Pre-perihelion Emergence of the CN Gas Coma in 3I/ATLAS Temporally and Spatially Resolved by the 7-Dimensional Telescope Astropy ( 4 ; 2 ; 3 ) , SciPy ( 56 ) , Source Extractor ( 8 ) SWarp ( 10 ) , SCAMP ( 9 ) , PSFEx ( 11 ) , GALFITM ( 29 ) , Gregory S. H. Paek, Myungshin Im, Mankeun Jeong, Hyeonho Choi, Yoonsoo P. Bach, Masateru Ishiguro, Bumhoo Lim, Seo-Won Chang, Ji Hoon Kim, Jooyeon Geem, Willem B. Hoogendam Abstract We present time-series medium-band ( R ∼ 20 R\sim 20 –40) observations of the third interstellar object 3I/ATLAS (C/2025 N1) obtained with the 7-Dimensional Telescope (7DT) which enables spatially resolved monitoring of its gas and dust activity from 2025 July to September. The m400-band image ( λ c = 400 \lambda_{c}=400 nm, Δ ​ λ ≈ 25 \Delta\lambda\approx 25 nm) reveals an emergence of a pronounced and spatially extended CN emission at a heliocentric distance of r h < 3 r_{h}<3 au. This onset is consistently identified across multiple diagnostics, including a break in the light-curve evolution, excess reflectance, inward expansion of annular excess beyond 10,000 10{,}000 – 20,000 20{,}000 km, growth of the coma half-light radius from ∼ 11,000 \sim 11{,}000 to ∼ 19,000 \sim 19{,}000 km, and a rapid rise in the CN production rate Q CN Q_{\rm CN} relative to A ​ f ​ ρ Af\rho . We further separate the CN-emitting and dust-scattered components via two-dimensional surface-brightness fitting into inner (dust) and outer (gas) components. We find that the outer component preserves a nearly constant profile shape, varying only in normalization, implying a relatively fast expansion of CN-bearing molecules. Together, these results reveal how a transition occurs in the optical from dust-dominated scattering at large heliocentric distances to volatile-driven, gas-dominated activity as 3I/ATLAS enters the inner Solar System. The timing and characteristics of the CN activation resemble the volatile enhancement observed in 2I/Borisov, suggesting that both known active interstellar objects exhibit comparable activation behavior at heliocentric distances of ∼ \sim 2–3 au. show]gregorypaek94@gmail.com show]myungshin.im@gmail.com I Introduction Interstellar objects (ISOs) offer a rare window into the planetesimal reservoirs of extrasolar planetary systems. Since the discoveries of 1I/’Oumuamua (1I; 41 ) and 2I/Borisov (2I; 27 ), the small ISO sample has already revealed striking diversity, from a coma-free object (1I) to a gas-rich comet (2I) with CN as the first and dominant optical volatile. The newly discovered 3I/ATLAS (C/2025 N1) on July 1, 2025, continues this diversity ( 54 ) . Rubin Observatory, ZTF, and NASA TESS precovery data revealed that 3I/ATLAS was already active at approximately 6.4 au as early as May 2025, with a dust-dominated coma and no detectable nucleus rotation, providing a crucial baseline for early activity prior to its discovery in July ( 15 ; 40 ; 23 ; 57 ) . The VLT/MUSE observations on 2025 July 3 (two days after discovery) showed a red, featureless continuum across 5000–9000 Å and no detected bands from CN, C 2 , NH 2 , or [O I], implying that the observed spectral reflectance at r_{h}\simeq 4.5 au is predominantly governed by scattering from the dust coma ( 46 ) . However, by mid-August, independent optical campaigns reported the onset of CN emission. A 10-night MDM program (Aug 8–17) captured the emergence and nightly strengthening of a weak CN feature near the 388.3 nm band head at r_{h}\simeq 3.2 – 2.9 au, deriving a CN production rate of \log Q_{\mathrm{CN}}\sim 23 (in molecules s -1 ) together with a stringent upper limit on the production-rate ratio of \mathrm{C}_{2} to \mathrm{CN} , \log[Q_{\mathrm{C}_{2}}/Q_{\mathrm{CN}}]<-1.05 , placing 3I among the class of carbon–chain–depleted comets ( 52 ) . Complementary spectrophotometric observations with UH 2.2-m/SNIFS and integral-field spectroscopy with Keck-II/KCWI subsequently detected CN and Ni i emission over multiple epochs between mid-August and early September, confirming a steady increase in CN production as the comet approached the Sun ( 31 ; 32 ) . Near-simultaneous VLT X-shooter/UVES spectra independently detected CN and multiple Ni i lines, likewise showing increasing production rates toward smaller heliocentric distances ( 50 ) . Given the non-detection in early July ( 1 ; 32 ) , these results indicate that CN outgassing switched on in mid-August. This timing coincides with the detection of a CO 2 -dominated coma: SPHEREx (Aug 12) detected strong CO 2 emission and water-ice absorption over a spatially extended coma ( 38 ) , and JWST/NIRSpec (Aug 18–19) revealed CO 2 with weak H 2 O and CO ( 18 ) . CN emission in comets originates primarily from photodissociation of parent molecules such as HCN released from the nucleus ( 24 ) . Observationally, CN is detected not only in the well-known violet system near 388 nm but also in a weaker red system at longer wavelengths near \sim 787 nm; these bands arise from distinct electronic transitions of the CN radical and together provide complementary diagnostic information on the spatial distribution and excitation conditions of CN in cometary comae (e.g., 24 ). However, spatially extended CN distributions observed in comets like 67P/Churyumov–Gerasimenko (67P) suggest an additional extended source, where CN parent molecules are preserved within dust grains and released via grain sublimation or fragmentation at distances >10^{4} km from the nucleus ( 45 ; 51 ) . This gas-dust interaction–where volatiles are stored in and released from dust particles–plays a critical role in shaping the spatial structure and temporal evolution of coma activity. Understanding the spatial structure of CN emission therefore provides key constraints on the mechanisms of volatile release and the coupling between gas and dust in cometary comae. Unlike 1I, which showed no detectable coma activity despite evidence for non-gravitational acceleration ( 42 ) , and 2I, for which CN emission was first detected only at relatively small heliocentric distances ( r_{h}\sim 2 au) during its inbound evolution ( 37 ) , 3I/ATLAS exhibits a distinct activity pattern characterized by the onset of CN-dominated emission at larger heliocentric distances and persistent dust scattering throughout its approach to the Sun. Beyond a simple census of dust and gas properties, it is essential to track how their temporal evolution as an ISO is progressively modified by the Solar-system environment, and to structurally resolve and deblend the coma and nucleus to isolate intrinsic activity from solar-system processing. However, previous studies of ISO have provided limited spatially resolved, time-series coverage. Therefore, time-series medium-band spectrophotometry with the 7-Dimensional Telescope (7DT; 36 ) offers a unique dataset to understand their nature. It provides simultaneous coverage of \sim 20–40 optical bands and a wide field of view, enabling low-resolution SEDs in what is effectively a “spectroscopic mode,” while spatially and temporally resolving the coma structure—an advantage over long-slit spectroscopy. Leveraging annular photometry, we report not only the temporal evolution of SED but also CN line-dominated regions in projection. We use 7DT observations of 3I/ATLAS from 2025 July 13 to September 18 ( r_{h}\simeq 4.11 – 2.02 au) to (1) measure temporal brightening slopes across wavelength, (2) identify wavelength-dependent excesses diagnostic of composition, and (3) map their radial distributions to trace the CN gas and dust emission. A pronounced m400-band enhancement, coinciding with the CN violet system near 388 nm (and potential C 3 near 405 nm), allows us to trace the origin and spatial distribution of this blue-end excess and its implications for volatile release in extrasolar planetesimals. II Observation We observed 3I/ATLAS with 7DT in spectroscopic mode over 14 epochs from 2025 July 13 to September 18 (UT). The system observes the target field with all 20 available medium-band filters and three broadband filters ( g , r and i ). The medium-band set consists of 20 filters with 25 nm-bandwidths, spaced by 25 nm in central wavelength from 400 to 875 nm; each filter is named by its central wavelength with the prefix “m” (e.g., m400, spanning 387.5–412.5 nm). Per epoch, we obtained 3–18 frames per filter with 100 s on-source integrations. 7DT was operated with the TCSpy program ( 16 ; 17 ) , which allows simultaneous control of the telescope units. Because the number of available medium-band filters exceeds the sixteen telescope units, a full filter set for a given epoch could not be obtained strictly simultaneously. Instead, subsets of filters were observed sequentially, resulting in a time offset of approximately \sim 5 minutes between successive filter groups. Given the characteristic rotation periods and variability timescales reported in previous studies ( \sim 16–17 hr; 21 ; 53 ; 54 ; 23 ), this offset is negligible for the purposes of our analysis. We therefore treat the photometry obtained within each epoch as effectively simultaneous. The observations were done with sidereal tracking, since the non-sidereal tracking mode of 7DT is still in the implementation stage. Nevertheless, 3I/ATLAS moves at about 2″ over a 100 s frame while the apparent angular extent of the coma is of order several arcseconds (Figure 1 ), enabling us to spatially resolve its extent. Table 1 shows the summary of the observation taken before the 3I/ATLAS’s approach near the Sun prohibited observations in mid-September. Figure 1: 2\times 2 -binned stacked images of 3I/ATLAS obtained with 7DT on 2025 August 11 and 24 and September 18 (UT), centered on the target. The top panel shows the m400 image, and the bottom panel shows the r -band image. Concentric apertures corresponding to projected physical radius of 5{,}000 km, 15{,}000 km, and 25{,}000 km are overplotted with dotted, dashed and solid lines for each; these were used to extract photometry at fixed physical scales given the changing geocentric distance. White arrows indicate north and east (all panels share the same orientation), and yellow and cyan arrows mark the directions toward the Sun and of the object’s motion, respectively. III Data III.1 Preprocessing Each exposure was pre-processed and calibrated by the dedicated GPU pipeline gpPy-GPU ( 47 ) , which performs bias/dark/flat correction, astrometry, and zeropoint (ZP) calibration. Because non-sidereal tracking was unavailable, we registered images using ephemerides from JPL HORIZONS ( 26 ) at mid-exposure times and generated two products per filter/epoch: (1) a standard star-aligned median stack and (2) a target-centered median stack (Figure 1 ). In crowded fields, we applied differential image analysis using a reference image from a pre-surveyed tile prior to stacking to mitigate background source confusion; both stacks share a consistent ZP solution. The UT date, MJD, and heliocentric distance ( r_{h} ) corresponding to each epoch are indicated at the top of each panel. III.2 Photometric Calibration and Annular Measurements Photometric ZPs were derived from clean, unsaturated field stars by matching instrumental magnitudes to synthetic magnitudes computed from Gaia XP spectra ( 43 ) convolved with 7DT-bandpasses. To minimize large-scale background systematics near the target, ZP calibrators were selected from within a circular aperture centered on the target, with a radius of \sim 10 ′ . Aperture photometry on the target-centered stacks was performed in fixed physical apertures with radii ( \rho ) sampled every 500 km from 1{,}500 to 25{,}000 km. Annular (projected-radius) photometry was constructed by differencing adjacent physical apertures; these annuli form the basis of our reflectance and line-proxy maps (Section III.4 ). Solar-normalized reflectance spectra were computed per annulus by dividing calibrated fluxes by synthetic photometry of the Sun spectrum for each band (a high-resolution solar spectrum convolved with the 7DT throughput) 1 1 1 https://www.stsci.edu/hst/instrumentation/reference-data-for-calibration-and-tools/astronomical-catalogs/solar-system-objects-spectra , and normalizing at m550. III.3 Geometry Normalization and Light-curve Evolution For inter-epoch comparison we normalized magnitudes to absolute magnitude H(t) using the H–G phase function ( 13 ) adopting a canonical G=0.15 given our modest phase coverage per-band. We fitted the derived H(t) with a weighted linear function to obtain per-band heliocentric and temporal slopes within each physical aperture and annulus. III.4 Line-proxy Construction and Activity Diagnostics To isolate the blue-end excess associated with the CN gas emission (388 nm), we extrapolated the local continuum near 400 nm using neighboring medium-bands (m425, m450, m475, and m500 when available). A weighted linear fit in f_{\nu} provided the continuum estimate at \lambda=400 nm; subtracting this from the calibrated m400 flux yielded a line-proxy flux density. We converted the photometry to f_{\lambda} and computed band-integrated line–proxy flux F_{\rm line} by integrating over a fixed 2.5 nm window centered on the CN(0–0) feature, rather than the full effective bandwidth of m400. We adopt 2.5 nm as a representative width because CN emission for 3I/ATLAS is typically 2–3 nm wide ( 50 ; 52 ; 32 ; 31 ) noting that this window captures only a subset of the full CN (0–0) band structure. The CN (0–0) violet band is dominated by a sharp P-branch band head at 388.3 nm, followed by a series of R-branch lines extending toward shorter wavelengths. Because the m400 filter bandpass (387.5–412.5 nm) only partially overlaps the CN (0–0) band, primarily covering the P-branch band head while excluding most of the R-branch emission shortward of \sim 387.5 nm, the measured line-proxy flux samples only a fraction of the total CN (0–0) band emission. As a consequence, the CN fluorescence efficiency adopted here, computed for the full CN (0–0) band, introduces a systematic scaling uncertainty when applied to this partially sampled bandpass. The resulting CN production rates should therefore be regarded as lower limits, and a systematic underestimation by a factor of a few relative to fully band-integrated spectroscopic measurements is expected. We derived the CN production rate, Q_{\mathrm{CN}} , from the measured line flux ( F_{\rm line} ) using a Haser model ( 28 ) . CN fluorescence efficiencies were computed via the Lowell Observatory fluorescence calculator API 2 2 2 https://asteroid.lowell.edu/docs/comet ; see 44 , which accounts for the Swings effect through the heliocentric radial velocity; for the adopted observing geometry, we used a CN fluorescence efficiency of (L/N)_{\rm CN}=2.6\times 10^{-13}\ {\rm erg\ s^{-1}\ molecule^{-1}} , scaled as r_{h}^{-2} . We adopted Haser model scale lengths of 1.3\times 10^{4} km for the parent molecule HCN and 2.1\times 10^{5} km for the daughter molecule CN at 1 au, as compiled by 6 . The scale lengths were scaled with heliocentric distance as r_{h}^{2} and describe the radial fall-off of the Haser coma profile. This scaling corresponds to the commonly adopted assumption of a constant gas expansion velocity (typically v_{\rm exp}=1 km s -1 ) in Haser-model-based analyses, enabling direct comparison with previously published CN production rates. At large heliocentric distances, the actual gas expansion velocity may differ from the canonical 1 km s -1 assumption and can vary with r_{h} and activity level. As a dust proxy we computed Af\rho ( 5 ) from the r -band continuum flux. Finally, we use Q/Af\rho as a gas-to-dust activity ratio for comparisons with cometary and ISO samples. IV Results The m400 and r bands exhibit distinct morphologies (Figure 1 ). In the r band, the dust coma is concentrated within \lesssim 15{,}000 km from the nucleus and shows an asymmetric, sunward-directed tail; both the coma and tail grow in physical extent over time. By contrast, the m400 morphology remains largely quiescent prior to the middle of August, but by September 18 it develops into an extended, relatively symmetric feature reaching out to 25{,}000 km from the nucleus. Figure 2: Phase– and distance–corrected absolute magnitudes H of 3I/ATLAS as a function of heliocentric distance r_{h} and MJD, measured within a projected 15{,}000 km radius aperture. The top axis shows the corresponding MJD. For clarity, each filter sequence is vertically offset. Filled circles (diamonds) denote detections ( S/N\geq 5 ) in medium bands (broad bands) with 1\sigma photometric uncertainties, while inverted triangles mark measurements with S/N<5 that are excluded from the fits. Dashed lines show weighted linear fits of H versus r_{h} for each filter, with slopes s (mag au -1 ) printed next to the sequences. For the m400 band, we additionally overplot a continuous broken–linear fit (solid line) with the break constrained to 2.5\leq x_{b}\leq 3.5 au and optimized from the data, yielding best–fit x_{b}\simeq 2.97 au, corresponding to the \sim MJD 60904.3 (vertical grey dashed line) and pre/post slopes ( s_{\rm pre} at larger r_{h} and s_{\rm post} at smaller r_{h} ) as annotated. Figure 3: Time-series reflectance spectra of the interstellar object 3I/ATLAS obtained with 7DT medium-band (blue squares) and broadband (orange diamonds) photometry within circular apertures of radius 15{,}000 km centered on the nucleus, obtained between 2025 July 13 and September 18. Each panel corresponds to one observing night, with medium-band points normalized at the m550 filter. Downward triangles indicate 5\sigma upper limits for non-detections. The shaded gray region represents the reflectance template of D-type asteroids from 22 for reference. Vertical colored-bands mark expected gas emission features: CN (violet and red), C 2 (green), C 3 (blue) and [O I] (yellow). IV.1 Multi-band Light-Curve Brightening Figure 2 shows the phase–distance corrected absolute magnitudes ( H ) of 3I/ATLAS as a function of heliocentric distance and time at different wavelengths, where s denotes the slope of the light curve, expressed in units of mag au^{-1} . As the object approached perihelion, the brightness increased monotonically in all filters at a rate of about 0.89\pm 0.24~\mathrm{mag~au^{-1}} ( -0.027\pm 0.007~\mathrm{mag~day^{-1}} ). This is in agreement with the post-transition activity slope ( -0.014 mag day -1 ) after MJD 60870 (2025 July 14 UT) reported by 55 . For the m400 band, a continuous broken–linear fit as a function of heliocentric distance ( r_{h} ) provides a significantly improved description of the data. In this model, the breakpoint x_{b} is defined as the heliocentric distance at which the slope of the light curve transitions between two linear regimes. The best–fit breakpoint occurs at r_{h}\simeq 2.97 au, corresponding to MJD 60904.3 (2025 August 17 UT), and separates two distinct brightening regimes. At larger heliocentric distances ( r_{h}>x_{b} ), the m400 brightness increases more gradually, with a slope of s_{\rm pre}=0.720~\mathrm{mag~au^{-1}} , whereas at smaller distances ( r_{h} 10, sampled across a 9\times 9 grid and normalized to unit flux at the comet’s position. Background sources were masked using photutils.segmentation ( 14 ) , excluding pixels above 1.5 \sigma significance while retaining a central 100 \times 100 pixel region to avoid over-masking the coma. Each epoch was modeled using two concentric exponential disk components (inner and outer), together with a constant sky background. The centers of the two components were fixed to coincide, and the scale length (h) of the outer component was constrained to be larger than that of the inner component. This double-exponential representation is adopted as a phenomenological description of the radial surface brightness profile, serving as an empirical approximation to the Haser distribution ( 28 ) over the spatial scales probed by 7DT. In this framework, we interpret the inner component (with a shorter scale length) as tracing the dust-scattered continuum associated with the nucleus and inner coma, while the outer component represents the spatially extended coma. In particular, for the m400 band, the outer component is associated with CN-emitting gas that gives rise to the observed excess relative to the dust-dominated continuum. The results of the fitting are shown in Figure 7 , which shows the temporal evolution of half-light radii ( r_{1/2} ), scale lengths ( h ), and the flux density ratio of the outer to inner components ( F_{\rm out}/F_{\rm in} ) We find that the r_{1/2} value increases for m400 while it is nearly constant for the r band, suggesting that m400 (the CN-dominated component) becomes more extended as the heliocentric distance decreases, whereas this behavior is not observed in the r band (the dust-scattered component). The middle panel shows the flux ratios ( F_{\rm in}/F_{\rm out} ) of the two components (outer to inner). For r band, it is nearly 0 , meaning that the extended, outer component is almost negligible for the dust-scattered light. On the contrary, the outer component becomes significant starting from ~\sim 3 au, and this coincides with the photometric excess identified in Section IV.1 . Therefore, we consider that the outer component represents the CN excess for m400. Regarding the scale lengths of the inner and outer components, the scale lengths ( h ) of the inner component are similar in the m400 and r bands ( h\lesssim 1{,}000 km), suggesting that the inner component traces the dust-scattered light distribution. However, the scale length of the outer component in the r band remains nearly constant, whereas that in the m400 band shows no marked evolution, indicating that the profile shape of the CN-emitting gas is conserved over the observed time span. These results demonstrate that the onset of CN emission is driven by an increase in the flux of the outer component, rather than by an continuous increase in its spatial extent. Figure 7: The series of structural parameters and flux ratios as a function of time and heliocentric distance from GALFIT in physical units. Top : half-light radius r_{1/2} from the double-exponential model. Middle : flux density ratio ( F_{\rm out}/F_{\rm in} ) between the two exponential components. Bottom : characteristic scale lengths h from the double-exponential model (inner: filled symbols; outer: open symbols). Colors denote filters: blue—m400 and red— r . The vertical gray dashed line marks MJD =60904 . For m400, a linear regression over the post-break interval (MJD >60904 ) is shown in the top panel as a black dashed line with a 1 \sigma confidence band (blue shading); the best-fit slope is dr_{1/2}/dr_{h}=-1.1\times 10^{4}~\mathrm{km\,AU^{-1}} . In the middle panel, the post-break median outer scale length for m400 is \sim 1.2\times 10^{4}\pm 1.5\times 10^{3} km. The upper horizontal axis indicates the heliocentric distance r_{h} (au). V Discussion V.1 Emergence of Dust Rich, CN Emission Coma and Its Implication on Origin of 3I/ATLAS A broader picture for the origin of 3I/ATLAS emerges when comparing 3I/ATLAS with the ensemble of 45 Solar System comets from 6 , as shown in Figure 5 . The CN production rates of 3I/ATLAS fall within the distribution spanned by typical Jupiter-family and long-period comets, indicating that its volatile output is not an outlier relative to the Solar System population. The heliocentric dependence of Q_{\rm CN} for 3I/ATLAS within a 15,000 km aperture follows a steep power-law slope of \gamma~\sim-5.5 , consistent with the wide range of CN activity scalings ( Q_{CN}\propto r_{h}^{\gamma} ) observed among Solar System comets ( 6 ) . In contrast, its dust production—as traced by Af\rho —is elevated by factors of several compared to most comets in the 6 sample, placing 3I/ATLAS among the more dust-rich comae at comparable heliocentric distances. Consequently, the ratio Q_{\rm CN}/Af\rho falls toward the lower envelope of the Solar System distribution: although not anomalously low, 3I/ATLAS exhibits a comparatively dusty coma for its level of CN production. Contemporaneous spectroscopic observations show strong carbon-chain depletion, with upper limits on the C 2 -to-CN ratio ( \log Q_{\mathrm{C}_{2}}/Q_{\mathrm{CN}}\lesssim-0.8 ), placing 3I/ATLAS among the most carbon-chain depleted comets known ( 52 ; 50 ) . This combination of chemical signatures—high dust production, low relative CN output, and extreme carbon-chain depletion—provides important leverage on the formation environment and evolutionary history of 3I/ATLAS. This behavior is consistent with the broader phenomenon of carbon-chain depletion seen in other comets, and discussed in detail by previous studies ( 6 ; 18 , e.g.,) . In those works, carbon-chain depletion is generally attributed to a combination of formation in cold, UV-irradiated and/or metal-poor regions of the protoplanetary disk, where the build-up of carbon-chain precursors is inefficient, and subsequent long-term processing by UV photons and cosmic rays during residence in the outer system or interstellar space. For 3I/ATLAS, its likely low-metallicity birth environment and cosmic-ray contamination during a long interstellar travel time make both effects plausible. Recent JWST observations, which reveal extreme CO 2 - and CO-rich volatile compositions and strong surface reddening, support the interpretation that the near-surface materials have been substantially altered by long-term cosmic-ray processing ( 39 ) . Additional constraints may come from the temporal evolution of volatile production and changes in the nucleus/coma structure around perihelion. V.2 Evolution of Spatial Distribution of CN and Dust Components The evolution of the spatial distributions of the CN and dust components shows a clear and physically interpretable segregation. Morphologically, the m400 band exhibits a nearly spherically symmetric coma extending beyond 25{,}000 km (Figure 7 ), while the r band displays a sunward-pointing dust tail. The structural properties of the m400-emitting gas morphology indicate that, although the spatial extent of the CN-emitting coma increases during the early phase of activation, its overall characteristic scale remains largely unchanged once CN emission becomes dominant, indicating that the large-scale structure of the gas coma is preserved over the later stages of our observations. This behavior suggests that the observed increase in the m400 flux is not driven by continued structural broadening of the extended component, but rather by an increase in the relative brightness of the CN-dominated outer component with respect to the inner dust-scattered component. The invariance of the radial profile shape of the extended outer component suggests that the expansion velocity of the CN parent molecules is relatively fast. If the outflow velocity were slow, the increasing volatile production toward smaller heliocentric distances would lead to a progressive buildup of material near the nucleus, resulting in a more centrally concentrated radial profile with a reduced scale length as 3I/ATLAS approaches the Sun. No such trend is observed. Instead, the persistence of a self-similar outer profile is consistent with a rapid expansion of the major parent species, including HCN and its daughter CN. This interpretation is independently supported by recent JCMT detections of HCN, which show a simple single-Gaussian line profile with an expansion velocity of v_{\rm exp}\approx 0.46 km s -1 at r_{h}\sim 2 au, indicative of a dynamically uniform and freely expanding coma ( 19 ) . Together, the optical CN morphology and millimeter HCN spectroscopy consistently point to relatively rapid gas expansion in the inner heliocentric regime ( \sim 0.5 –1 km s -1 ). While the persistence of a self-similar outer CN profile is consistent with rapid expansion, a distributed or diffuse source of CN, arising from parent molecules released from dust grains throughout the coma, could also provide an alternative explanation for the observed morphology. In this context, the expansion velocity measured from millimeter spectroscopy ( v_{\rm exp}\approx 0.46 km s -1 at r_{h}\sim 2 au) indicates that canonical constant-velocity assumptions commonly adopted in simple Haser models should be regarded as approximate. V.3 Flattening of the Continuum Slope and Resurfacing Process Beyond the spatial structure and kinematics of the gas coma, the optical data also reveal systematic evolution in the properties of the dust-scattered continuum. As shown in Table 1 and Figure 4 , both the blue- and red-wavelength reflectance slopes, together with the g-r color, evolve toward less red values as 3I/ATLAS approaches the Sun, while the r-i color remains approximately constant. This wavelength-dependent behavior suggests that the observed color evolution is not a uniform reddening or fading of the dust continuum, but instead reflects changes that preferentially affect shorter optical wavelengths. Such wavelength-dependent color evolution may reflect multiple physical processes, including changes in surface properties as well as variations in the characteristic dust particle size distribution, as grain size can influence the wavelength dependence of optical scattering. One possible interpretation is a resurfacing process, in which activity-driven dust emission exposes or blankets fresher, less-processed material, leading to a reduction of the red spectral slope, particularly at shorter wavelengths. Similar color evolution has been reported for Solar System bodies with irradiation-processed surfaces, where the onset of activity can diminish ultra-red spectral characteristics ( 33 ; 35 , e.g.,) . In this context, the contrast between the extremely red colors of the inactive 1I and the more comet-like colors observed for active ISOs such as 2I and 3I/ATLAS may reflect differences in surface processing and activity states. VI Conclusion We present a time-series analysis of the volatile and dust activity of 3I/ATLAS using medium-band photometry from the 7-Dimensional Telescope. Our data uniquely capture the emergence of the CN-dominated coma spatially and temporally. The m400-band excess, its large-scale symmetry, and the emergence of an extended m400 component out to \sim 25,000 km at heliocentric distances of r_{h}\lesssim 3 au together reveal a rapid transition. This volatile activation pattern is qualitatively closer to that of 2I than 1I. A comparison with Solar System comets shows that, while the CN production rate of 3I/ATLAS lies within the normal cometary distribution, its dust production ( Af\rho ) is enhanced by factors of several, placing it among the dustiest comets known. The resulting low Q_{\rm CN}/Af\rho ratio, combined with independently observed carbon-chain depletion, is consistent with previous findings for chemically depleted comets and offers a valuable opportunity to probe the origin as its volatile activity evolves through perihelion, underscoring the importance of continued 7DT monitoring. The 2D analyses of the m400 surface brightness profile indicate the excess m400 emission arises mostly in the outer part of the comet, with the radial profile of the CN-dominated gas component becoming progressively more prominent relative to the dust component. In particular, the decomposition of the dust-component and CN gas component is achieved by the 2-dimensional surface brightness fitting with GALFIT . The GALFIT result shows that the inner component follows the evolution of dust components as found in other wavelengths, while the outer component becomes brighter as the comet nears the sun but preserves the surface brightness shape during this change. This interpretation is consistent with the relatively rapid expansion velocity ( \sim 0.5 –1 km s -1 ) inferred for CN-bearing molecules and is independently supported by recent JCMT observations ( 19 ) . The extended spatial distribution of CN emission, combined with its temporal acceleration inside r_{h}\sim 2.5 au, provides strong evidence for a distributed source component in which CN parent molecules are released from dust grains rather than solely from the nucleus. Such gas–dust interaction mechanisms have been well established in several Solar System comets. In particular, similar distributed CN sources have been identified in 67P ( 45 ) , suggesting that this mechanism may represent a common process shared by ISOs and demonstrating that extended volatile sources are not confined to our Solar System. While our observations capture the pre-perihelion activity of 3I/ATLAS, continued post-perihelion monitoring will be essential for fully characterizing the longevity and drivers of its volatile activity. Tracking the evolution of the spatial distributions and relative contributions of gas and dust components beyond perihelion will provide critical insight into the physical and compositional properties of this interstellar visitor. In this context, the time-series, wide-field spectral mapping achieved here with the 7DT demonstrates a unique, IFU-like capability that enables high-cadence monitoring of large-scale emission-line and continuum morphology. Applying this capability to future surveys will enable us to trace the evolutionary histories of both Solar System comets and rare interstellar visitors. This work was supported by the National Research Foundation of Korea (NRF) grant, No. 2021M3F7A1084525, funded by the Korea government (MSIT). G.S.H.P. acknowledges support from the Pan-STARRS project, which is a project of the Institute for Astronomy of the University of Hawai’i, and is supported by the NASA SSO Near Earth Observation Program under grants 80NSSC18K0971, NNX14AM74G, NNX12AR65G, NNX13AQ47G, NNX08AR22G, 80NSSC21K1572, and by the State of Hawai’i. SWC acknowledges support from the Basic Science Research Program through the NRF funded by the Ministry of Education (RS-2023-00245013). J.H.K. acknowledges the support from the Institute of Information & Communications Technology Planning & Evaluation (IITP) grant, No. RS-2021-II212068 funded by the Korean government (MSIT). H.C. acknowledges the support from the National Research Foundation of Korea (NRF) grants, No. 2021M3F7A1084525 and RS-2025-00573214, funded by the Korean government (MSIT). B.Lim acknowledges the support from the National Research Foundation of Korea (NRF) grants funded by the Ministry of Education (RS-2025-25436385) W.B.H. acknowledges support from the National Science Foundation Graduate Research Fellowship Program under Grant No. 2236415. Data transfer from the host site was supported in part by the Korea Research Environment Open NETwork (KREONET) advanced research program funded by KISTI. This job has made use of the Python package GaiaXPy , developed and maintained by members of the Gaia Data Processing and Analysis Consortium (DPAC), and in particular, Coordination Unit 5 (CU5), and the Data Processing Centre located at the Institute of Astronomy, Cambridge, UK (DPCI). Table 1: Time-series m400 photometry of 3I/ATLAS. The complete photometric dataset for all filters is provided as an electronic table. Date MJD Filter r \Delta \alpha H m Refl. Slope blue Slope red g-r r-i (UT) (day) (au) (au) (deg) (mag) (AB mag) (%/1000Å) (%/1000Å) (mag) (mag) 2025-07-13 60869.2 m400 4.11 3.17 6.2 12.07 18.13\pm 0.33 1.43\pm 0.46 – – – 0.54\pm 0.09 2025-07-23 60879.5 m400 3.77 2.95 10.3 12.82 18.71\pm 0.17 0.61\pm 0.10 33.86\pm 3.13 14.03\pm 2.27 0.73\pm 0.03 0.33\pm 0.04 2025-07-25 60881.0 m400 3.72 2.92 11.0 12.54 18.40\pm 0.18 0.77\pm 0.13 14.75\pm 3.44 14.23\pm 2.38 0.68\pm 0.03 0.34\pm 0.04 2025-08-08 60895.0 m400 3.26 2.73 16.5 12.80 18.43\pm 0.54 0.43\pm 0.21 29.49\pm 6.25 6.85\pm 2.36 0.66\pm 0.04 0.23\pm 0.05 2025-08-11 60899.0 m400 3.14 2.69 18.0 12.35 17.91\pm 0.11 0.69\pm 0.07 14.34\pm 2.52 22.18\pm 1.75 0.72\pm 0.02 – 2025-08-13 60901.0 m400 3.07 2.67 18.7 12.17 17.70\pm 0.09 0.70\pm 0.06 23.30\pm 1.81 16.50\pm 1.38 0.71\pm 0.02 0.24\pm 0.03 2025-08-18 60905.1 m400 2.94 2.64 20.0 11.92 17.37\pm 0.10 0.93\pm 0.09 11.53\pm 2.36 20.93\pm 1.64 – – 2025-08-19 60907.0 m400 2.88 2.63 20.5 11.75 17.16\pm 0.06 0.97\pm 0.06 18.44\pm 1.72 15.41\pm 1.26 0.71\pm 0.01 0.32\pm 0.02 2025-08-24 60911.0 m400 2.76 2.61 21.5 11.80 17.13\pm 0.05 0.82\pm 0.04 24.78\pm 1.24 11.32\pm 0.86 0.67\pm 0.01 0.28\pm 0.02 2025-08-26 60913.0 m400 2.70 2.60 21.9 11.71 17.00\pm 0.05 0.89\pm 0.04 20.23\pm 1.36 13.72\pm 0.88 0.64\pm 0.01 – 2025-09-02 60921.0 m400 2.46 2.57 23.0 11.31 16.40\pm 0.05 1.07\pm 0.06 19.75\pm 1.47 10.26\pm 0.76 0.64\pm 0.01 0.26\pm 0.02 2025-09-06 60924.0 m400 2.37 2.56 23.2 11.20 16.21\pm 0.08 1.16\pm 0.09 14.38\pm 2.29 12.92\pm 0.95 0.64\pm 0.02 0.27\pm 0.02 2025-09-15 60933.0 m400 2.11 2.54 22.6 10.58 15.31\pm 0.02 1.44\pm 0.03 17.25\pm 0.85 4.81\pm 0.43 0.53\pm 0.01 0.25\pm 0.01 2025-09-17 60936.0 m400 2.03 2.54 22.1 10.40 15.02\pm 0.02 1.52\pm 0.04 14.43\pm 0.99 4.55\pm 0.49 0.49\pm 0.01 0.27\pm 0.01 2025-09-18 60936.5 m400 2.02 2.53 22.0 10.33 14.93\pm 0.02 1.53\pm 0.03 14.77\pm 0.81 3.11\pm 0.39 0.48\pm 0.01 0.25\pm 0.01