🔭 astrophysics

Estimating the Luminosities of Protostars with Limited Infrared Photometry

This study demonstrates that protostellar luminosities can be estimated with significantly improved accuracy using limited infrared photometry, particularly by combining multiple JWST filter measurements or utilizing far-infrared fluxes at 70–160 microns, which reduce uncertainty factors to 2 or lower compared to previous single-wavelength methods.

Original authors: Michael M. Dunham, Aina Palau, Nuria Huélamo, Eduard I. Vorobyov, Zach Yek, Sean Rand

Published 2026-04-14

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Michael M. Dunham, Aina Palau, Nuria Huélamo, Eduard I. Vorobyov, Zach Yek, Sean Rand

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The "Baby Star" Mystery

Imagine you are trying to figure out how much a newborn baby is eating and growing, but the baby is wrapped in a thick, fuzzy blanket that blocks your view. You can't see the baby directly, and you can't weigh it on a scale.

In astronomy, these "babies" are protostars (stars in their earliest stages of life). They are buried deep inside clouds of gas and dust. To understand how they grow, astronomers need to know their luminosity (how much energy they are putting out). This energy tells us how fast they are eating (accreting mass) and how heavy they are.

Traditionally, to get an accurate reading of a protostar's energy, astronomers had to build a complete "energy map" (called a Spectral Energy Distribution or SED). This meant collecting data from every possible color of light, from the shortest infrared waves to the longest radio waves. It's like trying to guess a person's weight by measuring their temperature, their heartbeat, their height, and their shoe size all at once. It's accurate, but it takes a massive amount of time and expensive telescope time.

The Problem: The "Fuzzy Blanket" is Too Thick

The problem is that for stars far away from us, we can't get all that data. The telescopes we used in the past (like Spitzer and Herschel) weren't sensitive enough to see the faint, distant stars clearly across all those different wavelengths. It's like trying to hear a whisper in a noisy stadium; you can only catch a few words.

The Solution: The "Shortcut" Recipe

This paper asks a simple question: Can we guess a protostar's total energy just by looking at a few specific colors of light?

The authors used a giant library of computer simulations (think of them as "digital twins" of baby stars) to test this. They simulated thousands of stars collapsing and glowing, then checked: If we only look at the light coming from one specific color, can we guess the total energy?

Here are their main findings, explained simply:

1. The "Magic Number" (70 Microns)

Previous studies found that if you look at a baby star at a specific infrared color called 70 microns (a very deep red, invisible to our eyes), it correlates very well with the star's total energy.

The Analogy: Imagine trying to guess how much a furnace is burning. You can't see the fire, but if you stick your hand near the vent and feel the heat at a specific spot, you can guess the furnace's power.
The Update: The authors confirmed this "magic number" works, but they found that the old recipes were underestimating the heat. The old formulas said the stars were about 2 to 3 times dimmer than they actually are. They have now updated the recipe to be more accurate.

2. The "Sweet Spot" Range

They didn't just stop at 70 microns. They tested 100 different colors of light.

The Finding: Any single measurement between 40 and 350 microns (the "Far-Infrared" range) gives a decent guess (within a factor of 3).
The Best Spot: The range between 70 and 160 microns is the "Goldilocks zone." If you measure the light here, your guess is usually within a factor of 2 of the real value. It's like finding the perfect spot on the furnace vent where the heat tells you exactly how hard it's working.

3. The New Super-Telescope (JWST) Challenge

Enter the James Webb Space Telescope (JWST). It is the most powerful infrared telescope ever built. It can see baby stars in distant galaxies that we've never seen before.

The Problem: JWST sees light at much shorter wavelengths (0.6 to 28 microns). The paper found that looking at just one of these short wavelengths is like trying to guess the furnace's power by looking at a single, flickering spark. It's very unreliable. The uncertainty is huge (you might be off by a factor of 4 to 10!).
The Reason: Short wavelengths are easily blocked or scattered by the dust blanket. They tell you more about the shape of the dust cloud than the actual power of the star.

4. The "Three-Filter" Trick

But wait! The authors found a clever workaround for JWST.

The Analogy: If you can't guess the weather by looking at a single cloud, maybe you can guess it by looking at the sky, the wind, and the humidity all at once.
The Discovery: If you take measurements from three different JWST filters (three different colors) at the same time, you can combine them to get a very accurate guess.
The Result: Using three specific JWST filters allows astronomers to guess the star's energy with the same accuracy as the old, expensive method of measuring the single "magic" 70-micron wavelength.

Why This Matters

This paper is like giving astronomers a new, faster, and cheaper way to do their job.

Speed: Instead of spending weeks gathering data for every single star, they can now use existing JWST data (which is being collected in huge surveys) to estimate the energy of thousands of baby stars instantly.
Scale: We can now study baby stars in distant galaxies that were previously too faint to measure. This helps us understand if stars form the same way everywhere in the universe.
Accuracy: They fixed the old math that was making stars look too dim, ensuring our understanding of how stars grow is correct.

The Bottom Line

You don't need to see the whole picture to understand the story. By using the right "clues" (specific wavelengths of light) and the right "math" (the new formulas), astronomers can now accurately guess the power of baby stars even when they are hidden deep inside their dusty nurseries. And thanks to the new "Three-Filter Trick," the James Webb Space Telescope is perfectly suited to solve this mystery for stars across the entire universe.

1. Problem Statement

Accurately determining the mass and mass accretion rates of protostars in their earliest evolutionary stages is critical for understanding star formation. However, direct measurement of protostellar mass requires detecting Keplerian rotation in circumstellar disks, which is observationally expensive and limited to a small number of nearby targets. Consequently, astronomers rely on bolometric luminosity ( $L_{bol}$ ) as an indirect tracer.

Traditionally, calculating $L_{bol}$ requires assembling a complete Spectral Energy Distribution (SED) from near-infrared to millimeter wavelengths. This is often infeasible for distant star-forming regions where current far-infrared (FIR) and submillimeter telescopes lack the sensitivity or angular resolution to resolve individual protostars. While previous studies (Dunham et al. 2008; Huard & Terebey 2017) established correlations between luminosity and flux at specific wavelengths (notably 70 $\mu$ m), these studies were limited by:

Narrow parameter spaces in their radiative transfer models.
A lack of applicability to the new, high-resolution capabilities of the James Webb Space Telescope (JWST), which operates primarily in the near- and mid-infrared (0.6–28 $\mu$ m).

2. Methodology

The authors utilized a comprehensive database of 9,990 evolutionary radiative transfer models derived from 23 hydrodynamical simulations of collapsing dense cores (Dunham & Vorobyov 2012). These models span a wide range of initial conditions, core masses, protostellar masses, and evolutionary stages (Stage 0 and Stage I).

Key Methodological Steps:

Model Weighting: To prevent bias from the specific simulation setup (which over-represents high-mass stars, low-inclination views, and high-luminosity accretion bursts), the authors assigned a weight ( $w_{total}$ $w_{t o t a l}$ ) to each model based on:
- IMF Weight ( $w_{IMF}$ ): Based on the Kroupa (2002) Initial Mass Function.
- Inclination Weight ( $w_{inc}$ ): Based on the solid angle subtended by the viewing angle.
- Time Weight ( $w_{time}$ ): Based on the duration of the timestep, correcting for the over-sampling of accretion bursts.
Correlation Analysis: The authors performed weighted linear least-squares fits in log-log space to correlate the internal luminosity ( $L_{int}$ ) with flux ( $F_\nu$ ) at various wavelengths.
JWST Filter Analysis: They tested single-wavelength correlations and multi-wavelength combinations using all 17 wide-band filters available on JWST's NIRCam and MIRI instruments.
- Two-Filter Fits: Used the form $\log(L_{int}) = c_0 + c_1 \log(F_{\lambda1}) + c_2 \log(F_{\lambda2})$ .
- Three-Filter Fits: Extended the fit to include a third flux term.
Correction for Systematic Bias: Recognizing that multi-dimensional fits might minimize residuals but fail to produce a 1:1 slope between estimated and intrinsic luminosity, the authors applied a secondary "Luminosity vs. Luminosity" (LvL) correction to ensure the estimated values are unbiased across the full luminosity range.

3. Key Contributions and Results

A. Re-evaluation of the 70 $\mu$ m Correlation

Confirmation and Revision: The study confirms a tight correlation between $L_{int}$ $L_{in t}$ and flux at 70 $\mu$ $μ$ m ( $F_{70}$ $F_{70}$ ). However, the authors demonstrate that previous relations (D08 and HT17) systematically underestimate luminosities.
- The new relation yields luminosities 1.9 times higher than HT17 and 3.2 times higher than D08 on average.
- This discrepancy is attributed to the broader parameter space (including lower-mass cores) covered by the new models.
Uncertainty: Using the updated 70 $\mu$ m correlation, the estimated $L_{int}$ has a $1\sigma$ uncertainty of a factor of 2.0 (0.302 dex).

B. Optimal Wavelengths for Single-Filter Estimates

The authors analyzed 100 wavelengths from 0.1 $\mu$ m to 36,000 $\mu$ m.
Best Single Wavelengths: The range 40 $\mu$ m – 350 $\mu$ m provides estimates with a $1\sigma$ uncertainty of a factor of 3 or lower.
Optimal Range: The range 70 $\mu$ m – 160 $\mu$ m reduces the uncertainty to a factor of 2 or lower.
Peak Performance: The best single wavelength is 100 $\mu$ m (Herschel PACS band), offering a $1\sigma$ uncertainty of a factor of 1.8 (0.254 dex), slightly better than 70 $\mu$ m.

C. JWST Observations: Single vs. Multi-Filter

Single Filter Limitations: Single JWST filters (0.6–27.9 $\mu$ m) correlate poorly with $L_{int}$ due to sensitivity to geometry and viewing angle. Uncertainties range from a factor of 4 (at 25.5 $\mu$ m) to a factor of 10 (at 0.7 $\mu$ m).
Two-Filter Strategy: Combining two JWST filters significantly improves reliability. The best combination (F356W and F2550W) reduces the uncertainty to a factor of 3.1 (0.493 dex).
Three-Filter Strategy: Using three filters simultaneously yields results comparable to single FIR measurements.
- Best Combination: F070W, F115W, and F2100W.
- Resulting Uncertainty: A factor of 2.1 (0.318 dex).
- This is nearly as accurate as the best single FIR flux measurements (100 $\mu$ m), despite using only near- and mid-infrared data.

4. Limitations and Future Work

Scattering Effects: The radiative transfer models used assume isotropic scattering. In reality, dust grains in dense cores preferentially forward-scatter light. The authors estimate this introduces an additional uncertainty of 0.1–0.2 dex at short wavelengths. Future work requires models with anisotropic scattering.
Mass Function: The simulation set lacks high-mass cores ( $>2 M_\odot$ ). While these would be down-weighted by the IMF, future studies should include higher-mass models to fully validate the correlations across the entire stellar mass spectrum.

5. Significance

This paper provides a crucial toolkit for the JWST era of star formation studies.

Enabling Large Surveys: It allows astronomers to estimate protostellar luminosities for thousands of objects in distant star-forming regions where FIR data is unavailable, simply by using limited JWST photometry (specifically 3 filters).
Correcting Historical Biases: It corrects the systematic underestimation of protostellar luminosities in previous literature, which impacts theories regarding episodic accretion and the "luminosity problem."
Methodological Framework: The demonstration that multi-filter near/mid-IR data can rival single FIR data accuracy opens new avenues for characterizing the physical properties (mass, accretion rate) of protostars in the most distant and crowded star-forming regions.