F8D1: Tidal Ultra-Diffuse Galaxy Case

• F8D1 is a low-surface-brightness ultra-diffuse galaxy in the M81 Group noted for its extended stellar stream and significant tidal disruption.
• Observations using double-Sérsic models reveal a complex stellar structure with substantial mass in tidal tails and evidence of late star-formation bursts.
• Spectroscopic analyses reclassify compact clusters from old globulars to young star clusters, emphasizing F8D1’s dynamic evolutionary history.

F8D1 is the designation of a nearby ultra-diffuse galaxy in the M81 Group that has become a reference case for tidal processing in low-mass systems. It is a very low-surface-brightness dwarf with a large effective radius, an extended stellar stream, and a recent star-formation history marked by late bursts followed by quiescence. Spectroscopy of a compact cluster projected on the galaxy shows that the observed object is a young star cluster rather than a classical old globular cluster, refining the interpretation of its cluster system. In a separate signal-processing literature, “F8D1” also denotes an algebraic-integer-based exact 8×8 forward 2-D discrete cosine transform architecture; that usage is unrelated to the galaxy (Žemaitis et al., 2022, Smercina et al., 17 Jul 2025, Forbes et al., 2024, Madanayake et al., 2015).

1. Astronomical identification and environment

F8D1 is a low-luminosity, very low-surface-brightness dwarf in the M81 Group and satisfies canonical ultra-diffuse galaxy criteria, with effective radius $R_e \ge 1.5\ \mathrm{kpc}$ and central surface brightness $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$ (Žemaitis et al., 2022). It was discovered by Caldwell et al. (1998) and lies in a region heavily contaminated by Galactic cirrus, a circumstance that long impeded integrated-light studies and motivated a resolved-star approach.

Its J2000 coordinates are $\mathrm{RA}=09{:}44{:}45.95$, $\mathrm{Dec}=+67{:}26{:}27.7$ (Žemaitis et al., 2022). A Subaru/HSC tip-of-the-red-giant-branch analysis gave $(m-M)_0 = 27.82 \pm 0.03$ and $D = 3.67 \pm 0.06\ \mathrm{Mpc}$, while later HST work adopted $D \approx 3.8\ \mathrm{Mpc}$ and $\mu \approx 27.9$ (Žemaitis et al., 2022, Smercina et al., 17 Jul 2025). In projection, F8D1 lies $1.91^\circ$ from M81 and $0.53^\circ$ from NGC 2976; the 2025 HST study describes it as $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$0 southwest of M81 and $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$1 from NGC 2976 in projection (Žemaitis et al., 2022, Smercina et al., 17 Jul 2025). The same study reports line-of-sight velocities of $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$2, $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$3, and $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$4 (Smercina et al., 17 Jul 2025).

The environmental context is central to its interpretation. Resolved-star maps and HST follow-up show that F8D1 is not an isolated diffuse dwarf but a tidally disrupting M81-group satellite with an extensive stream. This places it among the nearest and best-resolved examples of a UDG undergoing active transformation.

2. Structure, surface brightness, and stellar mass

CFHT/MegaCam surface photometry found that the main body of F8D1 is well fit by a low-$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$5 Sérsic profile,

$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$6

with $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$7 and $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$8 (Žemaitis et al., 2022). Best-fitting parameters are $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$9–$\mathrm{RA}=09{:}44{:}45.95$0 in $\mathrm{RA}=09{:}44{:}45.95$1, central surface brightnesses $\mathrm{RA}=09{:}44{:}45.95$2, $\mathrm{RA}=09{:}44{:}45.95$3, and $\mathrm{RA}=09{:}44{:}45.95$4, and effective radii $\mathrm{RA}=09{:}44{:}45.95$5 $\mathrm{RA}=09{:}44{:}45.95$6, $\mathrm{RA}=09{:}44{:}45.95$7 $\mathrm{RA}=09{:}44{:}45.95$8, and $\mathrm{RA}=09{:}44{:}45.95$9 $\mathrm{Dec}=+67{:}26{:}27.7$0 (Žemaitis et al., 2022). The galaxy is only mildly flattened, with ellipticity $\mathrm{Dec}=+67{:}26{:}27.7$1 and position angle $\mathrm{Dec}=+67{:}26{:}27.7$2.

Its mean colors within $\mathrm{Dec}=+67{:}26{:}27.7$3 are $\mathrm{Dec}=+67{:}26{:}27.7$4 and $\mathrm{Dec}=+67{:}26{:}27.7$5, with only weak color gradients (Žemaitis et al., 2022). The same work reported $\mathrm{Dec}=+67{:}26{:}27.7$6 from direct summation to $\mathrm{Dec}=+67{:}26{:}27.7$7 and $\mathrm{Dec}=+67{:}26{:}27.7$8 from Sérsic extrapolation, together with a main-body stellar mass of $\mathrm{Dec}=+67{:}26{:}27.7$9 (Žemaitis et al., 2022).

A later HST star-count analysis recast the structure in terms of a double Sérsic model for the stellar mass surface density, separating a bound inner spheroid from an outer stream (Smercina et al., 17 Jul 2025). The inner component has $(m-M)_0 = 27.82 \pm 0.03$0 and $(m-M)_0 = 27.82 \pm 0.03$1, while the outer stream component has $(m-M)_0 = 27.82 \pm 0.03$2 and $(m-M)_0 = 27.82 \pm 0.03$3 (Smercina et al., 17 Jul 2025). Under the assumption $(m-M)_0 = 27.82 \pm 0.03$4, the equivalent central surface brightness for the bound spheroid is $(m-M)_0 = 27.82 \pm 0.03$5, with $(m-M)_0 = 27.82 \pm 0.03$6 and $(m-M)_0 = 27.82 \pm 0.03$7 (Smercina et al., 17 Jul 2025). The same model yields $(m-M)_0 = 27.82 \pm 0.03$8 and $(m-M)_0 = 27.82 \pm 0.03$9, as well as $D = 3.67 \pm 0.06\ \mathrm{Mpc}$0 and a total progenitor stellar mass of $D = 3.67 \pm 0.06\ \mathrm{Mpc}$1 when the mapped northern stream and a possible southern counterpart are included (Smercina et al., 17 Jul 2025).

Taken together, these measurements place F8D1 securely in the UDG regime while showing that its present morphology cannot be understood from the inner body alone. The outer stellar debris is a structural component of comparable interpretive importance.

3. Stellar populations, metallicity, and star-formation history

Resolved-star color-magnitude diagrams show that F8D1 is presently quiescent but not exclusively ancient. Subaru/HSC data showed a prominent red giant branch and a substantial luminous asymptotic giant branch population above the TRGB, with no young main-sequence population (Žemaitis et al., 2022). HST imaging later reached below the Red Clump in a central field and in a parallel field along the major axis, allowing synthetic CMD reconstruction of the star-formation history back to $D = 3.67 \pm 0.06\ \mathrm{Mpc}$2 (Smercina et al., 17 Jul 2025).

The HST analysis finds that F8D1 formed $D = 3.67 \pm 0.06\ \mathrm{Mpc}$3 of its stars more than $D = 3.67 \pm 0.06\ \mathrm{Mpc}$4 ago, albeit with large uncertainties at the oldest ages (Smercina et al., 17 Jul 2025). Two younger episodes are well constrained. A substantial burst occurred $D = 3.67 \pm 0.06\ \mathrm{Mpc}$5–$D = 3.67 \pm 0.06\ \mathrm{Mpc}$6 ago: in the central field it was $D = 3.67 \pm 0.06\ \mathrm{Mpc}$7 the lifetime-average star-formation rate and formed $D = 3.67 \pm 0.06\ \mathrm{Mpc}$8 of the present-day stellar mass there, while in the outer inner-stream field it was $D = 3.67 \pm 0.06\ \mathrm{Mpc}$9 the lifetime-average rate and formed $D \approx 3.8\ \mathrm{Mpc}$0 of the local stellar mass (Smercina et al., 17 Jul 2025). A smaller burst followed at $D \approx 3.8\ \mathrm{Mpc}$1, at $D \approx 3.8\ \mathrm{Mpc}$2 the lifetime-average star-formation rate in both fields; in the central field this episode formed $D \approx 3.8\ \mathrm{Mpc}$3, of which $D \approx 3.8\ \mathrm{Mpc}$4 remains after stellar mass loss (Smercina et al., 17 Jul 2025). Cumulative SFHs imply $D \approx 3.8\ \mathrm{Mpc}$5 in the central field and $D \approx 3.8\ \mathrm{Mpc}$6 in the inner-stream field, with outer flanking fields consistent with global quenching $D \approx 3.8\ \mathrm{Mpc}$7–$D \approx 3.8\ \mathrm{Mpc}$8 ago (Smercina et al., 17 Jul 2025).

Metallicity estimates depend on method and spatial coverage. Photometric metallicity distribution functions from Subaru/HSC, based on 10 Gyr PARSEC isochrones, gave a peak $D \approx 3.8\ \mathrm{Mpc}$9 dex within $\mu \approx 27.9$0, with $\mu \approx 27.9$1 dex and no obvious metallicity gradient along the stream (Žemaitis et al., 2022). The HST CMD-based modeling instead found lifetime-average metallicities of $\mu \approx 27.9$2 in the central field and $\mu \approx 27.9$3 in the outer field, while stars formed in the last $\mu \approx 27.9$4 have $\mu \approx 27.9$5 centrally and $\mu \approx 27.9$6 in the outer field (Smercina et al., 17 Jul 2025). The radial metallicity gradient is described as mild.

These results establish F8D1 as a galaxy with a predominantly old stellar mass but a significant late-time evolutionary phase. The outer field being younger than the central spheroid demonstrates a real age gradient between bound body and stream.

4. Tidal debris and interaction history

The defining observational feature of F8D1 is its giant tidal stream. Subaru/HSC red-giant-branch star-count maps revealed an S-shaped stellar tail extending for more than a degree on the sky, corresponding to $\mu \approx 27.9$7 or about $\mu \approx 27.9$8 at $\mu \approx 27.9$9 (Žemaitis et al., 2022). Its average surface brightness is $1.91^\circ$0, well below practical integrated-light limits in a cirrus-dominated field (Žemaitis et al., 2022). The tail curvature is measurable: its locus lies $1.91^\circ$1 west of the inner body at $1.91^\circ$2–$1.91^\circ$3 and $1.91^\circ$4 east at $1.91^\circ$5–$1.91^\circ$6, giving a textbook S-shaped morphology (Žemaitis et al., 2022). The longitudinal density profile is fit by a Sérsic model with $1.91^\circ$7 and $1.91^\circ$8, and the outermost points follow an $1.91^\circ$9 decline (Žemaitis et al., 2022).

The fraction of light or stellar mass in the debris is large. From RGB flux summation and an assumed symmetric counter-tail, the visible northeast tail implies a total tail fraction of $0.53^\circ$0–$0.53^\circ$1 of the present-day luminosity (Žemaitis et al., 2022). The later HST analysis, based on the double-Sérsic decomposition, finds that $0.53^\circ$2–$0.53^\circ$3 of F8D1’s present-day stellar mass is in the tails, depending on how an unmapped southern counterpart is treated (Smercina et al., 17 Jul 2025).

The interaction history remains under active interpretation. The 2022 study argued that the most likely origin of the disruption is a recent close passage to M81, citing M81’s much larger halo mass, the $0.53^\circ$4 separation, the $0.53^\circ$5 3D separation from NGC 2976, and the lack of disturbance in NGC 2976’s outer stellar isophotes (Žemaitis et al., 2022). The 2025 study also treats tidal processing as dominant but allows that the most recent pericenter could plausibly have involved either NGC 2976 or M81, noting that traversing $0.53^\circ$6 in $0.53^\circ$7 requires $0.53^\circ$8, whereas $0.53^\circ$9 over the same interval requires $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$00 (Smercina et al., 17 Jul 2025). AGB/RGB ratios along the inner stream imply that F8D1 was globally star-forming until $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$01 ago, and that the inner portion of the stream contains stars formed as recently as $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$02 ago, constraining the most recent pericenter to $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$03 (Smercina et al., 17 Jul 2025).

The combination of extreme stream length, large stripped fraction, and recent star formation in stream material makes F8D1 a direct observational link between dwarf irregular progenitors and strongly disrupted satellite remnants.

5. Compact clusters and globular-cluster status

The status of compact clusters in F8D1 changed substantially with spectroscopy. Historically, one compact stellar system projected on F8D1 had been referred to as a “GC,” with a very uncertain radial velocity from Chiboucas et al. (2009), $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$04 (Forbes et al., 2024). The NASA/IPAC Extragalactic Database lists $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$05 for F8D1, and the 2024 study suggests that this may reflect the earlier cluster measurement rather than the galaxy’s systemic velocity (Forbes et al., 2024).

Keck/KCWI spectroscopy observed a single F8D1 compact cluster candidate, designated GC1, at $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$06, $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$07 (Forbes et al., 2024). The observation was obtained on 2021-01-13 with KCWI on Keck II, using the Medium slicer and BL grating centered at $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$08 with spectral resolution $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$09 and an exposure time of $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$10 (Forbes et al., 2024). Reduction used the standard KCWI pipeline following Gannon et al. (2020), including trimming, standard-star calibration, background subtraction through a surrounding annulus, 1D extraction, barycentric correction, and median combination; the F8D1 spectrum had low S/N, $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$11 per $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$12 (Forbes et al., 2024). Full spectral fitting employed pPXF with the Coelho (2014) synthetic stellar library across 256 input parameter combinations, with simultaneous modeling of emission lines when present (Forbes et al., 2024).

The object is associated with F8D1 but is not an old globular cluster. Its measured radial velocity is $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$13, its age is $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$14, and its metallicity is $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$15 (Forbes et al., 2024). Strong Balmer absorption lines in the blue KCWI spectrum indicate an A-type stellar population and corroborate the young age (Forbes et al., 2024). On that basis, the object was reclassified from a “known GC” to a young star cluster. Given that only this single compact stellar system was observed, F8D1 currently has zero confirmed classical old globular clusters in that study and one young star cluster; the specific frequency of old GCs is therefore effectively

$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$16

pending future confirmations (Forbes et al., 2024).

The later HST study reports that F8D1 contains a compact nuclear star cluster with spectroscopic age $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$17 and inferred stellar mass $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$18, and states that the $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$19 burst likely formed or rejuvenated the NSC (Smercina et al., 17 Jul 2025). This places the compact cluster population within the broader late-time star-formation history rather than within a rich ancient globular-cluster system.

6. Formation scenarios, analogs, and unresolved issues

The recent literature converges on tidal processing as the dominant driver of F8D1’s present state, but it does not reduce the system to tides alone. The 2025 analysis argues that the combination of a double-Sérsic structure, $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$20–$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$21 of the present-day stellar mass in tails, and late bursts at $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$22 and $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$23 is most naturally explained if F8D1’s dark-matter halo had previously been cored by bursty star formation, making the galaxy susceptible to tidal expansion (Smercina et al., 17 Jul 2025). Under that interpretation, $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$24–$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$25 stellar mass loss in the tails implies $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$26–$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$27 total mass loss, and a cored progenitor would have experienced a factor $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$28–$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$29 increase in $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$30, suggesting a progenitor half-light radius of $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$31–$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$32 rather than the current $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$33 (Smercina et al., 17 Jul 2025).

The same study compares F8D1 to Local Group systems and finds that it is consistent with a progenitor star-forming galaxy similar to NGC 6822 that is in the midst of a transition to a Sagittarius-like system (Smercina et al., 17 Jul 2025). The comparison is explicit: F8D1 resembles an intermediate stage in a sequence from star-forming dwarf irregular, through tidally expanded UDG undergoing disruption, to a more mature stream system akin to Sagittarius (Smercina et al., 17 Jul 2025). It is less disrupted than Sagittarius, with $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$34–$\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$35 of its mass in tails rather than $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$36, and it retains a younger star-forming stream segment (Smercina et al., 17 Jul 2025).

Alternative pathways are discussed but given less weight. High-angular-momentum or low-star-formation-efficiency scenarios may explain sustained gas reservoirs and late star formation, yet the pronounced late bursts and strong tidal features indicate that environment, rather than spin alone, shaped the current UDG appearance (Smercina et al., 17 Jul 2025). A “failed galaxy” channel is argued against because F8D1 shows late-time star formation and a seemingly typical globular-cluster population, with visual inspection yielding $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$37 candidates rather than an extraordinary GC richness (Smercina et al., 17 Jul 2025).

Several uncertainties remain material. The southern stream is unmapped, so total stripped mass carries an upper uncertainty that allows for a symmetric southern counterpart (Smercina et al., 17 Jul 2025). No internal velocity dispersion has yet been reported, so there is no direct dynamical mass and no direct core–cusp inference (Smercina et al., 17 Jul 2025). Older epochs of the SFH remain degenerate without access to the oldest main-sequence turnoff, and TP-AGB modeling contributes systematic uncertainty (Smercina et al., 17 Jul 2025). For the cluster system, only one compact stellar system has been spectroscopically observed, and no photometric structural parameters such as $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$38 are reported for that object (Forbes et al., 2024). A plausible implication is that F8D1 is now one of the clearest nearby laboratories for testing how tidal stripping, late star formation, and internal feedback jointly produce UDG-like structure.

7. Signal-processing usage of the name

In an unrelated engineering context, “F8D1” denotes a proposed 8×8 forward 2-D discrete cosine transform architecture based on algebraic-integer exact computation (Madanayake et al., 2015). The design performs the entire separable 2-D DCT in an AI basis without any intermediate final reconstruction step between column and row transforms, so all internal quantities remain exact in AI basis until the end (Madanayake et al., 2015). It uses a time-multiplexed row-parallel organization consisting of an input decimator and SerDes, one 8-point 1-D AI Arai column core, an AI transpose buffer, four parallel row cores, and sixty-four per-coefficient final-reconstruction pipelines (Madanayake et al., 2015).

The architecture is multiplierless in the usual FPGA sense: no general-purpose multipliers are used, and the final reconstruction relies on add–shift constant-multiplier networks or an expansion-factor scheme (Madanayake et al., 2015). Its stated advantage is that quantization noise is injected only once, in the final reconstruction step, and is independently tunable for each coefficient channel rather than leaking across channels as in conventional fixed-point 2-D DCT pipelines (Madanayake et al., 2015). On a Xilinx Virtex-6 XC6VLX240T FPGA, the best 8-bit implementation achieved $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$39, an 8×8 block rate of $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$40, an equivalent pixel rate of $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$41, and a frame rate of $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$42 for $\mu_g(0) \gtrsim 24\ \mathrm{mag\ arcsec^{-2}}$43 images (Madanayake et al., 2015).

This engineering usage is purely nominal and should not be conflated with the astronomical object of the same designation.