Tem per a ture and hu mid ity mon i tor ing to iden tify ideal pe ri ods for liq ue fac tion on Earth and Mars – data from the High An des
Akos KERESZTURI1, 2, *, Bernadett PAL1 and Akos GYENIS3, 4
1 Konkoly Thege As tro nom i cal In sti tute, MTA Cen tre for Ex cel lence, Re search Cen tre for As tron omy and Earth Sci ences, H-1121 Konkoly Thege 15-17, Bu da pest, Hun gary
2 Eu ro pean Astrobiology In sti tute, vir tual in sti tute hosted by the Eu ro pean Sci ence Foun da tion
3 Korall-Print Bt., H-7625, Dr. Majorossy 43. Pecs, Hun gary
4 Bu da pest In sti tute of Tech nol ogy and Eco nom ics, H-1111 Műegyetem rkp. 3, Bu da pest, Hun gary
Kereszturi, A., Pal, B., Gyenis, A., 2020. Tem per a ture and hu mid ity mon i tor ing to iden tify ideal pe ri ods for liq ue fac tion on Earth and Mars – data from the High An des. Geo log i cal Quar terly, 64 (4): 898–914, doi: 10.7306/gq.1559
Dur ing an al most two week-long field cam paign in the Atacama Desert high al ti tude re gion of Ojos del Salado vol cano, tem - per a ture (T) and rel a tive hu mid ity (RH) val ues were mon i tored on the sur face and <1–5 cm sized rocks, fo cus ing on the night-time val ues. The aim was to iden tify and eval u ate po ten tial tem po ral char ac ter is tics of daily T and RH changes, search - ing for ideal pe ri ods for del i ques cence that has re cently been pro posed for Mars. Al though the at mo spheric pres sure on Mars is much lower than on Earth, and the at mo sphere is drier in gen eral, the huge daily tem per a ture fluc tu a tion there could pro duce el e vated hu mid ity val ues at night-time; this as pect has thus been ana lysed on Earth at a desert lo ca tion, where be - cause of the high el e va tion night-time cool ing is very strong, just like on Mars. Dif fer ent nearby sur face lo ca tions showed the same tem po ral T/RH char ac ter is tics, but ev i dent vari a tions were ob served be tween dif fer ent days. Strong fluc tu a tions could be ob served on 10–20 min ute long tem po ral scales, that might in flu ence the del i ques cence pro cess, and should be ac - counted for in fu ture mis sions aim ing to ana lyse this pro cess on Mars. Night-time pe ri ods were fa vour able for del i ques cence.
Among the mod elled Mars-rel e vant salts [CaCl2, Ca(ClO4)2, Mg(ClO4)2, NaCl] the lon gest du ra tions of pos si ble del i ques - cence were for CaCl2, Ca(ClO4)2 and Mg(ClO4)2, ~7–12 hours for one day. The du ra tion for del i ques cence showed some in - crease along with the ris ing el e va tion, due to the de creas ing night-time tem per a ture. Thus de spite the low hu mid ity on Mars, the cold nights may cause el e vated RH to wards del i ques cence. The Atacama Desert lo ca tions ana lysed are a use ful an a - logue of the del i ques cence pro cess on Mars. Fluc tu a tion in RH was ob served in night-time, sug gest ing that sim i lar vari abil ity might be pres ent on Mars, and that should be con sid ered in the fu ture, in clud ing in eval u at ing how fast the mi cro scopic liq uid for ma tion pro gresses. Night-time slope winds ex pected on Mars might have a strong im pact on the lo cal T/RH con di tions. A more de tailed anal y sis in the fu ture should fo cus on iden ti fy ing and sep a rat ing re gions with and with out much of the ex pected night-time fluc tu a tion.
Key words: del i ques cence, hu mid ity, high moun tain, Mars, Mars an a logue.
INTRODUCTION
This work eval u ates the on-site mon i tor ing-based re sults of night-time T and RH mea sure ments at the Ojos del Salado re - gion of the High-An des, near the Atacama Desert. The del i - ques cence pro cess (by which a sub stance ab sorbs mois ture from the air un til it dis solves in the ab sorbed wa ter and forms a so lu tion), was found to be a re al is tic pos si bil ity pro duc ing mi cro - scopic liq uid wa ter on Mars to day, de spite the dry con di tions. A thin-layer of liq uid so lu tion could form on the top layer of such hy gro scopic salts that have the abil ity to ab sorb mois ture from
the air. Onboard the ExoMars 2020 rover the HABIT in stru ment (Mar tin-Torres and Zorzano, 2018) will be able to de tect this del i ques cence pro cess for the first time. While mod els of Mar - tian at mo spheric be hav iour are of ten used to eval u ate the pos - si bil ity of del i ques cence (Lewis, 2003), they usu ally con sider
“gen eral or av er age” con di tions, which in re al ity may be more vari able. To better un der stand the at mo spheric con di tions’ im - por tance for del i ques cence, this Earth-based Mars an a logue site was sur veyed to ob serve the tem po ral be hav iour of tem per - a ture and hu mid ity val ues (Farris and Davila, 2016), un der partly Mars-rel e vant (re duced at mo spheric pres sure and hu - mid ity) con di tions. The re sults gained there were com ple - mented with Mars-rel e vant at mo spheric mod el ling, as the joint eval u a tion of these two ap proaches can pro vide rel e vant in for - ma tion for the ex pected con di tions on Mars. The con tri bu tion of this work to our cur rent knowl edge is to pro vide some ex am ple mea sure ments of rel a tive hu mid ity and tem per a ture changes in night -time at a high-al ti tude dry and cold Mars an a logue site on the Earth, in or der to in form re search into Mars.
* Corresponding author, e-mail: kereszturi.akos@csfk.mta.hu Received: November 10, 2019; accepted: July 8, 2020; first published online: October 8, 2020
Be sides the oc cur rence of fa vour able con di tions for del i - ques cence, the pres ence of spe cific Mars-rel e vant salts is also im por tant; how ever, our study did not search for these, and this could be the tar get of a fu ture lon ger mis sion. How ever, ear lier in for ma tion ex ist on the min er al ogy of the Atacama re gion, mainly in situ though re cently also from re mote sens ing data (Har ris et al., 2016). These sur veys dem on strated that sev eral Mars-rel e vant hy gro scopic salts are pres ent there. The high est salt con cen tra tion oc curs on salt flats, of ten dom i nated by chlo - rides, es pe cially ha lite (Artieda et al., 2015), and gyp sum (Amundson et al., 2012), which have been con sid ered as use ful Mars an a logue ma te ri als (Osterloo et al., 2008) re gard ing their for ma tion and po ten tial bi o log i cal rel e vance (Weng et al., 2018). Ni trates and per chlor ates were also iden ti fied in vari able abun dance, which are most abun dant in the dri est Yungay re - gion among the sites sur veyed (McKay and Claire, 2016), partly be cause per chlor ates are ex tremely sol u ble, and can re main only at the dri est lo ca tions. Nat u ral per chlor ates (ClO4
-) have been iden ti fied at sev eral dry ar eas on the Earth (in clud ing south west ern United States, south ern Af rica, United Arab Emi - rates, China, Antarctica, and Chile) with con cen tra tions rang ing from 10–1 to 106 mg/kg (Jack son et al., 2015) – and this is abun - dant in the Atacama re gion. Oxychlorines have al ready been pro posed as a rea son for the lack or very low abun dance of or - ganic ma te rial in the soils of the Atacama Desert (Mont gom ery et al., 2019), thus they are im por tant tar gets. At the three tar get sites ana lysed in this study, evaporites oc cur in vari able amounts, al though no spe cific compositional anal y sis has yet been re al ized there.
The Atacama Desert has been con sid ered as a Mars an a - logue area on Earth be cause of its dry ness (Stepinski and Stepinski, 2005), oc cur rence of ox i dants (Catling et al., 2010;
Smith et al., 2014; Jack son et al., 2015) and ex treme hab i tats (Navarro-González et al., 2003; Heldmann et al., 2010; Wierz - chos et al., 2013; Azua-Bustos et al., 2018). The char ac ter is tics of tem per a ture and hu mid ity daily cy cles iden ti fied here, their re peat abil ity, and the rea sons for dif fer ences be tween var i ous daily cy cles, might have Mars-rel e vant con se quences as re - gards for mu lat ing ex pec ta tions for the del i ques cence pro cess there. In this work we mainly fo cus on night-time val ues mea - sured in the High An des; the day time val ues are much more un - cer tain and are noted only for con text. The data we have ob - tained may help to iden tify those tem po ral char ac ter is tics (like cer tain parts of daily cy cles, re quired tem po ral res o lu tion, scale of fluc tu a tions etc.) that should be a fo cus of fu ture ob ser va - tions and plan ning.
Dur ing the geo log i cal his tory of Mars there were pe ri ods when the planet likely had rivers, lakes (Cabrol and Grin, 1999;
Fassett et al., 2008; Czechowski et al., 2013; Ehlmann et al., 2016) and even large stand ing liq uid wa ter bod ies (Tokano et al., 2005). Liq uid wa ter can not ex ist in large vol umes there to - day, mainly be cause of the thin at mo sphere, low tem per a ture and gen eral dry ness. How ever, based on new re sults, it can emerge at a mi cro scopic scale on the sur face of hy gro scopic min er als (Reiss et al., 2010; Nikolakakos and Whiteway, 2018), if the cir cum stances are fa vour able (Losiak et al., 2015;
Pal and Kereszturi, 2017). The nec es sary con di tions, el e vated hu mid ity and tem per a ture can be met on Mars, as the me te o ro - log i cal ob ser va tions of the Rover En vi ron men tal Mon i tor ing Sta tion (REMS) in stru ment onboard Cu ri os ity rover showed a few years ago (Torres et al., 2015). Del i ques cence is an im por - tant step to ward the po ten tial emer gence of bulk liq uid brines (Farris and Davila, 2017). There are nu mer ous other stud ies in - di cat ing the ex is tence of liq uid wa ter or brines on Mars (Zorzano et al., 2009), that might fa cil i tate move ment on the Mar tian sur face in the form of streaks em a nat ing from Dark
Dune Spots (Kere szturi and Rivera-Valentin, 2012, 2016), Re - cur ring Slope Linae (McEwen et al., 2011; Ojha et al., 2015) while what look like drop lets on the leg of the Phoe nix lander have also been ob served (Renno et al., 2009).
BACKGROUND INFORMATION
The study site: the Ojos del Salado (6893 m, 27°06’34” S, 68°32’32” W) is the high est vol cano on Earth (Óscar, 1995), with an el e va tion of 6893 m in the Atacama Desert – Altiplano re gion, above the Puna de Atacama plain. The re gion is a desert with out veg e ta tion or soil cover (Fig. 1). Due to the un du - lat ing to pog ra phy and high al ti tude, per ma frost is pres ent (Cobos and Corte, 1990; Ahumada, 2002) with some gla cier rem nants (Oyarzun, 1987). The mor phol ogy of the sur face is dom i nated by rocks and boul ders; at a few lo ca tions rip ples (rarely dunes) are also ob serv able. Ojos del Salado has the high est al ti tude snow line (Clapperton, 1994), how ever >6500 m the ter rain is mostly (but not al ways) snow cov ered (Amman et al., 2001) and it hosts the high est al ti tude lakes and desert re - gion on Earth.
The cli ma tic con di tions are not well-known, partly be cause of the sto chas tic changes and the poor ob ser va tional cov er age.
From the re mote sens ing anal y sis of Gspurning et al. (2006), it is dif fi cult to de ter mine the snow cov er age, as fresh snow might sub li mate within hours. There is a char ac ter is tic night-time tem - per a ture drop, oc ca sion ally <–10°C even dur ing sum mer time, and above 6000 m it is of ten <–10°C even in day time. The cli - ma tic snow line is sit u ated ~7000 m, and a high-al ti tude desert can be found be low 6000 m (Ammann et al., 2001; Azócar and Brenning, 2010; Nagy et al., 2014a, b; Nagy, 2019). Pre cip i ta - tion is oc ca sion ally pres ent but only in the form of snow (no rain
>5000 m al ti tude), mainly dur ing lo cal win ter. The snow might sub li mate away, but a small amount of ephem eral run off can oc ca sion ally be gen er ated by the melt ing snow.
Tem per a ture and hu mid ity mon i tor ing have al ready been re al ized in the Atacama re gion at lower al ti tude lo ca tions, and at other Mars an a logue sites on the Earth. Azua-Bustos et al.
(2015) sur veyed the dri est Yungay re gion of the Atacama Desert and the even drier María Elena South lo ca tion, re port ing a mean at mo spheric rel a tive hu mid ity of 17.3%, a con stant 14%
at the depth of 1 m in the soil, and found a num ber of vi a ble bac - te rial spe cies there. Based on the cli mate and mois ture data of four years (1994–1998) re ported by McKay et al. (2003) there was only one sig nif i cant rain event of 2.3 mm, near mid night lo - cal time. This may have been a rain out of a heavy fog that pro - duced liq uid wa ter in the soil for 65–85 h over the 4 years. How - ever, all of these ear lier mea sure ments were re al ized at much lower al ti tudes (usu ally <1100 m el e va tion), thus it is im por tant to ex tend the anal y sis to higher lev els, as at a lower-pres sure en vi ron ment the at mo spheric vari abil ity might be larger, and closer to that rel e vant for the Mar tian sur face. Un for tu nately no re mote data re cord ing was re al is tic at this al ti tude, be cause the strong winds of ten re dis trib ute dust, and any un guarded de tec - tor may be cov ered with a thick sed i ment layer and pro duce false data. As a re sult only a rel a tively short in ter val of data re - cord ing was pos si ble here, dur ing a manned ex pe di tion.
For del i ques cence to take place on Mars, el e vated at mo - spheric hu mid ity and hy gro scopic salts need to be pres ent on the oth er wise gen er ally dry planet. Hu mid ity can be el e vated dur ing night-time be cause of the very low tem per a ture, and so the rel a tive hu mid ity could in crease sub stan tially. Un der the con di tions mea sured by the REMS in stru ment at Gale Crater on Mars, some chlo rine-con tain ing salts can hy drate and oth ers
de hy drate on daily cy cles based on lab o ra tory tests (Gough et al., 2019), how ever, at mo spher i cally based hydration of salts may be very slow. Aque ous so lu tions from the in ter ac tion of var i ous per chlor ates are ex pected to emerge around the Phoe - nix land ing site but not prob a ble at the Cu ri os ity land ing site of Gale crater (Primm et al., 2018); how ever, del i ques cence is ex - pected there (Rivera Valentín et al., 2018). Ana lys ing the Phoe - nix land ing site, RH was usu ally <5% in the day and >95% at night-time, the tran si tion be tween the two val ues be ing rapid (Zent et al., 2016). Ac cord ing to re cent re sults (Mar tin-Torres et al., 2015; Pal and Kereszturi, 2017) del i ques cence is a re al is tic pos si bil ity for mi cro scopic liq uid wa ter emer gence on Mars un - der cur rent con di tions. Al though in the High-An des re gion there are var i ous hy gro scopic salts, which might pro vide ideal sur - faces for the del i ques cence pro cess, in this work only the me te - o ro log i cal con di tions, and Mars-re lated the o ret i cal oc cur rence of del i ques cence, were ana lysed.
Per chlor ates and chlo rates were first iden ti fied by the Wet Chem is try Lab o ra tory ex per i ment on the Phoe nix Lander on Mars (Hecht et al., 2009; Kounaves et al., 2010; Toner et al., 2014a). The ex is tence of per chlor ates was fur ther in di cated by the Ther mal and Evolved Gas An a lyzer (Boynton et al., 2009), of the Sam ple Anal y sis at Mars in stru ment onboard the Cu ri os - ity rover (Glavin et al., 2013) and per chlor ates were also de - tected spec tro scop i cally by the Mars Re con nais sance Or biter
(Ojha et al., 2015). Per chlor ates and chlo rates are pop u lar among re search ers due to their abil ity to con sid er ably lower the freez ing point of wa ter, and to their hy gro scopic na ture. These prop er ties sug gest the po ten tial for liq uid wa ter on Mars in spite of the cold and dry, frigid con di tions (Gough et al., 2011; Nuding et al., 2014; Toner et al., 2014a, 2015a, b). Be cause of the great vari abil ity of me te o ro log i cal con di tions on Mars, sev eral as - pects of the Mar tian H2O cy cle are await ing dis cov ery (Mar ti nez et al., 2017).
METHODS
In this study, in situ tem per a ture and hu mid ity mea sure - ments were ana lysed and com pared to Mar tian at mo spheric mod el ling-based trends. At the field sites at mo spheric and shal - low subsurface tem per a ture and rel a tive hu mid ity were mea - sured us ing 8 sen sors per site. These were used at three dif fer - ent al ti tudes dur ing the field trip on the slopes of Ojos del Salado vol cano in the High An des and Atacama re gion for a to - tal pe riod of 12 days, at 3810, 4328 and 5260 m a.s.l.
(Kereszturi, 2019).
SHT21 Hu mid ity and Tem per a ture Sen sor IC in stru ments (4 x 1.5 x 0.5 cm) pro duced by Sensirion com pany, were used.
Fig. 1. Im ages of the area, show ing the gen eral out look of the bar ren, rocky de bris-cov ered sur face (A, B) and a map of the lo ca tions of mea sure ments (C)
These were equipped with ca pac i tive type hu mid ity sen sors, band gap tem per a ture sen sors and a spe cial ized dig i tal in te - grated cir cuit – all on a sin gle CMOS chip. The sen sors pro vide cal i brated, linearized sig nals in dig i tal I2C for mat. The res o lu - tion of SHT2x can be changed by com mand (8/12bit up to 12/14bit for RH/T). The RH res o lu tion of the in stru ment is better than 0.7%, but can reach up to 6% be low zero Cel sius; the tem - per a ture res o lu tion is better than 0.04°C (op er at ing range –40 to 125°C).
The de tec tors ac quired mea sure ments ev ery 5 min utes.
They were ex posed on the sur face to be able to re act quickly to
changes in the en vi ron ment, and ar ranged in a 1 m2 area <1 m dis tance apart at each site (Fig. 2). Four types of plac ing were used in or der to see pos si ble dif fer ences re lated to cm-scale top o graphic un du la tions (shad ow ing). The first plac ing type was a nearly smooth area di rectly on the sur face, the sec ond at the bot tom of a 12–14 cm di am e ter, 8–12 cm deep pit (de pres sion), the third next to a stone (min. 8 cm high) at its east and west sides, the fourth at the bot tom of a 8–12 cm deep pit as noted above, but cov ered with a stone from above (that al lows air to flow through the de pres sion; Fig. 2).
Fig. 2. The plac ing of de tec tors (yel low-green box with black rod)
A – ex posed on a bar ren sur face; B – in side a small pit; C, D – be side rocks; E, F – in side a pit cov ered by a rock;
the 5 cm scale at the bot tom is ap prox i mately valid for all in sets
Al though the mea sure ments tar geted the night-time pe ri - ods, some ob ser va tions from the day time dataset are in ter est - ing too. In or der to pro vide fur ther com pare with Mars-rel e vant as pects, we ana lysed the mea sure ments in con trast with Mar - tian Re cur ring Slope Lineae sites, Horowitz Crater and New ton Ba sin (McEwen et al., 2011). We ob tained sur face tem per a ture and rel a tive hu mid ity curves from model cal cu la tions by Laboratoire de Météorologie Dynamique Mars Gen eral Cir cu la - tion Model (LMDZ GCM), de tailed in For get et al. (1999), in - clud ing a wa ter cy cle as de scribed in Navarro et al. (2014).
To cal cu late the rel a tive hu mid ity on Mars, we need the wa - ter vapour vol ume mix ing ra tio at the sur face. Since a GCM does not model this at the Mar tian sur face, but at ~4 m above it, we as sume that the area be tween the sur face and this height layer is well-mixed (this as sump tion is roughly ac cept able as the best avail able ap proach, be cause of near-sur face mix - ing-in duced wind shear even in a rel a tively sta ble at mo sphere).
With this ap prox i ma tion, we con sider that the wa ter vapour vol - ume mix ing ra tio at the sur face is equal to the value at ~4 m.
The rel a tive hu mid ity is then de rived as:
RH Q
Qsat
= 0
where: Q0 is the es ti mated wa ter vapour vol ume mix ing ra tio at the sur face.
Ac cord ing to Mar tin-Torres et al. (2015) and Pal and Kereszturi (2017) del i ques cence is a re al is tic pos si bil ity for mi - cro scopic liq uid wa ter emer gence on Mars un der cur rent con di - tions. Al though in the High An des re gion there are var i ous hy - gro scopic salts which might pro vide the ideal sur faces for the del i ques cence pro cess, in this study only the me te o ro log i cal con di tions and Mars-re lated the o ret i cal oc cur rence of del i ques - cence were ana lysed.
We ex am ined the ex pected po ten tial be hav iour of 5 salts al - to gether, eval u at ing the model based re sults and the mea sured
pa ram e ters in or der to see whether these com po nents liq uefy on Mars and the ana lysed Earth an a logue lo ca tions: cal cium per chlor ate: Ca(ClO4)2 (Toner et al., 2014b), mag ne sium per - chlor ate: Mg(ClO4)2 (Möhlmann and Thom sen, 2011), cal cium chlo ride: CaCl2 (Davila et al., 2010), so dium chlo ride: NaCl and mag ne sium sul phate: MgSO4 (Chevrier et al., 2012). Mag ne - sium per chlor ate was iden ti fied by the Phoe nix lander (Hecht et al., 2009) and cal cium per chlor ate was also iden ti fied at the Mar tian sur face; these are the most prob a ble salts to liq uefy through del i ques cence un der cur rent con di tions, mostly be - cause of their low eutectic tem per a ture.
RESULTS
The nu mer i cal sum mary of the ob served tem per a ture and hu mid ity val ues can be seen in Ta ble 1, while some ob served typ i cal daily curves at the Ojos del Salado vol cano are shown in Fig ure 3. The tem per a ture and rel a tive hu mid ity curves ex hibit gen eral trends that are re lated mostly to the daily cy cles, with in - creased tem per a ture in the day and in creased hu mid ity at night-time, al though they showed some spe cific fea tures too.
These ob ser va tions sug gest that the pa ram e ters re corded are more site- (and lo cal con di tions-) de pend ent than re lated to any pos si ble spe cific char ac ter is tics of cer tain de tec tors. The tem - per a ture and hu mid ity curves are anticorrelated to each other, ac cord ing to ex pec ta tions.
An over view of val ues of the tem per a ture and hu mid ity mea sured are in di cated in Ta ble 1 where all de tec tors were used to cal cu late the av er age, min i mum and max i mum val ues for night-time tem per a tures, us ing all sen sors from 2 days at Laguna Santa Rosa, 2 days at Laguna Verde, and 4 days at Atacama Base Camp. For day time val ues our own mea sure - ments were com bined with other stud ies (Gspurning et al., 2006; Azócar and Brenning, 2010) to eval u ate the range and min i mum and max i mum val ues. Eval u at ing the num bers in Ta - ble 1, subzero tem per a tures might ap pear at all lo ca tions even
T a b l e 1 Sum mary of mea sure ments
Lo ca tion Laguna Santa
Rosa Laguna Verde Atacama Base Camp
Day time T [C°]
min 1.4 –2.7 –12.2
max 22 23 18
av er age 14.4 8.6 8.0
Day time RH [%]
min 3.0 6.7 0.7
max 78.5 98.4 89.7
av er age 11.4 20.8 12.4
Night time T [C°]
min –0.7 –3.9 –13.5
max 17.3 17.5 7.8
av er age 3.4 2.9 –2.3
Night time RH [%]
min 30.7 26.2 32.2
max 85.3 100.0 94.7
av er age 29.6 39.1 33.0
Day time tem per a tures are not part of the anal y sis
in day time too, how ever, even at night-time it was not nec es - sary to go be low zero. The day time and night-time av er age val - ues both de creased with in creased al ti tude. A high range of daily fluc tu a tion can be ob served for both tem per a ture and hu - mid ity. The rel a tive hu mid ity in par tic u lar could show ex treme val ues; in the day time it was usu ally very low, ~11–21% but rarely (dur ing snow fall) could reach sat u ra tion. At night-time, it never de creased be low ~26–30%, but could also reach sat u ra - tion. Ac cord ing to ex pec ta tions, the av er age night-time RH was higher than the av er age day time RH.
In Ta ble 1 we can see the over all day time and night-time tem per a ture and rel a tive hu mid ity val ues by mea sur ing lo ca - tion. As the al ti tude in creases the tem per a ture drops, which shows here in the day time and night-time min i mum tem per a - tures with an ~10°C de crease be tween the low est and the high - est mea sur ing lo ca tions. The same is ap par ent if we look at the day time and night-time av er age tem per a tures. As the pres sure de creases with in creas ing el e va tion, the amount of wa ter that the at mo sphere can hold also de creases. How ever, with lower amounts of wa ter vapour in the air, rel a tive hu mid ity lev els could be high be cause of the de creas ing tem per a ture. This might be the rea son for the ta ble show ing no clear trend in the rel a tive hu mid ity min i mum, max i mum or av er age lev els – in colder nights, a smaller ab so lute hu mid ity can pro duce el e vated rel a tive hu mid ity.
A gen eral over view of the re corded data and char ac ter is tics of daily curves are shown in Fig ure 3. The curves re corded by
dif fer ent de tec tors show sim i lar trends on the same days (com - par ing the curve sec tions be low and above each other). This shows that the lo ca tion types of the dif fer ent de tec tors cause much smaller dif fer ences than the dif fer ences be tween var i ous days. How ever dif fer ences be tween par tic u lar days could be eas ily iden ti fied. Vari abil ity on an hourly scale (the fin est ob - serv able fluc tu a tion of the curves) is also pres ent, es pe cially in the night-time hu mid ity val ues, for al most all nights.
There were strong fluc tu a tions in cer tain parts of the dataset. We com pared these fluc tu a tions be tween nearby de - tec tors with each other, to see if they all re corded the same vari - abil ity (see some ex am ples for the Laguna Santa Rosa site in the Fig ure 4A), while in the case of rel a tive hu mid ity mea sure - ments, the trends be tween dif fer ent de tec tors were sim i lar. The num bers by the curves show the iden ti fi ca tion num ber of each de tec tor there. The fluc tu a tions show the same trends at dif fer - ent de tec tors roughly 1 m apart, both on ex posed sur faces or in shad owed and cov ered lo ca tions. Sim i lar trends are vis i ble at all de tec tor lo ca tion types: a small pit (51, 54), next to a stone (58, 57), un der a stone (55). Since sim i lar trends are ob serv able at all de tec tor lo ca tion types, these small place ment dif fer ences (in side a pit, next to or un der a stone etc.) do not in flu ence the trends sub stan tially.
Whole-night ex am ple curves are vis i ble in Fig ure 4B. The night-time tem per a ture (lower) and rel a tive hu mid ity (up per) curve pairs of day 2 (thicker) and day 1 (thin ner) of de tec tor no.
52 are pro vided. They show ex am ples for a cloudy, less cold Fig. 3. Sum mary of all curves re corded where dif fer ent colours mark dif fer ent de tec tors (top right cor ner)
Please note that the last two days of the ob ser va tions have not been con sid ered dur ing the anal y sis as in these cases the de tec tors were sit - u ated close to the lo ca tion of site where melt ing was tak ing place and were cov ered by wet sed i ments; thus, they were not re al is tic. Plac ing of the de tec tors at each site was done ac cord ing to: 51 – ex posed on a bar ren sur face, 52 and 54 – in side a small pit, 55 and 56 – be side rocks, 57 and 58 – in side a pit cov ered by a rock
and windy (day 1) and a cloud less thus colder (day 2) night-time with much smaller fluc tu a tions in the sec ond case.
Night-time tem per a ture vari a tions could also be ob served.
These were usu ally not larger than 1–2°C dur ing an hour in night-time, but oc ca sion ally could reach up to 2°C rise/fall in 10 min utes. The tem po ral den sity and am pli tude of fluc tu a tions in - creased along with the top o graphic el e va tion. The level of fluc - tu a tion also showed dif fer ences be tween var i ous days, for ex - am ple in the Atacama Base Camp on day 2 the day time fluc tu a - tions were ~2–4°C while on day 1 dur ing >10 oc ca sions they reached >5°C.
Some spe cific events are worth high light ing. Rel a tively brief changes in tem per a ture and re lated in crease in rel a tive hu mid - ity could be ob served at sev eral sites and dur ing sev eral pe ri - ods. The rea son prob a bly re lates to the strong winds that can carry some what warmer or colder air masses to the lo ca tion of the de tec tor. A snow storm at Laguna Verde on day 2 pro duced sat u rated con di tions dur ing the night for 6–7 hours. At the Atacama Base Camp the daily fluc tu a tions were much stron ger on day 1 than on day 2, to gether with more el e vated RH on day 2 (av er age 40.6%) than day 1 (25.7%).
POSSIBILITY FOR DELIQUESCENCE
Fig ure 5 be low shows the pre sumed time in ter vals (as ver ti - cal blue bars), when the cir cum stances are fa vour able for del i - ques cence of cer tain salts. The cal cu lated rel a tive hu mid ity val - ues are shown on the left y axis, and the sur face tem per a tures on the right y axis. The eutectic tem per a tures and rel a tive hu - mid ity lev els needed for del i ques cence are in di cated with hor i -
zon tal lines de not ing the salt in ques tion. The ideal time pe ri ods are in di cated by light blue rect an gles show ing the be gin ning and end of the ideal win dow.
In Fig ure 5, we can see that the rel a tive hu mid ity stayed al - most con stant through out the night on the sec ond day, re sult ing in long and con tin u ous pos si ble pe ri ods suit able for del i ques - cence. On the third day the rel a tive hu mid ity fluc tu ated quite heavily dur ing the night, and this re sulted in shorter, more scat - tered ideal pe ri ods; the over all ideal time win dow did not change be tween the two days. Ta ble 2 gives a sum mary for two ex am - ple days at the Atacama Base Camp, which was the high est al - ti tude among the lo ca tions ana lysed, prob a bly the most Mars-rel e vant one among the sites sur veyed.
In Ta ble 2 one can see the two ex am ple days show ing dif - fer ent del i ques cence char ac ter is tics. On day 2 at the Atacama Base Camp all the salts ex cept Mg(SO)4 could liq uefy, and the ideal time pe ri ods are long and con tin u ous. On day 3 the rel a - tive hu mid ity lev els were fluc tu at ing through out the night and dropped be low 60% af ter ~2 am, which led to shorter and more scat tered ideal pe ri ods.
In Fig ure 6 one can see the to tal length of ideal time in ter - vals for del i ques cence. The mea sure ments were taken at Laguna Santa Rosa for 1 day, Laguna Verde for 2 days and at the Atacama Base Camp for 4 days. There were 7 dif fer ently placed de tec tors at each site and their mea sured ideal time win - dows were summed for each day. There fore there are more mea sure ment points at the Atacama Base Camp than at Laguna Santa Rosa for ex am ple, be cause the same de tec tor has 4 points at the for mer site, one for ev ery day with ideal time pe ri ods. An ex am ple of our cal cu la tions is as fol lows: at the third day at Atacama Base Camp, the de tec tor counted 51 (ex posed Fig. 4. Ex am ple of night-time rel a tive hu mid ity curves
A – 3rd day at the Atacama Base Camp of the dif fer ent de tec tors, to dem on strate that the change in rel a tive hu mid ity re corded by 5 dif fer ent de tec tors, lo cated close to each other, show the same RH val ues (ver ti cal axis) as time elapsed (hor i zon tal axis); B – night-time tem per a ture (lower) and rel a tive hu mid ity (up per) curve pairs of day 2 (thicker) and day 1 (thin ner) curves of de tec tor no. 52
Fig. 5. Daily tem per a ture and rel a tive hu mid ity curves of four Mars-rel e vant salts
The cal cu lated rel a tive hu mid ity val ues are shown on the left y axis, and the sur face tem per a tures on the right y axis. In the A, B, C and D in - sets the sec ond day of mea sure ment is shown at the Atacama Base Camp with 4 dif fer ent salts de noted. In in sets E, F, G and H the next day is shown at the same lo ca tion with the same salts. There is a no ta ble dif fer ence be tween the rel a tive hu mid ity be hav iour of the two days. On the third day (right col umn) it fluc tu ated dur ing the night, caus ing the ideal time pe ri ods to be di vided into shorter du ra tions
on the bar ren sur face) mea sured ideal con di tions for Ca(ClO4)2
be tween 19:00–21:40, 22:30–03:20 and 05:00–05:10 (the min i - mum ideal win dow was 10 min utes, shorter pe ri ods were ex - cluded). This summed up is 7 hours and 30 min utes, 7.5 hours, which is then shown as a sin gle point in Fig ure 6. Cal cium per - chlor ate [Ca(ClO4)2] and mag ne sium per chlor ate [Mg(ClO4)2] showed the lon gest to tal ideal time pe riod, with cal cium chlo ride (CaCl2) be hind them with a few hours in to tal. The same “or der of salts” could be iden ti fied at each of the three sites, sug gest - ing that ma te rial-re lated prop er ties of the salts could strongly in - flu ence the re sult at each site. Mag ne sium sul phate [MgSO4(H2O)] could only liq uefy through del i ques cence at Laguna Verde, while so dium chlo ride (NaCl) could liq uefy at all the lo ca tions mea sured; how ever, the to tal times were usu ally shorter than those for the per chlor ates.
Tak ing Ta ble 1 into con sid er ation, we can see that Laguna Santa Rosa shows shorter ideal time pe ri ods, most likely due to the lower av er age rel a tive hu mid ity. Out of the three lo ca tions, Laguna Santa Rosa has the high est tem per a tures, both av er - age and min i mum. How ever, the av er age rel a tive hu mid ity is the low est here. Laguna Verde has the high est rel a tive hu mid ity val ues ac cord ing to the data, but given the snow fall here, the rel a tive hu mid ity av er age is bi ased by the sat u rated de tec tors.
As we would as sume, the av er age tem per a tures are the low est at the high est al ti tude, at the Atacama Base Camp. We see no sig nif i cant vari a tion be tween Laguna Verde and the Base Camp. This is most likely due to the bal ance be tween their re - spec tive tem per a ture and rel a tive hu mid ity lev els; this also sug - gests that the true av er age rel a tive hu mid ity of Laguna Verde
should be some where be tween the av er ages of Laguna Santa Rosa and the base camp.
In Fig ure 7 one can see how the ideal time pe ri ods for del i - ques cence were ar ranged by lo cal time, lo ca tion and salt type.
The be gin nings and the ends of ideal pe ri ods are de noted by dif fer ently shaped and col oured sym bols ac cord ing to the spe - cific salts. At Laguna Santa Rosa we can see that al most ev ery ideal win dow is at least 5 hours long, start ing ei ther ~10 pm, or
~2 am. There were only 4 in stances when the ideal win dow was shorter. At Laguna Verde there is a sig nif i cant group start - ing ~7 pm and end ing be tween 3 and 10 am grad u ally, de - pend ing on the salt. An other group is from 8–9 pm to 10–11 am. There are few in stances dur ing the night, as well as a few short time win dows in the early morn ing hours. The third group is lo cated at the top right, start ing in the late af ter noon hours and end ing be fore mid night. Atacama Base Camp (the high est el e va tion site) shows the most ideal win dows, but this is also be cause there are 4 days of data, while there are only 2 days from Laguna Verde, and 1 day from Laguna Santa Rosa. On the graph there is an al most con tin u ous 45° line con sist ing of very short time win dows, pos si bly due to tem per a ture or rel a - tive hu mid ity fluc tu a tions close to the min i mum ET (eutectic tem per a ture) and WA (wa ter ac tiv ity) thresh olds – such fluc tu - a tions were less abun dant at the two other sites (Laguna Verda, Laguna Santa Rosa) than at the high est el e va tion. The group seen at Laguna Verde, start ing from ~7 pm and end ing be tween 3 and 10 am, is also vis i ble here. Note the empty re - gion be tween ~11 and 4 pm, when there seems to be no con di - tions for del i ques cence.
T a b l e 2 Sum mary of ex am ple salts’ del i ques cence pe ri ods on days at the Atacama Base Camp
Lo ca tion Salt type Be gin ning of ideal pe riod End of ideal pe riod To tal du ra tion Date Lo cal time Date Lo cal time [h]
Atacama Base Camp 2. day
CaCl2 2018/2/19 19:15 2018/2/20 06:10 10.9
Ca(ClO4)2 2018/2/19 19:05 2018/2/20 18:55 12.2
Mg(ClO4)2 2018/2/19 19:05 2018/2/20 18:55 11.9
Mg(SO)4 – 00:00 – 00:00 0.0
NaCl 2018/2/19 21:40 2018/2/20 05:55 8.3
Atacama Base Camp 3. day
CaCl2
2018/2/20
19:15
2018/2/20
19:25
7.2
19:35 21:35
22:40 23:00
23:50
2018/2/21
00:00
2018/2/21
00:10 00:50
01:10 02:15
02:25 05:10
Ca(ClO4)2
2018/2/20
19:00
2018/2/20 21:40
7.4
22:40 23:00
23:15
2018/2/21 03:20
2018/2/21 05:00 05:20
Mg(ClO4)2 2018/2/20
19:00
2018/2/20 21:35
8.6
22:40 23:00
23:25 2018/2/21 05:10
Mg(SO)4 – 00:00 – 00:00 0.0
NaCl 2018/2/20 20:30 2018/2/20 21:10 0.6
DISCUSSION
The T/RH val ues at the dif fer ent de tec tors, lo cated close to each other at dif fer ent al ti tudes, showed sim i lar trends (see Fig.
4A). This im plies that the small-scale (cm) dif fer ences in the spa tial dis tri bu tion of rocks, and lo ca tions ex posed or shel tered from the wind, did not over come the an a logue trends. Even be - low rocks at 1–3 cm depth, prob a bly be cause of the con stant wind, no sub stan tial dif fer ence could be iden ti fied com pared to the open air lo ca tions in the T/RH val ues re corded. This sug - gests that large dif fer ences (>1–2°C and 2.4% for RH) are not ex pected ac cord ing to the lo cal con di tions at sim i lar cm-scale lev els on Mars (above/next to/un der cm-sized stones). How - ever, due to the lower at mo spheric den sity, and re lated smaller heat trans port, on Mars, larger dif fer ences in T/RH pa ram e ters can not be ex cluded and there fore need con sid er ation there.
The re sults also sug gest that un der Mar tian sur face con di tions, with even lower at mo spheric pres sure, rel a tively large fluc tu a - tions are ex pected in T and RH pa ram e ters, that the mod els can not sim u late yet (see be low).
In agree ment with the ex pec ta tions, del i ques cence seems to be pos si ble un der the hyperarid con di tions of the Atacama Desert site ana lysed us ing only the avail able small wa ter vapour con tent of the at mo sphere. Sim i lar trends were ob -
served over all for the Atacama sites and mod el ling-based cal - cu la tions for Mars. Night-time del i ques cence was pos si ble at all of these sites, and Mars-rel e vant salts con sid ered are CaCl2, Ca(ClO4)2, Mg(ClO4)2 (sup port ing del i ques cence for 10–12 h du ra tion, al most all night long), while Mg(SO)4 and NaCl were less ideal for this pro cess with shorter pe ri ods, and for NaCl no del i ques cence may be pres ent at all. This large dif fer ence be - tween the spe cific salts points to the im por tance of hygro - scopicity, what makes a dif fer ence in sim i lar T and RH con di - tions. Al though the few days cov ered by the on-site sur vey was not long enough to ob tain sta tis ti cally valid re sults rel e vant for an nual du ra tions, be cause of el e vated T and RH fluc tu a tions in day 3 at the Atacama Base Camp, the to tal du ra tion of pe ri ods fa vour able for del i ques cence was de creased by 30–40% and di vided into sep a rate pe ri ods. If such fluc tu a tion also takes place on Mars, the mod el ling-based av er age re sults should be con sid ered care fully.
These fluc tu a tions in creased along with in creased el e va - tion, what may not be solely re lated to the lower at mo spheric pres sure and air den sity (e.g., the de creased ther mal in er tia), be cause for ex am ple at the Atacama Base Camp on day 2 the fluc tu a tions are sub stan tially smaller in gen eral than on day 1.
These fluc tu a tions were prob a bly caused by at mo spheric con - di tions, for ex am ple wind fluc tu a tions. Wind can carry air masses with dif fer ent tem per a tures from nearby lo ca tions, Fig. 6. The to tal amount of time when the cir cum stances were ideal for cer tain salts to liq uefy through del i ques cence
(dif fer ent salts marked by dif fer ent sym bols, see key at top right)
The y axis shows the to tal length of ideal time per day. There were al to gether 7 de tec tors, each of them po si tioned dif fer ently. Each sym bol shows the to tal amount of ideal time for 1 spe cific salt at 1 spe cific de tec tor summed for 1 day. The mea sure ments spanned 1 day at Laguna Santa Rosa, 2 days at Laguna Verde and 4 days at the Atacama Base Camp
which might have other sur face tem per a tures be cause of dif fer - ent day time ex po sure, albedo or ther mal in er tia. These fac tors are im por tant as the small night-time tem per a ture fluc tu a tions pro duce rel a tively large fluc tu a tions in rel a tive hu mid ity. These RH val ues might fluc tu ate be tween be ing be low and above the crit i cal del i ques cence rel a tive hu mid ity level on the same night – thus in flu enc ing the emer gence of mi cro scopic liq uid, and this pos si bil ity should be tar geted by fu ture anal y sis.
As the al ti tude in creases, the av er age tem per a ture and the pres sure de crease. With the at mo sphere be com ing more rarified at higher el e va tions, the to tal amount of wa ter vapour that the at mo sphere can po ten tially hold is de creased; how - ever, it is still pos si ble to reach high rel a tive hu mid ity val ues.
There is a del i cate bal ance be tween sur face tem per a ture and rel a tive hu mid ity val ues when it co mes to ideal cir cum stances for del i ques cence. Be fore the tem per a tures drop be low the eutectic tem per a tures of the ex am ined salts, the in creas ing al ti - tude can re sult in lon ger ideal pe ri ods of del i ques cence. We can see this if we com pare the mea sure ments re corded at Laguna Santa Rosa with Laguna Verde, but it is not as ob vi ous if we com pare Laguna Verde with Atacama Base Camp.
Based on the ex pe ri ence gained by this pro ject, the next step could be the in stal la tion of a real del i ques cence de tec tor there, re sem bling the HABIT onboard ExoMars 2020 rover (Mar tin-Torres and Zorzano, 2018), which is pro posed at lower al ti tude by Farris and Davila (2017).
MARS-RELEVANT ASPECTS
At mo spheric tem per a ture vari abil ity, es pe cially dur ing eve - ning hours (Ham il ton et al., 2014), as well as daily vari abil ity in hu mid ity (Mar ti nez et al., 2016) was ob served on Mars by the REMS in stru ment onboard Cu ri os ity rover. Pres sure fluc tu a - tions were also ob served by Cu ri os ity of up to 0.2 Pa in day time and 1 Pa at night-time over a mat ter of sec onds/min utes, pos si - bly re lated to upslope/downslope winds. In the night, top o - graphic slope winds con trib uted to pro duc ing si mul ta neous changes in sur face tem per a ture. Air tem per a tures fluc tu ated with am pli tudes of ~2 K, while the ground tem per a ture showed 0.2–0.5 K vari a tions prob a bly by forced con vec tion (Ullán et al., 2017).
In Fig ure 8A we can see the daily sur face tem per a ture and rel a tive hu mid ity curves at Ely sium Planitia, the land ing site of the In Sight mis sion. The boxes mark the be gin nings and ends of the ideal pe ri ods for Ca(ClO4)2 to liq uefy through del i ques - cence. The cir cum stances are ideal dur ing the night, sim i lar to what we can see in the Earth mea sure ments above. The ideal time is ex pected to be con tin u ous from ~8 pm un til al most 11 pm. There is an other short win dow just af ter 7 am. Dur ing the night, the sur face tem per a ture drops be low the eutectic tem per - a tures. In Fig ure 8B we can see an other type of be hav iour at the Phoe nix land ing site, Green Val ley. Here both cal cium per chlor - ate and mag ne sium per chlor ate could liquify ac cord ing to the model cal cu la tions. The ideal times are in the early morn ing hours from ~5:30 am un til 8:30 am, with a short (~30 min ute) ideal win dow for mag ne sium per chlor ate as well. The eve ning is ideal from ~9 pm ex tend ing to ~11:30 pm. The pro posed ideal time for mag ne sium per chlor ate is again ~30 min utes long in the eve ning.
The vari a tions in tem per a ture at the sur veyed High An des sites, con cen trated be tween 10–16 lo cal time, sug gest that they might be re lated to strong so lar in so la tion. A con nec tion with lo - cal wind is also pos si ble, but it is im por tant to note that on calm days, the wind starts to blow in the af ter noon, and on windy days there was con tin u ous strong wind all day, but still stron ger Fig. 7. The be gin ning and the end of each ideal time pe riod
for all of the salts ana lysed, sep a rated by al ti tude (lo ca tions) The be gin nings and ends of ideal pe ri ods are de noted by dif fer ently shaped and col oured sym bols ac cord ing to the spe cific salts. The x axis shows the be gin ning of the ideal time win dow, and the y axis shows its end, both ac cord ing to lo cal time. Each point rep re sents one in ter val of the spe cific salt at a given al ti tude
Fig. 8. Sim u lated daily sur face tem per a ture and rel a tive hu mid ity curves on Mars for com par i son
The boxes mark the be gin nings and ends of the ideal pe ri ods for Ca(ClO4)2 (light blue) and for Mg(ClO4) (dark blue) to liq uefy through del i ques cence. The land ing site of In Sight, Ely sium Planitia is at the top (A) with Green Val ley, the land ing site of Phoe nix rover at the bot tom (B). The ideal times for del i ques cence are de noted by blue rect an gles. The min i mum wa ter ac tiv ity and tem per a ture lev els with the name of the given salt are shown with hor i zon tal lines
Fig. 9. Com par i son be tween the daily sur face tem per a ture (red) and rel a tive hu mid ity (blue) curves at the Atacama Base Camp (dashed lines) and two RSL sites in Horowitz Crater and New ton Ba sin (solid lines) on Mars
At the top there is an ex am ple for mark edly fluc tu at ing rel a tive hu mid ity lev els (A), while at the bot tom there is one for a mostly con stant night-time rel a tive hu mid ity (B); the over all trends of the curves are sim i lar, but the sim u lated Mar tian val ues show
greater daily vari a tions, as ex pected
in the af ter noon. Thus it is not likely that the wind would be the main rea son for the fluc tu a tions on Earth.
In Fig ure 9 there is a com par i son be tween the Atacama Base Camp mea sure ments and model cal cu lated val ues at two re cur ring slope lineae (RSL) sites (McEwen et al., 2011) on Mars. These RSL-s are dark fea tures on the sur face of Mars that were pre vi ously con sid ered as ev i dence for liq uid sub surface flow of wa ter due to their changes in shape and col our. New re search in ter prets them as pos si ble gran u lar flows, where the slope is steep enough for grains of dust and sand to slide down (Dundas et al., 2017). The ques tion of whether the RSL are caused by gran u lar flows or liq uid wa - ter/brine ac tiv ity is still un der de bate to day. The over all trends are sim i lar, but the sim u lated Mar tian re sults show greater daily vari a tions in both cases, as ex pected. The Mar tian curves have a 1 hour tem po ral res o lu tion, and they were cal cu - lated by the LMDZ GCM de tailed in Forget et al. (1999), in clud - ing a wa ter cy cle (Navarro et al., 2014). The rel a tive hu mid ity val ues were de ter mined as de scribed in Pal and Kereszturi (2017). Both the daily tem per a ture and rel a tive hu mid ity fluc tu - a tions are much larger on Mars, but small-scale fluc tu a tions are vis i ble only on Earth as the Mar tian dataset can not modelthe pro cess on this scale. Such fluc tu a tions ob vi ously could in flu ence the tem po ral be hav iour of del i ques cence, as it could ex ist on Mars, even though the model data (Fig. 9) is not able to show it.
Al though the pres sure and tem per a ture con di tions barely over lap be tween Mars and these Earth lo ca tions, the style of T/RH fluc tu a tion from the Atacama re gion point to con di tions and pro cesses that might be rel e vant for Mars. For ex am ple the RH val ues on Mars at the Phoe nix land ing site were close to ~0% at –13.5°C day time (Farris et al., 2017), but thus cor re - spond with the low est tem per a ture re corded at the Atacama Base Camp. The tem po ral be hav iour shows rel e vance be - tween the two plan ets (very dry day time and oc ca sion ally sat - u rated night-time), though the ac tual val ues did not over lap.
The im por tance and Mars rel e vance of the T/RH be hav - iour at the site ana lysed is its high el e va tion that pro duced large daily fluc tu a tions, close to sat u ra tion, even in an oth er - wise dry re gion (sim i lar to Mars, where the 60–100°C daily fluc tu a tion could re sult in close-to-sat u ra tion con di tions de - spite the low to tal wa ter vapour abun dance). An other im por - tant an a logue as pect is the high vari abil ity, down to sev eral min utes, that was also partly ob served on Mars by the REMS in stru ment onboard the Cu ri os ity rover, though this is not yet in volved in the at mo spheric mod els de scrib ing the Mar tian near-sur face con di tions.
Fur ther an a logue as pects of the ana lysed High An des Atacama site make it im por tant for Mars-rel e vant an a logue work on the Earth in the fu ture. Be sides the gen eral dry ness and de creased at mo spheric pres sure plus el e vated UV ir ra di a - tion, is the oc cur rence of per ma frost (ground with tem per a tures
<0C° for at least two con sec u tive years; van Everdingen, 2005) in patchy dis tri bu tion >5200 m and con tin u ously >5600 m (Nagy et al., 2019). The per ma frost is ac com pa nied by some post-vol - ca nic ac tiv ity there (Moreno and Gib bons, 2007) while some gla ciers also ex isted there un til re cently (Oyarzun, 1987).
These char ac ter is tics, to gether with the con di tions to ana lyse the oc cur rence of liq uid wa ter in the form of del i ques cence, make this site highly Mars-rel e vant.
CONCLUSIONS
1. Dur ing the field work at the Ojos del Salado moun tain ous re gion in the High An des Atacama area, tem per a ture and rel a - tive hu mid ity mea sure ments were made with de tec tors ex - posed on the sur face and next to or be low some rocks, in or der to iden tify night-time changes in tem per a ture and rel a tive hu - mid ity val ues. The ex act place ment of the de tec tors did not in - flu ence the mea sure ments as much as did the chang ing of over all weather con di tions, and small-scale mi cro- or nano - climate ob ser va tions are worth mak ing on Mars, as the dif fer - ences in T/RH pa ram e ters could be larger there be cause of the lower at mo spheric den sity and re lated re duced heat trans port pos si bil ity. The pat terns re corded showed char ac ter is tic daily cy cles with de creased tem per a ture and el e vated rel a tive hu - mid ity dur ing the night; how ever, due to weather changes, dif - fer ences be tween cer tain days could be also ob served. The over all be hav iour and daily trends of rel a tive hu mid ity and sur - face tem per a tures, along with the pos si ble ideal time win dows, show over lap be tween the Atacama Desert and the Mar tian sites ana lysed.
2. Strong fluc tu a tion could be ob served at 10–20 min - ute-long tem po ral scales in the tem per a ture and rel a tive hu mid - ity val ues, that was not pres ent in the Mar tian model-based datasets, thus any fu ture de tec tors should be pre pared to wit - ness such fluc tu a tions on Mars. Con sid er ing dif fer ent Mars-rel - e vant salts, the lon gest du ra tion for del i ques cence (be tween 7 and 12 hours in a night) was found in the case of CaCl2, Ca(ClO4)2 and Mg(ClO4)2 salts. MgSO4 salt pro vided the short - est pe riod, while with NaCl salt al most no pe riod fa vour able for del i ques cence was found.
3. The ideal du ra tion for del i ques cence showed some in - crease with the ris ing el e va tion, be cause of the de creas ing night-time tem per a ture re sulted in larger rel a tive hu mid ity val - ues. This ob ser va tion sup ports the ex pec ta tions that de spite the low hu mid ity on Mars, the very low night-time tem per a tures could pro vide the nec es sary con di tions for del i ques cence. The lo ca tions sur veyed in the High An des desert pro vide in ter est ing con di tions for more de tailed anal y sis of the del i ques cence pro - cess on Mars in the fu ture, in clud ing test ing of in stru ments to iden tify the emer gence of liq uids, as pro jected for the HABIT in - stru ment onboard ExoMars 2020 rover mis sion.
Ac knowl edge ments. The field work, done by A. Kereszturi, was sup ported by the COOP_NN_116927 pro ject; some in stru - ment-re lated as pects and the nu mer i cal anal y sis were sup - ported by the GINOP-2.3.2-15-2016-00003 grant of the Hun - gar ian Na tional Re search, De vel op ment and In no va tion Of fice (NKFIH). The Mar tian mod el ling done by B. Pal was sup ported by the EXODRILTECH pro ject of ESA. The com ments by the jour nal re view ers are ap pre ci ated.
REFERENCES
Ahumada, A.L., 2002. Periglacial phe nom ena in the high moun - tains of north west ern Ar gen tina. South Af ri can Jour nal of Sci - ence, 98: 166–170.
Ammann, C., Jenny, B., Kammer, K., Messerli, B., 2001. Late Qua ter nary gla cier re sponse to hu mid ity changes in the arid An - des of Chile (18–29°S). Palae ogeora phy Palaeoclimatology Palaeo ec ol ogy, 172: 313–326.
Azócar, G.F, Brenning, A., 2010. Hy dro log i cal and geomorpho - logical sig nif i cance of rock gla ciers in the dry An des, Chile (27–33 S). Per ma frost and Periglacial Pro cesses, 21: 42–53.
Amundson, R., Barnes, J.D., Ew ing, S., Heimsath, A., Chong, G., 2012. The sta ble iso tope com po si tion of ha lite and sul fate of hyperarid soils and its re la tion to aque ous trans port. Geo - chimica and Cosmochimica Acta, 99: 271–286.
Artieda, O., Davila, A., Wierzchos, J., Buhler, P., Rodríguez Ochoa, R., Carmen Ascaso, J., 2015. Sur face evo lu tion of salt-en crusted playas un der ex treme and con tin ued dry ness.
Earth Sur face Processes and Land forms, 40: 1939–1950.
Azócar, G.F., Brenning, A., 2010. Hy dro log i cal and geomorpho - logical sig nif i cance of rock gla ciers in the dry An des, Chile (27–33°S). Per ma frost et Periglacial Pro cesses, 21: 42–53.
Azua-Bustos, A., Caro Lara, L., VicuÔa, R., 2015. En vi ron men tal Mi cro bi ol ogy Re ports, 7: 388–394.
Azua-Bustos, A., Fairén, A.G., González-Silva, C., Ascaso, C., Carrizo, D., Fernández-Martínez, M.Á., Fernández- Sam pe - dro, M., García-Descalzo, L., García-Villadangos, M., Mar - tin-Redondo, M.P., Sánchez-García, L., Wierzchos, J., Parro, V., 2018. Un prec e dented rains dec i mate sur face mi cro bial com - mu ni ties in the hyperarid core of the Atacama Desert. Sci en tific Re ports, 8, id. 16706.
Boynton, W.V., Ming, D.W., Kounaves, S.P., Young, S.M.M., Arvi - dson, R.E., Hecht, M.H., Hoffman, J., Niles, P.B., Hamara, D.K., Quinn, R.C., Smith, P.H., Sutter, B., Catling, D.C., Mor - ris, R.V., 2009. Ev i dence for cal cium car bon ate at the Mars Phoe nix Land ing Site. Sci ence, 335: 61–64.
Cabrol, N.A., Grin, E.A., 1999. Dis tri bu tion, clas si fi ca tion, and ages of Mar tian im pact crater lakes. Icarus, 142: 160–172.
Catling, D.C., Claire, M.W., Zahnle, K.J., Quinn, R.C., Clark, B.C., Hecht, M.H., Kounaves, S., 2010. At mo spheric or i gins of per - chlor ate on Mars and in the Atacama. Jour nal of Geo phys i cal Re search, 115: CiteID E00E11.
Chevrier, V.F., Rivera Valentin, E.G., 2012. For ma tion of re cur ring slope lineae by liq uid brines on pres ent day Mars. Geo phys i cal Re search Let ters, 39: L21202.
Clapperton, C.M., 1994. The Qua ter nary gla ci ation of Chile.
Revista Chilena de Historia Nat u ral, 67: 369–383.
Cobos, D., Corte, A., 1990. Geocryological ob ser va tions in Ojos del Salado, Cen tral An des, Lat. 27°. IGCP/UNESCO Pro ject 297, 2nd Meet ing, Geocryology of South ern Af rica, Rhodes Uni - ver sity, Grahams town.
Czechowski, L., Witek, P., Misiura, K., 2013. Dy nam i cal model of rivers on Mars. Eu ro pean Plan e tary Sci ence Con gress, ab stract EPSC2013–917.
Davila, A.F., Duport, L.G., Melchiorri, R., Jänchen, J., Valea, S., de los Rios, A., Fairén, A.G., Möhlmann, D., McKay, C.P., Ascaso, C., Wierzchos, J., 2010. Hy gro scopic salts and the po ten tial for life on Mars. Astrobiology, 10: 617–628.
Dundas, C.M., McEwen, A.S., Chojnacki, M., Milazzo, M.P., Byrne, S., McElwaine, J.N., Urso, A., 2017. Gran u lar flows at re cur ring slope lineae on Mars in di cate a lim ited role for liq uid wa ter. Na ture Geoscience, 10: 903–907
Ehlmann, B.L., Swayze, G.A., Milliken, R.E., Mus tard, J.F., Clark, R.N., Murchie, S.L., Breit, G.N., Wray, J.J., Gondet, B., Poulet, F., Carter, J., Cal vin, W.M., Benzel, W.M., Seelos, K.D., 2016. Dis cov ery of alu nite in Cross crater, Terra Sirenum, Mars: ev i dence for acidic, sul fu rous wa ters. Amer i can Min er al o - gist, 101: 1527.
Farris, H.N., Davila, Al., 2016. Del i ques cence-driven brine for ma - tion in the Atacama Desert, Chile. 47th Lu nar and Plan e tary Sci - ence Con fer ence, ab stract 2518.
Farris, H.N., Davila, A., 2017. Cal cium per chlor ate brine for ma tion in the Atacama desert, Chile and im pli ca tions for liq uid wa ter at the sur face of Mars. Astrobiology Sci ence Con fer ence, ab stract no. 3483.
Farris, H.N., Conner, N.B., Chevrier, V.F., Rivera-Valentin, E.G., 2017. Ad sorp tion driven regolith-at mo spheric wa ter va por trans - fer on Mars: an anal y sis of Phoe nix TECP data. Icarus, 308:
71–75.
Fassett, C.I., Head, J.W., 2008. Val ley net work-fed, open-ba sin lakes on Mars: dis tri bu tion and im pli ca tions for Noachian sur - face and subsurface hy drol ogy. Icarus, 198: 37–56.
For get, F., Hourdin, F., Fournier, R., Hourdin, C., Talagrand, O., Col lins, M., Lewis, S.R., Read, P.L., Huot, J.-P., 1999. Im - proved gen eral cir cu la tion mod els of the Mar tian at mo sphere from the sur face to above 80 km. Jour nal of Geo phys i cal Re - search, 104: 24155–24176.
Glavin, D.P., Freissinet, C., Miller, K.E., Eigenbrode, J.L., Brun - ner, A.E., Buch, A., Sutter, B., Ar cher, P.D., Atreya, S.K., Brinckerhoff, W.B., Ca bane, M., Coll, P., Conrad, P.G., Coscia, D., Dworkin, J.P., Franz, H.B., Grotzinger, J.P., Leshin, L.A., Mar tin, M.G., McKay, C., Ming, D.W., Navarro González, R., Pav lov, A., Steele, A., Sum mons, R.E., Szopa, C., Teinturier, S., Mahaffy, P.R., 2013. Ev i dence for per chlor - ates and the or i gin of chlo ri nated hy dro car bons de tected by SAM at the Rocknest ae olian de posit in Gale Crater. Jour nal of Geo phys i cal Re search, 118: id. 20144.
Gough, R.V., Primm, K.M., Rivera-Valentín, E.G., Martínez, G.M., Tolbert, M.A., 2019. Solid-solid hydration and de hy dra tion of Mars-rel e vant chlo rine salts: Im pli ca tions for Gale Crater and RSL lo ca tions. Icarus, 321: 1–13.
Gough, R.V., Chevrier, V.F., Baustian, K.J., Wise, M.E., Tolbert, M.A., 2011. Lab o ra tory stud ies of per chlor ate phase tran si tions:
Sup port for metastable aque ous per chlor ate so lu tions on Mars.
Earth and Plan e tary Sci ence Let ters, 312: 371–377.
Gspurning, J., Lazer, R., Sulzer, W., 2006. Re gional cli mate and snow/gla cier dis tri bu tion in South ern Up per Atacama (Ojos del Salado) – an in te grated sta tis ti cal, GIS and RS based ap proach.
Grazer Schriften der Geographie and Raumforschung, 41:
59–70.
Ham il ton, V.E., Vasavada, A.R., Sebastián, E., Torre Juárez, M., Ramos, M., Armiens, C., Arvidson, R.E., Carrasco, I., Chris - tensen, P.R., De Pablo, M.A., Goetz, W., Gómez-Elvira, J., Lemmon, M.T., Madsen, M.B., Martín-Torres, F.J., Martínez - -Frías, J., Molina, A., Palucis, M.C., Rafkin, S.C.R., Rich ard - son, M.I., Yingst, R.A., Zorzano, M.-P., 2014. Ob ser va tions and pre lim i nary sci ence re sults from the first 100 sols of MSL Rover En vi ron men tal Mon i tor ing Sta tion ground tem per a ture sen sor mea sure ments at Gale Crater. Jour nal of Geo phys i cal Re search, 119: 745–770.
Hecht, M.H., Kounaves, S.P., Quinn, R.C., West, S.J., Young, S.M.M., Ming, D.W., Catling, D.C., Clark, B.C., Boynton, W.V., Hoffman, J., DeFlores, L.P., Gospodinova, K., Kapit, J., Smith, P.H., 2009. De tec tion of per chlor ate and the sol u ble chem is try of Mar tian soil at the phoe nix lander site. Sci ence, 325: 64.
Heldmann, J.L., Conley, C.A., Brown, A.J., Fletcher, L., Bishop, J.L., McKay, C.P., 2010. Pos si ble liq uid wa ter or i gin for Ata - cama Desert mud flow and re cent gully de pos its on Mars. Icarus, 206: 685–690.
Jack son, W.A., Böhlke, J.K., Andraski, B.J., Fahlquist, L., Bexfield, L., Eckardt, F.D., Gates, J.B., Davila, A.F., McKay, C.P., Rao, B., Sevanthi, R., Rajagopalan, S., Estrada, N., Sturchio, N., Hatzinger, P.B., An der son, T.A., Or ris, G., Betancourt, J., Stonestrom, D., Latorre, C., Li, Y., Harvey, G.J., 2015. Global pat terns and en vi ron men tal con trols of per -
chlor ate and ni trate co-oc cur rence in arid and semi-arid en vi - ron ments. Geochimica et Cosmochimica Acta, 164: 502–522.
Jack son, W.A., Böhlke, J.K., Andraski, B.J., Fahlquist, L., Bex - field, L., Eckardt, F.D., Gates, J.B., Davila, A.F., McKay, C.P., Rao, B., Sevanthi, R., Rajagopalan, S., Estrada, N., Sturchio, N., Hatzinger, P.B., An der son, T.A., Or ris, G., Betancourt, J., Stonestrom, D., Latorre, C., Li, Y., Har ris, J.K., Cous ins, C.R., Claire, M.W., 2016. Spec tral iden ti fi ca tion and quan ti fi ca - tion of salts in the Atacama Desert. Pro ceed ings of the SPIE 10005: id. 100050I.
Kate, I.L., 2018. Or ganic mol e cules on Mars. Sci ence, 360:
1068–1069.
Hecht, M.H., Kounaves, S.P., Quinn, R.C., West, S.J., Young, S.M.M., Ming, D.W., Catling, D.C., Clark, B.C., Boynton, W.V., Hoffman, J., DeFlores, L.P., Gospodinova, K., Kapit, J., Smith, P.H., 2009. De tec tion of per chlor ate and the sol u ble chem is try of Mar tian soil at the phoe nix lander site. Sci ence, 325: 64.
Kereszturi, A., 2019. Unique and po ten tially Mars rel e vant flow re - gime and wa ter sources at a High-An des Atacama site. Astro - biology, 20: 723–740.
Kereszturi, A., Rivera-Valentin, E.G., 2012. Lo ca tions of thin liq uid wa ter lay ers on pres ent-day Mars. Icarus, 221: 289–295.
Kereszturi, A., Rivera-Valentin, E.G., 2016. Pos si ble wa ter lu bri - cated grain move ment in the circum-po lar re gion of Mars. Plan - e tary Space Sci ence, 125: 130–146.
Kereszturi, A., Aszalós, J., Heiling, B., Kapui, Zs., Kiraly, Cs., Leél-Össy, Sz., Nagy, B., Pal, B., Skulteti, A., Szalai, Z., 2019.
Cold, dry, windy, and UV ir ra di ated: sur vey ing Mars-rel e vant con di tions in Ojos del Salado Vol cano (An des Moun tains, Chile). Astrobiology, 20: 677–683.
Kounaves, S.P., Hecht, M.H., Kapit, J., Gospodinova, K., DeFlo - res, L., Quinn, R.C., Boynton, W.V., Clark, B.C., Catling, D.C., Hredzak, P., Ming, D.W., Moore, Q., Shusterman, J., Stroble, S., West, S.J., Young, S.M.M., 2010. Wet chem is try ex per i - ments on the 2007 Phoe nix Mars Scout Lander mis sion: data anal y sis and re sults. Jour nal of Geo phys i cal Re search, 115:
E00E10.
Lewis, S.R., 2003. Mod el ling the mar tian at mo sphere. As tron omy and Geo phys ics, 44: 4.6–4.14.
Losiak, A., Czechowski, L., Velbel, M.A., 2015. Ephem eral liq uid wa ter at the sur face of the mar tian North Po lar Re sid ual Cap: re - sults of nu mer i cal mod el ling. Icarus, 262: 131–139.
Mar tin-Torres, J., Zorzano, M.-P., 2018. The in stru ment HABIT (Hab it Abil ity, Brine Ir ra di a tion and Tem per a ture) on the ExoMars plat form. 42nd COSPAR Sci en tific As sem bly, Ab stract id. F3.3-6-18.
Mar tin-Torres, F.J., Zorzano, M.-P., Valentin-Serrano, P., Harri, A.M., Genzer, M., Kemppinen, O., Rivera-Valentin, E.G., Wray, J., Bo Madsen, M., Goetz, W., McEwen, A.S., Hard - grove, C., Renno, N., Chevrier, V.F., Mischna, M., Navarro - -Gon za lez, R., Mar ti nez-Frias, J., Conrad, P., McConnochie, T., Cockell, C., Berger, G., Vasavada, A., Sum ner, D., Vaniman, D., 2015. Tran sient liq uid wa ter and wa ter ac tiv ity at Gale Crater on Mars. Na ture Geoscience, 8: 357–361.
Mar ti nez, G., McConnochie, T., Renno, N., Meslin, P.-Y., Fischer, E., Vicente-Retortillo, A., Borlina, C., Kemppinen, O., Genzer, M., Harri, A.-M., de la Torre-Juárez, M., Zorzano, M.-P., Mar tin-Torres, J., Bridges, N., Maurice, S., Gasnault, O., Gomez-Elvira, J., Wiens, R., 2016. Di ur nal vari a tion of at - mo spheric wa ter va por at Gale crater: anal y sis from ground - -based mea sure ments. EGU Gen eral As sem bly, ab stract EPSC2016-9297.
Martínez, G.M., Newman, C.N., De Vicente-Retortillo, A., Fischer, E., Renno, N.O., Rich ard son, M.I, Fairén, A.G., Gen - zer, M., Guzewich, S.D., Haberle, R.M., Harri, A.M., Kemppinen, O., Lemmon, M.T., Smith, M.D., de la Torre- Juárez, M., Vasavada, A.R., 2017. The mod ern near-sur face Mar tian cli mate: a re view of in-situ me te o ro log i cal data from Vi - king to Cu ri os ity. Space Sci ence Re views, 212: 295–338.
McEwen, A.S., Ojha, L., Dundas, C.M., Mattson, S.S., Byrne, S., Wray, J.J., Cull, S.C., Murchie, S.L., Thomas, N., Gulick,
V.C., 2011. Sea sonal flows on warm Mar tian slopes. Sci ence, 333: 740.
McKay, C.P., Friedmann, E.I., Gómez-Silva, B., Cáceres- Villanueva, L., Andersen, D.T., Landheim, R., 2003. Tem per a - ture and mois ture con di tions for life in the ex treme arid re gion of the Atacama Desert: four years of ob ser va tions in clud ing the El NiZo of 1997–1998. Astrobiology, 3: 393–406.
McKay, L., Claire, M., 2016. The pres ence and dis tri bu tion of salts as a palaeoprecipitation proxy in Atacama soils. EGU Gen eral As sem bly, ab stract EPSC2016-212.
Mont gom ery, W., Jaramillo, E.A., Royle, S.H., Kounaves, S.P., Schulze-Makuch, D., Sephton, M.A., 2019. Ef fects of ox y - gen-con tain ing salts on the de tec tion of or ganic biomarkers on Mars and in ter res trial an a log soils. Astrobiology, 19: 711–721.
Moreno, T., Gib bons, W., 2007. The Ge ol ogy of Chile. Geo log i cal So ci ety Lon don.
Möhlmann, D., Thom sen, K., 2011. Prop er ties of cryobrines on Mars. Icarus, 212: 123–130.
Nagy, B., Mari, L., Kovács, J., Nemerkényi, Zs., Heiling, Zs., 2014a. En vi ron ment changes in the Dry An des – mon i tor ing re - search on the Ojos del Salado (in Hun gar ian). In: HUNGEO 2014 Mag yar Földtudományi (eds. T. Cserny, P. Kovács-Pálffy and Á. Krivánné Horváth): 53–62. 1103 szakemberek XII.
találkozója. Bu da pest: Magyarhoni Földtani Társulat.
Nagy, B., Mari, L., Kovács, J., Nemerkényi, Zs., Heiling, Zs.
2014b. Data from the subsurface of a high-moun tain desert: wa - ter and ice on the Ojos del Salado (in Hun gar ian), Egyetemi Meteorológiai Füzetek – Me te o ro log i cal Notes of Uni ver si ties, 25:123–128.
Nagy, B., Ignéczi, A., Kovács, J., Szalai, Z., Mari, L., 2019. Shal - low ground tem per a ture mea sure ments on the high est vol cano of the Earth, the Mt. Ojos del Salado, Arid An des, Chile. Per ma - frost and Periglacial Pro cesses, 30: 3–18.
Navarro, T., Mad eleine, J.-B., For get, F., Spiga, A., Millour, E., Montmessin, F., Määttänen, A., 2014. Global cli mate mod el ing of the Mar tian wa ter cy cle with im proved microphysics and radiatively ac tive wa ter ice clouds. Jour nal of Geo phys i cal Re - search, 119: 1479–1495.
Navarro-González, R., Rainey, F.A., Molina, P., Bagaley, D.R., Hollen, B.J., de la Rosa, J., Small, A.M., Quinn, R.C., Grunthaner, F.J., Cáceres, L., Gomez-Silva, B., McKay, C.P., 2003. Mars-like soils in the Atacama Desert, Chile, and the dry limit of mi cro bial life. Sci ence, 302: 1018–1021.
Nikolakakos, G., Whiteway, J.A., 2018. Lab o ra tory study of ad - sorp tion and del i ques cence on the sur face of Mars. Icarus, 308:
221–229.
Nuding, D.L., Rivera Valentin, E.G., Da vis, R.D., Gough, R.V., Chevrier, V.F., Tolbert, M.A., 2014 Del i ques cence and ef flo res - cence of cal cium per chlor ate: An in ves ti ga tion of sta ble aque - ous so lu tions rel e vant to Mars. Icarus, 243: 420–428.
Oyarzun, C.G., 1987. Inventario de Glaciares de los An des Chilenos desde los 180 a los 320 de Latitud Sur. Revista de Geografía Norte Grande, 14: 35–48.
Ojha, L., Wil helm, M.B., Murchie, S.L., McEwen, A.S., Wray, J.J., Hanley, J., Massé, M., Chojnacki, M., 2015. Spec tral ev i dence for hy drated salts in re cur ring slope lineae on Mars. Nat u ral Geoscience, 8: 829–832.
Óscar, G.F., 1995. Volcanes de Chile (in Span ish). Instituto Geográfico Militar Year book, San ti ago, Chile.
Osterloo, M.M., Ham il ton, V.E., Bandfield, J.L., Glotch, T.D., Baldridge, A.M., Christensen, P.R., Tornabene, L.L., An der - son, F.S., 2008. Chlo ride-bear ing ma te ri als in the south ern high lands of Mars. Sci ence, 319: 1651–1654.
Oyarzun, C.G., 1987. Inventario de Glaciares de los An des Chilenos desde los 180 a los 320 de Latitud Sur. Revista de Geografía Norte Grande, 14: 35–48.
Pal, B., Kereszturi, A., 2017. Pos si bil ity of mi cro scopic liq uid wa ter for ma tion at land ing sites on Mars and their ob ser va tional po ten - tial. Icarus, 282: 84–92.
Primm, K.M., Gough, R.V., Wong, J., Rivera Valentin, E., Mar ti - nez, G.M., Hogancamp, J.V., Ar cher, P.D., Ming, D.W., Tolbert, M.A., 2018. The ef fect of Mars rel e vant soil analogs on
