SPHEREx 0.75 to 5 $μ$m Spectra for a Sequence of Nearby Brown Dwarfs

Abstract: The SPHEREx all-sky survey has now measured the R$\sim$40-100 infrared spectra of thousands of nearby brown dwarfs in the chemically rich 0.75-5 $μ$m range. The survey's wide spectral coverage and high S/N permits flux measurements that capture several broadband molecular absorption features, and upwards of 80$\%$ of the total bolometric luminosity of most brown dwarfs. Atmospheric models are known to yield systematic disagreements in the inferred temperatures and radii of brown dwarfs, necessitating benchmarking against observations. In this work, we present SPHEREx spectra across a broad sequence of 37 nearby field brown dwarfs, ranging from L0 to Y4 ($\sim$2500-250 K) and compare them to theoretical expectations. We additionally compile spectra for separate low-gravity and low-metallicity objects, and show how they trend with constant spectral type. We fit the measured spectra to the well-known forward model grids Sonora Diamondback, Elf Owl, BT-Settl, ATMO2020 and ATMO2020++ and compare their goodness-of-fit as a function of wavelength, spectral type, and treatment of clouds and chemistry. We find that the models continue to struggle to simultaneously fit the J/H/K peaks and the 4 $μ$m opacity window, especially in L/T transition objects. The largest deviations appear around the chemistry-sensitive CO$2$ and CO features. Despite these offsets, the models broadly capture their trends across the L/T transition, with the observed sample of field dwarfs strongly preferring the weak vertical mixing ($k\mathrm{zz}$ = 10$^4$ cm$^2$s$^{-1}$) Elf Owl models over strong mixing. The spectra shown here along with future SPHEREx data will help guide improvements to models.

• The paper offers detailed 0.75–5 μm spectra for 37 nearby brown dwarfs, establishing empirical benchmarks for atmospheric models.
• It rigorously compares observations with five atmospheric model grids, revealing significant discrepancies in predicting molecular features and vertical mixing.
• Findings emphasize the need for improved models that integrate cloud physics and non-equilibrium chemistry to accurately capture brown dwarf atmospheres.

SPHEREx Infrared Spectroscopy of Nearby Brown Dwarfs: Benchmarks and Model Discrepancies

Introduction

This study presents a comprehensive sequence of 0.75–5 $μ$m spectra of 37 nearby brown dwarfs—spanning spectral types L0 to Y4—measured by the SPHEREx all-sky survey, and conducts a rigorous comparison against leading substellar atmospheric model grids. SPHEREx offers spectral resolution $R \sim 40$–$100$ and signal-to-noise ratios that enable unprecedented characterization of dominant molecular features shaping brown dwarf atmospheres, capturing upwards of 80% of their emergent bolometric luminosity. By uniting targets of typical field age and composition with sequences at low gravity and subsolar metallicity, this work establishes critical empirical benchmarks for atmospheric models, which are essential due to persistent systematic discrepancies in their derivation of brown dwarf parameters.

SPHEREx Survey and Spectral Scope

SPHEREx, a NASA Medium Explorer mission, provides all-sky infrared spectrophotometry in 102 channels from 0.75–5 $μ$m, achieving deep sensitivity and fine spectral sampling across its wavelength coverage. This vastly improves upon the broadband photometry of previous all-sky surveys (e.g., 2MASS, WISE) by resolving the molecular bands and continuum windows crucial for atmospheric diagnostics. The SPHEREx dataset enables robust mapping of core opacity features—H$_2$O, CH$_4$, CO, CO$_2$, and H$_2$ CIA—across the full field brown dwarf sequence, and includes coverage of rarely sampled 3–5 $μ$m bands that are sensitive to disequilibrium chemistry and vertical atmospheric mixing.

The dominance of various molecular species as a function of wavelength and their interplay in the SPHEREx spectral regime is evident:

Figure 1: The 0.75–5 $μ$m SPHEREx spectrum for a representative T7 field dwarf, illustrating primary molecular opacity contributors. H$R \sim 40$0O and CH$R \sim 40$1 bands alternate throughout, with CO and CO$R \sim 40$2 features emerging at longer wavelengths, and a newly apparent weak CH$R \sim 40$3 emission at 3.3 $R \sim 40$4m signaling possible thermal inversion.

Observational Characterization of Brown Dwarf Sequences

Spectral Trends in Field, Low-Gravity, and Low-Metallicity Subsamples

The field sequence from L0 to Y4 is systematically delineated by temperature-sensitive molecular band depth, interplay of opacity windows, and transitions of dominant chemical species. Across the L dwarfs, the development of CH$R \sim 40$5 absorption and sharp J/H/K window profile changes are observed, with a marked reversal of peak heights through the L/T transition—a fingerprint of changing atmospheric chemistry and cloud clearing:

Figure 2: The SPHEREx field sequence, normalized by the 3.7–4.7 $R \sim 40$6m flux, displays the evolution of spectral morphology and molecular features as a function of spectral type.

Low-gravity sequences (L0, L5) show reduced water feature amplitude and pronounced K-band peaks, indicative of broader, more isothermal photospheres dominated by cloud scattering. Low-metallicity objects, especially amongst L subtypes, manifest in significantly higher NIR flux ratios and the dominance of H$R \sim 40$7-H$R \sim 40$8 CIA opacity at higher pressures as metallicity decreases:

Figure 3: A progression in L0/L7 brown dwarfs with decreasing [Fe/H], emphasizing the enhanced NIR continuum and diminished molecular absorption in lower-metallicity objects.

Quantitative Model Comparison

A suite of five contemporary atmospheric model grids (ATMO2020, ATMO2020++, BT-Settl, Sonora Diamondback, Sonora Elf Owl) were used for forward-model comparison against the SPHEREx data. Despite matching nominal physical inputs (e.g., $R \sim 40$9, $100$0, [Fe/H]), substantial model-to-model discrepancies are found—especially in the 3–5 $100$1m windows that are critically sensitive to carbon chemistry and vertical mixing, with flux variations often reaching factors of $100$2:

Figure 4: Comparison of model spectra at identical physical parameters, revealing strong departures in mid-infrared (3–5 $100$3m) flux due to differences in chemistry and mixing assumptions.

Model fits across the sample routinely yield $100$4, even for high S/N spectra, emphasizing that parameter inferences are dominated by model systematics rather than observational errors.

Spectral Indices and Chemical Diagnostics

Custom spectral indices centered on CH$100$5, CO, and CO$100$6 features were defined relative to the 4 $100$7m window to extract chemical evolution trends with spectral type and to benchmark model capabilities:

Figure 5: Definition of CH$100$8, CO, and CO$100$9 indices on a T6 dwarf spectrum, normalized at the relatively continuum-like 4 $μ$0m window.

These indices exhibit a monotonic progression across the sequence, tightly correlating with spectral type and atmospheric temperature. Notably, the observed CO and CO$μ$1 indices decrease sharply between T6 and T7 (corresponding to $μ$21000–700 K), mapping the turnover to CH$μ$3-dominated chemistry. Model grids broadly reproduce the qualitative trend but differ significantly in amplitude and locus; notably, the Sonora Elf Owl models with weaker vertical mixing (log $μ$4) are favored by the data, in contrast to the higher mixing rates commonly inferred from other studies.

Figure 6: Observed and model-derived CH$μ$5 and CO index strengths, showing a sharp CO downturn near T6/T7, while early T/L types provide closer alignment with equilibrium chemistry assumptions.

Figure 7: A similar comparison for the CH$μ$6/CO$μ$7 indices reinforces the trend: CO$μ$8 absorption diminishes with advancing spectral subtype and is again poorly matched by all models beyond T6.

Degeneracies and Model Limitations

Model explorations across metallicity, gravity, C/O, and $μ$9 parameter axes demonstrate that the spectral indices are simultaneously sensitive to multiple atmospheric inputs, rendering interpretation highly degenerate—particularly regarding mixing rate versus metallicity effects on carbon abundances:

Figure 8: The relationship between CH$_2$0 and (CO–CO$_2$1) index illustrates the multidimensional degeneracy, with empirical sequences poorly reproduced by any single axis of model variation.

Implications and Outlook

This work underscores that despite advances in atmospheric modeling and laboratory molecular data, current substellar model grids are systematically unable to reproduce the field sequence of high S/N brown dwarf spectra at the precision offered by SPHEREx. The model deficiencies are especially acute at the L/T transition and in features linked to disequilibrium carbon chemistry. The strong preference for weak vertical mixing models in field dwarfs is not fully commensurate with expectations from population-level studies, highlighting a disconnect between model input physics and observed trends.

Practically, the SPHEREx spectral atlas provided can serve as a calibration resource for future substellar and exoplanet atmospheric studies, as well as a safeguard to prevent brown dwarfs from contaminating high-$_2$2 galaxy searches in deep survey data. From a theoretical standpoint, these benchmarks emphasize the need for integrated models that connect cloud microphysics, atmospheric mixing, and molecular opacities in a manner that self-consistently reproduces observed substellar SEDs.

Conclusion

The SPHEREx survey delivers benchmark-quality spectra for 37 nearby brown dwarfs over the 0.75–5 $_2$3m range, documenting empirical trends across temperature, surface gravity, and metallicity. Major atmospheric model grids fail to fit the detailed features in these spectra, especially where vertical mixing and cloud sedimentation affect mid-infrared chemistry-sensitive windows. The data suggest model improvements are urgently needed at the interface of non-equilibrium chemistry and cloud formation physics. As the SPHEREx mission continues, expanded samples and increased S/N will reshape empirical characterization and, by extension, direct model refinement, ultimately benefitting both substellar and exoplanetary atmospheric studies (2607.00543).