App"

Currently, we have a few detectors (like giant underwater tanks) that can spot these neutrinos. However, the software we use to simulate and study them was built for the explosion, not the warning signs.

Think of it like trying to use a hurricane tracker to predict a gentle breeze. The hurricane tracker is great for massive storms, but it's too clunky and slow for the subtle shifts in wind that happen days before the storm hits.

• The Old Way: Scientists had to use a "hacky" workaround. They would calculate the neutrino energy on a piece of paper, feed that number into the computer, and then manually try to match it up with other data. It was like trying to build a house by gluing bricks together with your teeth—possible, but messy and prone to errors.
• The Goal: They wanted a unified tool that could naturally handle these "pre-explosion" whispers, making it easy to simulate what our detectors would see.

2. The Solution: Upgrading the Tool (sntools)

The team, led by Ellie O'Brien, updated a piece of software called sntools. Think of sntools as a high-end video game engine for neutrinos. It's already famous for simulating the big "bang" of a supernova. Now, they've added a new feature: Pre-Supernova Mode.

The "Time-Binning" Analogy:
One of the biggest challenges was how to slice up time in the simulation.

• The Explosion: Happens in seconds. To study it, you need to look at time in tiny slices (milliseconds), like taking a photo with a super-fast shutter speed.
• The Pre-Explosion: Happens over days. If you tried to take a photo every millisecond for three days, you'd end up with millions of empty photos (because nothing much happens in a millisecond) and your computer would crash from boredom.
• The Fix: The team figured out the perfect "slice size." They decided to look at the star in 1-second intervals. It's like switching from a high-speed camera to a time-lapse camera. You capture the slow build-up of the star's final days without wasting computer power on empty moments.

3. Why Does This Matter?

If we can simulate these neutrinos accurately, we can build a real-time early warning system.

Imagine a lighthouse keeper who can smell the smoke before the fire starts. If our detectors spot these pre-supernova neutrinos, they could send an alert to astronomers: "Hey, that star is about to go boom! Point all your telescopes at it right now!"

This would give scientists a "golden window" to watch the actual explosion from the very first second, helping us understand:

• How stars die.
• Why some explode and others don't.
• How heavy elements (like the gold in your jewelry) are forged in the chaos.

4. The Current Status

The team has successfully built this new "Pre-Supernova Mode" into the software.

• Testing: They've run "stress tests" to make sure the new software agrees with older, trusted methods. It's like checking that a new recipe tastes the same as the classic one before serving it to guests.
• Future: They are currently refining the software (a "beta" version) and plan to use it soon to help the Hyper-Kamiokande project (a massive next-generation detector in Japan) figure out exactly how sensitive it needs to be to catch these faint signals.

In a Nutshell

This paper is about upgrading our cosmic toolkit. By teaching the computer software how to listen to the "whispers" of a dying star, the team is preparing us to catch the first-ever warning of a supernova. It's about moving from reacting to the explosion to predicting it, giving us a front-row seat to one of the universe's most dramatic events.

Technical Summary

1. Problem Statement

While neutrinos from core-collapse supernova (CCSN) bursts (e.g., SN1987A) have been detected, pre-supernova neutrinos—produced during the final burning stages of a massive star prior to collapse—have not yet been observed. Detecting these neutrinos is critical for:

• Providing an early warning system for imminent supernovae, allowing astronomers to mobilize resources for observation.
• Advancing the understanding of supernova explosion mechanisms and the effects of metallicity.

Technical Challenges Identified:

• Simulation Inefficiency: Previous sensitivity studies (e.g., for Super-Kamiokande) relied on complex, approximate methods. These involved feeding weighted positron energy spectra into simulation software and separately simulating 2.2 MeV photon events. This approach failed to correctly handle outgoing positron directions and used approximate Inverse Beta Decay (IBD) cross-sections.
• Time Binning Mismatch: Standard supernova burst simulations use ~1 ms time bins. However, pre-supernova neutrino fluxes evolve over days to weeks with an exponential rise just before collapse. Using 1 ms bins for pre-supernova data results in massive inefficiency (mostly empty bins) and increased computational time, while fixed wide bins fail to capture the rapid flux increase near collapse.
• Lack of Unified Framework: There was no integrated tool to generate pre-supernova events that could be directly fed into detector simulation software for current and next-generation experiments like Hyper-Kamiokande (HK).

2. Methodology

The authors present the integration of pre-supernova neutrino models into sntools, an existing neutrino event generator originally designed for supernova burst discrimination.

• Code Adaptation: The team modified sntools to support pre-supernova event generation, enabling it to handle the specific characteristics of pre-collapse neutrino fluxes.
• Time Binning Optimization:
  • Conducted extensive studies to determine the optimal time bin size.
  • Result: A default bin size of 1 second was selected. This balances the need to capture the exponential flux rise near collapse (which 1 ms bins miss efficiently) while avoiding the inefficiency of empty bins associated with much wider bins.
  • The implementation allows user override for flexibility, enabling small experiments sensitive only to the pre-collapse peak to use finer binning.
• Model Integration: The software now supports four families of pre-supernova models (Odrzywolek, Patton, Kato, and Yoshida) by reusing their implementations within the SNEWPY framework.
• Event Generation: The updated tool generates realistic, event-by-event data including:
  • Timing information.
  • Directional information.
  • Multiple interaction channels (specifically IBD).
  • Output files compatible with downstream detector simulation software.

3. Key Contributions

• Unified Framework: Created a single, robust framework for studying both supernova burst and pre-supernova neutrinos, replacing fragmented and approximate simulation methods.
• Optimized Time Binning: Solved the computational inefficiency of simulating long-duration, low-flux pre-supernova events by implementing a dynamic 1-second default binning strategy.
• Accurate Physics Implementation:
  • Enabled complete IBD interaction event generation.
  • Correctly modeled the direction of outgoing positrons.
  • Utilized precise IBD cross-sections rather than approximations.
• Model Diversity: Integrated four distinct theoretical models, allowing for sensitivity studies that can potentially constrain these models based on future detection data.

4. Results

• Validation:
  • Internal Consistency: The generated time profiles match the expected input distributions (Figure 2). Notably, the simulation correctly reproduces the "step" feature in the neutrino emission profile approximately 2–3 hours before collapse, caused by silicon shell burning.
  • External Consistency: Comparisons with Monte Carlo data used in a 2022 Super-Kamiokande sensitivity study (Figure 3) show good agreement in positron energy spectra.
  • Discrepancy Analysis: Minor differences observed in energy spectra were attributed to differing implementations of IBD cross-sections and rejection sampling methods in the legacy code, confirming the new implementation's distinct approach.
• Sensitivity Potential: Simulations (Figure 1) demonstrate that Hyper-Kamiokande can generate distinct event spectra for different pre-supernova models (e.g., Odrzywolek vs. Patton), suggesting that detection could place strong constraints on stellar evolution models.
• Software Release: A beta version (v1.5b1) was released in October 2025 for testing.

5. Significance

• Operational Readiness: The completion of the validation process will allow sntools to be used in the official software chain for Hyper-Kamiokande to generate signal Monte Carlo data for upcoming pre-supernova sensitivity studies.
• Scientific Impact: By providing a robust tool to simulate pre-supernova neutrinos, this work prepares the scientific community to capitalize on a "golden opportunity" if a nearby CCSN occurs. It enables the transition from theoretical prediction to practical early-warning systems.
• Future-Proofing: The modular design supports future updates, including variable bin sizes and expanded model support, ensuring the tool remains relevant for next-generation neutrino detectors.

In summary, this paper bridges a critical gap in neutrino astronomy simulation tools, moving from approximate, fragmented methods to a unified, optimized, and validated framework capable of supporting the detection and analysis of pre-supernova neutrinos.