Discovery of a New Spectral Transition in Swift J0243.6+6124 in the Sub-Eddington Regime

Technical Summary: Discovery of a New Spectral Transition in Swift J0243.6+6124

Problem Statement
Accreting X-ray pulsars (AXRPs) exhibit complex spectral and timing evolutions driven by changes in mass accretion rates. While theoretical frameworks predict characteristic transition luminosities—such as the Coulomb luminosity ($L_{\text{Coul}}$) and the critical luminosity ($L_{\text{crit}}$)—most sources display only one or two such transitions. Swift J0243.6+6124, the first Galactic ultraluminous X-ray pulsar, presents a unique laboratory with an accretion luminosity spanning six orders of magnitude ($10^{34}$–$10^{39}$ erg s$^{-1}$). Previous studies have identified several transitions in this source, yet discrepancies exist between narrowband (NICER) and broadband (Insight-HXMT) spectral analyses, and the physical origin of multiple transitions in the sub-Eddington regime remains unclear. This paper addresses the identification of a new spectral transition in the sub-Eddington regime and investigates the underlying magnetic field configuration required to explain the source's complex behavior.

Methodology
The authors conducted a detailed spectral analysis of Swift J0243.6+6124 using data from the Insight-HXMT and NICER telescopes during the decay phases of three outbursts (2018, 2019, and 2023).

• Data Selection: Observations were selected to cover the sub-Eddington regime ($L_X \sim 7.9 \times 10^{36}$ to $1.2 \times 10^{38}$ erg s$^{-1}$). The analysis adopted the latest Gaia DR3 distance of $5.2 \pm 0.3$ kpc.
• Spectral Modeling: Spectra were modeled using the constant*tbabs*(bbodyrad+cutoffpl) model in XSPEC. The study compared three data configurations: Insight-HXMT only (2–100 keV), NICER only (0.7–10 keV), and joint fits (0.7–100 keV).
• Statistical Analysis: Parameter uncertainties were estimated via Markov Chain Monte Carlo (MCMC) sampling. To characterize the spectral evolution, the authors fitted the blackbody radius–luminosity relation with both single-line and broken-linear models. Model comparison was performed using $\Delta$AICc, $\Delta$BIC, and Likelihood-Ratio Tests (LRT) with 10,000 bootstrap iterations.
• Physical Interpretation: The derived transition luminosity was compared against theoretical predictions for Coulomb deceleration and critical luminosity. The authors proposed a multipolar magnetic field model (dipole + quadrupole) to reconcile the existence of multiple transitions and the discrepancy between weak and strong magnetic field estimates.

Key Results

1. Discovery of a New Transition ($L_t$): The analysis reveals a distinct break in the spectral parameter evolution at a transition luminosity of $L_t \approx 4.46^{+0.53}_{-0.53} \times 10^{37}$ erg s$^{-1}$. This is the fifth transition luminosity identified in Swift J0243.6+6124.
2. Spectral Evolution Signatures: At $L_t$, the correlation between the blackbody radius and luminosity reverses from positive to negative. Specifically, the blackbody normalization (radius) exhibits a significant turnover: it increases with luminosity below $L_t$ and decreases above $L_t$. Concurrently, the photon index evolves from nearly constant to positively correlated with luminosity, and the cutoff energy ($E_{\text{cut}}$) evolution slows.
3. Instrumental Discrepancies: The study highlights that NICER-only fits (narrowband) yield non-physical results, drifting to low-temperature ($kT \sim 0.4$ keV) and large-radius ($R \sim 20$ km) solutions. In contrast, Insight-HXMT and joint fits constrain a high-temperature ($kT \sim 3$ keV) and small-radius ($R \sim 1$ km) blackbody component consistent with a polar cap hotspot. The authors conclude that narrowband fitting alone introduces intrinsic biases in continuum inference.
4. Magnetic Field Configuration: The observed $L_t$ is inconsistent with standard Coulomb luminosity predictions for typical neutron star parameters, which would require an implausibly low magnetic field ($B \sim 10^{10}$ G). Instead, the authors propose a multipolar magnetic field configuration.
   • Weak Pole: $L_t$ is interpreted as the critical luminosity of a "weak" magnetic pole with a surface field strength of $B_S \sim 2.8 \times 10^{12}$ G.
   • Strong Pole: The previously detected 146 keV cyclotron resonance scattering feature (CRSF) implies a "strong" pole with $B_N \sim 1.6 \times 10^{13}$ G.
   • Geometry: An axisymmetric superposition of a dipole ($6.6 \times 10^{12}$ G) and a quadrupole ($9.4 \times 10^{12}$ G) field can reproduce this asymmetry. This configuration explains why different poles enter the critical regime at different luminosities, generating multiple spectral transitions.

Significance and Claims
The paper claims that the discovery of $L_t$ significantly increases the complexity of the accretion physics in Swift J0243.6+6124, making it the source with the most identified transitions. The authors assert that the existence of five distinct transitions cannot be explained by standard dipole magnetic field models, which typically predict only $L_{\text{Coul}}$ and $L_{\text{crit}}$.

The primary significance of the work lies in the proposal that multipolar magnetic fields are the underlying mechanism governing the emission in this source. The study demonstrates that a combination of dipole and quadrupole fields can create asymmetric polar caps, allowing for distinct critical luminosities at each pole. This model successfully reconciles the low-field transition ($L_t$) with the high-field CRSF detection and the higher-luminosity transitions ($L_4, L_5$). Furthermore, the paper emphasizes that broadband observations are essential for correctly constraining spectral parameters in AXRPs, as narrowband data can lead to erroneous physical interpretations. The findings suggest that the local magnetic field of the neutron star can exceed $10^{13}$ G even if the effective field on the magnetospheric scale appears weaker, providing a new framework for understanding accretion columns in highly magnetized systems.