Varying linear polarisation in the dust-free gamma-ray burst 210610B

A&A, 690, A216 (2024) h ps://doi.o g/10.1051/0004-6361/202348572 c The Au ho s 2024 As onomy & As ophysics Va ying linea pola isa ion in he dus - ee gamma- ay bu s 210610B J. F. Agüí Fe nández1,?, A. de Uga e Pos igo2,3 , C. C. Thöne4, S. Kobayashi5, A. Rossi6, K. Toma7,8 , M. Jelínek4, D. A. Kann1,9, S. Co ino10 , K. Wie sema11, D. Ha mann12 , P. Jakobsson13, A. Ma in-Ca illo14, A. Meland i15,10, M. De Pasquale16, G. Pugliese17, S. Sa aglio6,18,19 , R. L. C. S a ling20, J. Š obl4, M. Della Valle21, S. de We 22, and T. Za a 23,24 (A ilia ions can be ound a e he e e ences) Recei ed 12 No embe 2023 /Accep ed 23 Feb ua y 2024 ABSTRACT Con ex . Long gamma- ay bu s s (GRBs) a e p oduced by he collapse o some e y massi e s a s, ha emi ul a- ela i is ic je s. When he je s collide wi h he in e s ella medium hey decele a e and gene a e he so-called a e glow emission, which has been obse ed o be pola ised. Aims. We s udy he pola ime ic e olu ion o he GRB210610B a e glow, a z=1.1341. This allows us o e alua e he ole o geome ic and/o magne ic mechanisms in he GRB a e glow pola isa ion. Me hods. We obse ed GRB210610B using imaging pola ime y wi h CAFOS on he 2.2 m Cala Al o Telescope and FORS2 on he 4 ×8.1m Ve y La ge Telescope. Complemen a y op ical spec oscopy was ob ained wi h OSIRIS on he 10.4m G an Telescopio Cana ias. We s udied he GRB ligh -cu e om X- ays o he op ical bands and he Spec al Ene gy Dis ibu ion (SED). This allowed us o s ongly cons ain he line-o - sigh ex inc ion. Finally, we s udied he GRB hos galaxy using op ical o NIR da a o i he SED and de i e i s in eg a ed p ope ies. Resul s. GRB210610B had a b igh a e glow wi h a negligible line-o -sigh ex inc ion. Pola ime y was ob ained a h ee epochs: du ing an ea ly pla eau phase, a he ime when he ligh cu e b eaks, and a e he ligh cu e s eepened. We obse e an ini ial pola isa ion o ∼4% ha goes o ze o a he ime o he b eak, and i hen again inc eases o ∼2%, wi h a change in he posi ion angle o 54 ±9deg. The spec um shows ea u es wi h e y low equi alen wid hs. This indica e a small amoun o ma e ial in he line o sigh wi hin he hos . Conclusions. The lack o dus and he low amoun o ma e ial in he line o sigh o GRB210610B allowed us o s udy he in insic pola isa ion o he GRB op ical a e glow. The GRB pola isa ion signals a e consis en wi h o de ed magne ic ields in e eshed shock o /and hyd odynamics- scale u bulen ields in he o wa d shock. Key wo ds. pola iza ion – echniques: pola ime ic – gamma- ay bu s : indi idual: GRB210610B 1. In oduc ion Gamma- ay bu s s (GRBs) a e among he mos ene ge ic elec- omagne ic explosions ha ha e been obse ed in he Uni e se. These e en s ha e wo main emission episodes: he p omp - emission and he a e glow-emission phases. The p omp emis- sion ep esen s he i s obse able elec omagne ic emission and is domina ed by gamma- ay pho ons las ing seconds o minu es a e he onse o he bu s . The a e glow has a synch o on spec- um anging om adio o gamma- ays and e ol es in ime du - ing much longe ime spans. GRBs a e ypically classi ied as sho o long acco ding o hei measu ed T901du a ion in gamma- ays (Kou elio ou e al. 1993). Sho GRBs (sGRB) a e associa ed wi h he coales- cence o wo compac objec s, hei T90 du a ion is ypically sho e han 2seconds, and hei X- ay spec um is ha d. The disco e y o a sGRB associa ed wi h he g a i a ional wa e (GW) de ec ion GW170817 de ini i ely linked a sGRB wi h he me ge o wo neu on s a s (Abbo e al. 2017). On he o he hand, long GRBs (lGRB) ha show T90 longe han 2seconds and ha e a so e spec a in he p omp emission phase. These a e ca aclysmic e en s associa ed wi h he collapse o massi e s a s. They a e also associa ed wi h he de ec ion o b oad-line ?Co esponding au ho ; [email p o ec ed] 1The T90 is he ime span wi hin which a GRB eleases be ween 5% and 95% o he o al ene gy du ing he p omp phase. (BL) ype Ic supe no ae (SNe) (see e.g. Galama e al. 1998; Hjo h e al. 2003). La ely, se e al e en s de ec ing a kilono a (KN) emission in a bu s wi h a du a ion o dozen seconds ha e cas doub s on he 2-seconds di ision as he unique c i e- ion o dis inguish be ween sGRBs and lGRBs (Ras inejad e al. 2022;T oja e al. 2022;Yang e al. 2022;Gompe z e al. 2023; Le an e al. 2023,2024). The p omp emission in a GRB is powe ed by a newly o med compac objec ha is ed by he su ounding ma e ial. Acc e ion on o his compac objec launches ul a- ela i is ic je s, in which he p omp emission is gene a ed h ough in e - nal dissipa ion p ocesses including in e nal shocks (see e.g. Rees & Mesza os 1994;Sa i & Pi an 1997;Kobayashi e al. 1997). This p omp high-ene gy emission eleases iso opic- equi alen ene gies ha can each up o 1055 e g (see e.g. Bu ns e al. 2023), al hough he eal eleased ene gy can be se - e al o de s o magni ude lowe due o he je collima ion. Pola ime y is an essen ial ool o explaining GRB physics. Using his echnique, we can es models ha include magne ic ields and he geome ical cha ac e is ics ha a e a play, and we can de e mine how hey e ol e h oughou he di e en phases o he GRB. The s udy o he p omp emission pola isa ion and i s empo al e olu ion can help us o unde s and how he je is powe ed by he cen al engine. The long and ex emely b igh GRB221009A was obse ed du ing he p omp - and a e glow- emission phase by he Imaging X- ay Pola ime y Explo e Open Access a icle, published by EDP Sciences, unde he e ms o he C ea i e Commons A ibu ion License (h ps://c ea i ecommons.o g/licenses/by/4.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. This a icle is published in open access unde he Subsc ibe o Open model.Subsc ibe o A&A o suppo open access publica ion. A216, page 1 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) (IXPE) and only uppe limi s we e ob ained (Neg o e al. 2023). S udies o he p omp -emission pola isa ion wi h ins umen cal- ib a ed o pe o m pola ime y a e a e so a . The As osa mis- sion wi h he on-boa d Cadmium-Zinc-Tellu ide Image (CZTI) ins umen shows he p omp emission as highly pola ised, while o POLAR, GRBs a e lowly pola ised o no pola ised a all (see e.g. Gill e al. 2021). The ma e ial ejec ed h ough he je s e en ually collides wi h he ci cumbu s ma e ial, decele a ing as in e ac s wi h i . These o wa d shocks gene a e a b oadband synch o on emission ha is known as he a e glow. This b igh shock can be obse ed du ing days o e en mon hs in he case o adio equencies. In ce ain cases, a e e se shock ha p opaga es backwa ds wi hin he ela i is ic je can be obse ed a ea ly imes (see e.g. Mészá os & Rees 1997;Pi an 1999). Di e en models p edic his emission o be pola ised. G uzino & Waxman (1999) p o- posed a model in which he a e glow would show pa ches wi h locally o de ed magne ic ields andomly o ien ed bu wi h many o hem sha ing a common di ec ion leading o a global low le el o pola isa ion in he GRB a e glow. The numbe o s udies using op ical linea imaging pola ime- y and spec opola ime y echniques is s ill e y small (see e.g. Co ino & Go z 2016). The i s GRB wi h a pola isa ion de ec- ion in he a e glow was GRB990510. I s pola isa ion deg ee (PD) was 1.7%, 0.7days a e de ec ion (Co ino e al. 1999; Wije s e al. 1999). Many e o s ha e been made o inc ease he sample o pola ime ic measu emen s, bu he ain ness o hese objec s, hei as e olu ion, and he long obse ing imes equi ed o a good signal- o-noise a io (S/N) lea e us wi h less han 20 GRBs wi h a measu emen abo e 3σ. Fas eac- ion is c ucial o obse ing he i s GRB phases in which a highe PD migh be expec ed. The highes measu ed alue was o GRB120308A wi h a PD o 28% 5 minu es a e he bu s , which apidly dec eased o 16% as he a e glow e ol ed (Mundell e al. 2013). A high PD o ∼10% was also measu ed o GRB020405, GRB090102 and GRB091208B (Be sie e al. 2003;S eele e al. 2009;Ueha a e al. 2012). All o hem we e obse ed ea lie han 0.01 days a e he GRB de ec ion. Obse - a ions a la e s ages shows his sou ces as lowly-pola ised wi h some deg ee o a iabili y h oughou hei ligh -cu e e olu ion. The mos de ailed example is GRB030329 (G eine e al. 2003) whose PD e ol es somewha andomly om 0.91% a 0.5days o 1.4% 37.5days a e he bu s . A e glow pola isa ion mus be unde s ood oge he wi h he su ounding en i onmen . Dus in he line o sigh can change he obse ed PD and p e en he measu emen o he in insic pola isa ion (Lazza i e al. 2003). This measu emen by i sel is also incomple e when i is no ollowed by he s udy o he ligh - cu e e olu ion. I is c ucial o dis inguishing be ween he mod- els ha ha e been p oposed o explain he GRB je physics o unde s and whe he he e is a b eak in he ligh cu e and how he pola isa ion beha es be o e, du ing and a e his b eak (see e.g. Co ino & Go z 2016). We p esen a comp ehensi e s udy o GRB210610B, i s a e glow emission and i s en i onmen using di e en ech- niques. Pola ime ic, spec al and pho ome ic obse a ions we e secu ed o his b igh long GRB. We also obse ed he pu a- i e hos galaxy in o de o cha ac e ise i and place he GRB in o con ex . This wo k is s uc u ed as ollows: In Sec ion 2we p esen he obse a ions o bo h, he a e glow and he unde ly- ing galaxy. In Sec ion 3we p esen he esul s o he analysis o he linea pola ime y o he a e glow, he ligh cu e and i s spec um, as well as he analysis o he galaxy. In Sec ion 4 we discuss he esul s by se ing he amewo k in o which GRB210610B is embedded. We also place he esul s o he GRB a e glow pola ime y measu emen s in o con ex . Finally, in Sec ion 5we p esen he conclusions. Th oughou his s udy, we desc ibe he spec al and empo al e olu ion o he da a using he con en ion by which Fν∝ −αν−β. We adop a cosmological model wi h H0=67.3 km s−1Mpc−1, ΩM=0.315 and ΩΛ=0.685 (Planck Collabo a ion XVI 2014). 2. Obse a ions 2.1. High-ene gy da a GRB210610B was i s de ec ed by he Gamma- ay Bu s Mon- i o (GBM) on boa d he Fe mi obse a o y a 19:51:05.05UT o 10 June 2021 wi h a T90 =55.04±0.72s and a luence o (1104.2±0.5) ×10−5e gcm−2in he 10keV o 10MeV band (Malaca ia e al. 2021; on Kienlin e al. 2020). A he edshi o he GRB (see Sec . 3.3), he compu ed iso opic ene gy is Eiso, es =4.17+0.02 −0.02 ×1053 e g in he 0.1keV o 10MeV band. This alue is ully consis en wi h he so-called Ama i ela- ion (Ama i e al. 2002;Ama i 2006) o long GRBs. The bu s was de ec ed ∼22s la e by he Bu s Ale Telescope (BAT; Ba helmy e al. 2005) on boa d he Neil Geh els Swi Obse a- o y wi h coo dina es RA=16h15m45s, Dec= +14◦2302900 and an unce ain y o 30. The X-Ray Telescope (XRT; Bu ows e al. 2005) s a ed obse ing he sou ce 89.9s a e BAT and quasi-simul aneously, he Ul a iole /Op ical Telescope (UVOT; Roming e al. 2005) obse ed he ield in he Whi e band. The image showed a new b igh sou ce wi hin he XRT posi ion wi h a magni ude in he Swi /UVOT na i e sys em o 13.70±0.14, unco ec ed o galac ic ex inc ion (Page e al. 2021). The la e analysis, p esen ed in Siegel e al. (2021), upda ed his alue o 13.63±1.10 and he sou ce loca ion o RA=16h15m40.40s, Dec= +14◦23056.900 wi h an unce ain y o 000 .42. In he e ined analysis om he BAT da a, he bu s had a T90 du a ion o T90 =69.38±2.53s in he 15 o 350keV band wi h an Epeak =339.3 ±218.6 keV(K imm e al. 2021). The bu s showed a ha dness a io (HR)2o 1.78±0.04. This alls in he uppe pa o he bulk o he HR dis ibu ion o long GRBs (see e.g. Lien e al. 2016). The bu s was also de ec ed by Konus- Wind obse a o y wi h a du a ion ∼100s (F ede iks e al. 2021). The HR o he bu s , i s T90, and i s iso opic ene gy elease clas- si y GRB210610B as a long GRB. 2.2. Linea pola ime y imaging We obse ed he GRB210610B ollowing he Swi /BAT ale wi h wo di e en ins umen s in linea pola ime y mode. We i s obse ed i wi h he Cala Al o Fain Objec Spec o- g aph (CAFOS) moun ed on he 2.2me e elescope a Cala Al o Obse a o y3. The obse a ions s a ed 0.08days a e he Swi /BAT de ec ion wi h an exposu e pe hal -wa e pla e (HWP) posi ion angle (see below) o 900s in Rband. A second obse a ion was ob ained wi h he FOcal Reduce and low dis- pe sion Spec og aph (FORS2) moun ed on he Uni Telescope 1 o he Ve y La ge Telescope (VLT) o he Eu opean Sou h- e n Obse a o y (ESO)4. The VLT/FORS2 obse a ions s a ed 2The ha dness a io o spec al ha dness (Kou elio ou e al. 1993) o a GRB is he a io be ween he luence emi ed du ing he p omp emis- sion in he 50–100keV band o e he 25–50keV band. 3Obse a ions we e ob ained wi h p og amme F21-2.2-021 (PI: Agüí Fe nández, J. F.). 4Obse a ions we e ob ained wi h p og amme 106.21T6.003 (PI: Tan- i , N.). A216, page 2 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) 0.24days a e he GRB ale in RSpecial,bHigh, and zSpecial bands wi h 180s exposu e ime pe image and wi h wo cycles wi h he same con igu a ion. Fo bo h ins umen s, obse a ions we e ob ained using he HWP in ou di e en o a ion angles o 0.0◦, 22.5◦, 45.0◦, and 67.5◦and wi h a Wollas on p ism o spli he ligh in o he o di- na y (o) and ex ao dina y (e) beams. A mask was se o a oid o e lapping o he ligh om he wo beams. To calib a e he CAHA/CAFOS obse a ions, one high- and one ze o-pola ised s anda d s a we e obse ed ollowing he same p ocedu e as o he GRB. The la - ields in he co espond- ing band we e ob ained wi h he ull op ical se up in he ligh pa h. Fo he VLT/FORS2 da a, calib a ion and s anda d obse - a ions we e pe o med as speci ied in he FORS2 Use Manual (Ande son 2015). We ob ained a second epoch wi h VLT/FORS2 in RSpecial ∼ 1day. We inc eased he exposu e ime pe image up o 300s o accoun o he ading o he sou ce by pe o ming wo cycle o his obse a ions. 2.3. Pho ome y Pho ome ic obse a ions we e pe o med using se e al ins u- men s a di e en obse a o ies and in mul iple bands. In his sec ion, we desc ibe he de ails o each o he obse a ions. The measu ed pho ome y is compiled in Table A.1. 2.3.1. Small Binocula Telescope The Small Binocula Telescope (SBT), loca ed a he Ondˇ e- jo Obse a o y, obse ed GRB210610B a dusk wi h he wo 20cm New onian as og aphs moun ed on a common moun . The main de ec o s ha e 4096×4096pixel CCDs ha p o ide a ield o iew o 3.5◦×3.5◦wi h a 3.1400 sampling. Ope a ions a e designed such ha he eadou ime o one came a equals he exposu e ime o he o he came a o a oid any blind ime du ing he obse a ion. The SBT obse a ions s a ed ∼860s a e he bu s . Du ing he ea ly ollow-up phase, 12s exposu es we e aken wi hou a il e . A e wa ds, he obse a ions consis ed o up o 34 expo- su es o 120s each in SDSS- 0band (see Table A.1 o u he de ails). Obse a ions we e in e up ed a 23:47UT, almos 4 h a e he GRB de ec ion. The a e glow was no de ec ed in single exposu es, and we he e o e s acked he images o ob ain an accep - able (S/N). We used mon age o ob ain a weigh ed image co-addi ion simila ly o Mo gan e al. (2008), which op i- mises o he a iable backg ound caused by dusk and hick ci us. As o he images in 0band, pho ome ic calib a ion was pe o med agains he A las ca alogue, which uses he Pan- STARRS ca alogue o ain a ge s. Images in he clea band we e calib a ed agains he 0band and a polynomial co ec ion in ol ing pho ome ic colou s g0− 0, 0−i0, and i0−z0was pe - o med. Pho ome ic measu emen s ob ained his way a e shi ed compa ed o he o iginal s anda d 0band by a co ec ion ha depends on he objec colou . This co ec ion can be compu ed om he i ed ela ion when he pho ome ic colou s o he a e - glow a e known. By i ing he comple e pho ome ic se wi h an empi ical b oken powe law we, de e mined his co ec ion o be kc− =−0.0045mag. The e o e, unde he assump ion ha he a e glow shows no colou e olu ion, we a e able o con- e he un il e ed alues om SBT in o 0by simply sub ac ing kc− . 2.3.2. FRAM-ORM elescope The 25cm elescope FRAM-ORM, is ope a ed as pa o CTA-N a Roque de los Muchachos a La Palma, Spain. The elescope is equipped wi h a 1024 ×1024 pixel CCD de ec o and a Bessel il e se , which p o ides a ield o iew o 260×260wi h a pixel scale o 1.500. FRAM-ORM obse ed he GRB s a ing a 20:49 UT (1 h a e igge ) and ob ained a o al o 83×60s images in Rband using a o al o 1.5h o obse ing ime. Images we e educed using s anda d p ocedu es and combined in o g oups in o de o p o ide a good (S/N). Pho ome y was pe o med in a simila way as o he SBT un il e ed images, using 0band and 0−i0 band ca alogue alues. The de i ed co ec ion ac o is kR− = −0.036mag. 2.3.3. Telescope D50 The D50 Telescope is a 0.5m New onian obo ic elescope loca ed a Ondˇ ejo Obse a o y. I has a 1024 ×1024 EMCCD came a wi h a ield o iew 200×200, scaled a 1.1800 pe pixel. The elescope is equipped wi h an SDSS il e se . D50 s a ed obse ing 0.06days a e he bu s de ec ion, s a ing wi h SDSS- 0band and ollowed by a se o obse a ions in SDSS-g0, 0,i0, and z0bands using di e en exposu e imes (see Table A.1 o u he de ails). Du ing he i s nigh , he ele- scope spen 3.2h on a ge and s opped a 01:05UT. The e we e addi ional obse a ions in nigh 2, 4, 5, and 7, collec ing 16.75h o u he ollow-up da a. All images we e p ocessed in a s anda d manne pe o ming da k sub ac ion, la - ield co ec ion, and inge emo al o i0 and z0, and he images we e co-added when necessa y. Pho ome- y was pe o med using he SDSS ca alogue in he co espond- ing band. The GRB a e glow emission is de ec ed e en a he la es epoch. 2.3.4. CAHA/CAFOS In addi ion o he pola ime ic obse a ion, we pe o med u he imaging o he a e glow in he second nigh wi h CAFOS on he 2.2m elescope o he Cala Al o Obse a o y. The images we e co ec ed using bias and la ields using s anda d p ocedu es in IRAF. The obse a ions we e pe o med wi h he RC il e and calib a ed wi h espec o PanSTARRS ield s a s, o which a il e co ec ion was used o de i e AB magni udes in he RC band. These magni udes we e hen ans o med in o he 0band using he colou in o ma ion ha we ha e om he a e glow. The inal alues a e shown in Table A.1. 2.3.5. GTC/HiPERCAM The ield o GRB210610B was obse ed in ou epochs using he HiPERCAM mul i-band image (Dhillon e al. 2021) moun ed on he 10.4m G an Telescopio de Cana ias (GTC) a Roque de los Muchachos Obse a o y (La Palma, Spain) using p og amme GTCMULTIPLE2C-21A (PI: de Uga e Pos igo). HiPERCAM simul aneously ob ains obse a ions in he i e SDSS il e s (u0,g0, 0,i0,z0) using e icien dich oic beam- spli e s and mul iple came as. The las o ou obse a ions was ob ained almos wo mon hs a e he bu s when he emission was domina ed by he hos galaxy (see Fig. 1). The da a educ ion was pe o med using an au oma ic shell sc ip ha inds and o ganises he iles, calls commands om he HiPERCAM pipeline o pe o m bias and la co ec ions and A216, page 3 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) Fig. 1. Colou images o he ield using he g0, 0and i0bands o HiPER- CAM a 1day and 58 days a e he bu s , espec i ely. In he i s image he a e glow is s ongly de ec ed, in he second image, i has aded and he hos galaxy domina es he emission. 2.7a csec no h-wes o he hos galaxy lies a companion galaxy a he same pho ome ic edshi o he hos (see Sec . 3.4). con e s he HiPERCAM one-dimension i s iles in o classical wo-dimension i s images. Fu he IRAF p ocedu es allow us o ob ain an e en backg ound om he di e en quad an s o each de ec o . Finally, he images a e egis e ed and combined using SWARP (Be in 2010). Pho ome y was pe o med wi h ape u e pho ome y using e e ence ield s a s om he PanSTARRS ca - alogue. Fo he las epoch, we used he same ape u e o compa e he esul s o he es o he da a. Addi ionally, we pe o med pho ome y o he comple e hos galaxy adap ing he ape u e o i s ligh , and in a simila way, we ob ained pho ome y o a nea by objec no h-wes o he hos , which we iden i ied as a companion galaxy a he same edshi (see Sec . 3.4). 2.3.6. Pe ek 2 m elescope The Pe ek 2.0m elescope a he Ondˇ ejo Obse a o y obse ed GRB210610B a e glow 6days a e he bu s in SDSS-g0band. The pho ome ic came a o his elescope has a 1092×736pixel CCD wi h a ield o iew o 50×70scaled a 0.400/pixel. A e s anda d imaging da a educ ion and a e imaging co-addi ion, pho ome y was pe o med as o he D50 elescope. The GRB a e glow is well de ec ed. 2.3.7. GTC/EMIR La e- ime nea -in a ed (NIR) obse a ions we e pe o med on 19 Feb ua y 2022 in sea ch o he hos galaxy o GRB210610B wi h he EMIR ins umen (Ga zón e al. 2022) moun ed on he 10.4m GTC elescope wi h p og amme GTCMULTIPLE2H- 21B (PI: de Uga e Pos igo). The obse a ion consis ed o a o al exposu e o 349 ×3s in Hband. The da a educ ion was pe - o med using a sel -made pipeline ha co ec s la ields, does backg ound sub ac ion, bad-pixel masking, alignmen and com- bina ion o he ames. The hos galaxy was no de ec ed in he inal ame down o a 3-σlimi ing magni ude o 22.9mag. 2.4. Spec oscopy We obse ed GRB210610B using he long-sli mode o he Op ical Sys em o Imaging and low Resolu ion In eg a ed Spec- oscopy (OSIRIS) (Cepa e al. 2000) moun ed on he 10.4m GTC. The obse a ion consis ed o 3×900s exposu es wi h g ism R1000B and a sli wid h o 100 o ien ed a he pa allac ic angle5. The mean epoch o he obse a ion was 11 June 2021 a 01:59:14UT (6.12972h a e he Swi igge ) a a mean ai mass o 1.13. Flux calib a ion was pe o med ela i e o an obse a ion o he Ross 640 (Oke 1974) spec opho ome ic s anda d, obse ed a he beginning o he same nigh and using he same g ism. The a e glow spec um shows a e y s ong con inuum, wi h a median (S/N) o ∼100 pe dispe sion elemen and weak abso p- ion ea u es. 3. Resul s 3.1. Linea pola ime ic analysis Fo all linea pola ime y da a, we pe o med egula imaging- educ ion p ocesses. Images we e bias sub ac ed and la - ield co ec ed using PyRAF asks (Science So wa e B anch a STScI 2012). The la ield co ec ion o VLT/FORS2 was pe - o med as speci ied in González-Gai án e al. (2020). Fo CAHA/CAFOS, we combined he la ields using PyRAF, and we used a cus omized Py hon sc ip o sepa a e he oand e beams in o wo sepa a e images. We no malised each beam o he co esponding median alue and c ea ed he oand ebeam combined la . We ollowed he same p ocedu e o each image. We pe o med his co ec ion since la ields we e ob ained wi h he ull op ics on he ligh pa hway. This led o a a he di e - en coun a e in each beam wha would ha e led o an inac- cu a e no malisa ion. Finally, we spli all he educed images in o an oand eimage and applied he L.A. Cosmic algo i hm ( an Dokkum 2001) o emo e cosmic ays. We used PyRAF o measu e he on- ame ull wid h a hal maximum (FWHM) in each image including he high- and ze o- pola ised s anda d s a s. The FWHM was measu ed indepen- den ly o he oand ebeam since he shape o he poin sp ead unc ion (PSF) can a y, especially o sou ces wi h a high deg ee o pola isa ion. To ob ain e e ence ield s a s we used a sou ce de ec ion algo i hm based on DAOS a Finde in Pho u ils and applied i o he backg ound-sub ac ed oand eimages sepa a ely. The selec ed sou ces we e hose wi h a h eshold abo e he median plus h ee imes he s anda d de ia ion o he backg ound. The s a is ics we e calcula ed pe beam and pe angle posi ion o he HWP using a sigma clipping o he masked image. F om hese sou ces, we disca ded hose ha we e clea ly ex ended and hose ha lay oo close o pa ially wi hin he ins umen al mask edges. We also checked whe he he sou ce was sa u a ed in any o he beams. We ended up wi h i e sou ces, including GRB210610B, in he FORS2 images and i e sou ces in he CAFOS images. The sou ces om FORS2 and CAFOS images a e di e en due o sa u a ion o because sou ces all a he mask edges. We pe o med ape u e pho ome y using ci cula ape u es wi h a adius equal o he FWHM and applied in ini e-ape u e co ec ions using a cus om Py hon sc ip o he FORS2 da a. In o de o a oid con amina ion by a nea by spu ious sou ce, we measu ed he lux o he CAHA/CAFOS da a using a ixed ape u e o h ee imes he FWHM pe image and pe beam and sub ac ed he sky o an annulus a ound he sou ce wi h an inne and ou e adius o ou and i e imes he FWHM, espec- i ely. In he FORS2 images, his spu ious sou ce was ou side he 5The obse a ions we e ob ained unde he p og am GTCMULTIPLE2C-21A (PI: de Uga e Pos igo). A216, page 4 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) measu ed egion. The e o s o he ape u e pho ome y we e ob ained conside ing each beam as an unique image, ha is, we sepa a ed he o dina y s ipes om he ex ao dina y s ipes. Wi h he measu ed lux alues pe sou ce and pe beam, o and e, we ob ained he S okes pa ame e s Qand Udesc ib- ing he linea pola isa ion (see e.g. Pa a & Romaniello 2006; Bagnulo e al. 2009) o each o ou images. We used he no - malised S okes pa ame e o he linea pola isa ion. Q I=q=2 N N−1 X i=0 Ficos iπ 2(1) U I=u=2 N N−1 X i=0 Fisin iπ 2(2) whe e Nis he numbe o posi ions o he hal -wa e pla e, Iis he in ensi y, and Fiis he no malised lux di e ence pe HWP posi ion angle, Fi= o,i− e,i o,i+ e,i·(3) Following he equa ions abo e, we can ob ain he pola isa- ion deg ee Plin, Plin =qq2+u2.(4) Fo he posi ion angle, we ollowed he o malism as used in Bagnulo e al. (2009). We hen co ec ed o he e ec s o op ics and de ec o in he pola isa ion images. To do his, ze o-pola ised s anda d s a s we e obse ed in he co esponding bands wi h VLT/FORS2 ollowing he p ocedu es de ailed in Ande son (2015). Fo CAHA/CAFOS, obse a ions o s anda d s a s we e ca ied ou in he same nigh . In he case o VLT/FORS2, b- and z-band s anda ds we e comple ely sa u a ed, and no subsequen s an- da ds we e ound in he ESO a chi e a ound he ime o he obse a ion. Fo he Rband, he s anda ds we e also sa u- a ed. We hen used WD1620−391 o he ze o-pola ised s an- da d s a , obse ed 18days a e he GRB, o VLT/FORS2, and BD+332642 (Tu nshek e al. 1990) o CAHA/CAFOS obse ed in he same nigh as GRB210610B. We inally emo ed he e ec on he pola isa ion induced by he dus in he Milky Way (MW). To do his, we measu ed he qand upa ame e s o he ield s a s in ou images. Howe e , FORS2 is known o ha e a adial p o ile in which he pola isa ion a ies ac oss he ield om he op ical axis owa ds he edges o he de ec o s (see e.g. González-Gai án e al. 2020). To accoun o his e ec , we applied he co esponding co ec ion by using he q,ubackg ound co ec ion and he ins umen al pola isa- ion maps p esen ed in González-Gai án e al. (2020). We hen measu ed he Galac ic in e s ella pola isa ion (Galac ic ISP, o GISP) using h ee me hods, ollowing Wie sema e al. (2012): Fi s , we measu ed he mean alues o he q,upa ame e s om he ield s a s. Then we pe o med a one-dimensional Gaussian i o he q,u alues o he ield s a s by gene a ing a no mal dis- ibu ion o alues o he q,uwi hin hei co esponding e o s, and we hen i ed a Gaussian o each S okes pa ame e . Finally, we pe o med a wo-dimensional Gaussian i o he q,u, adap - ing he p ocedu e om he one-dimensional Gaussian i (see Fig. 2 o an example o he wo-dimensional Gaussian i ). All he q,u i ed pa ame e s a e consis en wi hin he e o s, and we he e o e ec o ially sub ac ed he mean alue om he q,u alues o he GRB. The mean sky alues a e lis ed in Table 1, Fig. 2. Two-dimensional Gaussian i o he ield s a s q,upa ame- e s. Le : No mal dis ibu ion o andomly gene a ed da a a ound he q,u alues. The ampli ude o gene a e he da a is he S okes pa ame e e o s. Cen al panel: wo-dimensional Gaussian model i ed o he da a poin s. Righ : Residuals. which a e all consis en wi h o close o ze o. Bo h qand ushow some a ia ion om one epoch o he nex , which we assumed is due o di e en sky backg ound ligh be ween he epochs6. Fo he CAFOS images, he same analysis was pe o med, al hough we we e unable o co ec o backg ound pola isa ion and ins umen al pola isa ion ac oss he images because we lack a cha ac e isa ion as we ha e o FORS2. Howe e , luna illumi- na ion was close o 0%, and we do no expec a high ins umen al pola isa ion (Pa a & Taubenbe ge 2011) o CAFOS. Indeed, he wo-dimensional Gaussian i o he ield s a s (see Fig. 2) shows i o be consis en wi h no pola isa ion om he ISM and implici ly, no pola isa ion om he CAFOS ins umen . A e he pola isa ion induced by he MW dus was emo ed we calcula ed he inal pola isa ion om bo h ins umen s in all he il e s. Nex , we needed o conside he con ibu ion o he pola isa ion om he dus in he hos galaxy i sel . Since he in o ma ion is mo e limi ed han o he dus in he MW we assumed a Se kowski law (Se kowski e al. 1975) o he hos galaxy. F om he spec al ene gy dis ibu ion (SED) i o he GRB ligh cu e, we ob ained he colou excess on he line o sigh (see Sec . 3.2). Assuming his alue as he ex inc ion o he a e glow a he hos galaxy, using PISP(%) ≤9.0×E(B−V), we ind ha he con ibu ion om he hos galaxy could be as high as PHos ISP ≤0.09%. We no e ha in ex agalac ic sigh lines, his may no be applicable because he dus p ope ies may di e om hose o he MW (Nagao e al. 2022). This alue is well below he PD measu ed o he a e glow, and we he e- o e conside ed he hos con ibu ion o be negligible. Finally, we co ec ed o he pola isa ion de ini ion bias wi h he modi- ied asymp o ic es ima o (MAS) Plaszczynski e al. (2014). The co ec ed alues o he pola isa ion deg ee a e shown in Table 1 be o e and a e he bias co ec ion. We also de e mined he pola isa ion posi ion angle (PA) o he GRB a e glow. The FORS2-measu ed aw PA was co - ec ed o ch oma ici y using he abula ed alues pe bandpass p esen ed in Ande son (2015). We hen co ec ed o i using he s anda d s a Hil ne 652, obse ed on he 19, July 2021, al hough he PA co ec ion is e y small wi h ∆θ=0.5±0.9. The high pola isa ion s anda d in RSpecial band closes in ime o he GRB was also sa u a ed in a leas one beam. We co ec ed he de i ed PA alue o he s anda d o he alue p esen ed in Ciko a e al. (2017). As he measu ed b-band alue o he PA in Hil ne 652 is e y close o he Bband alue in Ciko a e al. (2017) we use Bband o co ec o he b-band alue. Fo he z band, no obse a ions we e ound so a o his s anda d, and we he e o e used he PA o he Iband in Ciko a e al. (2017). The high pola isa ion s anda d obse ed o CAFOS, Hil ne 960, 6ESO wea he log. A216, page 5 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) Table 1. Measu ed alues o he linea pola isa ion and PA o GRB 210610B. Epoch Bandpass Ins umen /Telescope qsky usky PLin PLin,Debiased θS/N − 0(day) (%) (%) (◦) 0.1205 RCCAHA/CAFOS 0.2±2.4 –0.01±0.31 4.50±1.45 4.27±1.45 183±9>500 0.2418 RSpecial VLT/FORS2 –0.30 ±0.32 –0.29 ±0.20 0.28 ±0.20 – 267 ±19 >710 0.2593 RSpecial VLT/FORS2 –0.79 ±0.28 0.81 ±0.22 0.60 ±0.24 – 17±11 >690 0.2698 bHigh VLT/FORS2 – – 0.18±0.16 – 187±24 >395 0.2803 zSpecial VLT/FORS2 – – 0.23±0.28 – 199±35 >180 1.2674 RSpecial VLT/FORS2 –0.13 ±0.22 0.06 ±0.07 2.28 ±0.22 2.27 ±0.22 237 ±3>330 1.2766 RSpecial VLT/FORS2 –0.36 ±0.39 0.06 ±0.07 1.72 ±0.27 1.69 ±0.27 238 ±5>300 No es. When he s anda d co ec ed PA is nega i e, i was co ec ed o posi i e alues by sub ac ing i om 360◦. We measu ed he (S/N) in each beam and a each HWP angle image. The alue we p esen he e is he lowes we measu ed in one beam in one image a a ce ain HWP angle. was co ec ed o he “ heo e ical” alue ollowing Schmid e al. (1992). We de ec pola isa ion a &3σsigni icance o he i s and las epochs, while he second epoch is consis en wi h ze o pola - isa ion. We de ec 1σpola isa ion in he Rband on he second epoch. We conside his measu emen , oge he wi h he Rband obse a ion a ∼0.26days o he band zband as limi , howe e . Since he MAS co ec ion is no comple ely applicable when PD/σP<3.0, we did no apply his co ec ion o he men ioned limi s. In his PD egime, he PA would beha e e a ically and, we he e o e canno ea i as a limi . 3.2. Ligh -cu e analysis We i s modelled he op ical and X- ay ligh cu es simul a- neously wi h a smoo hly b oken powe law (Beue mann e al. 1999), F=(Fκ 1+Fκ 2)−1/κ, whe e FX= b eak( / b eak)−αX, b eak being he lux densi y a he b eak ime b eak,κis he b eak smoo hness pa ame e , and he subsc ip s 1,2 indica e p e- and pos -b eak, espec i ely. We did no conside da a be o e 0.06days (∼5ks) because hey a e s ill domina ed by he ini- ial apid decay. E en be o e any modelling was pe o med, i is clea ha he op ical ligh cu e is ini ially la a leas un il ∼0.25days (∼20 ks; see Fig. 3), while a he same ime he X- ay beha iou is di icul o disce n, hough he wo g oups o obse a ions a 4ks and 7ks show possible ading, in con as o he op ical. No knowing he p ecise e olu ion o he ea ly X- ay da a, we allowed he ini ial X- ay decay o be di e en om he op ical. No e ha a di e en decay implies a colou e olu ion be ween X- ays and he op ical. We ind a shallow b eak wi h b eak,op =0.326 ±0.011 d (27.7±1.2 ks), a la op ical decay wi h α1,op =0.00 ±0.01, and an op ical o X- ay decay index α2,b eak =1.85 ±0.04, wi h a b eak smoo hness κ=1.4±0.1, wi h χ2/d.o. .=2.6. A e he b eak, we analysed he ull op ical o X- ay SED when we he co e age in bo h equency egimes was bes . To build his SED, we i s c ea ed he XRT spec a using he ime- slice ool in he XRT eposi o y (E ans e al. 2007,2009) in he ange 30ks−50ks a a mid- ime o 1.2days. We hen shi ed he op ical GTC/HIPERCAM ug iz da a a 1.097days close o hese imes using he decay indices ound abo e. We modelled he a e glow SED om op ical o X- ay e- quencies using Xspec 12.13.0 (A naud 1996). The edshi was ixed o z=1.1341, and we ixed he Galac ic hyd ogen col- umn densi y o NH=3.94 ×1020 cm−2(Willingale e al. 2013). The da a a e bes modelled by a single powe -law (χ2/d.o. .= 264.5/298) wi h β=0.869+0.003 −0.007, in insic abso p ion NH= 18.3+6.5 −5.9×1020 cm−2(using he Tübingen-Boulde ISM abso p- Fig. 3. GRB op ical o X- ay ligh cu e i o a ailable pho ome y (see Table A.1) and Swi X- ay da a. The g ey shaded egion is no aken in o accoun o he i as i may be con amina ed by he p omp emission. The clea cyan egions ma k he imes we chose o de i e he SED o he ligh cu e. The da a a e shown wi h an o se in lux o be e isibili y. ion model; Wilms e al. 2000) and E(B−V)<0.01mag, using he Small Magellanic Cloud (SMC) ex inc ion law (Pei 1992). We also ob ained an op ical o X- ay SED a 0.079days (6871s) aking he X- ay da a be ween 6000 and 8000s and using he op ical ligh cu e abo e o shi he co esponding g iz pho ome y. A e ixing he edshi , ex inc ion, and abso p- ion as o he la e epoch and ollowing he same i ing p o- cedu e, we ind ha he da a a e bes modelled by a b oken powe law (χ2/d.o. .=143.3/197) wi h βop =0.43+0.046 −0.046, and βX=βop +0.5, which indica es he p esence o a spec al b eak, like he cooling b eak o synch o on emission. We no e ha he spec al index o he high- equency b anch is consis en wi h he alue ound a la e ime, sugges ing ha he b eak shi ed o lowe equencies wi h ime, which is expec ed o an ISM en i- onmen wi hin he i eball model and be o e he je -b eak occu s (Sa i e al. 1998). Howe e , he shallow decay o he ea ly ligh cu e canno be explained wi hin he s anda d i eball model, unless we conside a mo e complex scena io such as an ene gy injec ion (e.g., Zhang e al. 2006). 3.3. Op ical a e glow spec um The a e glow spec um obse ed by OSIRIS ∼0.25days a e he bu s shows se e al ansi ions o Fe ii, Mg ii, and Mg i(see Table 2), all o hem a a common edshi o z=1.1341±0.0004 A216, page 6 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) Table 2. Equi alen wid hs in he obse e ame measu ed in he a e - glow spec um. Obse ed λFea u e z0EW (Å) (Å) 5001.69 Feii 2344.21 1.1336 0.49±0.07 5066.45 Feii 2374.46 1.1337 0.23±0.06 5084.79 Feii 2382.77 1.1340 0.61±0.07 5520.91 Feii 2586.65 1.1344 0.25±0.07 5549.64 Feii 2600.17 1.1343 0.53±0.07 5976.16 Mgii 2796.35 1.1341 3.29±0.10 Mgii 2803.53 1.1345 6090.41 Mgi2852.96 1.1348 0.64±0.07 4353.92 Mgii 2796.35 0.5570 0.39±0.10 4366.46 Mgii 2803.53 0.5575 0.23±0.08 (see Fig. 4and Table 2). This alue is a lowe limi o he bu s edshi due o he non-de ec ion o ine-s uc u e lines exci ed by he GRB i sel . The de ec ion o he a e glow in he blues band om Swi /UVOT se s an uppe limi o z=1.7, using he so-called Lyman d op-ou echnique, ollowing (Jakobsson e al. 2012). Conside ing his limi and he lack o abso p ion lines common o GRBs (Ch is ensen e al. 2011) a highe edshi , we hence adop ed z=1.1341 as he edshi o GRB210610B. We also de ec ed wo abso p ion ea u es ha co espond o he Mgii double in an in e ening sys em a z=0.5572 ±0.0002. The equi alen wid hs (EWs) o he de ec ed abso p- ion lines we e measu ed using he GRBspec da abase ools (de Uga e Pos igo e al. 2014;Blažek e al. 2020) (see Table 2). We ollowed de Uga e Pos igo e al. (2012) o compa e hese alues wi h he common end o long GRB sigh -line en i- onmen s. The line s eng h pa ame e (LSP) measu ed o GRB210610B is ex emely low, LSP =−2.17 ±1.13, implying ha his line o sigh has weake ea u es han 99.85% o he GRBs in he a o emen ioned sample. The line-s eng h diag am in Fig. 4shows ha he FeII lines a e pa icula ly weak and a e only de ec ed h ough he e y high signal- o-noise a io o he spec um. The magnesium ea u es a e s onge , bu s ill among he weakes in he sample. The low EW alues imply a low col- umn densi y and hence a low amoun o gas and possibly dus in he line o sigh consis en wi h a negligible dus -induced pola - isa ion. 3.4. Hos galaxy We obse ed he loca ion o GRB210610B ∼58 days a e he bu s using GTC/HiPERCAM and ∼253days a e wi h GTC/EMIR. In he HiPERCAM images, we de ec an unde ly- ing objec a he GRB posi ion and we conside his objec o be he hos . We also ind a pu a i e companion galaxy owa ds no h-wes o he hos candida e (see Fig. 1) a a dis ance o ∼200 .7, which would co espond o a dis ance o 22kpc a he ed- shi o he sys em. The pho ome y o he hos galaxy is shown in Table A.1, oge he wi h he alues o he companion. We pe o med a SED analysis o he wo galaxies assum- ing ha bo h o hem a e a he GRB edshi . Fo his analysis, we used he SED i ing code CIGALE7(Bu ga ella e al. 2005; Noll e al. 2009;Boquien e al. 2019) in i s la es e sion. To i he s a o ma ion his o y (SFH), we chose a delayed s a o ma- ion his o y wi h an age o he main s ella popula ion anging 7h ps://cigale.lam. / Fig. 4. Abso p ion spec oscopy and line p ope ies o GRB210610B. Top: Spec al ea u es de ec ed a he edshi o he GRB, plo ed in ed, and he ea u es om he in e ening sys em a e shown in blue. The dashed blue line shows he e o spec um. Bo om: Line-s eng h dia- g am o he spec al ea u es in he hos galaxy. The diag am compa es he ea u es measu ed in ou spec um ( ed) wi h he a e age alues o a la ge sample (black) (see Sec . 3.3). om 0.1 o 13Gy and a la e bu s wi h an age a ying om 20−500My . We allowed he code o a y he co esponding mass ac ion om he o al galaxy mass o he la e bu s om 0, which implies a single decaying exponen ial o he SFH mod- elling, up o 0.6, ha is, 60% o he o al galaxy mass. We use he ini ial mass unc ion (IMF) as desc ibed in Chab ie (2003) wi h aB uzual & Cha lo (2003) s ella popula ion model, assuming a s ella sub-sola me allici y (Z∗) o he galaxy (see Sec s. 3.3 and 4) ha was se o a y om 0.004 o 0.008, acco ding o he scheme used in B uzual & Cha lo (2003). The nebula emis- sion was modelled conside ing he same me allici y alues o he gas. Fo he a enua ion law, we conside ed he modi ied Calze i e al. (2000) law implemen ed in CIGALE, assuming a Small Magellanic Cloud (SMC) ex inc ion law o he a en- ua ion o he emission lines. We did no obse e a colou e olu ion o he Galac ic-dus -co ec ed (Schla ly & Finkbeine 2011) magni udes, and we do no de ec he hos galaxy in he H band down o 22.9 magni udes. This migh indica e a low le el o dus emission due o low dus hea ing om UV massi e s a pho ons, indica ing a low numbe o massi e s a s o an in insic lack o dus in his sys em. Since he amoun o de ec ed Mg, ypically o med in he explosion o massi e s a s, is la ge com- pa ed o he amoun o Fe in he aced sys em, al hough s ill low o common lGRBs sigh -lines (see Fig. 4), his a ou s he scena io o low in insic ex inc ion a he han he absence o massi e s a s (see Sec . 4 o an ex ended explana ion). The e- o e, we chose he colou excess o he nebula lines o be lowe han 0.1. The slope o he a enua ion cu e (Boquien e al. 2019) was allowed o a y be ween −0.6 and 0.6. Fo he dus emission, we selec ed he Dale e al. (2014) models and allowed he exponen ha con ols he adia ion ield dis ibu- ion o he e-emi ed ene gy by dus hea ing o a y be ween 1.0 and 3.0. We applied he same SED i o bo h galaxies. Howe e , he companion galaxy shows colou excess in he HiPERCAM obse a ions, and we he e o e allowed he colou excess o he nebula lines o ange up o 0.4. The esul s o he com- A216, page 7 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) Table 3. Fi ed physical p ope ies o he pu a i e hos galaxy o GRB210610B and i s pu a i e companion. Hos galaxy Hos companion log10(M∗) (M) 9.10+0.40 −0.20 9.60+0.55 −0.24 log10(SFR) (M/y ) 1.06+0.12 −0.10 0.47+0.32 −0.10 sSFR(Gy −1) 9.26±6.02 0.76±0.68 AV(mag) 0.19±0.10 0.51±0.33 Z∗0.006±0.002 0.006±0.002 Reduced χ20.49 0.50 Fig. 5. GRB hos SED wi h he co esponding esiduals modelled using he CIGALE i ing code (Bu ga ella e al. 2005;Noll e al. 2009; Boquien e al. 2019) o he pu a i e hos galaxy o GRB210610B. The e ical blue a ow indica es he Hband uppe limi . pu ed physical p ope ies om he SED i can be ound in Table 3and he bes i is shown in Fig. 5. The SED o he companion galaxy allowed us o de e mine a pho ome ic ed- shi ha is consis en wi h he edshi o GRB210610B. The companion is a a dis ance o 2.67a csec, co esponding o a physical dis ance o 22.6kpc a a edshi o z=1.1341. We he e o e assumed ha hese wo galaxies a e pa o a g oup. 4. Discussion A b oad s udy o he GRB p omp and a e glow emission oge he wi h i s hos galaxy is c ucial o be e cons ain s o he cha ac e is ics o GRB210610B. The bu s p omp emission p esen s he GRB as a ha d bu s , as obse ed by Swi , posi- ioning i in he op pa o he long GRB egion o he ha d- ness a io e sus T90 ela ion (see e.g. Lien e al. 2016). The iso opic equi alen ene gy, oge he wi h he obse ed peak ene gy, shows ha GRB210610B is ully consis en wi h he Ama i ela ion (Ama i e al. 2002;Ama i 2006) o long GRBs. The p omp , high-ene gy emission o GRB210610B has been analysed wi h Fe mi da a by Chen e al. (2022). They epo ed ha he p omp emission is bes i by a hyb id je model (Gao & Zhang 2015) in which a ho i eball componen domi- na es he emission in he beginning, while a Poyn ing lux com- ponen supe sedes a la e imes. Chen e al. (2022) desc ibed hese esul s as consis en wi h he magne a model as a plau- sible cen al engine (Me zge e al. 2011). 4.1. Hos galaxy The SED analysis o he GRB210610B hos e eals a galaxy wi h a low s ella mass, consis en wi h a dwa galaxy ha is ac i ely o ming s a s and has low ex inc ion. A he same ed- shi , he companion is mo e massi e han he GRB hos , wi h a highe s ella mass bu lowe SFR. The ex inc ion o his galaxy is also highe han ha o he GRB hos assuming he same ex inc ion law. F om he a e glow spec um, we ind ha GRB210610B is embedded in an en i onmen wi h low amoun s o Feii and somewha highe alues o Mgii and Mg i. The low amoun o Fe could be indica i e o a low numbe o SNIa in he hos , since hese explosions a e he main sou ces o Fe (Pagel 2009). This could mean ha ei he he e is an in insic lack o his ype o s ella explosions nea he abso be si e o ha he sys em i sel is oo young o ha e been Fe-en iched ia SNIa. Howe e , he absence o ine-s uc u e lines in he spec um does no allow us o de e mine he dis ance o he abso bing clouds o he explo- sion si e and, he e o e, he a e glow spec um could be ac- ing gas in he ex e nal pa s o he hos galaxy, whe e we would expec i o be less en iched. Ne e heless, we ind a highe el- a i e alue o he EWs o Mg, al hough i is s ill low compa ed o he mean alue ound o GRB si es (de Uga e Pos igo e al. 2012). Mg is eleased in o he ISM h ough he explosion o mas- si e s a s (Pagel 2009). Toge he wi h he low amoun o Fe, his migh sugges ha he hos galaxy is a e y young sys em. This is also suppo ed by he ex inc ion we measu e om he SED i . The AV alue may indica e ha he hos galaxy has a low amoun o dus , which is expec ed o a low me allici y and, he e o e, o a young sys em. 4.2. A e glow The GRB a e glow ollows a decay-pla eau-decay beha iou wi h an ini ial decay ha is well i ed by a b oken powe law wi h an op ical spec al slope a ∼0.08days o βop =0.43+0.05 −0.05 and a βXR ∼0.83. A e wa ds, he ligh cu e en e s a pla eau phase ha las s ∼0.247days o inally change o he inal decay a ∼0.326days a e bu s . This inal decay is be e i ed by a powe law wi h β=0.869+0.003 −0.007, consis en wi hin e o s wi h βXR a ∼0.08days a e GRB. The change in he spec al slope a ∼0.08days migh be indica ing a spec al b eak a he beginning o he pla eau phase. The SED i shows negligible ex inc ion on he line o sigh owa ds he GRB and a low X- ay hyd ogen col- umn densi y, as compa ed o NH alues o ela i ely low edshi GRBs (Campana e al. 2010). This low NH oge he wi h he low E(B−V), con adic s wha would be expec ed o low- edshi bu s s, whe e a highe dus - o-gas a io is expec ed o lowly NH X- ay abso bed bu s s (Campana e al. 2010). 4.3. Pola isa ion Ou pola isa ion obse a ions ma ch h ee e y impo an s ages o he GRB210610B a e glow ligh -cu e. The i s pola ime- y measu emen s we e pe o med 0.1205days a e bu s , igh a e he ligh cu e en e ed he pla eau. The second epoch a ∼0.26days is close o he end o his pla eau phase, almos con- sis en wi h he b eak o he op ical ligh cu e a 0.326days a e he GRB. A e his, he ligh cu e unde goes i s inal decay, whe e a inal pola ime y epoch was obse ed. The a e glow is pola ised a he beginning and owa ds he end o he ligh - cu e e olu ion, bu no a ound he op ical b eak a 0.326days a e bu s , whe e he pola isa ion d ops o ze o wi h a small ise o 0.61% in he nex obse a ion, which is sligh ly abo e he A216, page 8 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) PHos ISP limi , bu only a a 2.5-σle el. In he inal decay, he pola isa ion inc eases o 2.27% and hen dec eases o 1.69% as he a e glow ades away. We ind ha band zbands show pola - isa ion alues consis en wi h ze o a a close epoch o he 2.5-σ R-band measu emen . The e o e, he mul i-band obse a ions do no allow us o asses ch oma ici y/ach oma ici y in he a e glow pola isa ion. The PA a ies by ∼54◦be ween he 3σde ec ions. The measu ed pola isa ion is well consis en wi h p io measu e- men s o linea pola ime y o GRB a e glows, as shown in Fig. 6. One impo an aspec when de e mining he in insic pola i- sa ion is o cons ain any con ibu ion om he dus in he hos galaxy. A possible pola isa ion om he MW was emo ed du - ing he analysis (see Sec . 3.1). The SED i o he GRB ligh cu e esul s in a negligible alue o he a e glow ex inc ion on he line o sigh and he in e ed uppe limi o he GRB hos ISP is a he small compa ed o e o s o he measu ed 3- σpola isa ion de ec ions (see Sec . 3.1). This means ha ei he he pola isa ion con ibu ion o he hos galaxy along he line o sigh is well below he PHos ISP limi o ha his con ibu ion would cancel he a e glow pola isa ion ou . The ela i ely high alues we measu ed o he 3σde ec ions and he e y low limi o he hos galaxy pola isa ion led us o assume ha he hos con ibu ion is negligible. This low ex inc ion also suppo s he scena io in which he pola isa ion is in e p e ed as in insic o he GRB a e glow. This is u he con i med by he pola isa ion non-de ec ion in he second epoch, which is an indi ec measu e- men o he hos ISP. 4.4. Theo e ical In e p e a ions o he pola isa ion signals The i s pola ime y obse a ions show a a he high linea pola isa ion deg ee o ∼4% a ∼0.1205day. Conside ing ha he obse a ions we e ca ied ou du ing a shallow decay/pla eau phase, a non-negligible ac ion o op ical pho ons migh o igi- na e om e eshed shocks in he o iginal ejec a om he cen- al engine. As p e ious pola ime y s udies o he ea ly a e - glow indica e ha ejec a om he cen al engine con ain la ge- scale o de ed magne ic ields, a leas o a subg oup o GRBs i no o all (e.g. Mundell e al. 2013), he e eshed shock emission can be pola ised due o he o de ed magne ic ields in he ejec a. The combina ion o he pola ised e eshed shock emission and unpola ised o wa d shock emission is likely o gi e low-/in e media e-pola isa ion signals. Fo GRB191016A, Sh es ha e al. (2022) epo ed he de ec ion o pola isa ion sig- nals wi h P∼5−15% which a e coinciden wi h he s a o he pla eau phase. An ene gy injec ion model was discussed o explain he coincidence. The op ical ligh cu e s a s o decline a ∼0.2days (see he op panel o Fig. 7). This indica es ha he ene gy injec- ion s ops a ound ha ime and ha he op ical band is dom- ina ed by he o wa d-shock emission well be o e he second pola isa ion epoch is conduc ed a ∼0.24day. The mag- ne ic ields in he o wa d-shock egion ( he shocked ambien medium) a e con en ionally assumed o be gene a ed locally by mic oscopic ins abili ies in shocks (see e.g. Med ede & Loeb 1999), and hey a e expec ed o be highly angled. The PD o he o wa d-shock emission is expec ed o be ze o when he line o sigh does no un along he je edge. The low pola isa ion a ∼0.24−0.28days can be explained na u ally when he op ical emission is domina ed by he o wa d-shock emission. Due o he ela i is ic beaming e ec , he obse e can see only a small isible egion (a small pa ch wi h an angula size o 1/Γ, loca ed a ound he poin a which he line o sigh in e - sec s he je ) ins ead o he en i e su ace o a shock on . The isible egion appea s as a ing due o a ela i is ic limb- b igh ening e ec (G ano e al. 1999). Synch o on emission om each small segmen o he ing can be pola ised when he andom magne ic ields pa allel and pe pendicula o he shock no mal ha e di e en a e aged s eng hs. Howe e , he ne PD is ze o because o he symme y o he isible egion. As he o wa d shock slows down, he angula size o he isible egion g ows as 1/Γ∝ 3/8(ISM) o 1/4(wind medium). E en ually, a pa o he ing is loca ed ou side he je edge and he emission egion becomes asymme ic. This migh ha e occu ed by he hi d pola isa ion epoch a ∼1.27day. Be ween he second and hi d epochs, he angula size o he isible egion can g ow by a ac o o ∼1.7 (ISM) o 1.4 (wind medium). Op ical linea pola isa ion measu emen s ha e been ca ied ou o many la e GRB a e glows ypically se e al hou s o a ew days a e he p omp gamma- ay emission (e.g. Co ino & Go z 2016). In his pe iod, a je b eak is expec ed o occu . The de ec- ion o uppe limi s o he linea PD a e gene ally low (less han a ew pe cen ), which migh indica e ha he shock-gene a ed andom magne ic ields pa allel and pe pendicula o he shock no mal ha e simila a e aged s eng hs. The pola isa ion signals P∼2% a ∼1.27days migh be explained in his geome - ical model. I he la ge-scale magne ic ields in he ejec a a e o oidal, he PA change be ween ∼0.24 days and ∼1.27 days is expec ed o be 0 o 90◦. Howe e , he la ge-scale magne ic ields in he ejec a can be la gely dis o ed be o e hey injec ene gy in o he o wa d shock (o he o iginal magne ic s uc u e can be e y di e en om he o oidal con igu a ion). The posi- ion angle change can be any alue. Howe e , his model p e- dic s a s eepe decline a la e imes. E en in he non-sp eading je model, he expec ed decay index is α=3(p−1)/4+3/4∼2.05 (ISM) o (3p−1)/4+1/2∼2.30 (wind medium) o p=2.74, which is s eepe han he obse ed alue α=1.85 ±0.04. I we ely on he a he high alue o pob ained om he SED mod- elling, we can ule his geome ical model ou . The na u e o magne ic ields ha a e gene a ed in shock ins abili ies is no ye ully unde s ood. The mic oscopic-scale angled magne ic ields may decay so apidly in he downs eam o he shock ha hey could no accoun o he obse ed syn- ch o on lux (e.g. Si oni e al. 2015). Al e na i ely, he o wa d- shock egion could ha e magne ic ield u bulence on la ge scales, compa able o he wid h o shocked egion ∼R/16Γ(e.g. Si oni & Goodman 2007). In his case, he PD and PA empo- ally change in a andom manne , and PD ∼70%/√N, whe e Nis he numbe o pa ches wi h a cohe en magne ic ield wi hin he angula scale 1/ΓG uzino & Waxman (1999). Kuwa a e al. (2023) cons uc ed a semi-analy ic model wi h a a ying la ge- scale u bulen ield in he o wa d-shock egion, o which hey pe o med nume ical calcula ions in he case o iso opic u bu- lence and ze o iewing angle. They ob ained a andomly a y- ing PD on a imescale o hou s a a le el o ∼1−3% and PA wi h changes ha we e no limi ed o 90◦. These p ope ies appea o be consis en wi h ou da a o GRB210610B. When hyd odynamic-scale u bulen magne ic ields a e assumed, we ha e wo possible scena ios: (1) he ∼4%, ∼0.2%, and ∼2% pola isa ion signals a e all due o u bulen magne ic ields, o (2) he ∼4% pola isa ion signal is due o pola ised e eshed shock emission, and he o he wo signals a e due o u bulen magne ic ields. Fo non-sp eading op-ha je s wi h mic oscopic-scale an- gled ields, he PD ligh cu e would ha e wo maxima a ound a je b eak, wi h he pola isa ion PA changing by 90◦be ween he i s and second maximum (Ghisellini & Lazza i 1999;Sa i A216, page 9 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) Appendix B: Linea pola isa ion measu emen s on GRBs a e glow emission. Table B.1. Measu ed alues o he linea pola isa ion and PA o GRB a e glows om he li e a u e. GRB Redshi Tmid PLin θRe (days) (%) (◦) GRB990510 1.62 0.7708 1.7±0.2 101±3 [1] 0.8583 1.6±0.2 96±4 [2] 1.8083 2.2+1.1 −0.9112+15 −17 " GRB990712 0.43 0.44 2.9±0.4 121.1±3.5 [3] 0.70 1.2±0.4 116.2±10.1 [3] 1.45 2.2±0.7 139.1±10.4 " GRB020405 0.695 1.2292 1.50 ±0.40 172±8 [4] 1.3208 9.89±1.30 180±4 [5] 2.2682 1.96±0.33 154±5 [6] 3.8792 1.47±0.43 168±9 " GRB020813 1.35 0.21528 2.22 ±0.07 157.6±1.0 [7]∗ 0.26181 1.98±0.04 153.4±1.7 " 0.34167 1.96±0.07 152.0±1.2 " 0.89792 1.07±0.22 154.3±5.9 [8] 0.93750 1.42±0.25 137.0±4.4 " 0.97542 1.11±0.22 150.5±5.5 " 1.01625 1.05±0.23 146.4±6.2 " 1.11667 1.43±0.44 155.8±8.5 " 1.97958 1.26±0.34 164.7±7.4 " GRB021004 2.33 0.37 1.72 ±0.56 187.7±8.3 [9] 0.38 2.09±0.60 173.0±7.9 " GRB030329 0.17 0.5321 0.92±0.10 86.13±2.43 [10]∗∗ 0.5492 0.86±0.09 86.74±2.40 " 0.5671 0.87±0.09 88.60±2.64 " 0.5850 0.80±0.09 91.12±2.88 " 0.6921 0.66±0.07 78.52±2.94 " 0.7129 0.66±0.07 76.69±2.89 " 0.7342 0.56±0.05 74.37±3.11 " 1.5204 1.97±0.48 83.20 " 1.5500 1.37±0.11 61.65±2.38 " 1.5800 1.50±0.12 62.29±2.44 " 1.6700 1.07±0.09 59.41±2.51 " 1.7000 1.09±0.08 66.07±2.45 " 1.7200 1.02±0.08 67.05±2.60 " 1.7400 1.13±0.08 70.56±2.51 " 2.6800 0.52±0.06 30.76±5.04 " 2.7000 0.52±0.12 12.55±4.63 " 2.7200 0.31±0.07 24.50±6.94 " A216, page 16 o 17 Agüí Fe nández, J. F., e al.: A&A, 690, A216 (2024) Table B.1. con inued. GRB Redshi Tmid PLin θRe (days) (%) (◦) 3.5400 0.57±0.09 53.85 ±4.08 " 3.5600 0.53±0.08 57.08 ±4.06 " 3.5800 0.42±0.10 62.21 ±6.10 " 5.6600 1.68±0.18 66.32 ±3.38 " 7.6400 2.22±0.28 75.16 ±3.32 " 9.5900 1.33±0.14 70.91 ±3.31 " 13.6000 2.04±0.57 1.16 ±7.64 " 22.5000 0.58±0.10 42.7 ±9.26 " 37.5000 1.48±0.48 25.42 ±9.41 " XR080109 0.007 3.6416 0.95±0.20 114.9 ±5.9 [11] 5.5552 0.85±0.28 106.1 ±9.4 " 20.6279 1.05±0.06 135.3 ±1.7 " 20.6443 1.28±0.06 132.5 ±1.4 " 52.5578 1.42±0.46 139.0 ±9.1 " GRB080928 1.6919 1.7 4.49+1.16 −0.96 41.3±6.3 [12] GRB090102 1.55 0.0025 10.1±1.3 – [13] GRB091208B 1.063 0.0042 10.4 ±2.5 92±6 [14] GRB091018 0.971 0.2461 1.07±0.30 179.2 ±16.1 [15] 0.4548 1.44±0.32 2.2 ±12.6 " 1.1394 1.73±0.36 69.8 ±11.7 " 1.1552 3.25±0.35 57.6 ±6.1 " 1.1735 1.99±0.35 27.6 ±10.0 " 1.1893 1.42±0.36 114.6 ±14.0 " 1.3918 0.97±0.32 32.8 ±17.8 " 1.4493 1.08±0.35 88.7 ±17.9 " 2.3902 1.45±0.37 169.0 ±14.3 " GRB120308 2.22 0.0033 28+4 −434±4 [16] 0.0042 23+4 −444±6 " 0.0052 17+5 −451±9 " 0.0062 16+7 −440±10 " 0.0081 16+5 −455±9 " GRB121024A 2.298 0.2194 4.09 ±0.2 163.7±2.8 [17] 0.2302 4.83±0.2 160.3 ±2.3 " 0.2782 3.82±0.2 182.7 ±3.0 " 0.2928 3.12±0.19 175.3 ±3.5 " 0.3088 3.39±0.18 178.0 ±2.9 " 0.3252 3.49±0.18 180.3 ±3.0 " 0.3412 3.2±0.18 174.5 ±3.3 " 1.2995 2.66±0.6 83.0 ±12.6 " GRB191221B 1.148 0.121 1.4±0.1 68 ±5 [18] 0.417 1.0±0.1 57 ±5 " 2.525 1.3±0.1 62 ±6 " GRB210610B 1.1345 0.0973 4.27±1.45 183±9 This wo k 0.2407 0.22±0.20 267 ±19 " 0.2688 0.03±0.17 – " 0.2793 0.15±0.28 – " 1.2655 2.27±0.22 237 ±3 " 1.2829 1.69±0.27 238 ±5 " GRB210619B 1.937 0.1057 2.2±0.7 22.0 ±10.0 [19] 0.1070 2.6±0.8 2.0 ±8.0 " No es. We include only he alues om li e a u e o which we calcula e P/σP>3.0 wi h Tmid in obse e ame. (∗) F om he spec opola ime ic measu emen s in (Ba h e al. 2003), since we do no conside ch oma ici y in he pola isa ion, we show he median alue o he measu ed pola isa- ion on he di e en wa eleng h bins, o he h ee epochs hey p esen s. (∗∗) Fo GRB030329 we made use o he da a p esen ed in Co ino & Go z (2016) and, speci ically, he esul s p esen ed in his e iew aken om G eine e al. (2003) and Magalhaes e al. (2003). Re e ences o alues om li e a u e a e: [1] (Co ino e al. 1999), [2] (Wije s e al. 1999), [3] (Rol e al. 2000), [4] (Mase i e al. 2003), [5] (Be sie e al. 2003), [6] (Co ino e al. 2003), [7] (Ba h e al. 2003), [8] (Go osabel e al. 2004), [9] (Rol e al. 2003), [10] (Co ino & Go z 2016), [11] (Go osabel e al. 2010), [12] (B i io e al. 2022), [13] (S eele e al. 2009), [14] (Ueha a e al. 2012), [15] (Wie sema e al. 2012), [16] (Mundell e al. 2013), [17] (Wie sema e al. 2014), [18] (U a a e al. 2023), [19] (Manda akas e al. 2023) A216, page 17 o 17