VARIABILITY IN SSTC2D J163134.1-240100, A BROWN DWARF WITH QUASI-SPHERICAL MASS LOSS

Ve sion Ap il 24, 2024 P ep in ypese using L A T EX s yle openjou nal . 09/06/15 VARIABILITY IN SSTC2D J163134.1-240100, A BROWN DWARF WITH QUASI-SPHERICAL MASS LOSS Aleks Scholz SUPA, School o Physics & As onomy, Uni e si y o S And ews, No h Haugh, S And ews, KY16 9SS, Uni ed Kingdom, email: [email p o ec ed] Ko aljka Muzic Ins i u o de As o ísica e Ciências do Espaço, Faculdade de Ciências, Uni e sidade de Lisboa, Ed. C8, Campo G ande, 1749-016 Lisbon, Po ugal Vic o Almend os-Abad Is i u o Nazionale di As o isica (INAF) – Osse a o io As onomico di Pale mo, Piazza del Pa lamen o 1, 90134 Pale mo, I aly An onella Na a School o Cosmic Physics, Dublin Ins i u e o Ad anced S udies, 31 Fi zwilliam Place, Dublin 2, I eland Da y Ruiz-Rod iguez Na ional Radio As onomy Obse a o y, 520 Edgemon Rd., Cha lo es ille, VA 22903, USA Lucas A. Cieza Nucleo de As onomia, Facul ad de Ingenie ia, Uni e sidad Diego Po ales, A . Eje ci o 441, San iago, Chile C is ina Rod iguez-Lopez Ins i u o de As o ísica de Andalucía (IAA-CSIC), Glo ie a de la As onomía s/n, 18008 G anada, Spain Ve sion Ap il 24, 2024 Abs ac We epo on a sea ch o a iabili y in he young b own dwa SST1624 (∼M7 spec al ype, M∼0.05 M⊙), p e iously ound o ea u e an expanding gaseous shell and o unde go quasi-sphe ical mass loss. We ind no a iabili y on imescales o 1-6 hou s. Speci ically, on hese imescales, we ule ou he p esence o a pe iod wi h ampli ude >1%. A pho ome ic pe iod in ha ange would ha e been e idence o ei he pulsa ion powe ed by Deu e ium bu ning o o a ion nea b eakup. Howe e , we see a 3% dec ease in he K-band magni ude be ween wo consecu i e obse ing nigh s (a 10 σ esul ). The e is also clea e idence o a ia ions in he WISE ligh cu es a 3.6 and 4.5 µm on imescales o days, wi h a en a i e pe iod o abou 6 d (wi h a plausible ange be ween 3 and 7 d). The bes explana ion o he a ia ions o e days is o a ional modula ion due o spo s. These esul s dis a ou cen i ugal winds d i en by as o a ion as mechanism o he mass loss, which, in u n, makes he al e na i e scena io – a he mal pulse due o Deu e ium bu ning – mo e plausible. 1. INTRODUCTION Pho ome ic moni o ing has p o en o be a use ul obse a ional ool in he explo a ion o b own dwa s. When hey a e young, subs ella objec s o en show a i- abili y due o magne ic ac i i y, acc e ion, and disks, analogous o he mo e massi e T Tau i s a s (Scholz e al. 2009; Moo e e al. 2019). Fo e ol ed b own dwa s, pho ome ic a iabili y can signi y he p esence and e o- lu ion o clouds in he a mosphe e (e.g. Me che e al. 2015). Va iabili y s udies ha e been able o measu e sub- s ella o a ion pe iods as a unc ion o age (Bou ie e al. 2014), as well as cons ain a mosphe ic p ope ies and ea ly e olu iona y p ocesses in b own dwa s. Thanks o nume ous moni o ing s udies in he pas wen y yea s, we know ha young b own dwa s show o- a ion pe iods anging om a ew hou s o se e al days, whe eas old e ol ed b own dwa s gene ally ha e pe i- ods sho e han one day (Moo e e al. 2019; Vos e al. 2022). A young ages, he sho es measu ed pe iods a e a ound he physical limi – he pe iod co espond- a Xi :2307.14087 2 [as o-ph.SR] 23 Ap 2024 2 Scholz e al. ing o b eakup eloci y, whe e cen i ugal and g a i a- ional o ces a e in balance a he equa o . Ro a ion nea o a ound b eakup in young b own dwa s has been epo ed by a ious au ho s, including Zapa e o Oso io e al. (2003); Caballe o e al. (2004); Scholz & Eislö el (2005); Rod íguez-Ledesma e al. (2009). Such as o- a ion will ha e signi ican e ec s on in e nal s uc u e and ea ly e olu ion, which needs o be accoun ed o in models (Yoshida 2023). The b eakup eloci y is calcula ed using = p2GM/3R, ollowing Po e (1996). The ac o 2/3ac- coun s o he ac ha an objec o a ing a b eakup will be obla e. Fo con enience, his co esponds o P= 0.14205×R1.5/M0.5. Fo subs ella masses and ages 1-5 My , his yields b eakup pe iods >5h, as shown in Figu e 1 o wo adii, ypical o young b own dwa s. As he objec s age and con ac , he b eakup pe iod d ops wi h ∼R1.5, bu he o a ion pe iod (assuming angula momen um conse a ion) changes wi h ∼R, i.e. wi h inc easing age he b eakup pe iod will d op u he be- low he o a ion pe iod. As a esul , ield b own dwa s which some imes ea u e e y sho o a ion pe iods o ∼1h (Tannock e al. 2021) a e s ill signi ican ly abo e b eakup. Ano he physical p ocess ha can be p obed by pho- ome ic moni o ing is pulsa ions. B own dwa s ex- hibi Deu e ium bu ning du ing he i s ew My s o hei exis ence, p edic ed o lead o a luminosi y bu s (Salpe e 1992) and o adial o non- adial pulsa ions. Palla & Ba a e (2005) show in a s abili y analysis o Deu e ium bu ning b own dwa s he p esence o un- damen al modes wi h pe iods be ween 1 and 5 h. Fo 0.1 M⊙s a s, Rod íguez-López e al. (2012) ind pe i- ods o modes exci ed by Deu e ium bu ning o 4.2-5.2 h. A dedica ed obse a ional sea ch o pulsa ion pe iods down o mmag p ecision, by Cody & Hillenb and (2014), howe e , emained unsuccess ul. In p inciple, some o he sho es pho ome ic pe iods measu ed so a o b own dwa s (see abo e) could be caused by pulsa ions. Mos au ho s p e e , howe e , o in e p e he pe iodic a i- abili y in young b own dwa s as a esul o o a ion combined wi h magne ic su ace spo s. I pulsa ions we e con i med, i would open a new ou e o de e mine and asce ain undamen al p ope ies in b own dwa s du ing he phase o con ac ion. In Figu e 1 we include he es ima ed pulsa ion pe iods om Palla & Ba a e (2005) o ages o 1-2 My and sub- s ella masses. As can be app ecia ed om his igu e, he gap be ween pulsa ion pe iods and b eakup pe iods widens wi h dec easing mass and inc easing adius. A pe iod below 4 h o such a young e y low mass objec would be e y di icul o in e p e as o a ion pe iod, o e ing a window o sea ch o pulsa ions. 2. THE TARGET SSTc2d J163134.1-240100 (he ea e SST1624) is a young b own dwa in he Ophiuchus s a o ming com- plex, iden i ied o iginally in he Spi ze ’Co es o Disks’ 0.02 0.03 0.04 0.05 0.06 0.07 0.08 Mass( M ) 2.0 4.0 8.0 16.0 32.0 Pe iod (h) P b eak R = 0.5 R P b eak R = 1.0 R mass o a ge Pulsa ion 1-3My Fig. 1.— Es ima ed o a ion pe iod a b eakup eloci y (lines) and pulsa ion pe iod (do s) o young b own dwa s, as a unc ion o mass. The b eakup pe iods a e shown o wo di e en adii, consis en wi h he expec ed alues o in la ed, young subs ella objec s. Pulsa ion pe iods a e he alues gi en by Palla & Ba a e (2005) o ages be ween 1 and 2 My ( hei Table 1). Fo masses o 0.06 and 0.08 M⊙ hese au ho s lis se e al pulsa ion pe iods, o ages anging om below 1 o 3 My . The es ima ed mass o ou a ge is ma ked by a dash-do ed line. Fig. 2.— 2MASS JHK colou image o he ield a ound ou a ge . Ma ked in g een is he 4.9′×4.9′SOFI ield. The a ge is ma ked in o ange; he ou s a s used as compa ison s a s in yellow. No h is up and eas is le . su ey (E ans e al. 2009) as a ain , ed young s ella objec . A 2MASS image o he a ea a ound he a - ge is shown in Figu e 2. In ecen ALMA obse a ions as pa o he ODISEA p og am (Cieza e al. 2019), i was ound o ha e a sphe ical, obla e en elope wi h a diame e o 200-300 AU, which shows he kinema ics o an expanding shell (Ruíz-Rod íguez e al. 2022). The shell has been de ec ed in 12CO, bu no in he mm con- inuum. A e co ec ing o ex inc ion, he objec has Va iabili y in SSTc2d J163134.1-240100 3 an in a ed spec al ene gy dis ibu ion consis en wi h a diskless pho osphe e. This is e y unusual – so a e y ew young b own dwa s ha e been de ec ed in CO wi h ALMA, and he ones ha a e, usually ha e an ob- ious disk (Ricci e al. 2014). On he o he hand, nu- me ous young b own dwa s ha e been de ec ed in he submm/mm con inuum (Tes i e al. 2016; Sanchis e al. 2020). We a e hus con on ed wi h an en i ely new phe- nomenon ha equi es new ideas, con inued explo a ion, and ollow-up obse a ions. Ruíz-Rod íguez e al. (2022) de e mine he na u e o he sou ce based on KMOS spec a, and a e ca e ul o ule ou al e na i e explana ions, such as an AGB s a in he backg ound o a collapsing molecula co e. P ope mo ions con i m ha he objec is indeed a membe o he young popula ion in Ophiuchus, wi h an age o ∼2My . Mos likely, i is a 0.05 M⊙b own dwa , wi h mid/la e M spec al ype, o e luminous, and seen h ough signi ican ex inc ion o AV∼42 mag. The bes explana ion o he gaseous expanding shell is s ong, uncollima ed mass loss. The age o he shell has been es ima ed kinema ically o be ∼104yea s, a small ac- ion o he b own dwa ’s age, and i s gas mass (unde plausible assump ions) is 1.5M⊕. The 4σuppe limi in he con inuum obse a ions co esponds o a dus mass o <1M⊕. A his poin he o igin o he expanding shell is un- known. Ruíz-Rod íguez e al. (2022) sugges ha he mass loss may be he esul o a he mal pulse p oduced by he onse o Deu e ium bu ning, analogous o he he mal pulses ha d i e mass loss in AGB s a s Ke - schbaum e al. (2017). Ano he op ion o a b own dwa o ejec a gaseous shell, o iginally sugges ed by Scholz & Eislö el (2005), is h ough non-s eady cen i ugal winds caused by ex emely as o a ion close o b eakup speed – analogous o as o a ing OB s a s (Po e 1996). This would lead ini ially o a ing (a ’dec e ion disk’), expand- ing and dispe sing o a quasi-sphe ical s uc u e, consis- en wi h he obla e shape o he shell o SST1624. As poin ed ou in Sec ion 1, bo h pulsa ion and as o a ion can cause a pe iodic lux change de ec able by pho ome ic moni o ing, wi h he plausible pe iods o pulsa ions being sho e han hose o as o a ion nea b eakup (see Figu e 1). Tes ing o a iabili y in SST1624 is he e o e a p omising way o explo e he o i- gin o i s shell. This is he pu pose o his pape . Fo he eco d, Ruíz-Rod íguez e al. (2022) also dis- cuss ano he op ion o explain he shell, he engul men and consump ion o a plane a y companion. They deem his scena io as mo e exo ic and less p obable han he he mal pulse. As his idea would no p oduce a a i- abili y signa u e, we will no discuss i u he he e. 3. OBSERVATIONS AND DATA REDUCTION 3.1. K-band moni o ing We ha e moni o ed SST1624 o e wo nigh s, May 26 h and May 27 h 2023, wi h ESO/NTT and he in a ed came a SOFI, unde p oposal ID 111.24GN. O e all, his obse ing un comp ised h ee nigh s, bu he hi d one was no usable due o clouds. In each o he wo nigh s, we s ayed on he a ge o 5-6 h, using he KS il e , DIT o 20 s, NDIT o 5 o 10, and employing andom di he s. In bo h nigh s, he condi ions we e s able, wi h clea skies, low humidi y (8-26%), mode a e wind speed (5- 11 ms−1, well below he poin ing limi o 14 ms−1 o his elescope), and empe a u e be ween 11 and 16 deg ees Celsius, all acco ding o he ESO Me eo moni o . The seeing in ou da a is be ween 1 and 2 a csec, wi hou any apid changes. The da a educ ion includes c oss- alk co ec ion, la ielding, sky sub ac ion, and bad-pixel co ec ion. In- di idual exposu es we e combined, whe e necessa y, o achie e a uni o m o al exposu e ime pe ame o 200 s. On each image, we pe o med ape u e pho ome y using pho u ils (B adley e al. 2023) using a cons an ape u e o 10 pixels, which ypically co esponds o 2-3 FWHMs o he seeing-limi ed PSF. This includes backg ound sub- ac ion, whe e he backg ound was measu ed in an an- nulus a ound he sou ce. We op ed no o adap he ape u e o he changing FWHM, o main ain consis- ency ac oss he ligh cu e. The ape u e adius was chosen o op imise he signal- o-noise in he ligh cu e, a e expe imen ing wi h a wide ange o adii. The aw ligh cu e o SST1624 and all o he s a s in he ield show subs an ial a ia ions due o he change in ai mass o e he cou se o he nigh . We chose ou s a s in he same ield ha a e 2-3 mag b igh e han ou a ge (ma ked yellow in Figu e 2), a e aged hei ligh cu es, and sub ac ed his a e age om he a ge ligh cu e (in uni s o magni udes) o co ec o a mosphe ic e - ec s. The inal ligh cu e o SST1624 o hose wo nigh s is shown in Figu e 3. Judged by he noise in he compa ison s a ’s ligh cu es, he ypical e o o indi- idual pho ome y poin s o he a ge is 0.01-0.02 mag (o 1-2%). We made su e ha he ligh cu e does no depend sig- ni ican ly on he choice o compa ison s a s, by checking he esul wi h only a subse o hem. We also checked he compa ison s a s agains each o he , and do no ind any signs o a iabili y in hem. We no e ha he same calib a ion p ocess also co ec s o o he en i onmen al in luences on he pho ome y, like changes in seeing. Us- ing he compa ison s a s o calib a ion, we de e mine ha SST1624 has a K-band magni ude o 14.08 ±0.05 in he 2MASS sys em, consis en wi h i s 2MASS alue o 14.06. 3.2. A chi al WISE ligh cu es We also ob ain he WISE ligh cu e a 3.6 (W1) and 4.5 µm (W2) o SST1624 om he ’ALLWISE Mul iepoch Pho ome y Table’ (Cu i e al. 2021) and he ’NEOWISE-R Single Exposu e (L1b) Sou ce Ta- ble’ (Mainze e al. 2014). The ALLWISE able con- ains 23 epochs o his sou ce wi hin 200 d, s a ing a MJD55254. Wi h one excep ion, all o hese measu e- men s a e concen a ed in wo 24 h spans, sepa a ed by 4 Scholz e al. 0.05 0.10 0.15 0.20 0.25 MJD - 60091.0 -0.10 -0.05 0.00 0.05 0.10 Rela i e K-band magni ude 1.00 1.05 1.10 1.15 1.20 1.25 MJD - 60091.0 Fig. 3.— K-band ligh cu e o SSTc2d J163134.1-240100, measu ed o e wo nigh s in May 2023. The magni udes a e measu ed ela i e o he a e age o ou ield s a s. The blue dashed lines ma k he a e age o a gi en nigh . The ypical e o o indi idual da apoin s is 0.01-0.02 mag. 4.6 4.8 5.0 5.2 5.4 5.6 MJD - 55250.0 11.6 11.8 12.0 12.2 12.4 ALLWISE Magni ude W1 W2+1.0 186.2 186.4 186.6 186.8 187.0 187.2 187.4 MJD - 55250.0 Fig. 4.— ALLWISE ligh cu e o SSTc2d J163134.1-240100, comp ising 22 o he 23 da apoin s in he a chi e. The W2 magni udes a e o se by 1.0 mag o cla i y. Va iabili y in SSTc2d J163134.1-240100 5 0 500 1000 1500 2000 2500 3000 3500 MJD - 56700.0 11.6 11.8 12.0 12.2 12.4 12.6 12.8 NEOWISE Magni ude W1 W2+1.0 Fig. 5.— NEOWISE-R ligh cu e o SSTc2d J163134.1-240100, comp ising 237 da apoin s. A ew ou lie s a e no shown he e. The W2 magni udes a e o se by 1.0 mag o cla i y. The ypical pho ome ic e o is 0.03 mag. ∼180 d, and shown in Figu e 4. NEOWISE-R p o ides 237 good de ec ions (as o Ap il 2024), o e a du a ion o ∼3500 d, s a ing a MJD 56700.0, wi h 20 day-long spans co e ed by abou 10 da - apoin s each. Again he measu emen s a e a ailable o he wo bands W1 and W2. We no e ha his ligh cu e ends a MJD 60167 which co esponds o Augus 2023, a ew mon hs a e ou dedica ed K-band moni o ing. The e a e, howe e , no da apoin s coinciding wi h ou obse a ions. The ull ligh cu e is shown in Figu e 5. Bo h he ALLWISE and NEOWISE a chi es p o ide e o ba s o indi idual pho ome y poin s. Those e o s a e de e mined om an assessmen o he epea abili y o he measu emen s as a unc ion o magni ude and galac- ic la i ude (W igh e al. 2010). The nea es sou ce is a b igh s a 36 a csec o he sou heas . This sou ce is he eason he WISE a chi e epo s a isk o con amina ion. The pho ome ic quali y in he ALLWISE pho ome y is obus wi h a signal- o-noise a io >10 (i.e. he ph_qual lag is ’AAAA’). 4. LIGHTCURVE ANALYSIS 4.1. The Max-F es Fo he analysis p esen ed in his pape , we de ised a simple, obus es o check o he p esence o a pe iod in noisy da a by compa ing he a iance in he ligh cu e be o e and a e sub ac ing a pe iodici y, a p ocedu e he ea e called Max-F. In sho , he ou ine es s o wha ex en he noise in he ligh cu e could be caused by a pe iodici y wi h speci ic p ope ies. We de ine a ange o es pe iods o a gi en da ase . Fo each es pe iod we calcula e he phases o he da apoin s (be ween 0.0 and 1.0). Then we sub ac sinecu es o he gi en es pe iod, a ying he ampli- ude and ze opoin o he phase. The la e is changed om 0.0 o 1.0 wi h a s epsize o 0.01. We eco d he s anda d de ia ion o he ligh cu e o each sub ac ed sinecu e, σP,i, and he a iance a io F=σ2/σ2 P,i. He e σis he s anda d de ia ion o he o iginal ligh cu e. We ind he maximum alue o F o each pe iod, and plo ha as a unc ion o pe iod. This p ocedu e gi es us a measu e o he maximum ac ion o he noise ha could be explained by he p es- ence o a sinusoidal pe iod. The ou ine cons i u es in essence an F- es o he equali y o a iances, a s an- da d s a is ical es . A simila es – bu only o e - i y a p e iously adop ed pe iod – was used o example by Scholz & Eislö el (2004) in he ligh cu e analysis o young e y low mass sou ces. Fo no mally dis ibu ed da ase s, he quan i y Fshould ollow an F-dis ibu ion (hence he name). 4.2. K-band ligh cu e The ligh cu e we ob ained om he SOFI obse a- ions appea s un ema kable. By eye, i shows noise a ound a la mean, in each o he wo consecu i e nigh s, as can be app ecia ed om Figu e 3. The e is no ob ious s uc u e, in pa icula no ob ious pe iodic a iabili y. The s anda d de ia ion is 1.5 and 1.6%, o nigh 1 and nigh 2 espec i ely. To check o g adual ends, we i he da a in indi id- ual nigh s wi h a s aigh line. In he i s nigh , he bes i ing slope is -0.006 mag/day, which is consis en wi h ze o. In he second nigh , he i gi es 0.026 mag/day o 0.010 mag/day when igno ing he e y i s ou lying da - apoin . Again his would be consis en wi h ze o o e a ime window o 6 h. The e o in he slope o bo h nigh s amoun s o 0.03 mag/day, conside ably la ge han he slope i sel , as expec ed o a null esul . The e is, howe e , a signi ican o se be ween he mean magni ude in he wo nigh s; he objec dimmed by 3% o e he cou se o 1 d, a 10 σde ec ion gi en he e o o he mean in he ligh cu es in indi idual nigh s. This shi is no seen in indi idual compa ison s a s; he di - e ence be ween he nigh ly a e ages o he ou com- pa ison s a s is 0.55%, 0.12%, 0.60%, and 0.07%. Gi en ha he obse ing se up was iden ical in bo h nigh s, and he ange o ai masses was he same, oo, his dimming is likely caused by in insic a iabili y in ou a ge . We no e ha his o se , in combina ion wi h he linea slopes measu ed o indi idual nigh s, could be consis en wi h a sinusoidal a ia ion – in ha case he objec would be nea maximum in he i s nigh ; he pe iod would be on he o de o days. We ca ied ou he Max-F es desc ibed in Sec ion 4.1, o a pe iod ange om 0.1 h o 6 h in s eps o 0.05 h. The ampli udes we e a ied be ween 0.001 and 0.03 mag, in s eps o 0.001 mag. The esul s a e shown in Figu e 6. Fo he i s nigh , we ind ha he maximum alue o Fis 1.16, o a pe iod o 3.0 h. Fo he second nigh , he maximum Fis 1.12, o pe iods sho e han 1 h; o longe pe iods he maximum is 1.04. Fo compa ison, o N∼80, a F- alue o >1.45 would mean ha he wo a iances be o e and a e sub ac ing he sinewa e a e signi ican ly di e en wi h a alse ala m p obabili y < 6 Scholz e al. 0123456 Pe iod (h) 1.0 1.05 1.10 1.15 1.20 F 0123456 Pe iod (h) Fig. 6.— Resul s o he Max-F es o ou K-band ligh cu e, o nigh 1 on he le , and nigh 2 on he igh . The igu e show he alue o Fas a unc ion o es pe iod, as desc ibed in Sec ion 4.1 and 4.2. 5%. Taken oge he , hese esul s a e consis en wi h he a iances being equal be o e and a e pe iod sub ac ion. I also means ha a mos a ew pe cen o he a iance can be explained by a po en ial pe iod. To es whe he o no an ac ual pe iod wi h an ampli- ude >0.01 mag (1%) can be eco e ed wi h his me hod, we injec ed a i icial pe iods in o ou ligh cu e. We chose a ange o pe iods om 1 o 5 h, and a cons an ampli ude o 1%. Fo hose pa ame e s, he maximum Fwould be ∼1.3, depending sligh ly on pe iod, wi h a clea ly de ined maximum a he injec ed pe iod alue. The injec ed pe iod is hen also isible by eye in he ligh cu e. This demons a es ha pe iods wi h hese cha ac e is ics a e easily eco e ed by ou Max-F es . We can he e o e ule ou he p esence o pe iods o 1-5 h wi h ampli udes la ge han 1% in he K-band ligh cu e. 4.3. WISE ligh cu es In he ALLWISE and NEOWISE-R ligh cu es, SST1624 shows clea a ia ions, domina ed by modula- ions on imescales o days, wi h ypical peak- o-peak ampli udes o 0.1-0.3 mag (see Figu es 4 and 5). These a ia ions a e signi ican ly la ge han he ypical e o , which is 0.03 mag o a single epoch. The e o e i is plausible o assume he sou ce is a iable in he WISE ligh cu es. Taken he wo da ase s oge he , he a i- abili y is sus ained o e mo e han 13 yea s. O e long imescales, he b igh ness o his objec appea s o be mos ly s able wi hin he ma gins o he in a-nigh a i- a ions. The NEOWISE-R da ase allows a mo e de ailed anal- ysis. I consis s o 20 se s o poin s, each measu ed wi hin a couple o days, and sepa a ed by abou 180 d (see Fig- u e 5). Gi en his sampling, he NEOWISE-R ligh cu e is no sensi i e o he sho imescales co e ed wi h ou K-band moni o ing, bu can be use ul o pick up a i- abili y on imescales anging om days o yea s. We ca ied ou he same Max-F es we desc ibed in Sec ion 4.1, o sea ch o a possible pe iodici y. Fo pe i- ods anging om 1 o 10 d, and o he ull NEOWISE-R ligh cu es, he maximum Fis 1.08 in W1 ( o a pe iod o 6 d) and 1.12 in W2 ( o a pe iod o 8 d). We no iced ha he maximum Finc eases signi ican ly i we un he es only on he i s hal o he da ase . Fo band W1, we ob ain peaks a pe iods o 3.6 and 6.1 and >9.0d, wi h F > 1.25. Fo W2, he highes peak is o P= 6.0d, wi h F∼1.25. The esul s o he i s hal o he da ase a e shown in Figu e 8. As a eminde , his means ha he a iance is educed o 80% (1/1.25) when sub ac - ing his speci ic pe iodici y. Howe e , o his alue o F he e is s ill a subs an ial p obabili y (11%) ha he wo a iances be o e and a e sub ac ing he pe iod a e equal (assuming no mally dis ibu ed samples). In Figu e 7 we show he i s hal o he NEOWISE-R da apoin s plo ed in phase o P= 6.0d. As can be app ecia ed om his igu e, he pe iodici y looks plau- sible by eye. I is he e o e concei able ha pa s o all o he a iabili y seen in WISE da a is caused by a lux modula ion wi h a pe iod o abou 6 d and an ampli ude o ±0.05 mag. We show he Max-F esul s in Figu e 8, wi h he pu a i e pe iod o 6.0 d ma ked. Va iabili y in SSTc2d J163134.1-240100 7 0.0 0.2 0.4 0.6 0.8 1.0 Phase -0.2 -0.1 0.0 0.1 0.2 W1 ela i e magni ude 0.0 0.2 0.4 0.6 0.8 1.0 Phase -0.2 -0.1 0.0 0.1 0.2 W2 ela i e magni ude Fig. 7.— The i s hal o he NEOWISE-R da ase , plo ed in phase o a pe iod o P= 6.0d. O e plo ed is he sinecu e wi h an ampli ude o 0.05 mag o guide he eye. The ypical pho ome ic e o is 0.03 mag. To quan i y ypical imescales o a iabili y om he NEOWISE-R ligh cu e, we calcula ed he di e ence be- ween each pai o da apoin s ∆M, and plo his o e hei ime di e ence ∆ (see Figu e 9). This analysis was ca ied ou o he wo bands sepa a ely, bu hey bo h gi e e y simila esul s. As can be seen in he ig- u e, he o e all ange o ∆Mis no a s ong unc ion o ime di e ence ∆ . The median o ∆M(shown as o ange squa es) is 0.04-0.05 o he sho es ∆ (hou s o days), sligh ly la ge han he pho ome ic e o o 0.03 mag. The median o ∆Minc eases g adually o 0.07-0.10 up o ∆ ∼1000 d, whe e i pla eaus. The 80% pe cen ile o ∆M(shown wi h o ange iangles poin ing downwa ds) is by and la ge cons an a 0.1 mag o W1 and close o 0.2 mag o W2. O e all, he e is no s ong e idence o long- e m a iabili y om he NEOWISE-R da ase . 5. SUMMARY AND DISCUSSION SST1624 is a b own dwa wi h quasi-sphe ical mass loss, a unique objec ha may gi e us insigh s in o im- po an aspec s o he ea ly e olu ion o subs ella ob- jec s. Two possible mechanisms o explain he mass loss a e a he mal pulse om he onse o Deu e ium bu ning o cen i ugal winds due o e y as o a ion. In bo h cases, we may expec a iabili y in sho imescales, ei- he due o pulsa ion o o a ion. We p esen he e in a ed ligh cu es o SST1624 om newly ob ained high-cadence K-band ESO/NTT obse - a ions and om he publicly a ailable ALLWISE and NEOWISE-R a chi es. O e all, he e is clea e idence o a iabili y in SST1624, on imescales o >6h o days, measu ed a 2-4.5 µm in all a ailable da ase s. On he o he hand, we ule ou a iabili y, and speci ically he p esence o a pho ome ic pe iod on imescales <6h, o ampli udes >1%. We do iden i y a en a i e pho- ome ic pe iod o 6 d (wi h a plausible ange o 3-7 d) in he i s hal o he NEOWISE-R da ase , seen a 3.6 and 4.5 µm. The e may be addi ional a iabili y on imescales o yea s. We no e ha ou a ge was included in he a iabil- i y sea ch by Pa k e al. (2021) using he i s 6.5 y o NEOWISE da a – no a iabili y was ound. This s udy, howe e , was only ocused on long- e m a ia ions o e imescales longe han days. Also, he NEOWISE-R da ase is now signi ican ly longe . The a iabili y on day-long imescales, and he en a- i e pe iod o 6 d, is bes explained by magne ic ac i i y, i.e. spo s on he su ace co- o a ing wi h he objec . This implies ha he o a ion pe iod o his objec is in he o de o a ew days. Pe iod and ampli ude a e compa- able o wha has been ound o young b own dwa s – pe iods ypically ange om a ew hou s o se e al days, ampli udes a e 2% o la ge (Scholz & Eislö el 2005; Moo e e al. 2019). The WISE ligh cu e is e y clumpy, wi h gaps o 6 mon hs be ween clumps o da a. O e such long imescales, he spo dis ibu ion can change, lead- ing o phaseshi s and ampli ude changes. Fo imescales a ound 1 d, he numbe o da apoin s is e y low. These wo cha ac e is ics in combina ion hampe any pe iod sea ch and can easily explain why he pe iod is no ap- pa en in he en i e da ase . To obus ly con i m he en- 8 Scholz e al. 0246810 Pe iod (d) 1.0 1.1 1.2 1.3 F 0246810 Pe iod (d) Fig. 8.— Resul s o he Max-F es o he NEOWISE ligh cu e (W1 on he le , W2 on he igh ; in bo h cases only he i s hal o he da apoin s was used). The igu e show he alue o Fas a unc ion o es pe iod, as desc ibed in Sec ion 4.1 and 4.3. a i e pe iod o 6 d as he o a ion pe iod, a dedica ed moni o ing campaign o e a leas one week, i possible wi hou day ime gaps, would be needed. The only o he concei able explana ion o he ob- se ed a iabili y is he p esence o some esidual ci cum- sub-s ella dus close o he objec ha causes a iabili y h ough obscu a ion (Scholz e al. 2009) – howe e , he lack o an in a ed excess and he lack o a mm con inuum de ec ion ende s his in e p e a ion unlikely. Coming back o he opic o he quasi-sphe ical mass loss, he unique ea u e o his objec : Wi h ou dedi- ca ed K-band moni o ing, we se ou o ind pe iods on imescales o hou s, o dis inguish be ween pulsa ions (1- 4 h) and o a ion nea b eakup (>5h). Since we did no ind pe iods on hese imescales, he a ailable da a does no gi e us a de ini i e answe on he cause o he mass loss. Ha ing said ha , he ac ha he e is clea a iabili y o e imescales o days, bes explained by o- a ional modula ions, would imply ha he objec is in ac no o a ing close o b eakup. Ro a ion pe iods o se e al days a e a he high end o he pe iod dis ibu- ion o young b own dwa s, see Figu e 13 in Moo e e al. (2019). This ende s he scena io whe e he mass loss is due o he onse o Deu e ium bu ning, a ou ed and wo ked ou by Ruíz-Rod íguez e al. (2022), mo e likely. The absence o a pulsa ion pe iod in he ligh cu es could mean ha he objec has s opped pulsa ing since he ejec ion o he shell, o ha he pulsa ions happen a ampli udes below 1%, as i is likely he case o pulsa - ing M dwa s (Rod íguez e al. 2016; Rod íguez-López 2019). This b own dwa dese es u he ime-se ies obse - a ions o a) e i y he o a ion pe iod and b) sea ch o pulsa ions wi h highe p ecision. As he i s b own dwa wi h obse ed quasi-sphe ical mass loss, his sou ce may p esen an oppo uni y o lea n mo e abou he hi h- e o unexplo ed aspec s in he ea ly e olu ion o b own dwa s. Deu e ium bu ning is a common ea u e in all young b own dwa s, and i his leads o mass loss, ei- he sus ained, o in bu s s, i would a ec he subs ella mass unc ion and angula momen um e olu ion. Ex- plo ing he p e alence and du a ion o subs ella mass loss is an impo an ask o u u e wo k. We hank he e e ee o a p omp and ho ough epo on his pape . We hank he ESO eam a he NTT o he suppo p io and du ing he obse - ing un. This publica ion makes use o da a p od- uc s om he Wide- ield In a ed Su ey Explo e , which is a join p ojec o he Uni e si y o Cali- o nia, Los Angeles, and he Je P opulsion Labo a- o y/Cali o nia Ins i u e o Technology, unded by he Na ional Ae onau ics and Space Adminis a ion. AS ac- knowledges suppo om he UKRI Science and Tech- nology Facili ies Council h ough g an ST/Y001419/1/. KM acknowledges suppo om he Fundação pa a a Ciência e a Tecnologia (FCT) h ough he CEEC- indi idual con ac 2022.03809.CEECIND and esea ch g an s UIDB/04434/2020 and UIDP/04434/2020. REFERENCES As opy Collabo a ion, Robi aille, T. P., Tolle ud, E. J., e al. 2013, A&A, 558, A33 As opy Collabo a ion, P ice-Whelan, A. M., Sipőcz, B. M., e al. 2018, AJ, 156, 123 Va iabili y in SSTc2d J163134.1-240100 9 0 500 1000 1500 2000 2500 3000 Time di e ence (d) 0.02 0.05 0.1 0.2 0.4 W1 di e ence 0 500 1000 1500 2000 2500 3000 Time di e ence (d) 0.02 0.05 0.1 0.2 0.4 W2 di e ence Fig. 9.— Di e ences be ween NEOWISE-R magni udes, as a unc ion o ime di e ence, o he wo bands sepa a ely (W1 in he igh panel, W2 in he le panel). The ed squa es show he median o an in e al o ±10 d a ound a gi en ime di e ence. Simila ly, he upwa ds and downwa ds poin ing o ange iangles show he 20% and 80% pe cen iles o he same in e als. The ypical e o in he magni ude di e ence is 0.04 mag. Bou ie , J., Ma , S. P., Mohan y, S., e al. 2014, P o os a s and Plane s VI, 433. doi:10.2458/azu_uap ess_9780816531240-ch019 B adley, L., Sipőcz, B., Robi aille, T., e al. 2023, Zenodo Caballe o, J. A., Béja , V. J. S., Rebolo, R., e al. 2004, A&A, 424, 857. doi:10.1051/0004-6361:20047048 Cieza, L. A., Ruíz-Rod íguez, D., Hales, A., e al. 2019, MNRAS, 482, 698. doi:10.1093/mn as/s y2653 Cody, A. M. & Hillenb and, L. A. 2014, ApJ, 796, 129. doi:10.1088/0004-637X/796/2/129 Cu i, R. M., W igh , E. L., Con ow, T., e al. 2021, VizieR Online Da a Ca alog, II/328 E ans, N. J., Dunham, M. M., Jø gensen, J. K., e al. 2009, ApJS, 181, 321. doi:10.1088/0067-0049/181/2/321 Ke schbaum, F., Mae cke , M., B unne , M., e al. 2017, A&A, 605, A116. doi:10.1051/0004-6361/201730665 Mainze , A., Baue , J., Cu i, R. M., e al. 2014, ApJ, 792, 30. doi:10.1088/0004-637X/792/1/30 Me che , S. A., Heinze, A., Apai, D., e al. 2015, ApJ, 799, 154. doi:10.1088/0004-637X/799/2/154 Moo e, K., Scholz, A., & Jayawa dhana, R. 2019, ApJ, 872, 159. doi:10.3847/1538-4357/aa 5c Palla, F. & Ba a e, I. 2005, A&A, 432, L57. doi:10.1051/0004-6361:200500020 Pa k, W., Lee, J.-E., Con e as Peña, C., e al. 2021, ApJ, 920, 132. doi:10.3847/1538-4357/ac1745 Po e , J. M. 1996, MNRAS, 280, L31. doi:10.1093/mn as/280.3.L31 Ricci, L., Tes i, L., Na a, A., e al. 2014, ApJ, 791, 20. doi:10.1088/0004-637X/791/1/20 Rod íguez-Ledesma, M. V., Mund , R., & Eislö el, J. 2009, A&A, 502, 883. doi:10.1051/0004-6361/200811427 Rod íguez-López, C., MacDonald, J., & Moya, A. 2012, MNRAS, 419, L44. doi:10.1111/j.1745- 3933.2011.01174.x10.5479/ADS/bib/1912LicOB.7.102C Rod íguez, E., Rod íguez-López, C., López-González, M. J., e al. 2016, MNRAS, 457, 1851. doi:10.1093/mn as/s w033 Rod íguez-López, C. 2019, F on ie s in As onomy and Space Sciences, 6, 76. doi:10.3389/ spas.2019.00076 Ruíz-Rod íguez, D. A., Cieza, L. A., Casassus, S., e al. 2022, ApJ, 938, 54. doi:10.3847/1538-4357/ac8 5 Sanchis, E., Tes i, L., Na a, A., e al. 2020, A&A, 633, A114. doi:10.1051/0004-6361/201936913 Salpe e , E. E. 1992, ApJ, 393, 258. doi:10.1086/171502 Scholz, A. & Eislö el, J. 2004, A&A, 419, 249. doi:10.1051/0004-6361:20034022 Scholz, A. & Eislö el, J. 2005, A&A, 429, 1007. doi:10.1051/0004-6361:20041932 Scholz, A., Xu, X., Jayawa dhana, R., e al. 2009, MNRAS, 398, 873. doi:10.1111/j.1365-2966.2009.15021.x Tannock, M. E., Me che , S., Heinze, A., e al. 2021, AJ, 161, 224. doi:10.3847/1538-3881/abeb67 Tes i, L., Na a, A., Scholz, A., e al. 2016, A&A, 593, A111. doi:10.1051/0004-6361/201628623 Vos, J. M., Fahe y, J. K., Gagné, J., e al. 2022, ApJ, 924, 68. doi:10.3847/1538-4357/ac4502 W igh , E. L., Eisenha d , P. R. M., Mainze , A. K., e al. 2010, AJ, 140, 1868. doi:10.1088/0004-6256/140/6/1868 Yoshida, S. 2023, MNRAS, 518, 1484. doi:10.1093/mn as/s ac3143 Zapa e o Oso io, M. R., Caballe o, J. A., Béja , V. J. S., e al. 2003, A&A, 408, 663. doi:10.1051/0004-6361:20030987 This pape was buil using he Open Jou nal o As- ophysics L A T EX empla e. The OJA is a jou nal which p o ides as and easy pee e iew o new pape s in he as o-ph sec ion o he a Xi , making he e iewing p o- cess simple o au ho s and e e ees alike. Lea n mo e a h p://as o. heoj.o g.