ALMA/NICER observations of GRS 1915+105 indicate a return to a hard state
AAstronomy & Astrophysics manuscript no. 39581corr © ESO 2021February 4, 2021
ALMA/
NICER observations of GRS 1915 +
105 indicate a return to ahard state
K. I. I. Koljonen (cid:63) and T. Hovatta
The Finnish Centre for Astronomy with ESO, University of TurkuAalto University Metsähovi Radio Observatory, Metsähovintie 114, FI-02540 Kylmälä, FinlandReceived ; accepted
ABSTRACT
Context.
GRS 1915 +
105 is a transient black hole X-ray binary consistently emitting 10–100% of the Eddington luminosity in theX-ray band over the last three decades until mid-2018 when the source luminosity suddenly decreased by an order of magnitude. Thisphase was followed by a change to a state with even lower average X-ray fluxes never seen before during the outburst but presentingrenewed flaring activity at di ff erent wavelengths, albeit with mean fluxes still in decline. Aims.
GRS 1915 +
105 has the longest orbital period known among low-mass X-ray binaries, the largest accretion disk size, andtherefore the largest mass supply for accretion. The high inclination of the disk allows the study of geometrical e ff ects of the accretionflow such as changes in the height-to-radius ratio or the e ff ect of accretion disk winds on the intrinsic emission that is expected duringthe outburst decay. In addition, the transient jet is expected to change to a compact, self-absorbed, steady jet. Methods.
We conducted two full polarization
Atacama Large Millimeter Array observations to study the jet properties during theoutburst decay by analyzing the spectral, polarization, and intra-epoch variability for both observation epochs. In addition, we analyzedalmost daily
Neutron Star Interior Composition Explorer pointing observations, modeling X-ray power spectral densities, spectralenergy distributions, and light curves with a physically motivated model to follow the changing accretion disk properties throughoutthe outburst decay and relating them to the jet emission.
Results.
We show that the X-ray and millimeter (mm) spectral, timing, and polarization properties are consistent with those of atypical decaying X-ray binary outburst and that GRS 1915 +
105 has descended into the low-luminosity hard X-ray state. The jetemission in the mm is consistent with a compact, steady jet with ∼
1% linear polarization, and the magnetic field is likely aligned withthe jet position angle. Relating the mm emission to the X-ray emission reveals that the source has changed from a higher radio / X-raycorrelation index to a lower one; L radio ∝ L . X . Key words.
Accretion, accretion disks – binaries: close – stars: black holes – stars: winds, outflows – X-rays : binaries
1. Introduction
GRS 1915 +
105 is a transient, black hole X-ray binary (XRB)that started its decades-long outburst in 1992 (Castro-Tiradoet al. 1994). During the outburst, it was one of the brightestXRBs, consistently emitting 10–100% of the Eddington lumi-nosity in the X-ray band. However, in mid-2018, the sourceluminosity decreased suddenly by an order of magnitude andhas diminished since. GRS 1915 +
105 hosts a K-type giant star(Greiner et al. 2001) with a mass of ∼ +
105 was the first Galactic source known toexhibit apparent superluminal motion of the jet components(Mirabel & Rodríguez 1994). The source presented both tran-sient and steady jet phases during its outburst. The transient jetphases occurred approximately once every year and consistedof bright events at infrared and radio wavelengths (Pooley &Fender 1997; Klein-Wolt et al. 2002). In between the transientjet phases, the steady jet phases were periods of prolonged hardX-ray emission (Foster et al. 1996; Dhawan et al. 2000; Fuchs (cid:63) e-mail: [email protected] et al. 2003) and a low radio flux density of ∼
10 mJy (Ogley et al.2000) indicating emission from a compact self-absorbed jet. Thetransient jet phase in XRBs often coincides with a time when theX-ray spectrum changes rapidly. This spectral evolution is takento arise from a physical change in the accretion disk structurewhere the optically thin and geometrically thick disk is replacedby an optically thick and geometrically thin disk due to increasedmass accretion rate (Gallo et al. 2003). During this time, strongradio flares are observed, and bright, optically thin synchrotron“blobs” or internal shocks can be seen emanating from the cen-tral source in radio interferometric images (Fender et al. 1999).Due to the strong radio emission and trackable jet com-ponents, the jet and system parameters of GRS 1915 +
105 areknown relatively well. The jet component velocities range be-tween 0.65 c and 0.98 c , jet inclination is 60-70 deg to the line ofsight, the mass of the black hole is 10-14 solar masses, and thedistance is 7-10 kpc (Fender et al. 1999; Reid et al. 2014). Dueto the high inclination, we are looking at the accretion disk rela-tively edge-on. This allows us to potentially study the evolutionof the accretion flow geometry, such as changes in the height-to-radius ratio of the accretion flow or the e ff ect of accretion diskwinds on the intrinsic emission (e.g. Lee et al. 2002; Miller et al.2016; Neilsen et al. 2018). Article number, page 1 of 18 a r X i v : . [ a s t r o - ph . H E ] F e b & A proofs: manuscript no. 39581corr
In July 2018, GRS 1915 +
105 entered an unusually extendedlow-flux X-ray phase followed by a change to a state with evenlower average X-ray fluxes never seen before during the outburstbut presenting renewed flaring activity at di ff erent wavelengths,most notably in the radio (Koljonen et al. 2019; Trushkin et al.2019; Motta et al. 2019, 2021). Detailed X-ray observations sug-gest that diminished and obscured accretion might be feedingthe variable jets (Koljonen & Tomsick 2020; Miller et al. 2020;Neilsen et al. 2020; Motta et al. 2021; Balakrishnan et al. 2020).The obscuring matter is likely located close to the X-ray sourceeither ejected from the accretion disk by a disk wind or originat-ing from a pu ff ed-up or warped accretion flow.In this paper, we present two full polarization Atacama LargeMillimeter Array (ALMA) and nearly daily
Neutron Star Inte-rior Composition Explorer ( NICER ) monitoring observations ofGRS 1915 +
105 that were taken during the outburst decay beforethe source entered to the obscured state. The observations anddata reduction processes are described in Section 2. In Section 3,we go through the detailed timing and spectral analyses of both
NICER and ALMA data sets. We show that the X-ray and mil-limeter (mm) spectral and timing properties of GRS 1915 + +
105 has descended to a low-luminosity hardstate. The outburst decay of GRS 1915 +
105 o ff ers us a detailedview of the physical processes in the accretion flow and the jetleading from the outburst towards quiescence and to the anoma-lous obscured state. In Section 4, we discuss the accretion diskand jet properties during the di ff erent phases of the outburstdecay, and speculate on accretion disk properties in a scenariowhere the outburst has ended. Finally, we conclude in Section 5.
2. Observations
We conducted two band 3 (90–105 GHz) full polarizationALMA observations in Cycle 6 on November 1, 2018, andMarch 21, 2019, during the outburst decay of GRS 1915 + ∼
10 min), polar-ization calibrator ( ∼ ∼ − + + Common Astron-omy Software Applications (CASA) package version 5.4.0-70 byrunning the calibration script provided with the data. We flaggedthe first ten channels in the second spectral window due to badD-terms for both epochs. In addition, antennas DA63 and DA41for the first and second epochs, respectively, were flagged for allspectral windows due to outlier values in gain amplitude. Also,for the second epoch, we flagged antennas DA42 and DA48 forthe first spectral window due to outlier values in the X- and Y-polarized phase di ff erence. The resulting images show that thesource is a point source within the beam ( ∼ (cid:48)(cid:48) and ∼ (cid:48)(cid:48) inthe first and second epochs, respectively) without any observedstructure.Flux densities were compared with both imaging methodsand a delta function was fitted to the uv-plane using uvmultifit ff ects, we did the followingtests. We examined the gain amplitudes from the phase calibra-tion to verify that any changes in the amplitudes are not corre-lated with the flux density variations. We also examined the wa-ter vapor radiometer data of each antenna to detect any correlatedchanges in atmospheric conditions and the target flux density, butnone were present. The mean precipitable water vapor contentduring the first and second epochs was 1.3 mm and 0.9 mm, re-spectively. Combined with the more extended configuration dur-ing the first epoch, we expect a larger scatter in the flux densitydue to changing atmospheric conditions during the first epoch.This is indeed seen in our third check where we divided the arrayinto two individual sub-arrays (DV and DA antennas separately)and repeated the flux density estimation. During the first epoch,the flux density estimates from the sub-arrays vary up to 0.2 mJy,and during the second epoch they vary by less than 0.1 mJy. We downloaded all
NICER data from the High Energy Astro-physics Science Archive Research Center (HEASARC) using atime range of MJD 58238–58634 (May 2018 – May 2019) be-ginning from the outburst decay phase and ending in the descentto the obscured phase. We disregarded observations with expo-sure times of less than 500 seconds. This resulted in 87 pointings,shown in Table A1.We reduced the observations using nicerdas version 6a withparameters nicercal _ filtexpr= "EVENT_FLAGS =bx x cor _ range= "4-" to remove high particleradiation intervals associated with the Earth’s auroral zones.We selected PI energy channels between 30 and 1200 (0.3–12keV). We extracted the X-ray spectra using xselect v g . Weused the photon redistribution matrix (RMF) and the on-axisaverage ancillary response file (ARF) for all 52 detectorscombined from HEASARC’s Calibration Database (CALDB).For the background we use a public background file available inHEASARC ( nixtiback pi ).For the timing analysis, we calculated the power spectraldensities (PSDs) directly from the cleaned event files using abinning of 2 − s (or 2 − s in case of the data in the obscuredphase), an energy band 1–10 keV, and a segment length of 16 s(or 256 s in case of the data in the obscured phase). We normal-ize all the PSD to rms variability. All the segments were furtheraveraged over the whole pointing and binned geometrically by afactor of 1.05 before importing them to the Interactive SpectralInterpretation System ( isis ; Houck & Denicola 2000) for modelfitting.For the spectral analysis, we binned the data adaptively inthe ranges 0.3-1.5 keV to S / N = / N = / N = / N = isis . We estimated the errors on the parameter valuesand fluxes through Monte Carlo analysis. For those spectra with1-10 keV flux densities below 0.3 × − erg s − cm − , an addi-tional unabsorbed power-law component was needed to nicelyfit the low-energy data below 1 keV that could arise from imper-fect background subtraction. However, the normalization of thiscomponent is very small, containing only 0.003-0.006% of the Article number, page 2 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + Feb
18 12
May
18 20
Aug
18 28
Nov
18 8
Mar
19 16
Jun F - k e V [ M AX I pho t on s c m − s − ] A L M A E p . A L M A E p . F - k e V / F - k e V decay Re-brightening Obscured F - k e V [ − e r g s − c m − ] Q P O fr e qu e n c y [ H z ] MJD a b c d ef g
Fig. 1.
Top : MAXI / GSC 2–20 keV daily light curve of GRS 1915 + Middle / top : MAXI / GSC10–20 keV / Middle / bottom : NICER ff erent decay phases are indicated andshown as an alternating shading scheme. Bottom : QPO frequencies asdetermined from modeling the
NICER
PSD. The exponential and lin-ear decay phases show di ff erent rates of decay for the QPO frequency(shown as dotted lines). The seven lettered observations correspond toexample PSD and spectra shown in Figs. 3, 7, and 6. total flux and a ff ecting only the soft X-rays below 1 keV with afairly steep spectral slope of Γ ∼
3. Results
The ALMA observations coincided with a stable low-luminosityX-ray state where the observed luminosity is much lower andX-ray hardness higher than the usual outburst values. This statewas superseded by an even lower luminosity state with a higher F keV /F keV F - k e V [ M AX I pho t on s c m − s − ] t MJD <5832858328 MJD <58605t MJD >58605FlaresSoft StateALMAL Edd Edd Edd Edd Fig. 2. MAXI / GSC hardness-intensity diagram of GRS 1915 + 105 fromdaily monitoring observations since August 2009. The blue data points(both dark triangles and light blue squares) indicate the low-luminositystate since August 2018 with increased spectral hardness. The light bluesquares correspond to the obscured decay phase with even lower fluxdensities and harder spectra with occasional, strong X-ray flares (reddiamonds) and highly variable radio emission, in addition to a softerstate in August-September 2020 with the flux densities back to the levelof the low-luminosity state (yellow dots). The numbered green boxescorrespond to the ALMA epochs that took place during the linear andrebrightening decay phases of the outburst decay. X-ray hardness ratio that is most likely an e ff ect of obscura-tion (Koljonen & Tomsick 2020; Miller et al. 2020; Neilsenet al. 2020; Motta et al. 2021; Balakrishnan et al. 2020). Fig-ure 1 shows the Monitor of All-sky X-ray Image / Gas Slit Cam-era ( MAXI / GSC) 2–20 keV count rate, MAXI / GSC hardness ra-tio, NICER + 105 exhibited an increase influx density and QPO frequency departing from the linear decaytrend, which we denote as a rebrightening phase.Figure 2 shows the hardness-intensity diagram from the MAXI / GSC daily observations of GRS 1915 + 105 since Aug2009. The three decay states separate easily from each other witha change in the countrate and the X-ray hardness. Assuming thatGRS 1915 + 105 is accreting at the Eddington limit for the high-est count rates ( ∼ − s − ) —a reasonable argumentbased on both observations (e.g., Done et al. 2004; but takinginto account the e ff ect of the smaller distance estimate in Reidet al. 2014) and simulations (e.g., Truss & Wynn 2004)— theoutburst phase corresponds to luminosities 0 . < L / L Edd < . ∼ Edd , and the obscured phaseto ∼ Edd (which agrees with values obtained by Koljo-nen & Tomsick 2020 and Miller et al. 2020). During the ob-scured phase, there were prominent flares with X-ray luminosi- Article number, page 3 of 18 & A proofs: manuscript no. 39581corr ties reaching ∼ Edd , and more recently, in August-September 2020, a softer state with X-ray luminosities around ∼ Edd . Assuming a black hole mass of 12 M (cid:12) (Reid et al.2014) the Eddington luminosity is L Edd = . × erg s − . In this section, we describe the X-ray spectral and timing prop-erties during the outburst decay in detail. As described above,we divide the data in four separate phases with distinct behaviorin the X-ray light curve as well as in X-ray spectral and timingproperties. We fit the individual pointing spectra from the NICER datasetwith a model consisting of an absorbed power law and reflec-tion model components ( relxill ; García et al. 2014; Dauseret al. 2014) modified by Gaussian emission and / or absorptionline components when necessary. In addition, for spectra with1-10 keV fluxes below 0.3 × − erg s − cm − we added an un-absorbed power-law model component to take into account thee ff ect of imperfect background subtraction as mentioned in Sec-tion 2.2.For the photoelectric absorption component, we chose vphabs , which allow the abundances of the elements of the neu-tral absorber to be changed. GRS 1915 + 105 is known to besurrounded by local, cold material likely in the form of an ac-cretion disk wind that absorbs the X-ray emission in additionto the interstellar absorption (Lee et al. 2002; Martocchia et al.2006). Following Martocchia et al. (2006), we grouped elementsthat have either a small e ff ect in the energy range in question orwhose origin is likely to be the same into the following groups:(1) H, He, C, N, O, Ne, Na, (2) Mg, (3) Al, (4) Si, (5) S, (6)Cl, Ar, Ca, and (7) Cr, Co, Ni, and Fe. Depending on the qualityof the spectrum, we grouped the abundances to an even smallernumber of groups (the smallest division consisting of two groupswith one for lighter elements than aluminum and one for heavierelements). In the fitting process, we found that the iron abun-dance tended towards unphysically low values, and therefore wedecided to fix it to the solar value. In addition, we added emis-sion and / or absorption lines of neutral and ionized iron (Fe I K α at 6.40 keV, Fe XXV K α complex at 6.7 keV, Fe XXV K β at7.80 keV, Fe XXVI Ly α at 6.97 keV, and Fe XXVI Ly β at 8.27keV), sulfur (2.3 keV), argon (3.0 keV), and calcium (3.7 keV)when necessary, mainly for the better quality spectra. Thus, thetotal model can be described as: vphabs × ( relxill + lines ) + powerlaw .In the fitting procedure, we first fitted the spectra with thelargest exposure times to pinpoint the values of some of the relxill parameters that were then kept fixed when modeling thelower quality spectra. These included the inner radius of the ac-cretion disk R in , spin of the black hole a , and the reflection factorR f of relxill . However, we found that the spin is not constrained,and the inner radius presents large values indicating that the X-ray spectra are not sensitive to relativistic e ff ects. Therefore, wedecided to fix the spin to zero. We further fixed the inclination to70 degrees (Reid et al. 2014; Mirabel & Rodríguez 1994), andiron abundance to the solar value. Other parameters were left asdefault values and fixed, except the power-law photon index ( Γ ),ionization parameter (log ξ ), and model normalization, whichwere allowed to vary for all spectral fits. We fit all the PSDs with up to three Lorentzians: a Lorentziancentered at zero frequency for the band-limited noise and twoLorentzians for the QPO and its upper harmonic. In addition tothe Lorentzian components, a power-law component is neededto model the higher luminosity PSD during the high-luminositypart of the exponential decay. We also model the Poisson noisepresent in the PSD as a constant power component. In Fig. 1, both the NICER ∼ MJD58330. We take this change as the approximate transition time ofthe two decay phases. The NICER PSDs during the exponentialand linear decay phases suggest a typical hard state PSD simi-lar to what is observed in plateau and radio-quiet states with aband-limited noise component and a type-C QPO. The QPO fre-quency is tightly correlated with the X-ray luminosity decreasingapproximately linearly from 4 Hz to 2 Hz during the exponentialdecay phase and further to ∼ panels b–c ). The total rms ofthe Lorentzians correlates with the X-ray hardness ratio start-ing at ∼ 20% at the beginning of the exponential decay phaseand reaching ∼ 35% in the linear decay phase. The rms evolutionis mainly due to increasing rms of the zero-centered Lorentzianwhile the rms of the QPO and harmonic are on average 7 ± ± > panel a in Fig. 3). It shows de-creasing rms variability during the decay and presents a power-law index of Γ ∼ . 0. All model parameters for each PSD aretabulated in Table A1.We also checked whether the QPO frequencies presented anytime lags between a soft (2–4 keV) and a hard X-ray (5–10 keV)band (see Fig. 4). On average, the main QPO frequency showssoft lags (soft photons lagging hard photons) and the upper har-monic hard lags (hard photons lagging soft photons) of about 10msec when the QPO frequency is above 2 Hz. When the QPOfrequency descends below 2 Hz; that is, during the linear decayphase, the lag at the QPO frequency averages to zero lag, butthe upper harmonic frequency shows hard lags. This behavior isconsistent with earlier studies during the outburst phase (Reiget al. 2000; Qu et al. 2010; Pahari et al. 2013; Zhang et al. 2020),where the phase lag has a log-linear relationship with the QPOfrequency and changes sign approximately at 2 Hz.During the exponential decay phase, the X-ray spectra showa slowly hardening and diminishing Comptonization component.The power-law photon index of the relxill component decreasesfrom Γ = Γ = ξ = ξ = α and S K α . In the linear decay phase,the continuum model parameters settle to a more stable statewith Γ= ± ξ = ± Γ= Γ > Article number, page 4 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + − − a) b) c) − − − d) − e) − f) P o w e r x F r e qu e n c y ( r m s m ea n ) Frequency [Hz] Fig. 3. PSD from six di ff erent epochs labeled in Fig. 1 ( bottom panel ). The PSDs are fitted with a model consisting of a zero-frequency Lorentzianmodeling the broadband noise, one or two narrow Lorentzians modeling the QPO (fundamental and first upper harmonic components), an addi-tional power-law component needed for PSD in panels a , d , e , and f , and Poisson noise as a constant power component (removed from the PSDsshown here). NICER pointingsas shown in Fig. 3. After a gap in the monitoring data of GRS 1915 + 105 (MJD58460–58550, see Fig. 1), the flux density and QPO frequencyhad departed from the decay profile as the source entered therebrightening phase. During this phase, the flux density and theQPO frequency are not correlated although both present an in-crease of about a factor of two with the QPO frequency rising upto 3 Hz but vanishing completely during the flux drop leading tothe obscured phase.In the X-ray PSD, a broad-band low-frequency componentis visible in addition to a band-limited noise component and atype-C QPO ( panels d–f in Fig. 3). As above, we model the ad-ditional component with a power-law model. The rms variabilityof the power-law component increases, and the power-law indexpresents higher values up to Γ ∼ . 5. On the contrary, the rms ofthe zero-centered Lorentzian decreases slightly.In the X-ray spectra, the power-law photon index andthe ionization parameter present elevated but steady values of Γ= ± ξ = ± α being the most dominant line with a resolvedFWHM of 160-180 eV (corresponding to a velocity of 7000-8000 km / s) and equivalent width of about 100 eV. The line centeris also gradually redshifted from 6.70 keV (corresponding to azero redshift) to 6.59 keV (corresponding to a redshift of ∼ / s). In addition, Fe XXV K β and Fe XXVI Ly β absorptionlines are also visible and possibly Ni K α line in one pointing atMJD 58608.5, although the energy of the line is o ff by 0.1 keV.During the rebrightening phase, there is a clear evolution ofthe ratio of Fe XXVI Ly β and Fe XXV K β column densities(estimated according to Lee et al. (2002); their Eq. 1) indicatinga decreasing ionization factor from log ξ ∼ ξ ∼ ξ ∼ Article number, page 5 of 18 & A proofs: manuscript no. 39581corr D en s i t y Lag Main (QPO > 2 Hz) Main (QPO < 2 Hz) Harmonic (QPO > 2 Hz) Harmonic (QPO < 2 Hz) Fig. 4. Lag distributions of the QPO frequencies between 2-4 keV and5-10 keV bands. The main QPO frequency and the first upper harmonicfrequency display, on average, soft and hard lags, respectively, when themain QPO frequency is above 2 Hz. When the QPO frequency is below2 Hz, the average lag at the main QPO frequency vanishes, and the firstupper harmonic QPO frequency shows hard lags. The change to the obscured state occurs likely at MJD 58608 orsoon after. When the source transits to the obscured phase, thereis a rapid drop in Γ , log ξ , and flux, and a change from the highlyionized iron absorption lines to emission lines. In the obscuredphase, the PSD is consistent with a pure power-law noise (Fig. 7)indicating that all the intrinsic timing information is lost leavingonly the scattered component with Γ ∼ . NICER pointings did not provide good fits.Instead, we used the modeling results of Koljonen & Tomsick(2020), and added a neutral lower ionization reflection compo-nent to the model ( xillver ) with all parameters tied to the relxill component except normalization. Thus, the model can be nowdescribed as: vphabs × ( relxill + xillver + lines ) + powerlaw .We further set the ionization parameter to 1, fixed the reflectionfactor to − xillver , but also fitted separately with a gaussian line to be ableto compare values to other lines shown in Table 1). The equiv-alent widths of the emission lines vary from ∼ 50 eV to ∼ α lineseems to be in both absorption and emission (see Fig. 8). In theprevious observation, taken half a day earlier, the absorption isnot visible. While the absorption of the Fe XXV K α line occurspreferably in the resonant line (6.70 keV), the emission is char- N H ( c m − ) Mg Al Si S Cl/Ar/Ca251020 N X ( c m − ) S K α Cl K α Ar K α Ca K α Fe I K α Fe XXV K α Fe XXV Ly α Fe XXV K β Fe XXVI Ly β Emission/Absorption L i n e e n e r gy [ k e V ] Γ ξ χ d . o . f decay Linear decay Re-brightening Obscured a b c d e f g MJD F - k e V ( − e r g s − c m − ) Fig. 5. Spectral fit results of the NICER data. From top to bottom, themodel parameter evolution is shown for the abundance of lighter el-ements (lighter than magnesium all fixed to the same value), heavierelements (Mg, Al, Si, S, and Cl / Ar / Ca displayed with di ff erent colors),line energies for the emission / absorption lines (displayed with di ff erentcolors), the photon power law index ( Γ ), ionization parameter (log ξ ),the corresponding fit quality ( χ / d.o.f), and the absorbed 1-10 keV fluxdensity with the colored letters corresponding to spectra shown in Fig.6. The alternate shading shows the times of the di ff erent decay phases. acterized by the forbidden line (6.64 keV), especially for highcolumn densities (Bianchi et al. 2005). Therefore, the observa-tion of the absorption and emission line indicates at least twoscattering components during this time. Extrapolating the lineardecay to the obscured state (see Fig. 5, bottom panel) shows thatthe observed flux density during the obscured state is a factorof 1.2-8.8 lower than the linear decay profile. The observations Article number, page 6 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + − − − − a)b)c)d)e)f)g) E n e r gy f ux [ k e V pho t on s c m − s − ] Energy [keV] D a t a / M od e l Fig. 6. NICER spectra and best-fit models from the seven di ff erentepochs labeled in Fig. 1 ( bottom panel ) and Fig. 5 ( bottom panel ). Thecoloring scheme is according to the colored letters shown in Fig. 5 ( bot-tom panel ). − − − − g) P o w e r x F r e qu e n c y ( r m s m ea n ) Frequency [Hz] Fig. 7. Pure power-law PSD from the obscured epoch labeled in Fig. 1( bottom panel ). The Poisson noise is subtracted. presenting the lowest fluxes during the obscured phase show asimilar trend with the linear decay, but with 0.045 × − erg s − cm − of flux removed, meaning that at most approximately 80%of the intrinsic flux density is absorbed and / or scattered in the1–10 keV band. The two ALMA epochs coincided with the linear decay phaseand the rebrightening phase (Fig. 1). The corresponding meanStokes I flux densities of these epochs were 2.5 mJy and 2.9mJy. A typical radio-quiet level for GRS 1915 + 105 is < 20 mJy(Muno et al. 2001; Klein-Wolt et al. 2002). The flux density risein the rebrightening phase is consistent with the rise in the X-ray flux indicating a connection between the two. As the spectralwindows in the ALMA band 3 are separated into two sidebandswith a 10 GHz break in between, 90–93 GHz and 102–105 GHz,we were able to estimate the in-band source spectra for the twoepochs shown in Fig. 9. In both cases the spectra is flat with F e I K α F e XXV K α F e XXV I L y α N i I K α F e XXV K β F e XXV I L y β MJD A r b it r a r y un it s ( l og s ca l e ) Energy [keV] Fig. 8. Evolution of the iron lines from absorption to emission whenGRS 1915 + 105 changed from the rebrightening phase (top seven spec-tra) to the obscured phase (bottom three spectra). Two spectra from theobscured phase (MJD 58622 and MJD 58623) are overlaid to highlightthe possible absorption feature of Fe XXV K α line. The best-fit modelsare shown as solid lines. spectral indices α = − . ± . 05 and α = . ± . 05 ( S ν ∝ ν α ), which are typical values for a self-absorbed synchrotron jet.The source was weakly polarized during both epochs withpolarization fractions of 1.3-1.6 %. We could only detect signif-icant Stokes U polarization indicating linear polarization with aposition angle close to − 45 degrees. The values for the Stokesfluxes from uv-plane fitting and imaging analysis are presentedin Table 2. We note that ∼ 1% linear polarization is a typicalvalue for self-absorbed synchrotron jet polarization in hard-stateXRBs. The optically thick, steady jets in XRBs show low levelsof linear polarization, from undetectable to a few percent (e.g.,Corbel et al. 2000; Russell et al. 2015). This also agrees with ear-lier results from the steady jet phase with the polarization factorof the stationary core of 1-2% (Fender et al. 2002).The jet position angle in GRS 1915 + 105 was previously de-termined by tracking the jet components: -36.7 ◦ ± ◦ (northwestdirection) and 146.5 ◦ ± ◦ (southeast direction; Miller-Joneset al. 2007), 130 ◦ ± ◦ (Reid et al. 2014), 133 ◦ –157 ◦ (Dhawanet al. 2000), 142 ◦ ± ◦ (Fender et al. 1999), and 151 ◦ ± ◦ (Ro-dríguez & Mirabel 1999). As the position angle is degeneratewith ± ◦ , our values also correspond to 136 ◦ –147 ◦ , whichagrees well with the values of the jet position angle at similarangular scales. Thus, the polarization position angles of ALMAobservations are consistent with being parallel to the jet. Dueto absorption e ff ects, the intrinsic electric vector position angleof the steady jet is expected to align parallel to the magneticfield, which is expected to be parallel to the jet axis. If this is thecase, this would mean that the e ff ect of Faraday rotation wouldbe quite small. All the above results point to the fact that GRS1915 + 105 exhibits a canonical compact steady optically thickXRB jet during both ALMA epochs. Article number, page 7 of 18 & A proofs: manuscript no. 39581corr Table 1. Neutral and ionized absorption and emission iron line modeling results of the NICER data during the descent to the obscured state. Thecorresponding X-ray spectra and models are plotted in Fig. 8. The line widths of all lines except in a few cases for Fe XXV K α line are belowdetector resolution ( ∼ MJD: 58555.7 58569.4 58583.0 58589.2 58596.6 58604.4 58608.5 58622.5 58623.0 58631.1 Fe I K α Energy (keV) – – – 6.31 + . − . – – – 6.400 ± ± ± + − – – – 138 ± ± ± Fe XXV K α Energy (keV) 6.70 ± ± ± ± ± ± ± ± / ± − ± − ± − ± − + − − + − − + − − + − ± ± / − ± 14 154 ± σ (keV) 0.01 0.01 0.01 0.01 0.07 + . − . ± + . − . / cm − ) 2.6 4.3 3.8 5.8 10 16 – – – – Fe XXVI Ly α Energy (keV) 6.97 ± ± ± ± ± ± + . − . + . − . ± − ± − ± − ± − ± − ± − ± + − ± 16 150 ± 12N (10 cm − ) 3.7 5.7 5.9 8.0 6.9 7.3 – – – – Fe XXV K β Energy (keV) – – – 7.84 ± ± ± ± + . − . EW (eV) – – – − ± − ± − ± − ± 11 – – 33 ± Fe XXVI Ly β Energy (keV) – – – 8.14 ± ± ± ± + . − . EW (eV) – – – − ± − ± − + − − ± 14 – – 31 ± Table 2. Polarized flux densities of ALMA observations. All Stokes fluxes are shown for both epochs and spectral windows from uv-plane fitting,but the position angle and polarization fraction are only calculated for the full band. We also show the Stokes I and U from full band imaginganalysis for comparison. UVMultiFit ImagingEpoch Spw I Q U PA p I U(mJy) (mJy) (mJy) (deg) (%) (mJy) (mJy)Ep. 1 All 2.499 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± As the ALMA data consists of several scans per epoch, in addi-tion to time-averaged analysis, we also studied the intra-epochvariability of the mm emission from GRS 1915 + + ff ect of a more extended array configuration andworse atmospheric conditions during the first epoch make theestimates of short-term variability less reliable. However, dur-ing the second epoch, the phase calibrator is very stable, andwe conclude that the variability seen on timescales of minutesis intrinsic to the source, demonstrating usefulness of ALMA instudying fast variability in XRBs. Radio / X-ray correlation is one of the most important pieces ofobservational evidence in connecting the mass accretion rateonto the compact object during an outburst event to the mass-loading of the jet that seems to be at work in both XRBs (e.g.,Hannikainen et al. 1998; Corbel et al. 2003; Gallo et al. 2003)and AGNs (e.g., Merloni et al. 2003; Falcke et al. 2004). Duringlow-luminosity hard X-ray states in XRBs, the logarithmicallyscaled X-ray and radio luminosities present a tight relation of L radio ∝ L ∼ . . This relation can be explained by assuming thatthe X-ray emission arises from Compton scattering in advection-dominated accretion flow (ADAF) and the radio emission fromthe optically thick synchrotron emission in the jet (Heinz & Sun-yaev 2003).To investigate the radio / X-ray correlation of GRS 1915 + / plateau states to compare them with our ALMA data. Weused the 350 GHz James Clerk Maxwell Telescope (JCMT)observations from Ogley et al. (2000) that were taken duringa radio-quiet state and 94 GHz Nobeyama Millimeter Array(NMA) observations from Ueda et al. (2002) that were takenduring a radio plateau state, both of which should present a com-pact steady jet. The corresponding 3-10 keV X-ray fluxes were Article number, page 8 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + ALMA Obs. 1(GRS 1915+105)F ~ ν -0.09±0.05 Pol. Cal. (J1924-2914)F ~ ν -0.58 Phase Cal. (J1922+1530)F ~ ν -0.87 ALMA Obs. 2(GRS 1915+105)F ~ ν Pol. Cal. (J1924-2914)F ~ ν -0.62 85 90 95 100 105 11080859095100 Phase Cal. (J1905+0952)F ~ ν -0.93 Frequency [GHz] F l ux d e n s it y [ m J y ] Fig. 9. ALMA band 3 spectra of GRS 1915 + 105 and calibrator sourcesfrom the two epochs. estimated using quasi-simultaneous archival RXTE data takenfrom HEASARC and reduced according to standard procedures.The resulting mm / X-ray correlation can be seen in Fig. 11 witha correlation coe ffi cient of 0.58 ± / X-ray correlation (gray trian-gles; same X-ray band, but radio luminosity is from 8.4 GHz;from Koljonen & Russell 2019). Both have similar correlationcoe ffi cients (assuming flat spectrum), reinforcing the hard-statenature of GRS 1915 + 105 during ALMA observations and thedecay phase. ALMA Ep. Phase Cal. (J1922+1530) ALMA Ep. F l ux d e n s it y [ m J y ] Time since obs. start [min] Phase Cal. (J1905+0952) Fig. 10. Band 3 light curve from the two ALMA epochs in time binsof one scan length ( ∼ 10 min). The top two panels show the data fromEpoch 1 and the bottom two panels from Epoch 2 for GRS 1915 + + + 4. Discussion To briefly summarize the above results, we find that GRS1915 + 105 exhibits typical XRB hard-state properties during theexponential and linear decay phases. These include decreasingX-ray luminosity down to ∼ 1% of the Eddington luminosity,PSD with a band-limited noise profile and a type-C QPO witha harmonic, an absorbed power-law spectrum that can be fittedwith a Comptonization model, and a weakly polarized, compact,optically thick jet. The mm / X-ray correlation shows similar co-e ffi cient to that of canonical low-luminosity XRBs in the radia-tively ine ffi cient track, further suggesting that the source is in thecanonical hard state. The X-ray light curve profile showing firstan exponential decay followed by a linear decay is a hallmarkof a viscous and irradiation-controlled decay observed duringthe end stages of transient XRB outbursts (e.g., Tetarenko et al.2018). However, the following rebrightening phase displayedsimilar X-ray and radio properties to the preceding exponentialand linear decay phases, with elevated X-ray and radio luminos-ity marking a departure from the linear decay trend. In addition,the PSDs show an additional red noise power-law component in-creasing in rms, and the X-ray spectra show high-ionization ab-sorption lines and modeling results indicate increasing absorp-tion. When the source transits to the obscured state, the X-rayproperties change markedly with an order of magnitude drop Article number, page 9 of 18 & A proofs: manuscript no. 39581corr J C M T N M AA L M A A L M A L (erg s − ) L R ( e r g s − H z − ) GRS 1915+105 ( ξ RX = ± )GX 339-4 ( ξ RX = ± ) Fig. 11. GRS 1915 + 105 mm / X-ray correlation (blue points). The high-luminosity data are from JCMT (Ogley et al. 2000) and NMA (Uedaet al. 2002) together with quasi-simultaneous RXTE observations. Incomparison, GX 339-4 hard-state radio / X-ray correlation is shown(gray triangles; same X-ray band, but radio luminosity is from 8.4GHz; from Koljonen & Russell 2019). Both have similar correlationslopes (assuming flat spectrum), reinforcing the hard-state nature ofGRS 1915 + 105 during ALMA observations. in the X-ray flux, the PSDs show a pure red noise profile, andthe X-ray spectra becomes much harder with prominent high-ionization emission lines. During the outburst, we divided the low-X-ray-flux states ofGRS 1915 + 105 into two di ff erent states depending on thestrength of the radio flux density: a radio-quiet hard ( χ ) state witha radio flux density of a few mJy, and a plateau state with theradio flux density presenting higher, typically 50-100 mJy fluxdensities (Foster et al. 1996; Pooley & Fender 1997; Fender et al.1999). The former is close to a ‘normal’ XRB hard state, whilethe latter is probably a high-luminosity excursion from a veryhigh state. For both states, the jet properties are similar (apartfrom the flux density) and correspond to a steady, flat-spectrum,optically thick compact jet. Selecting observations at the plateaustate which presents a compact steady jet, Rushton et al. (2010)were able to find a radio / X-ray relation of L radio ∝ L ∼ . ± . or L radio ∝ L ∼ . ± . if only taking the coronal emission into accountwhen estimating the X-ray luminosity (Peris et al. 2016).Many XRBs show a steeper radio / X-ray correlation duringhigh-luminosity phases of the hard state (e.g., Coriat et al. 2011).It is unclear whether the change depends on the source beingin outburst rise or decline (Koljonen & Russell 2019; Islam &Zdziarski 2018) or there exists multiple correlations dependingon the physical qualities of the systems (Gallo et al. 2012). Thedi ff erent physical mechanisms for multiple correlations havebeen suggested to arise from di ff erences in radio emission prop-erties (Casella & Pe’er 2009; Espinasse & Fender 2018), di ff er-ences in X-ray emission properties (Meyer-Hofmeister & Meyer 2014; Coriat et al. 2011; Xie & Yuan 2016), or inclination e ff ectson X-ray emission properties (Heil et al. 2015; Muñoz-Dariaset al. 2013; Motta et al. 2018; Petrucci et al. 2001; Nied´zwiecki2005) and radio emission properties (Soleri & Fender 2011;Motta et al. 2018). Whatever the case, the change of the corre-lation coe ffi cient of GRS 1915 + 105 to L radio ∝ L ∼ . indicates achange in the evolution of the outburst to a low-luminosity hardstate.Both plateau and χ states show similar X-ray timing prop-erties to those observed during the exponential and linear decaywith a PSD presenting a flat-top noise profile and a type-C QPO.However, the radio-quiet state shows QPOs with central frequen-cies greater than ∼ ∼ ff ect hasbeen attributed to a pivoting lag-energy spectrum with the lagincreasing with energy for progressively lower QPO frequenciesand decreasing for progressively higher QPO frequencies (Pa-hari et al. 2013; Zhang et al. 2020).Several explanations for the phase lag behavior have beendiscussed in the literature. van den Eijnden et al. (2017) relatethe phase lags to the inclination of the system. Soft lags are ob-served from high-inclination systems, and hard lags are observedfrom low-inclination systems for QPO frequencies exceeding 2Hz. In particular, for GRS 1915 + ff ering in physical picture, the un-derlying assumption in both scenarios is the same, presentingtwo (or more) regions located at di ff erent radii and having a dif-ferent spectral response. The di ff erence in the location producesthe change in the QPO frequency, while the di ff erent spectralresponse produces the change in the phase lag. E ff ectively, thismeans that for high QPO frequencies the spectral response of theinner part of the accretion flow is harder than that of the outerpart, while for low QPO frequencies the outer part presents aharder response. The zero time lag with ∼ / radio-quiet state of the outburst phase means that theplateau is a hard X-ray state with a compact hot inner flow whilethe radio-quiet state presents more expanded hot inner flow.However, the behavior of the radio emission is di ff erent, with theradio flux density staying at a low level of 1-3 mJy throughoutthe linear decay phase (Motta et al. 2021) while the plateau statepresents much higher radio flux densities. However, this can beexpected, as the mass accretion rate during the outburst phase isat least an order of magnitude higher (see Fig. 2). Assuming the observed light-curve profile obeys the standardXRB outburst decay profile, we fit the NICER light curve witha model used in Powell et al. (2007), Heinke et al. (2015), andTetarenko et al. (2018) corresponding to an exponential decay Article number, page 10 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + on a viscous timescale and a linear decay on an irradiation-controlled timescale. The full model can be written as: f X = (cid:40) ( f t − f ) exp ( − ( t − t break ) /τ e ) + f if t ≤ t b f t (1 − ( t − t b ) /τ l ) if t > t b , (1)where τ e is the viscous (exponential) timescale, τ l is theirradiation-controlled (linear) timescale, t b is the transition time, f t is the flux density at the transition, and f is the asymptoticflux density of the exponential decay. The fit resulted in the fol-lowing parameters: τ e = ± τ l = ± t b = ± f t = . ± . × − erg s − cm − ,and f = . ± . × − erg s − cm − .Taken at face value, the timescale of the exponential decay(58 days) appears to be relatively low. The approximate viscoustime of the whole disk is at least t visc ≈ α − ( H / R ) − t dyn ∼ α = . 2, thescale height to radius ratio as H / R = . , and t dyn ∼ ∼ × cm). However, the average viscosity parame-ter of the whole disk is likely much lower as only the irradiatedpart of the disk is in the hot state, which would make the vis-cous timescale even longer. Possible mechanisms to decrease theviscous time include increasing the value of the average viscos-ity parameter and / or increasing the scale height to a radius ratiofrom the canonical values. In addition, the outflowing wind re-duces the viscous timescale by a factor of (1 − e w ) , which for thestrong disk wind of GRS 1915 + 105 can be a sizable e ff ect.Typically, observation of the exponential decay requires thatthe disk has been completely ionized and R irr > R out . As GRS1915 + 105 harbors such a large disk, this is likely not the case.Instead, it appears that the outburst decay behaves as if GRS1915 + 105 has only a disk size of R out = R irr ∼ × cm. Alter-natively, the exponential decay may arise from another processthat reduces the mass accretion rate to the black hole. Cannizzo(2000) discusses a case where the evaporation of the thin accre-tion disk into a hot inner flow close to the compact object is dom-inant over the viscous evolution. Strong evaporation leads to ane-folding decay rate associated with the loss of material from theinner accretion disk that can be ten times faster than the viscousevolution. Indeed, the critical luminosity for the disk evapora-tion scheme to hold is considered to be approximately 5% of theEddington luminosity (Meyer-Hofmeister 2004), which matcheswith the luminosities observed at the end of the exponential de-cay. Based on the canonical scenario, the X-ray spectra showing aslowly hardening and diminishing Comptonization componentcorresponds to the reducing radiative cooling of the opticallythin and hot inner accretion flow as the optically thick accre-tion disk recedes or evaporates to larger radii. If the power-lawcomponent in the PSD during the exponential decay is flickernoise (with Γ ∼ 1) from the accretion disk (Lyubarskii 1997),this supports the receding or evaporating disk. However, a diskcomponent is not statistically needed to fit the X-ray spectra, butas the absorption towards GRS 1915 + 105 is high, the modelingof the curved soft X-ray part of the spectra can be degenerate anda low-temperature disk could still produce some of the soft X-rays. Similarly, the rms of the zero-centered Lorentzian increasesfrom 10% to 20%, which implies that the dilution from the diskis not impacting the rms in the linear decay phase. In the linear decay phase, the spectral evolution stops and settles into a stablestate with slowly decreasing flux density and QPO frequency. Ifthe type-C QPO frequency is tied to the size of the precessinghot inner flow (Ingram et al. 2009), this means that the disk isstill receding, but does not present enough cooling of the innerflow for the spectral evolution to remain constant.In the rebrightening phase, the evolution of both the spectraland timing properties have departed from the linear decay. TheX-ray power-law photon index and ionization parameter fromthe continuum modeling, in addition to the X-ray and radio fluxdensities, present elevated values. Thus, it seems that the massaccretion rate has increased slightly to the compact object. Atthe same time, the X-ray spectra show high-ionization absorp-tion lines and the power-law component steepens and becomesprominent in the PSD during the latter part of the rebrighteningphase, indicating an increased amount of absorbing and scatter-ing material along the line of sight. The redshifted Fe XXV K α could arise from inflowing dense clumps in the disk atmosphereans / or corona (Kubota et al. 2018). However, this scenario doesnot explain why Fe XXVI Ly α line is not redshifted. Alternativeto a redshifted line, several ionization zones that produce severalabsorbing charge states can broaden the absorption line withoutinvoking a fast wind (Miller et al. 2020). This scenario wouldindicate a relatively dense medium of N H = × cm − inthe line of sight.Based on Chandra data, Miller et al. (2020) argued that theobscuration is caused by a “failed wind” that arises close tothe black hole and is unable to escape from the system. Theirfirst Chandra observation coincided with the latter part of therebrightening phase (MJD 58603) where there was evidence ofobscuration through photoionized absorption. Alternatively, theemerging obscuration could arise in a radially stratified, pu ff ed-up outer disk as discussed in Neilsen et al. (2020). Based on pho-toionization modeling of NICER spectra during an X-ray flare inthe obscured state, these latter authors argued that the X-ray ab-sorption takes place further out in a vertically extended outerdisk (R ∼ few × cm). The pu ff ed-up outer disk could eitherarise from changes in the inner accretion flow increasing thetemperature and scale-height of the outer disk by irradiation, ora structural change of the outer disk connected to the ending ofthe outburst and a switch to the quiescent state.Based on these scenarios and our modeling of the NICER data, the failed wind or pu ff ed-up outer disk would graduallyform during the rebrightening phase. This can be seen both inthe continuum modeling, in the absorption line evolution, andin the X-ray PSD evolution. In the continuum, local absorptionfrom lighter elements, as well as the increase of the reflectionfactor, indicates increasing scattering in the surrounding mat-ter with time. At the same time, the equivalent widths of thehighly ionized iron absorption lines increase, and the ratio of FeXXVI / Fe XXV decreases, indicating the presence of increasedcooler material in the line of sight. Also, the steepening power-law component in the PSD as well as the dilution of the rmsof the zero-centered Lorentzian is consistent with a scatteringmedium slowly entering the line of sight. The additional lumi-nosity seen in the rebrightening phase with respect to the lineardecay could arise from reaccreting the failed wind or some in-teraction with the wind, such as for example backscattering ofemission from the far side of the scattering cloud, or a struc-tural change of the inner accretion flow resulting in an increasein the mass accretion rate. Although the X-ray observations mayindicate an emerging wind in the rebrightening phase, which inprinciple could work as a depolarizing medium for the mm emis-sion from the jet, the polarization fraction of the mm emission Article number, page 11 of 18 & A proofs: manuscript no. 39581corr increases in the rebrightening phase. This could either mean thatthe depolarization of the putative wind does not present a mea-surable e ff ect on the polarization or that the mm emission line ofsight does not coincide with the disk wind.The change to the obscured state occurs likely at MJD 58608or soon after. The spectra during the NICER pointing still showhighly ionized absorption lines but present a clear drop in thephoton power law index, ionization parameter, and flux densityof the continuum departing from the levels of the rebrighteningphase. The rest of the NICER pointings are well into the ob-scured phase. Extrapolating the flux decay from the linear decayphase shows that ∼ 80% of the observed flux is absorbed or scat-tered. The fate of the GRS 1915 + 105 outburst is currently unclear.Nevertheless, a return to a “regular” hard state, not been seen be-fore during the outburst, can be established based on the presentwork, as detailed above. If the outburst of GRS 1915 + 105 istaken to be similar to other XRBs, a return to a hard state heraldsthe impending end of the outburst and an eventual return to qui-escence. On the other hand, recent work on the obscured state byBalakrishnan et al. (2020), Motta et al. (2021), and Neilsen et al.(2020) shows that GRS 1915 + 105 has exhibited both strong X-ray and radio flaring, indicating a significant mass accretion rateto the compact object. However, due to the heavily modified X-ray spectra by the obscuring material, the intrinsic X-ray lumi-nosity is di ffi cult to estimate accurately with values ranging from1% (Koljonen & Tomsick 2020; Miller et al. 2020) to 10% (Bal-akrishnan et al. 2020; Neilsen et al. 2020) and possibly reaching100% of the Eddington ratio occasionally (Balakrishnan et al.2020). In addition, it is not altogether clear whether the radiativee ffi ciency of the jet remains constant. If there is matter expelledin the direction of the jet, it can interact with the jet materialproducing shocks and more e ffi cient dissipation of the jet kineticenergy.Possible clues as to the underlying mass accretion rate maycome from the much softer state observed during September2020. Based on MAXI / GSC data (Motta et al. 2021, see also Fig.2), the X-ray spectrum can be fitted with a thermal model with aflux corresponding to the level observed in the linear decay phase(assuming no obscuration). Excursions to lower hardness ratiosduring the outburst decays of XRBs are seen, for example duringthe 2002 / − ff ected by absorption and the source couldalso present larger intrinsic luminosities (Motta et al. 2021).As discussed in Miller et al. (2020) and Motta et al. (2021),GRS 1915 + 105 may be experiencing a phase of high obscura-tion under which the source continues to accrete at the samerate as before. At some point in the future, when the obscuringmatter is lifted, we will see the rise in the apparent luminosity,which could be what happened in 1992 at the “onset” of the out-burst. On the other hand, given the results presented above, itis reasonable to assume that GRS 1915 + 105 has reached a low-luminosity hard state and could be on its way to quiescence. Asthe accretion disk of GRS 1915 + 105 is very large, we are seeingthe return to quiescence play out in slow motion as compared toother XRBs with much smaller disks, and this process can there-fore take years. ∼ 30 year outburst In this section, we entertain the possibility that the outburst be-gan in 1992 and that the source is heading towards quiescencein the near future, and what that would imply in terms of systemparameters.Truss & Done (2006) give estimates on the outburst dura-tion based on the assumed amount of matter accreted during thewhole outburst. These latter authors estimate that for the disksize of R disk = × cm the outburst duration would exceeda thousand years assuming that all the matter in the disk is ac-creted and that the surface density throughout the disk is equal tothe critical surface density required to trigger an outburst throughthe thermal-viscous instability. However, using a more realisticsurface density profile and a mass loss due to a disk wind, Truss& Done (2006) were able to reduce this to 76-160 years (wherethe range comes from assuming zero or Eddington mass loss forthe disk wind). This timescale is still too long for a 30-year out-burst.The duration can be further decreased assuming that only apart of the disk participates in the outburst. Truss & Done (2006)give a range of t outburst / (1 + e w ) = − 23 years depending onthe mass loss of the disk wind e w = ˙ M w / ˙ M Edd = − , andusing a radius of influence of the incident X-rays of R irr , = , which fits with the assumed duration of 30 years with e w ∼ . M w ≈ g s − (Neilsen et al. 2012; Milleret al. 2016). Assuming that during the outburst GRS 1915 + M Edd = . × g s − for a 12 solar massblack hole, giving e w = ˙ M w / ˙ M Edd ≈ . , which is compatiblewith the value estimated from the outburst duration.A more rigorous study with smooth particle hydrodynamicssimulations including the e ff ects of the thermoviscous instabil-ity, tidal torques, irradiation by central X-rays, and wind mass-loss resulted in a similar outburst length of t out = − 40 yrfor an irradiation e ffi ciency of (cid:15) ∼ − (Deegan et al. 2009).Truss & Done (2006) give a slightly larger value for the irradi-ation e ffi ciency of (cid:15) = . × − for the above radius R irr andgiven Eddington luminosity and a standard radiation e ffi ciencyof η = . 1. However, both studies concluded that the irradiatione ffi ciency (on average) cannot be higher as it would make theoutburst length much longer.Based on the assumed outburst duration and the small valueof the irradiation e ffi ciency, it seems clear that only a part of thedisk has participated in the outburst. In the following, we esti-mate the fraction of the disk mass accreted during the outburstof GRS 1915 + (cid:104) ˙ M disk (cid:105) = (cid:104) ˙ M Edd (cid:105) + (cid:104) ˙ M w (cid:105) − ˙ M , (2)where values for ˙ M Edd and ˙ M w are given above. The mass-transfer rate from the companion ˙ M can be estimated from Dee-gan et al. (2009) and Ritter (1999): − ˙ M ∼ . × − (cid:32) M M (cid:12) (cid:33) . (cid:32) P orb (cid:33) . M (cid:12) yr − , (3)which gives − ˙ M ∼ . × g s − for system parame-ters of GRS 1915 + Article number, page 12 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + light curve, f = . × − erg s − cm − , which for the dis-tance of 8.6 kpc and a standard radiative e ffi ciency of η = . ∼ × g s − when taking into account a bolo-metric correction of 10, which is typical for a hard-state XRB(Koljonen & Russell 2019). This is of similar order to the valueestimated from Eq. 3. Inserting − ˙ M = . × g s − to Eq.2 results in ˙ M disk ∼ . × g s − , which for the assumedoutburst duration of 30 years is 2.3 × g mass lost. To regainthis mass via the mass-transfer rate from the companion wouldtake approximately 560 years.An approximate or upper limit for the total mass of the diskbefore the onset of the outburst can be roughly estimated from: M disk = (cid:90) R out π R Σ ( R ) dR , (4)and assuming that the surface density, Σ , at all radii, R , isequal to the critical surface density, Σ max , to trigger an outburstvia the thermal-viscous instability (Hameury et al. 1998): Σ max = . α − . c (cid:32) M M (cid:12) (cid:33) − . (cid:32) R cm (cid:33) . g cm − , (5)where α c is the cold viscosity parameter. Inserting Eq. 5to Eq. 4, and using α c = . 02 and M / M (cid:12) = 12 gives M disk = × R . g. Assuming that the disk can at least reachthe circularization radius, which for GRS 1915 + 105 is approxi-mately 2 × cm, gives M disk = . × g, resulting in 5% ofthe total mass being accreted or lost during outburst. Similarly,the radius which corresponds to the lost disk mass is approxi-mately R = × cm.Using the irradiation law from Dubus et al. (2001), T = (cid:15) L Edd / πσ SB R , the irradiated temperature needed to fully ion-ize an accretion disk layer, T H ∼ K (Dubus et al. 1999), theirradiation e ffi ciency of (cid:15) = − , and Eddington luminosity for L bol = . × erg s − cm − results in R irr ∼ × cmfor the irradiation radius, which is much less than the size of thedisk ( ∼ × cm), but close to the radius where mass equatesthe mass lost from the disk in the outburst as determined above.Thus, assuming the outburst is nearing its end, we can concludethat the disk region that participated in the outburst is boundedby the irradiation radius, which is on the order of 5–8 × cm,and contains about 5% of the total disk mass. 5. Conclusions We conducted two full polarization ALMA observations to-gether with almost daily NICER pointing observations to studythe jet and accretion disk properties during the outburst decay in2018–2019 (Sect. 2). We divided the outburst decay into four dis-tinct phases: an exponential decay phase, a linear decay phase, arebrightening phase, and an obscured phase (Sect. 3.1). The firsttwo phases commonly occur during a decaying XRB outburst,and we show that the X-ray spectral and timing properties ofGRS 1915 + 105 are indeed typical for a decaying XRB outburst(Sect. 3.2). In addition, the jet emission in the mm is consistentwith a compact, steady jet showing ∼ 1% linear polarization, andthe magnetic field likely aligned with the jet position angle (Sect.3.3). Together with archival mm observations in the hard state,we formed a mm / X-ray correlation that revealed a correlationcoe ffi cient of 0.6 between the logarithmically scaled luminosi-ties (Sect. 3.3.2). The latter two decay phases are anomalous and present evidence of increased absorption and scattering (Sects.3.2.3 and 3.2.4) likely in the form of an accretion disk wind or apu ff ed-up outer disk (Sect. 4.3).Due to the large mass reservoir of the accretion disk in GRS1915 + ffi ciency leading to the irradiated part of the disk beingmuch less than the size of the disk. This is in direct discrepancywith the exponential decay profile, which is typically linked tothe viscous decay of fully irradiated disks. We speculate that e ffi -cient evaporation of the inner accretion disk could be responsiblefor the e-folding decay profile (Sect 4.2). Of course, all of theseproblems could be solved by assuming that the outburst is stillcontinuing and we have only witnessed a peculiar transit first toa canonical low-luminosity hard state and secondly to a heavilyobscured but intrinsically bright accretion state (Sect 4.4).Since the beginning of its outburst, GRS 1915 + 105 hasshown remarkable behavior in emission from the accretion diskand the jet, and the recent observations of a variety of new ac-cretion states show no exception. The peculiarities likely arisefrom the large disk size and our near-edge-on viewing angle tothe disk, allowing us to study the geometrical e ff ects of the ac-cretion flow. In addition, due to the long viscous time of theaccretion disk in GRS 1915 + + 105 in the near future. Acknowledgements. K. I. I. K. was supported by the Academy of Finland project320085. T. H. was supported by the Academy of Finland projects 317383,320085, and 322535. We thank the anonymous referee for useful comments.In addition, we thank the sta ff of the Nordic ARC Node for help in analyzingthe reliability of the fast variability in our ALMA data. This paper makes use ofthe following ALMA data: ADS / JAO.ALMA / NRAO and NAOJ. This research has made use ofdata and / or software provided by the High Energy Astrophysics Science ArchiveResearch Center (HEASARC), which is a service of the Astrophysics ScienceDivision at NASA / GSFC. References Balakrishnan, M., Miller, J. M., Reynolds, M. T., et al. 2020, arXiv e-prints,arXiv:2012.15033Belloni, T., Homan, J., Casella, P., et al. 2005, A&A, 440, 207Bianchi, S., Matt, G., Nicastro, F., Porquet, D., & Dubau, J. 2005, MNRAS, 357,599Cannizzo, J. K. 2000, ApJ, 534, L35Casella, P., Belloni, T., & Stella, L. 2005, ApJ, 629, 403Casella, P. & Pe’er, A. 2009, ApJ, 703, L63Castro-Tirado, A. J., Brandt, S., Lund, N., et al. 1994, ApJS, 92, 469Corbel, S., Fender, R. P., Tzioumis, A. K., et al. 2000, A&A, 359, 251Corbel, S., Nowak, M. A., Fender, R. P., Tzioumis, A. K., & Marko ff , S. 2003,A&A, 400, 1007Coriat, M., Corbel, S., Prat, L., et al. 2011, MNRAS, 414, 677Dauser, T., Garcia, J., Parker, M. L., Fabian, A. C., & Wilms, J. 2014, MNRAS,444, L100Deegan, P., Combet, C., & Wynn, G. A. 2009, MNRAS, 400, 1337Dhawan, V., Mirabel, I. F., & Rodríguez, L. F. 2000, ApJ, 543, 373Done, C., Wardzi´nski, G., & Gierli´nski, M. 2004, MNRAS, 349, 393Dubus, G., Hameury, J. M., & Lasota, J. P. 2001, A&A, 373, 251Dubus, G., Lasota, J.-P., Hameury, J.-M., & Charles, P. 1999, MNRAS, 303, 139Espinasse, M. & Fender, R. 2018, MNRAS, 473, 4122Falcke, H., Körding, E., & Marko ff , S. 2004, A&A, 414, 895Fender, R. P., Garrington, S. T., McKay, D. J., et al. 1999, MNRAS, 304, 865 Article number, page 13 of 18 & A proofs: manuscript no. 39581corr Fender, R. P., Rayner, D., Trushkin, S. A., et al. 2002, MNRAS, 330, 212Foster, R. S., Waltman, E. B., Tavani, M., et al. 1996, ApJ, 467, L81Fuchs, Y., Mirabel, I. F., & Claret, A. 2003, A&A, 404, 1011Gallo, E., Fender, R. P., & Pooley, G. G. 2003, MNRAS, 344, 60Gallo, E., Miller, B. P., & Fender, R. 2012, MNRAS, 423, 590García, J., Dauser, T., Lohfink, A., et al. 2014, ApJ, 782, 76Greiner, J., Cuby, J. G., McCaughrean, M. J., Castro-Tirado, A. J., & Mennick-ent, R. E. 2001, A&A, 373, L37Hameury, J.-M., Menou, K., Dubus, G., Lasota, J.-P., & Hure, J.-M. 1998, MN-RAS, 298, 1048Hannikainen, D. C., Hunstead, R. W., Campbell-Wilson, D., & Sood, R. K. 1998,A&A, 337, 460Heil, L. M., Uttley, P., & Klein-Wolt, M. 2015, MNRAS, 448, 3348Heinke, C. O., Bahramian, A., Degenaar, N., & Wijnand s, R. 2015, MNRAS,447, 3034Heinz, S. & Sunyaev, R. A. 2003, MNRAS, 343, L59Houck, J. C. & Denicola, L. A. 2000, in Astronomical Society of the Pacific Con-ference Series, Vol. 216, Astronomical Data Analysis Software and SystemsIX, ed. N. Manset, C. Veillet, & D. Crabtree, 591Ingram, A., Done, C., & Fragile, P. C. 2009, MNRAS, 397, L101Islam, N. & Zdziarski, A. A. 2018, MNRAS, 481, 4513Kallman, T. & Bautista, M. 2001, ApJS, 133, 221Klein-Wolt, M., Fender, R. P., Pooley, G. G., et al. 2002, MNRAS, 331, 745Koljonen, K., Vera, R., Lahteenmaki, A., & Tornikoski, M. 2019, The As-tronomer’s Telegram, 12839, 1Koljonen, K. I. I. & Russell, D. M. 2019, ApJ, 871, 26Koljonen, K. I. I. & Tomsick, J. A. 2020, A&A, 639, A13Kubota, M., Odaka, H., Tamagawa, T., & Nakano, T. 2018, ApJ, 868, L26Lee, J. C., Reynolds, C. S., Remillard, R., et al. 2002, ApJ, 567, 1102Lin, D., Smith, I. A., Liang, E. P., & Böttcher, M. 2000, ApJ, 543, L141Lyubarskii, Y. E. 1997, MNRAS, 292, 679Martí-Vidal, I., Vlemmings, W. H. T., Muller, S., & Casey, S. 2014, A&A, 563,A136Martocchia, A., Matt, G., Belloni, T., et al. 2006, A&A, 448, 677Merloni, A., Heinz, S., & di Matteo, T. 2003, MNRAS, 345, 1057Meyer-Hofmeister, E. 2004, A&A, 423, 321Meyer-Hofmeister, E. & Meyer, F. 2014, A&A, 562, A142Miller, J. M., Raymond, J., Fabian, A. C., et al. 2016, ApJ, 821, L9Miller, J. M., Zoghbi, A., Raymond, J., et al. 2020, ApJ, 904, 30Miller-Jones, J. C. A., Rupen, M. P., Fender, R. P., et al. 2007, MNRAS, 375,1087Mirabel, I. F. & Rodríguez, L. F. 1994, Nature, 371, 46Motta, S., Williams, D., Fender, R., et al. 2019, The Astronomer’s Telegram,12773, 1Motta, S. E., Casella, P., & Fender, R. P. 2018, MNRAS, 478, 5159Motta, S. E., Kajava, J. J. E., Giustini, M., et al. 2021, arXiv e-prints,arXiv:2101.01187Muñoz-Darias, T., Coriat, M., Plant, D. S., et al. 2013, MNRAS, 432, 1330Muno, M. P., Remillard, R. A., Morgan, E. H., et al. 2001, ApJ, 556, 515Neilsen, J., Cackett, E., Remillard, R. A., et al. 2018, ApJ, 860, L19Neilsen, J., Homan, J., Steiner, J. F., et al. 2020, ApJ, 902, 152Neilsen, J., Petschek, A. J., & Lee, J. C. 2012, MNRAS, 421, 502Nied´zwiecki, A. 2005, MNRAS, 356, 913Nobili, L., Turolla, R., Zampieri, L., & Belloni, T. 2000, ApJ, 538, L137Ogley, R. N., Bell Burnell, S. J., Fender, R. P., Pooley, G. G., & Waltman, E. B.2000, MNRAS, 317, 158Pahari, M., Neilsen, J., Yadav, J. S., Misra, R., & Uttley, P. 2013, ApJ, 778, 136Peris, C. S., Remillard, R. A., Steiner, J. F., et al. 2016, ApJ, 822, 60Petrucci, P. O., Merloni, A., Fabian, A., Haardt, F., & Gallo, E. 2001, MNRAS,328, 501Pooley, G. G. & Fender, R. P. 1997, MNRAS, 292, 925Powell, C. R., Haswell, C. A., & Falanga, M. 2007, MNRAS, 374, 466Qu, J. L., Lu, F. J., Lu, Y., et al. 2010, ApJ, 710, 836Reid, M. J., McClintock, J. E., Steiner, J. F., et al. 2014, ApJ, 796, 2Reig, P., Belloni, T., van der Klis, M., et al. 2000, ApJ, 541, 883Remillard, R. A. & McClintock, J. E. 2006, ARA&A, 44, 49Ritter, H. 1999, MNRAS, 309, 360Rodríguez, L. F. & Mirabel, I. F. 1999, ApJ, 511, 398Rushton, A., Spencer, R., Fender, R., & Pooley, G. 2010, A&A, 524, A29Russell, T. D., Miller-Jones, J. C. A., Curran, P. A., et al. 2015, MNRAS, 450,1745Soleri, P. & Fender, R. 2011, MNRAS, 413, 2269Steeghs, D., McClintock, J. E., Parsons, S. G., et al. 2013, ApJ, 768, 185Tetarenko, B. E., Lasota, J. P., Heinke, C. O., Dubus, G., & Sivako ff , G. R. 2018,Nature, 554, 69Trushkin, S. A., Nizhelskij, N. A., Tsybulev, P. G., Bursov, N. N., & Shevchenko,A. V. 2019, The Astronomer’s Telegram, 12855, 1Truss, M. & Done, C. 2006, MNRAS, 368, L25Truss, M. R. & Wynn, G. A. 2004, MNRAS, 353, 1048Ueda, Y., Yamaoka, K., Sánchez-Fernández, C., et al. 2002, ApJ, 571, 918van den Eijnden, J., Ingram, A., & Uttley, P. 2016, MNRAS, 458, 3655van den Eijnden, J., Ingram, A., Uttley, P., et al. 2017, MNRAS, 464, 2643Xie, F.-G. & Yuan, F. 2016, MNRAS, 456, 4377Zhang, L., Méndez, M., Altamirano, D., et al. 2020, MNRAS, 494, 1375 Article number, page 14 of 18. I. I. Koljonen and T. Hovatta: ALMA / NICER observations of GRS 1915 + Table .1. NICER timing analysis parameters. Exponential decay phase Model: (P)owerlaw (L)orentzian (0) Lorentzian (1) = QPO Main freq. Lorentzian (2) = QPO 1st harmonic NICER Date Exp. Γ σ rms ν Q rms ν Q rms obs. MJD (s) (Hz) (%) (Hz) (%) (Hz) (%)A137 58238.609 5491 PL012 0.8 + . − . ± ± ± + . − . ± + . − . + − ± + . − . + − ± ± ± ± ± + − + − A139 58243.886 1386 L012 – 7 ± ± ± + − ± + . − . + − ± + . − . + − + − ± + − + . − . ± + − ± ± + . − . ± + − + . − . – – –A143 58261.847 1141 PL012 1.0 + . − . ± + − ± ± + . − . ± + − + − A144 58262.504 764 L012 – 5 ± ± ± + − + . − . ± + − ± + − ± ± ± + . − . + . − . + − + − A146 58264.674 709 L01 – 7 ± ± ± + − + . − . – – –A147 58265.962 873 L012 – 7 ± + − ± + − + . − . ± + − + − A148 58266.669 862 L012 – 4 ± ± ± + − ± + . − . + − ± + . − . ± ± + − + . − . ± + − ± + . − . ± ± + − ± ± + − + − A152 58271.366 1787 L012 – 4.9 ± ± ± ± ± ± + − ± ± ± ± + − ± ± + − ± ± ± ± + − ± + . − . + − + − A155 58274.583 965 L01 – 6 + − + − ± + − ± ± ± ± ± + . − . ± + − ± ± + . − . ± + . − . ± ± + . − . ± ± + . − . ± ± + . − . + . − . + − ± ± ± ± + − + . − . ± + − ± Linear decay phase Model: (P)owerlaw (L)orentzian (0) Lorentzian (1) = QPO Main freq. Lorentzian (2) = QPO 1st harmonic NICER Date Exp. Γ σ rms ν Q rms ν Q rms obs. MJD (s) (Hz) (%) (Hz) (%) (Hz) (%)A160 58308.830 719 L012 – 6 ± + − + . − . > ± ± > + − A161 58309.147 2705 L012 – 5.1 ± + . − . ± + − ± + . − . + − + − A162 58310.839 812 L01 – 5.3 + . − . ± ± + − ± + − ± + . − . > + − ± + − + − A164 58313.539 929 L012 – 3.9 + . − . + − ± + − ± ± + − + − A165 58314.695 806 L01 – 6 ± ± + . − . > ± ± ± ± > 14 7 ± + . − . > 11 6 + − A167 58316.560 537 L01 – 6 + − + − + . − . + − + − – – –A168 58326.992 779 L012 – 4 ± ± + . − . > + − ± + − + − A169 58327.056 735 L012 – 1.3 + . − . ± ± + − + − + . − . > . + − A170 58328.021 1561 L012 – 4.7 ± ± + . − . > 11 6.7 + . − . ± + − + − A171 58329.051 1612 L012 – 2.9 + . − . + − ± + − + . − . + . − . + − + − A172 58330.015 2974 L012 – 3 ± ± ± ± ± ± + − + − A174 58332.663 558 L01 – 6 + − ± ± + − ± ± ± + . − . + − ± + − + − ± + − ± + . − . + − + − A180 58343.136 928 L01 – 5.1 + . − . ± + . − . > ± + . − . + − ± + − ± ± + − + − A182 58350.534 1170 L01 – 6.0 + . − . ± + . − . + − ± ± ± + . − . + − ± ± + − ± + − + . − . ± + − + − A190 58362.268 1261 L012 – 7 + − + − ± + − ± + . − . > ± ± ± ± + − ± ± > + − A193 58378.221 1287 L012 – 5.4 ± ± ± + − ± ± + − + − A195 58415.814 1526 L012 – 1.1 + . − . + − ± + − ± + . − . + . − . + − A197 58420.314 3858 L012 – 4.6 + . − . + . − . ± + − ± ± > 20 3.2 + . − . A198 58424.756 5647 L01 – 4.2 ± + . − . ± + − ± ± + − + . − . A199 58425.013 6025 L012 – 4.2 ± + . − . + . − . + − ± + . − . + − + − A201 58426.946 1720 L012 – 3 ± ± ± + − ± ± + − + − A206 58438.095 636 L012 – 5 + − + − + . − . + − + − + . − . > + − A208 58440.027 856 L01 – 4.7 + . − . ± ± > + − – – –A209 58449.092 3065 L012 – 1.3 + . − . + − ± + − + − + . − . + . − . ± ± ± + . − . > + − – – –A211 58451.022 2391 L012 – 0.6 + . − . ± + . − . > 10 5 + − + . − . + . − . ± & A proofs: manuscript no. 39581corrA212 58452.888 798 L01 – 5 ± ± ± + − ± ± + − ± + − + − – – –A214 58455.205 850 L01 – 5 ± + − + . − . + − + − – – –A215 58456.234 2826 L012 – 1.2 + . − . + − ± ± ± + . − . + . − . ± ± ± + . − . > 15 7 ± ± > 10 5 + − A217 58458.229 1185 L012 – 4 ± + − ± > ± ± > ± ± + − + . − . + − + − ± + − > ± + − + . − . > ± ± ± ± + − + − – – –A222 58463.251 510 L01 – 6 ± ± ± + − + − – – – Rebrightening phase Model: (P)owerlaw (L)orentzian (0) Lorentzian (1) = QPO Main freq. Lorentzian (2) = QPO 1st harmonic NICER Date Exp. Γ σ rms ν Q rms ν Q rms obs. MJD (s) (Hz) (%) (Hz) (%) (Hz) (%)B101 58548.692 3818 PL012 0.7 + . − . ± ± ± + − ± + . − . + − ± + . − . ± + − ± + − ± ± + − ± + . − . ± ± ± + − ± + . − . + − ± + . − . ± + − ± + − ± ± + − ± + . − . + − + − + . − . + − ± + . − . > ± + . − . ± + − ± + − ± + . − . + − ± + . − . ± ± ± ± ± ± + − + − B801 58596.557 3766 PL012 1.3 ± ± ± ± + − ± + . − . + − ± ± ± + − ± + − ± + . − . + − ± ± ± + − ± Obscured phase Model: (P)owerlaw (L)orentzian (0) Lorentzian (1) = QPO Main freq. Lorentzian (2) = QPO 1st harmonic NICER Date Exp. Γ σ rms ν Q rms ν Q rms obs. MJD (s) (Hz) (%) (Hz) (%) (Hz) (%)C202 58622.518 14421 P 1.3 ± + . − . – – – – – – – –C204 58624.065 6830 P 1.9 ± ± ± + . − . – – – – – – – –C208 58628.057 4627 P 2.3 + . − . – – – – – – – –C209 58629.024 1532 – – – – – – – – – –C301 58631.089 15866 P 1.7 + . − . – – – – – – – –C302 58632.247 8146 P 2.0 + . − . – – – – – – – –C303 58633.601 4790 P 1.7 ± / NICER observations of GRS 1915 + Table .1. NICER spectral analysis parameters. Exponential decay phase NICER N(H) N(Mg) N(Al) N(Si) N(S) N(Ca) Γ log ξ R f R in F abs F unabs χ obs. 10 atoms / cm − erg / s / cm − erg / s / cm A137 2.40 ± ± ± ± ± ± ± ± + . − . + − ± ± ± ± ± ± ± ± + . − . + . − . ± ± ± ± ± ± ± ± + . − . + . − . ± ± ± ± ± ± ± ± ± + . − . ± ± ± + . − . ± ± ± + . − . ± + . − . ± ± ± ± ± ± + . − . ± + . − . ± ± ± ± ± ± + . − . ± ± + . − . ± ± ± + . − . ± ± ± + . − . ± ± ± ± ± ± + . − . + − ± ± ± ± ± ± ± ± ± ± + . − . ± ± ± ± ± ± ± ± ± ± + . − . ± ± ± ± ± ± ± ± ± + . − . ± + . − . ± ± ± ± ± ± ± + . − . ± + . − . ± ± ± ± ± ± ± + . − . ± + . − . + . − . ± ± + . − . ± ± ± + . − . ± ± + . − . ± ± ± ± ± ± ± ± ± + . − . ± ± ± ± ± ± + . − . ± + . − . ± ± ± ± ± ± ± + . − . ± ± + . − . ± ± ± ± ± ± ± ± ± + . − . + . − . + − ± ± ± ± ± ± ± ± ± + . − . ± ± ± = N(H) 3.4 ± = N(Al) = N(Al) = N(Al) 2.00 ± + . − . ± ± Linear decay phase NICER N(H) N(Mg) N(Al) N(Si) N(S) N(Ca) Γ log ξ R f R in F abs F unabs χ obs. 10 atoms / cm − erg / s / cm − erg / s / cm A160 2.21 + . − . = N(H) 2.8 ± = N(Al) = N(Al) = N(Al) 2.02 + . − . ± ± ± ± = N(H) 3.2 ± = N(Al) = N(Al) = N(Al) 2.03 ± + . − . ± ± + . − . = N(H) 3.1 + . − . = N(Al) = N(Al) = N(Al) 2.02 ± ± ± ± ± = N(H) 2.9 + . − . = N(Al) = N(Al) = N(Al) 2.05 ± ± ± ± + . − . = N(H) 2.9 + . − . = N(Al) = N(Al) = N(Al) 2.01 + . − . + . − . ± ± ± = N(H) 3.0 + . − . = N(Al) = N(Al) = N(Al) 2.04 ± + . − . ± ± + . − . = N(H) 3.4 + . − . = N(Al) = N(Al) = N(Al) 2.08 ± ± ± ± ± = N(H) 3.2 ± = N(Al) = N(Al) = N(Al) 1.97 + . − . + . − . ± ± ± = N(H) 3.5 + . − . = N(Al) = N(Al) = N(Al) 1.98 ± + . − . ± ± ± = N(H) 3.1 + . − . = N(Al) = N(Al) = N(Al) 1.93 ± + . − . ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.98 ± + . − . ± ± ± = N(H) 3.4 ± = N(Al) = N(Al) = N(Al) 2.00 ± + . − . ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.99 ± + . − . ± ± ± = N(H) 3.3 ± = N(Al) = N(Al) = N(Al) 2.00 + . − . + . − . ± ± ± = N(H) 2.8 ± = N(Al) = N(Al) = N(Al) 1.97 ± ± ± ± ± = N(H) 3.3 ± = N(Al) = N(Al) = N(Al) 1.96 ± + . − . ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.91 ± ± ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.89 ± ± ± ± ± = N(H) 2.9 ± = N(Al) = N(Al) = N(Al) 1.88 ± + . − . ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.86 ± + . − . ± ± ± = N(H) 3.4 ± = N(Al) = N(Al) = N(Al) 1.97 ± ± ± ± ± = N(H) 3.3 ± = N(Al) = N(Al) = N(Al) 2.00 + . − . ± ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.98 ± ± ± ± ± = N(H) 3.2 + . − . = N(Al) = N(Al) = N(Al) 1.98 ± + . − . ± ± ± = N(H) 3.4 + . − . = N(Al) = N(Al) = N(Al) 2.01 + . − . + . − . ± ± + . − . ± ± ± ± ± ± + . − . ± ± ± ± ± ± ± + − + . − . + . − . ± ± ± ± ± ± ± ± ± + . − . ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.81 ± ± ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.93 ± ± ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.93 ± ± ± ± ± = N(H) 3.3 ± = N(Al) = N(Al) = N(Al) 1.89 ± ± ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.90 ± ± ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.92 ± ± ± ± ± = N(H) 2.9 ± = N(Al) = N(Al) = N(Al) 1.92 ± + . − . ± ± ± = N(H) 3.4 ± = N(Al) = N(Al) = N(Al) 1.94 ± + . − . ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.91 + . − . + . − . ± ± & A proofs: manuscript no. 39581corrA215 2.21 ± = N(H) 2.9 ± = N(Al) = N(Al) = N(Al) 1.87 ± + . − . ± ± ± = N(H) 2.9 ± = N(Al) = N(Al) = N(Al) 1.85 ± ± ± ± ± = N(H) 2.6 ± = N(Al) = N(Al) = N(Al) 1.85 ± + . − . ± ± ± = N(H) 3.0 ± = N(Al) = N(Al) = N(Al) 1.82 ± ± ± ± ± = N(H) 3.1 ± = N(Al) = N(Al) = N(Al) 1.90 ± + . − . ± ± ± = N(H) 2.6 ± = N(Al) = N(Al) = N(Al) 1.85 ± + . − . ± ± ± = N(H) 2.7 ± = N(Al) = N(Al) = N(Al) 1.91 ± + . − . ± ± Rebrightening phase NICER N(H) N(Mg) N(Al) N(Si) N(S) N(Ca) Γ log ξ R f R in F abs F unabs χ obs. 10 atoms / cm − erg / s / cm − erg / s / cm B101 2.19 ± ± ± ± ± ± ± + . − . ± ± ± ± ± ± + . − . ± + . − . ± ± ± + . − . ± ± ± ± ± ± + . − . ± ± ± ± + − + . − . + . − . ± + . − . + . − . ± ± ± ± + − ± ± + − ± + . − . ± ± ± ± ± ± ± ± ± + . − . ± ± + . − . ± ± ± ± ± ± + . − . ± ± ± ± ± ± ± ± ± + . − . + − . 100 0.141 ± ± ± + . − . ± ± + . − . ± ± ± + − . 100 0.120 ± ± Obscured phase NICER N(H) N(Mg) N(Al) N(Si) N(S) N(Ca) Γ log ξ R f R in F abs F unabs χ obs. 10 atoms / cm − erg / s / cm − erg / s / cm C001 2.21 ± ± ± ± ± ± + . − . + . − . − ± ± ± + . − . + ± ± = N(H) 1.44 ± + . − . − ± ± + . − . ± + + . − . ± = N(H) 1.64 + . − . + . − . − ± ± + . − . ± ± ± ± = N(H) 1.69 + . − . ± − ± ± + . − . ± ± ± ± = N(H) 1.53 ± + . − . − ± ± ± + − + + . − . ± = N(H) 1.64 + . − . ± − ± ± + . − . ± + + . − . = N(H) = N(H) 1.51 + . − . + . − . − ± ± ± ± ± 10 2.6 ± = N(H) = N(H) 1.20 + . − . + . − . − ± ± + . − . + − + ± = N(H) = N(H) 1.63 ± + . − . − ± ± ± ± ± ± = N(H) = N(H) 1.63 + . − . + . − . − ± ± ± = N(H) = N(H) = N(H) = N(H) = N(H) 1.20 + . − . ± − ± ± ± = N(H) 19 + = N(H) = N(H) = N(H) 1.3 ± + . − . − ± ± + . − . = N(H) = N(H) = N(H) = N(H) = N(H) 1.2 + . − . + . − . − ± ±±