Little red dot variability over a century reveals black hole envelope via a giant Einstein cross

Abstract

"Little red dots" (LRDs) represent a new population of astronomical objects uncovered by JWST whose nature remains debated. Although many LRDs are suspected as active galactic nuclei (AGN), they show little variability on days-years timescales. We report the discovery of two gravitationally lensed LRDs at redshift $\sim$4.3 behind the cluster RXCJ2211-0350, one of which (RX1) is quadruply imaged with time delays spanning $\sim$130 years. RX1 exhibits intrinsic color and brightness variations of up to 0.7 magnitude among its images. These changes are consistent with blackbody-temperature variations of a photosphere, indicating long-term variability analogous to Cepheid-like pulsations but in a far more extended ($R \sim 2000$ AU) and massive ($M \gtrsim 10^6 \, M_{\odot}$) systems. These results suggest LRDs as a distinct class of AGN with stellar-like envelopes.

Figures (13)

• Figure 1: JWST NIRCam color composite image of the galaxy cluster RXC J2211--0350. Thanks to strong gravitational lensing, two multiply imaged LRDs, dubbed RX1 and RX2, are identified, both with photometric redshifts of $z=4.3$. The four images of RX1 are marked by red circles (1.1-1.4), and the five images of RX2 are marked by orange hexagons (2.1-2.5). The left (red) and right (orange) panels show zoom-in cutouts for each of these lensed images. In each cutout, the image name is given in the top left, the magnification ($\mu$; negative value indicates that the image parity is reversed and the image orientation is flipped) calculated from our cluster mass model is in the top right, the measured F444W-band magnitude (not corrected for magnification) is in the bottom left, and the inferred time delay ($\Delta t$ in days) is in the bottom right. The uncertainties of $\mu$ and $\Delta t$ are typically $\lesssim$ 5%.
• Figure 2: Brightness and color variability of R2211-RX1.(A) De-magnified light curves of R2211-RX1 in each JWST band, incorporating the gravitational lens time delays $\Delta t_{\rm grav}$ given in Table \ref{['tab:lens_param']}. $t_0$ is the VENUS observation date. Dotted lines show the mean magnitude of each band, and the error bars include magnification uncertainties. We show the maximal $\Delta m$ and $\chi_{\nu}^2$ values assuming no variability for each band. (B) Multi-band color evolution of R2211-RX1, with x-axis the same as (A). The y-axis shows the color difference relative to image R2211-RX1.2. (C) S/N distribution of brightness and color (F300M$-$F444W) variability between all epochs pairs for R2211-RX1 and the other 12 multiply imaged sources with $z_\mathrm{phot}$$>3$. Sources with $z_\mathrm{phot}$$\sim 4.3$ in Figure \ref{['fig:sourceplane_pos']} are marked with blue squares. For each epoch pair, the brightness variability S/N is measured as the median S/N of $\Delta \rm \,mag$ across the F200W and redder bands to reflect the overall variability of the SED. In the top/right panels, red lines indicate R2211-RX1, and the gray histogram represents the other sources. The two R2211-RX1 pairs with the shortest observed-frame intervals (10 and 36 yr) have S/N $\sim1$ and are shown as open stars, while all others have S/N$>2$. The S/N of the other sources are broadly consistent with a one-sided normal distribution and mostly below 1, suggesting that the measured variability of R2211-RX1 is not biased by systematic errors in magnification.
• Figure 3: SED modeling, light curve modeling, and $T_{\rm eff}$-$L_{\rm BB}$ evolution of R2211-RX1.(A) Intrinsic SEDs and best-fit models for the four multiple images of R2211-RX1. Filled/open circles mark fitted/excluded (potentially affected by strong emission lines) bands. The dashed colored lines show the best-fit single-temperature blackbody + shared power law (black dotted line). (B)$T_{\rm eff}$ versus $L_{\rm BB}$ for the four images of R2211-RX1. The solid gray line connects the four images, ordered by their relative time delays. The dashed black line indicates the best-fit power-law scaling $L_{\rm BB} \propto T_{\rm eff}^{4.9\pm1.0}$. Colored dots denote the evolutionary sequence derived from the light curve fitting in (C), showing a counterclockwise progression in chronological order. We also show the $T_{\rm eff}$-$L_{\rm BB}$ circle of a Cepheid with mean $T_{\rm eff}$ and $L_{\rm BB}$ scaled and color matched to the phase of the LRD circle in the bottom right. Dashed gray curves denote loci of constant blackbody radius. (C) Light curve (delensed flux in nJy) fitting results of R2211-RX1. Observed light curves (gray points) and best models (orange curves) are shown, with best-fit amplitude, phase, and $\chi^2_{\nu}$ marked in each band. The total $\chi^2_{\nu}$ is indicated in the upper left of the F444W band panel. Vertical lines in the top panel mark the 40 epochs sampled at one-year intervals, with colors corresponding to those in (B).
• Figure 4: Conceptional models of LRD dense gas envelopes under two scenarios.(A) Under the photospheric pulsation interpretation, we expect the dense H i gas to undergo similar pulsation as the partly ionized region, producing velocity shift of certain absorption lines among multiple images and periodicity in the light curve that can be obtained through long-term spectrophotometric monitoring. (B) Under the variable accretion state interpretation, we expect the variation of broad emission line strength correlated with the underlying continuum while no obvious velocity shift in absorption lines, and the resultant long-term light curve will appear stochastic. Expected outcomes from future spectroscopic (left) and photometric (right) monitoring are indicated.
• Figure S1: The SED and images of R2211-RX1 (A) and R2211-RX2 (B). The left panels show the stacked observed photometry and $2\sigma$ upper limits from VENUS JWST bands and RELICS HST F435W/F606W/F814W data (dark red diamonds). The best-match LRD template is overplotted in purple, and the best-fit redshifts ($z_{\rm map}$) with their 5th-95th percentile ranges from EAZY, CIGALE, and Prospector are indicated. The photometry from individual images (corrected for magnification $\mu$) is shown as lighter/semi-transparent markers in the background. The inset panels display the full redshift probability distributions $p(z)$ from the same three codes. The right panels show 3$^{\prime\prime}$$\times$ 3$^{\prime\prime}$ cutouts in the available JWST bands of R2211-RX1.4 and R2211-RX2.5, as well as the stacked HST F435W/F606W/F814W images. We show these two images because they have high magnifications and no nearby contamination. For R2211-RX1, the HST stack combines all multiple lensed images of the same source, while for R2211-RX2, the stack includes only images R2211-RX2.3, R2211-RX2.4, and R2211-RX2.5. The HST stacked images are smoothed with a Gaussian kernel ($\sigma=1$ pixel).