SN 2023ixf: The Closest Supernova of the Decade

Abstract

Supernova 2023ixf occurred on 18 May 2023 in the nearby galaxy Messier 101 ($D\approx 6.85$ Mpc), making it the closest supernova in the last decade. Following its discovery, astronomers around the world rushed to observe the explosion across the electromagnetic spectrum in order to uncover its early-time properties. Based on multi-wavelength analysis during its first year after explosion, Supernova 2023ixf is a type II supernova that interacted with dense, confined circumstellar material in its local environment—this material being lost from its red supergiant progenitor in the final years before explosion. In this article, we will review the findings of >80 studies already published on this incredible event and explore how the synthesis of SN 2023ixf observations across the electromagnetic spectrum can be used to constrain type II supernova explosion physics in addition to the uncertain mass loss histories of red supergiant stars in their final years.

1. Discovery

The discovery of Supernova (SN) 2023ixf was first reported to the Transient Name Server (TNS) by Koichi Itagaki [1] on 19 May 2023 17:27:15 (MJD 60083.73). The discovery magnitude by Itagaki was 14.9 mag in the clear filter, and the SN was located in the southeastern spiral arm of the Pinwheel Galaxy (Messier 101, NGC 5457; Figure 1) at a distance of $6.85\pm 0.15$ Mpc [2]. Then, on 19 May 2023 22:23:45 (MJD 60083.93), SN 2023ixf was classified as a type II supernova (SN II) with “flash ionization lines of H, He, C, and N″ [3]. Following its discovery and classification, numerous astronomers reported serendipitous detections and non-detections of SN 2023ixf to better estimate the time of first light [4]. Among these observations, the most constraining were reported in [5] with the last deep upper limit being >20.4 mag (<−8.78 mag) on MJD 60082.66, followed by the first detection of $17.1\pm 0.1$ mag (−12.1 mag) in r-band on MJD 60082.85. Consequently, the time of first light is conservatively placed at MJD $60082.757\pm 0.097$. However, tighter constraints on the time of first light have been made through model fits to the earliest photometry (e.g., see [6]).

2. Spectroscopic Properties

2.1. CSM Interaction Phase

The earliest spectrum of SN 2023ixf was obtained at a phase of ∼1.1 days using the SPRAT on the Liverpool Telescope [3]. As shown in Figure 2, the early spectra of SN 2023ixf show emission lines of H i, He i/ii, C iv, and N iii/iv/v, which are superimposed on a hot blue spectral continuum [8,9,10,11]. Such spectral lines are known as “flash” or “IIn-like” features and arise from the persistent photo-ionization of dense circumstellar material (CSM) ahead of the cooling forward shock [12,13,14,15,16,17]. Similar to type IIn SNe, these emission profiles are composed of both a narrow line core and Lorentzian “wings” that extend to ∼1000 $\mathrm{km}{\mathrm{s}}^{-1}$ that result from electron scattering in the ionized, optically thick CSM [14,18,19]. However, it should be noted that the velocity of the narrow line core is immediately influenced by radiative acceleration following shock breakout (SBO) and may not trace the exact progenitor wind velocity [20]. For SN 2023ixf, the earliest high-resolution ($R\approx$ 70,000) spectroscopy was obtained at $\delta t=1.51$ days and was used to derive a CSM wind velocity of ∼25 $\mathrm{km}{\mathrm{s}}^{-1}$ from the He i emission line [21]. This value is comparable to the narrow line core measurement of SN II 1998S (∼30 $\mathrm{km}{\mathrm{s}}^{-1}$ [22]), but lower than type IIn and type Ibn SNe (∼100–1000 $\mathrm{km}{\mathrm{s}}^{-1}$ [23]). Continued monitoring with high-resolution spectrographs revealed increasing equivalent widths of the narrow line cores that were attributed to either radiative acceleration of the pre-shock gas [20,21] or CSM asymmetries [24]. For the latter, the strongest evidence for asymmetric CSM came from spectropolarimetry observations that began at $\delta t=1.4$ days, the earliest phase that such observations had ever been conducted for a SN to date [25]. The first spectropolarimetric epoch showed high continuum polarization ($p\approx 1\%$), which rapidly decreased over the first week, in addition to a dramatic shift in position angle coincident with the fading of the IIn-like features [26].

The duration and evolution of IIn-like features in CSM-interacting SNe II is a direct probe of the CSM structure and shock physics [27]. Notably, the earliest spectra of SN 2023ixf at $\delta t<2$ days show electron-scattering-broadened emission lines of He i and N iii, which fade in strength by ∼3 days as higher ionization species such as He ii, N iv/v, and C iv become increasingly strong [28]. This detection of rising ionization in SN 2023ixf is coincident with increasing blackbody temperature and a red-to-blue color evolution, all of which are attributed to extended shock breakout from dense CSM [6,29]. As shown in past studies (e.g., [27,30,31]), the duration of the electron-scattering line profiles can be used to identify the radius at which the pre-shock CSM becomes optically thin to electron scattering. For SN 2023ixf, this transition occurred at $\delta t\approx 7$ days, which corresponds to a shock radius ∼6 $\times {10}^{14}$ cm for a shock velocity of ∼${10}^4\mathrm{km}{\mathrm{s}}^{-1}$ [8,9,10,29,32]. At this point, as shown in Figure 2, the spectra of SN 2023ixf show only narrow emission in H i transitions in addition to broad absorption “troughs,” indicating that the photosphere had receded into the swept-up material present in the fast-moving dense shell (i.e., shocked CSM). Then, after this transition phase, P-Cygni profiles typical of SNe II emerged from the fastest-moving SN ejecta below the dense shell. As multi-wavelength observations in the X-ray and radio revealed, the CSM in SN 2023ixf extends beyond the shock radii when both narrow and electron-scattering-broadened emission lines disappear. Estimates on CSM mass, density, and extent and progenitor mass loss rate are discussed in Section 7.3.

While the majority of the spectroscopic monitoring was in optical wavelengths, SN 2023ixf represented the first successful opportunity to capture the near-ultraviolet (NUV) spectra of a CSM-interacting SN within days of first light. The earliest NUV spectrum was obtained by the Neil Gehrels Swift Observatory at $\delta t=0.7$ days and showed no obvious spectral features, likely due to the extremely low resolution of the Swift-UVOT grism [10,29]. Then, as shown in Figure 3, NUV spectroscopy with the Space Telescope Imaging Spectrograph (STIS) on the Hubble Space Telescope (HST) was obtained beginning at $\delta t=3.6$ days [29]. These observations represented the first time that UV spectra of a young SN II were obtained during the CSM interaction phase. Intriguingly, the NUV spectrum of SN 2023ixf is relatively featureless—the strongest predicted IIn-like features are in the far-UV, which is not accessible with the HST/STIS CCD instrument. However, there are detections of narrow N iv $\lambda 1718$ and C iii $\lambda 2297$ emission, the latter of which persists until ∼8.7 days as a combined narrow P-Cygni profile plus broad absorption, similar to H$\alpha$ at the same phase. Additionally, far-UV spectra were obtained as early as +6.9 days using the UltraViolet Imaging Telescope (UVIT) on AstroSat [10]. Future UV missions such as the Ultraviolet Explorer (UVEX) will enable routine far- and near-UV spectroscopy of CSM-interacting SNe II during the first days after explosion [33].

2.2. Photospheric Phase

Following the CSM interaction phase, SN 2023ixf evolved to display Doppler-broadened P-Cygni profiles typical of SNe II during and after maximum light. As discussed in [34], the photospheric velocities of SN 2023ixf are overall consistent with samples of SNe II. Notably, the photospheric phase spectra of SN 2023ixf show some complexities, specifically in the H$\alpha$ profile, such as shallower absorption, high velocity features, and boxy redward emission. These observables are proposed to be related to the formation of a cold dense shell (CDS) during the CSM interaction phase as the forward shock swept up the dense, confined CSM, which produced the IIn-like signatures [35,36]. Furthermore, the boxy emission on the red side of H$\alpha$ is likely the result of shock power injected into the CDS from continued ejecta interaction with more extended CSM [37]. The presence of underlying shock power in SN 2023ixf is also validated by the detection of Mg ii $\lambda 2800$ emission in UV spectra during the photospheric phase [38].

2.3. Nebular Phase

SN 2023ixf began to transition to its nebular phase following the end of the light curve plateau at $\delta t\approx 80$ days. At this point, forbidden emission lines of [O i] and [Ca ii] emerged as the SN photosphere receded into the innermost layers of ejecta. Given its brightness, it was possible to track the polarization evolution of SN 2023ixf out to $\sim 120$ days, i.e., after the light curve plateau end and at the beginning of the nebular phase. Interestingly, SN 2023ixf displayed increasing polarization at the end of the photospheric phase as the photosphere traces the inner, iron-rich layers of the SN ejecta [34]. This rise in polarization was attributed to an asymmetric distribution of ⁵⁶Ni and/or a bipolar explosion [26,39]. These findings are broadly consistent with the blueshifted [O i] emission, which was interpreted as being the result of asymmetric distributions of O-rich ejecta [40].

The other notable feature of the late-time spectra in SN 2023ixf is the boxy, asymmetric emission observed in H$\alpha$, which becomes most prominent at $\delta t>200$ days [41,42,43]. This underlying boxy profile traces emission from the CDS that is located at ∼8000 $\mathrm{km}{\mathrm{s}}^{-1}$, and which is powered by persistent ejecta interaction with distant intervening CSM at >${10}^{15}$ cm. Additionally, the attenuation of redward flux leading to an asymmetric emission in H$\alpha$ is a signature of dust formation in CDS and/or inner SN ejecta. The presence/formation of dust in SN 2023ixf is supported by strong molecular CO emission in the NIR spectra and the presence of SiO emission in the James Webb Space Telescope NIRSpec and MIRI spectra (private communication).

3. Photometric Properties

Multi-band photometry of SN 2023ixf began ∼hours after first light and was able to capture the emergence of the SN shock from within the dense, confined CSM within ∼$2R_{★}$. One of the most compelling features of the SN 2023ixf light curve is the multi-component rise wherein the first ∼12 h of evolution requires a separate power law function fit, while the rise to peak brightness at $\delta t>12$ h can be described by a typical $F_{\nu }\propto t^2$ model [7]. By utilizing a multi-component light curve model formalism, ref. [6] derives a time of first light of MJD $60082.{788}_{-0.05}^{+0.02}$. Furthermore, during the first ∼day, SN 2023ixf displayed a dramatic red-to-blue color evolution (e.g., Figure 4) that appears to be common in CSM-interacting SNe II with IIn-like features (e.g., [31,44]) and is likely connected to shock breakout from the most confined CSM. This behavior is coincident with a nearly constant blackbody radius and rising blackbody temperature in SN 2023ixf during its first days [29,37]. Furthermore, ref. [6] attributes the reddened colors of SN 2023ixf in its first hours to dust destruction during the CSM shock breakout. Interestingly, this scenario is also consistent with the IR excess observed in SN 2023ixf at $\delta t=3.6$ days by the serendipitous observations with the Near-Earth Object Wide-field Infrared Survey Explorer (NEOWISE-R) [45].

SN 2023ixf rose to maximum brightness in optical bands within ∼5–6 days and reached an absolute magnitude in B-band of $M=-18.5$ mag [8,34,46]. This enhanced peak luminosity is typical of CSM-interacting SNe II (e.g., see Figure 4), which can be 1–2 magnitudes brighter than SNe II without IIn-like features. Given the linear decline of SN 2023ixf’s light curve plateau, it was given the “type II-L” SN distinction, which is often physically related to a smaller progenitor star hydrogen envelope mass [37,47]. SN 2023ixf has a measured pseudo-bolometric light curve plateau duration of $t_p\approx 82$ days, which places it within the sub-class of “short plateau” SNe II [48,49]. Multiple measurements of the ⁵⁶Ni mass were made for SN 2023ixf, with a favored value of ∼$0.06{\mathrm{M}}_{\odot }$ estimated from modeling of the post-plateau light curve [29,34,50]. Additionally, the light curve “tail” of SN 2023ixf declines faster than expected for radioactive decay power with complete $\gamma$-ray trapping, and modeling of this evolution recovers a trapping timescale of ∼250–300 days [34,50]. Presently, at $\delta t>600$ days, the optical light curve of SN 2023ixf has flattened, deviating from the expected decline rate of radioactive decay power, as shock power becomes dominant. This phenomenon was also observed at earlier phases in UV and X-ray wavelengths (Jacobson-Galán et al., in prep.).

4. X-Ray Observations

SN 2023ixf was first observed at X-ray wavelengths by Swift-XRT (0.3–10 keV) beginning ∼1 day after first light. Additionally, high signal-to-noise X-ray spectra were obtained using NuSTAR (3–79 keV) at $\delta t$ = 4.4–58.4 days, Chandra (0.5–8 keV) at $\delta t$ = 11.5–86.7 days, and XMM-Newton (0.3–10 keV) at $\delta t$ = 9.0–58.2 days. The exceptional coverage of SN 2023ixf in both soft and hard X-ray bands allowed for robust constraints on the temperature of the X-ray emitting plasma—this being the first time that such observations were possible for a SNe II with IIn-like features [51,52,53]. The X-ray emission observed in SN 2023ixf during the first ∼90 days was dominated by the forward shock and is best modeled as an absorbed bremsstrahlung spectrum [52,54]. As shown in Figure 5, SN 2023ixf rose to a peak X-ray luminosity of ∼${10}^{40}$ erg ${\mathrm{s}}^{-1}$ in ∼1 week, and represents one of the most luminous SN II-P/L observed to date. The detection of prominent neutral Fe K$\alpha$ emission combined with large intrinsic neutral hydrogen column densities from X-ray spectral modeling confirms the absorption of X-rays by dense, confined CSM within the first days to a week after explosion [51,54]. The rise to maximum X-ray brightness is attributed to decreasing photoelectric absorption as the CSM densities ahead of the forward shock also decrease in time. Currently, SN 2023ixf continues to be detected with X-ray telescopes, and the evolution of the X-ray spectrum may indicate the emergence of the radiative reverse shock (private communication).

5. Radio Observations

A high-cadence, multi-frequency campaign was also carried out in the radio with observations beginning as early as $\delta t=2.6$ days with the Submillimeter Array (SMA). However, radio emission was not detected until $\delta t=29.2$ days using the Karl G. Jansky Very Large Array (VLA) [55,56]. Additional observations were made with the Giant Metrewave Radio Telescope (GMRT), LOw-Frequency ARray (LOFAR), Northern Extended Millimetre Array (NOEMA), Japanese and Korean VLBI Networks, and European VLBI Network (EVN) [54,57,58,59]. The radio emission in SN 2023ixf is attributed to non-thermal synchrotron radiation from the forward shock that is suppressed by free–free absorption, which decreases in time as the density of the CSM also decreases [54].

6. Neutrinos, $\mathit{\gamma }$-Rays, and Gravitational Waves

SN 2023ixf allowed for some of the best constraints on neutrino, $\gamma$-ray, and gravitational wave (GW) emission from a SN to date, despite the non-detection of all multi-messenger signals. Specifically, the CSM interaction in SN 2023ixf provided a framework for the production of high-energy cosmic rays (CRs), $\gamma$-ray, and neutrinos. Overall, non-detection limits from Fermi-LAT and IceCube proved consistent with the expectations for neutrino and $\gamma$-ray generation using the CSM and explosion parameters inferred from X-ray through radio observations [60,61,62,63,64,65]. Lastly, LIGO–Virgo–KAGRA did not detect any GWs coincident with SN 2023ixf, but those observations were able to place the most stringent constraints to date on core-collapse SN GW energy/luminosity and proto-NS ellipticity [66].

7. Progenitor System

7.1. Pre-Explosion Imaging

Given its occurrence in M101, SN 2023ixf had a significant amount of associated pre-explosion imaging with both HST and Spitzer, which extended more than two decades before first light. As shown in Figure 6, a progenitor star was clearly detected in some optical HST filters and Channels 1 and 2 of Spitzer. Additional J- and K-band detections of the progenitor star were made through the Near-Infrared Imager (NIRI) on the Gemini-North Telescope and the MMT Observatory telescope. Based on the pre-explosion spectral energy distribution (SED), it was clear that the progenitor star was a red supergiant (RSG) that was enshrouded in a dust shell, as evidenced by the highly reddened SED (e.g., Figure 6). However, estimates on Zero Age Main Sequence (ZAMS) based on the progenitor luminosity and temperature are wide-ranging: $11\pm 1{\mathrm{M}}_{\odot }$ [67], $17\pm 4{\mathrm{M}}_{\odot }$ [68], $17\pm 2{\mathrm{M}}_{\odot }$ [69], $20\pm 4{\mathrm{M}}_{\odot }$ [70], $13\pm 1{\mathrm{M}}_{\odot }$ [71], $18\pm 1{\mathrm{M}}_{\odot }$ [72], $\sim 8-10{\mathrm{M}}_{\odot }$ [73], $12\pm 2{\mathrm{M}}_{\odot }$ [74], and $17\pm 3{\mathrm{M}}_{\odot }$ [75]. Analysis of the progenitor SED showed no detection of a binary companion, but given the non-detection limits in optical filters, only secondary stars with masses $>6.4{\mathrm{M}}_{\odot }$ could be ruled out. Intriguingly, NIR imaging with Spitzer showed significant variability with an estimated pulsation period of $\sim 1000$ days [67,68,70]. Given the strong CSM interaction during the early evolution of SN 2023ixf, comprehensive searches were performed to look for precursor events, which have been detected prior to some CSM-interacting SNe (e.g., [76,77]). However, no optical ground-based survey detected any precursor emission for SN 2023ixf, which effectively ruled out any outbursts or eruptions from the RSG progenitor star in the final years before explosion [53,75,78,79,80,81].

7.2. Light Curve and Spectral Modeling

Progenitor mass estimates were also made through modeling of the light curve and nebular spectra of SN 2023ixf. Using hydrodynamical modeling of the bolometric light curve during and after the plateau, ref. [82] favors a $M_{\mathrm{ZAMS}}=12{\mathrm{M}}_{\odot }$ progenitor with a radius of $720{\mathrm{R}}_{\odot }$ that explodes with a kinetic energy of 1.2 B. Similarly, refs. [34,83] prefer the explosion of a $M_{\mathrm{ZAMS}}=10{\mathrm{M}}_{\odot }$ progenitor. Alternatively, refs. [48,49] favor a larger progenitor ($M_{\mathrm{ZAMS}}=17{\mathrm{M}}_{\odot }$) with an inflated radius that had lost significant mass throughout its lifetime, leading to a depleted H envelope mass. In addition to light curve modeling, comparison with spectral models from [84,85] at nebular phases was also used to estimate the ZAMS mass of the SN 2023ixf RSG progenitor star. However, model matching to nebular spectra also yielded a variety of preferred progenitor masses: <12 ${\mathrm{M}}_{\odot }$ [41], 12–15 ${\mathrm{M}}_{\odot }$ [86], <15 ${\mathrm{M}}_{\odot }$ [43], 15.2–16.3 ${\mathrm{M}}_{\odot }$ [40], 15–19 ${\mathrm{M}}_{\odot }$ [50], 10–15 ${\mathrm{M}}_{\odot }$ [42], and 12.5–15 ${\mathrm{M}}_{\odot }$ [87].

7.3. CSM Structure and Mass Loss Mechanisms

Multi-wavelength observations of SN 2023ixf throughout its evolution enabled separate estimates on the circumstellar density profile and the mass loss rate of the RSG progenitor in the final years to decades prior to explosion. The presence of transient, high-ionization emission lines with electron-scattering wings in the early-time SN 2023ixf spectra necessitates CSM optical depths of $\tau \approx$ 3–10, which corresponds to a CSM density of ∼${10}^{-12}$ g cm⁻³ at ${10}^{14}$ cm. Furthermore, applying an IIn-like feature timescale of ∼7 days and a shock radius of ${10}^4\mathrm{km}{\mathrm{s}}^{-1}$ requires that the transition region from optically thick to thin pre-shock CSM occurs at ∼$6\times {10}^{14}$ cm; this corresponds to a lookback time of ∼8 years before explosion for a wind velocity of $25\mathrm{km}{\mathrm{s}}^{-1}$. As shown in Figure 2, the early-time spectral evolution is well matched by CMFGEN model spectra generated for RSG explosion interacting with an enhanced mass loss rate of ${10}^{-2}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$($v_w=50\mathrm{km}{\mathrm{s}}^{-1}$), confined to a radius of <${10}^{15}$ cm, with a total CSM mass in the range of $0.04–0.07{\mathrm{M}}_{\odot }$. Notably, the spectral models can match all optical emission line species present in SN 2023ixf without CNO enrichment in the CSM. Additionally, ref. [26] models the spectropolarimetry observations of SN 2023ixf using 2D polarized CMFGEN models to find that the early-time polarization can be explained by confined CSM with a pole-to-equator density contrast of ∼3.

Similar to spectral models, light curve modeling of SN 2023ixf also required confined, high-density CSM to explain the fast rise to a luminous peak brightness. For example, ref. [88] uses hydrodynamical modeling of the bolometric light curve to derive a best-fit mass loss rate of $3\times {10}^{-3}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$, confined to $<8\times {10}^{14}$ cm, and with a wind acceleration parameter of $\beta =5$. Modeling of the multi-band light curve by [34,83] suggests a multi-component density profile with higher mass loss $\dot{M}={10}^{-2}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$ within $R<5\times {10}^{14}$ cm and lower mass loss of ${10}^{-4}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$ at $R>5\times {10}^{14}$ cm. A similarly large mass loss rate of $5\times {10}^{-2}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$ within $R_{\mathrm{CSM}}\approx 5\times {10}^{14}$ cm is found from a hybrid shock cooling plus CSM-interaction analytic model employed by [6]. A similar combined shock cooling and CSM-interaction model was employed in [89], which used a Monte Carlo method to simulate radiative diffusion in order to match the earliest photometry of SN 2023ixf with $\dot{M}={10}^{-2}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$ and $R_{\mathrm{CSM}}={10}^{15}$ cm. Furthermore, model light curves associated with the best-matched CMFGEN models discussed above can reproduce the light curve peak across filters but likely require additional CSM mass directly above the stellar surface to reproduce the fast rise time observed in SN 2023ixf. Additionally, some light curve models require a significant amount of CSM; e.g., [90] finds $\dot{M}=0.1–1$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$ for a continuous mass loss scenario and $M_{\mathrm{CSM}}=0.3–1{\mathrm{M}}_{\odot }$ for an eruption scenario. Similarly, radiative transfer light curve models in [91] require $0.5–0.9{\mathrm{M}}_{\odot }$ of confined CSM to match the SN 2023ixf light curve. However, as shown by studies such as [30], IIn-like spectral features cannot form from SN ejecta interaction with such a large CSM mass within a confined radius.

Beyond UV/optical observations, modeling of the multi-epoch X-ray spectra was used to estimate a progenitor mass loss rate of ∼$3\times {10}^{-4}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$ beginning at shock radii of ∼$3\times {10}^{14}$ cm [52,53,54]. Furthermore, radio SED fitting confirmed a similar mass loss rate that remained consistent with a wind-like density profile out to radii >${10}^{16}$ cm [54]. Overall, both X-ray and multi-band radio observations modeled in [54] are in agreement with the mass loss rates inferred from the modeling of the progenitor star SED [67,68,70,72]. Intriguingly, the earliest high-frequency radio observations presented in [55] are inconsistent with the CSM densities inferred from the earliest X-ray detections. Similarly, the presence of IIn-like features in optical spectroscopy necessitates a larger CSM density at the same epoch than is inferred from the X-rays. This inconsistency is potentially reconciled through CSM asymmetries such as a clumpy progenitor wind. The complete CSM density profile constructed from all multi-wavelength observations is shown in Figure 7, and all CSM properties presented in the literature are summarized in

Table 1. Estimates from the literature for SN 2023ixf progenitor mass, mass loss rate, CSM radius, and CSM velocity.

Ref.	Method	${\mathit{M}}_{\mathbf{ZAMS}}$	$\dot{\mathit{M}}$	${\mathit{R}}_{\mathbf{CSM}}$	${\mathit{v}}_{\mathbf{CSM}}$
		(${\mathbf{M}}_{\odot }$)	(${\mathbf{M}}_{\odot }$ ${\mathbf{yr}}^{-\mathbf{1}}$)	(cm)	($\mathbf{km}{\mathbf{s}}^{-\mathbf{1}}$)
[67]	Pre-SN imaging	$11\pm 1$	$(1.3\pm 0.1)\times {10}^{-6}$	$(5.8\pm 0.6)\times {10}^{14}$	50
[68]	Pre-SN imaging	$17\pm 4$	$(0.3-3)\times {10}^{-4}$	>4 $\times {10}^{14}$	10
[69]	Pre-SN imaging	$17\pm 2$	2 $\times {10}^{-4}$	>2 $\times {10}^{15}$	115
[68]	Pre-SN imaging	$17\pm 4$	$(0.3-3)\times {10}^{-4}$	>4 $\times {10}^{14}$	10
[70]	Pre-SN imaging	$20\pm 4$	$(2-4)\times {10}^{-4}$	–	–
[71]	Pre-SN imaging	$13\pm 1$	–	–	–
[72]	Pre-SN imaging	$18\pm 1$	$(3.3\pm 0.26)\times {10}^{-4}$	>1015	50
[73]	Pre-SN imaging	8–12	–	–	–
[74]	Pre-SN imaging	$12\pm 2$	$(6-9)\times {10}^{-6}$	$(1.7-8.1)\times {10}^{15}$	70
[75]	Pre-SN imaging	$17\pm 3$	–	–	–
[78]	Pre-SN imaging	9–14	${10}^{-5}$	$5\times {10}^{14}$	10
[82]	LC model	12	–	–	–
[88]	LC model	12	$3\times {10}^{-3}$	$8\times {10}^{14}$	115
[83]	LC model	10	${10}^{-3}-{10}^{-2}$	$(0.6-1)\times {10}^{15}$	10
[34]	LC model	10	${10}^{-2}$	<5$\times {10}^{14}$	10
[34]	LC model	10	${10}^{-4}$	>5$\times {10}^{14}$	10
[48]	LC model	>17	${10}^{-1}-{10}^0$	$(0.4-1)\times {10}^{14}$	115
[90]	LC model	–	${10}^{-1}-{10}^0$	$(3-7)\times {10}^{15}$	115
[10]	LC model	–	${10}^{-3}$	$7\times {10}^{14}$	10
[49]	LC model	>17	–	–	–
[6]	LC model	–	$5\times {10}^{-2}$	$5\times {10}^{14}$	100
[89]	LC model	–	${10}^{-2}$	${10}^{15}$	75
[29]	LC Model	–	${10}^{-1.96}$	$2\times {10}^{14}$	30
[65]	LC Model	${12}_{-1}^{+2}$	–	$(3.6\pm 0.5)\times {10}^{15}$	55
[8]	Spectra/LC Model	–	${10}^{-2}$	$(0.5-1)\times {10}^{15}$	50
[9]	Spectra Model	–	${10}^{-3}-{10}^{-2}$	$5\times {10}^{14}$	50
[38]	Spectra Model	–	${10}^{-3}-{10}^{-5}$	$(1-6)\times {10}^{15}$	50
[11]	Spectra Model	–	$6\times {10}^{-4}$	$7\times {10}^{14}$	55
[26]	Spectra Model	–	${10}^{-2}$	$8\times {10}^{14}$	50
[21]	Spectra Model	–	${10}^{-2}$	$8\times {10}^{14}$	25
[51]	X-ray Model	–	$3\times {10}^{-3}$	<1015	50
[52]	X-ray Model	–	$5\times {10}^{-4}$	$(0.6-4)\times {10}^{15}$	115
[53]	X-ray Model	–	<5 $\times {10}^{-4}$	<4 $\times {10}^{15}$	50
[54]	X-ray/Radio Model	–	${10}^{-4}$	$(0.04-2)\times {10}^{16}$	25
[58]	Radio Model	–	${10}^{-6}-{10}^{-4}$	$(0.2-1)\times {10}^{16}$	115
[55]	Radio Model	–	>10−2	<1015	115

.

8. Conclusions

In this article, we have reviewed the multi-wavelength observations and modeling of type II SN 2023ixf, located at ∼6.85 Mpc in Messier 101. Pre-explosion imaging confirmed that the progenitor star was a dust-enshrouded red supergiant, but the exact ZAMS mass remains uncertain. Early-time spectroscopy revealed narrow, transient, high-ionization emission lines that result from SN ejecta interaction with dense, confined CSM that was created by the red supergiant in the final ∼3–6 years before explosion. The most local CSM at <5 $\times {10}^{14}$ cm is high density and could be described by a mass loss rate as large as ∼${10}^{-2}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$. At larger radii, all multi-wavelength (X-ray through radio) observations appear to confirm a similar wind-like density profile with a mass loss rate of ∼${10}^{-4}$ ${\mathrm{M}}_{\odot }$ ${\mathrm{yr}}^{-1}$. Because of the lack of detected outbursts in pre-explosion imaging, mechanisms capable of producing such high-density CSM in the final years before explosion could include a convection-driven chromosphere [92,93] and/or binary interaction. Today, SN 2023ixf remains sufficiently bright for detection by ground-based instruments and has plateaued in luminosity as shock power dominates over radioactive decay. The unprecedented multi-wavelength dataset of SN 2023ixf has solidified this event as the prototype for CSM-interacting SNe II, which will continue to be studied for many years to come.