Geode ma 438 (2023) 116636
A ailable online 11 Augus 2023
0016-7061/© 2023 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/).
Exploi a ion o he SoilPRO® (SP) appa a us o measu e soil su ace
e lec ance in he ield: Fi e case s udies
Eyal Ben Do
a
,
*
, Amihai G ano
a
, Rony Wallach
b
, Nicolas F ancos
a
, Daniela Helle Pea ls ein
a
,
Ba E a i
a
, Luboˇ
s Bo ů ka
c
, Asa Gholizadeh
c
, Thomas Schmid
d
a
Remo e Sensing Labo a o y, Geog aphy Depa men , Po e School o he En i onmen and Ea h Sciences, Facul y o Exac Sciences, Tel A i Uni e si y, Tel A i
699780, Is ael
b
Soil and Wa e Depa men , The Heb ew Uni e si y o Je usalem, Is ael
c
Depa men o Soil Science and Soil P o ec ion, Facul y o Ag obiology, Food and Na u al Resou ces, Czech Uni e si y o Li e Sciences P ague, Kamycka 129, Suchdol,
P ague 16500, Czech Republic
d
Depa men o En i onmen , Cen o de In es igaciones Ene g´
e icas, Medioambien es y Tecnol´
ogicas (CIEMAT), A da. Complu ense, 40, 28040 Mad id, Spain
ARTICLE INFO
Handling Edi o : Jingyi Huang
Keywo ds:
Soil e lec ance
SoilPRO
Field measu emen s
Soil su ace p ope ies
ABSTRACT
The SoilPRO® (SP) is an assembly designed o acqui e soil e lec ance in o ma ion in he ield wi hou dis u bing
he soil su ace, and ega dless o a mosphe ic and sola adia ion condi ions. This pape summa izes i e case
s udies in which he SP assembly was used o di e en applica ions. The case s udies consis ed o : (1) gene a ing
su ace spec al measu emen s unde any a mosphe ic condi ion; (2) compa ing he pe o mance o he SP o he
adi ional ba e ibe me hod o ica ious calib a ion o hype spec al sa elli e senso s; (3) assessing wa e
epellency o a soil su ace go e ned by o ganic ma e hyd ophobici y; (4) spa ial p edic ion o he a e o wa e
in il a ion in o he soil p o ile as go e ned by he soil su ace seal; and (5) using he SP appa a us o measu e soil
su ace e lec ance in Sou h She land Island, An a ica unde se e e wea he condi ions. The case s udies
included calcula ion o spec al quali y, p edic ion accu acy and measu emen s abili y. The pape discusses each
o he cases in de ail and concludes ha he SP (o simila assembly) is he bes way o measu e he e lec ance o
he o iginal soil su ace in he ield. In he i s case s udy, he spec um collec ed by he SP unde daily changing
illumina ion was shown o be s able ela i e o he adi ional measu emen me hods o con ac p obe o ba e
ibe . The second case s udy indica ed ha use o he SP o ica ious calib a ion is much mo e e icien (in e ms
o ime and s abili y) han g ound- u h p ac ice o e a la ge a ea, and in he hi d case s udy, he SP was able o
assess a soil su ace p ope y go e ned by o ganic ma e hyd ophobici y be e han he con ac p obe, which
des oys he soil su ace o ganic seal. A simila achie emen was gained in he ou h case s udy, p o iding a
be e assessmen o he wa e -in il a ion a e in o he soil. In he i h case s udy, he SP demons a ed
imp essi e high-quali y acquisi ion o soil su ace e lec ance wi h a e y low sun angle o e he Sou h Pole.
Based on hese case s udies and he high quali y o he da a gene a ed by he SP in he ield, we sugges building,
in pa allel o he classical soil spec al lib a ies gene a ed in he labo a o y, ield soil spec al lib a ies ha will
p ese e he soil su ace p ope ies scanned in he ield. We an icipa e he de elopmen o mo e applica ions o
he SP assembly based on he capabili ies shown in his pape .
1. In oduc ion
Soil e lec ance measu emen ac oss he isible–nea in a ed–-
sho wa e in a ed (VIS–NIR–SWIR; 350–2500 nm) egion is one o he
mos p ac ical means o bo h p oximal and emo e sensing o soil a -
ibu es (Visca a Rossel e al., 2016; Behe a e al., 2022). I has become
a common echnique o p edic ing se e al soil p ope ies in he
labo a o y—mo e apidly han he adi ional we chemis y me hods
wi h eliable accu acy (Ben-Do and Banin, 1995; Visca a Rossel e al.,
2006; S enbe g e al., 2010). Acco dingly, i is widely used no only in
he labo a o y bu also in he ield. In gene al, soil spec al in o ma ion
is ga he ed in o a soil spec al lib a y (SSL) da abase consis ing o bo h
spec al and chemical in o ma ion. Such lib a ies a e mainly con-
s uc ed in he labo a o y wi h spec al measu emen s pe o med unde
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (E.B. Do ).
Con en s lis s a ailable a ScienceDi ec
Geode ma
jou nal homepage: www.else ie .com/loca e/geode ma
h ps://doi.o g/10.1016/j.geode ma.2023.116636
Recei ed 18 Ma ch 2023; Recei ed in e ised o m 3 Augus 2023; Accep ed 4 Augus 2023
Geode ma 438 (2023) 116636
2
con olled condi ions (e.g. geome y, g ain size, illumina ion, and
mois u e con en ). Soil e lec ance measu emen s in he ield a e
s ongly a ec ed by a mosphe ic a enua ion, changes in he sun’s
ele a ion, use s abili y and expe ience, spec al p o ocol con igu a ion,
su ace a ia ions, uns able geome y, and he measu emen means
(Mil on e al., 2009). These lead o unce ain ies in he da a, and lowe
accu acy unde ou doo condi ions (Ch is y, 2008); Gomez e al., 2008;
Wenjun e al., 2014; Rodiono e al., 2015). Minimizing hese e ec s is
he e o e essen ial and equi es equen calib a ion, esul ing in a high
olume o measu emen s o a gi en a ge , oge he wi h ex ensi e
documen a ion o he en i onmen al condi ions du ing he spec al
acquisi ion, esul ing in a small numbe o samples pe uni ime. These
e ec s he e o e hinde accu a e compa isons be ween spec a and,
u he , do no allow quan i a i e analysis o he acqui ed spec al
in o ma ion.
Today, high olumes o ield measu emen s a e pe o med mos ly
wi h a ba e ibe (BF) o e op ic, which su e s om all o he abo e-
lis ed issues. Ano he me hod is o measu e soil e lec ance in he ield
using a con ac p obe (CP) de ice (Wenjun e al., 2014) a common as-
sembly used in he labo a o y ha ouches he soil su ace wi h an a ea
o abou 12 cm
2
. Howe e , his op ion su e s om poo ep esen a i e
a ea, poo con ac wi h i egula su aces, sensi i i y o holding s abil-
i y, and in some cases, i can ha m he agile soil su ace seal due o
p obe p essu e on he soil su ace. The e a e o he ways o ex ac soil
spec al in o ma ion in he ield wi hou ha ming he soil su ace, such
as deploying a goniome e o cons uc ing a s able and ixed o e-op ic
geome y wi h a i icial illumina ion (e.g. Wise and Ma s, 2022).
The las ew yea s ha e seen he de elopmen o assemblies ha can
o e come some o he limi a ions inhe en in spec al acquisi ion unde
ield condi ions (Mil on, 2010; Mil on e al., 2009). Some o hese sel -
de eloped ools make use o an in e nal illumina ion sou ce, ei he in-
side he p obe, as a modi ica ion o a ailable de ices (Kusumo, 2018), o
in an isola ed chambe , o su ace and subsu ace use (Rodiono e al.,
2015), o ensu e s able adia ion in ensi y wi hou any a mosphe ic
a enua ion. Howe e , hese ools a e hea y and canno be ca ied by an
o dina y ope a o ; hey equi e a ac o ow, which may demolish he
agile su ace seal, and a e no sui able o ield and g ound- u h wo k.
In gene al, a s anda d me hod o measu ing soil su ace e lec ance
in he ield should be obus , apid, ep esen a i e, and as ep oducible
and as close o he measu emen s pe o med in he labo a o y as
possible (i.e., enabling ex ac ion o he same spec um o e he same
a ea wi h no change in i s na u e, hus allowing he exploi a ion o he
many SSLs in he labo a o y). Acco dingly, al e na i e me hods and
p o ocols o measu ing soil spec al in o ma ion in he ield a e s ongly
equi ed. The echnique mus ep esen he soil su ace p ecisely as i is
seen om a emo e domain (i.e., a na u al su ace), co e a sizeable
ep esen a i e a ea, and o e come all he o he a o emen ioned p ob-
lems (i.e., use e ec , geome y s abili y, su ace cha ac e is ics, a mo-
sphe ic a enua ion (plus clouds) and changes in sun angle). In his
con ex , Ben-Do e al. (2017) de eloped a simple soil ield p obe as-
sembly named “SoilPRO®” (SP) (U.S. Pa en O ice Se ial No.
15407295). This de ice can be hooked up o any po able spec ome e
u nished wi h a ibe -op ic cable and has al eady demons a ed
ema kable esul s in ieldwo k scena ios. Ben-Do e al. (2017) showed
ha he SP pe o ms spec al measu emen s wi h labo a o y quali y,
which a e s able, e en when aken by di e en non-skilled use s. We
hus exploi ed his assembly in di e en applica ions, whe e he soil
su ace con ols ield p ope ies, and unde di e en a mosphe ic con-
di ions in he ield. This pape epo s on i e case s udies in which he
soil’s hin su ace mus be p ese ed o accu a ely app oxima e he soil
su ace p ope ies by spec oscopy. The SP was used by se e al inde-
penden use s, and he esul s a e epo ed he ein. Ou in en ion is
mainly o open he way o o he applica ions equi ing high-quali y soil
e lec ance p oduc s unde ex eme measu emen condi ions.
2. Gene al ma e ials and me hods
2.1. Soil ield p obe
The SP appa a us was used in all o he p esen ed case s udies. I was
designed a Tel A i Uni e si y and is pa en ed unde U.S. Pa en O ice
Se ial No. 15407295. I is composed o an aluminum cylinde (pain ed
ma e “Kodak” black inside) wi h a diame e o 24 cm and heigh o 25
cm, weighing 1.6 kg. The de ice is u nished wi h a ibe -op ic holde
ha main ains a s able geome y o bo h illumina ion and iewing
angle (Ben-Do e al., 2017). The iewing angle is 45◦, enabling mea-
su emen o an a ea o abou 1800 cm
2
wi h a BF o 25◦ ield o iew
(FOV). The assembly’s a ge dimensions we e designed o align wi h
he commonly used 25 cm ×25 cm (10 ×10 squa e inch) Spec alon
(Labsphe e®) whi e e e ence panel, o op imal pe o mance o whi e
e e ence measu emen s as ela i e e lec ance o each measu emen .
On op o he cylinde , a s abilized ungs en halogen lamp uni o mly
illumina es he su ace a a nadi angle. To use he SP illumina o lamp
(Analy ical Spec al De ices (ASD) P oLamp® and Ushio® bulb model
JC14.5V-50 WC) in he ield, we a ached a powe supply adap e (12 V
DC o 15 V DC and 220 V AC o 15 V DC acco ding o he bulb speci i-
ca ions). The adap e p o ides a s able inpu ol age o he bulb om
any g id o po able powe sou ce du ing ope a ion and indica es any
powe ins abili y o aul y ba e y. Fo he SP p o o ype, we used a
ligh weigh po able li hium-ion ba e y (11.1 V, 31.2 Ah, 1.98 kg),
which could ope a e o abou 5 h and, in con as o o he commonly
used CP ools, i does no consume p ecious powe om he ield spec-
ome e ’s ba e y. The ibe op ic om he ASD FieldSpec® (model FSP
350-2500P) is connec ed ia an adap e o a moun a ached o he
conical aluminum side addi ion o he chambe o posi ion i s o e op ic
a a 45◦angle o maximum a ge a ea. This con igu a ion is designed o
main ain minimal specula e lec ion and maximal Lambe ian adia ion
collec ion om he a ge below, as well as o simula e he geome y o
he ASD CP (ASD Inc., n.d.) appa a us and he illumina ion uni ’s
ope a ing p o ocol as sugges ed by ASD Inc. (2012). Fig. 1 p o ides he
SP scheme and all o he sys em’s accesso ies, and Fig. 2 illus a es he
ope a ing p inciple as demons a ed by a single ope a o .
The ope a ing scheme o he SP is simila o ha o labo a o y
spec al da a measu emen s. The spec ome e has o be wa med up o
60 min, he bulb o 15 min and he im o he SP pe iphe y has o be
cleaned wi h a so whi e pape . Fo con igu a ion and whi e e e ence
calib a ion, he assembly has o be deployed on a 10 squa e inch whi e
e e ence (Spec alon) (see Fig. 2c). The spec al con igu a ion consis s
o 30 eadings o samples, whi e e e ence and black cu en . Op imi-
za ion has o be es ablished, jus as in he labo a o y, as he measu e-
men scheme e lec ance is ela i e o he whi e e e ence, and a s able
and s aigh e lec ance line o 100% has o be achie ed o he whi e
e e ence sample. The SP beha es as a eal da k chambe as no SP signal
occu s du ing a sunny day and high sun ele a ion condi ion while he
bulb was shu o . Soil measu emen is done by gen ly placing he SP
su ace im on he soil su ace o a oid any leakage o ligh om he
inne SP olume.
2.2. S a is ical e alua ion
To e alua e he spec al simila i y in ou case s udies, we conduc ed
s a is ical analyses o oo mean squa e e o (RMSE, Eq. (1), a io o
pe o mance o in e qua ile dis ance (RPIQ, Eq. (2), and he a e age
sum o de ia ions squa ed (ASDS, Eq. (3).
RMSE =
∑
n
i=1
(Yi−
Yi)2
n
√
√
√
√
√(1)
RPIQ =IQ/RMSE (2)
E.B. Do e al.
Geode ma 438 (2023) 116636
3
ASDS =∑n
i=1((Yi/
Yi) − 1)2
n(3)
whe e ˆ
Y is he p edic ed alue, Y is he obse ed alue o he numbe o
da a poin s, and IQ is he in e qua ile dis ance o he measu ed alue.
These s a is ics we e used acco ding o each o he case s udies’
equi emen s.
3. Case s udies
Sec ions 3.1–3.5 p esen i e case s udies in which he SoilP o® (SP)
was u ilized. I is no ewo hy ha , in all ins ances, p ecise measu emen
o he uppe mos laye o delica e soil, unde undis u bed ou doo
condi ions such as su ace and adia ion, was c ucial. Each case is
accompanied by a ho ough desc ip ion o he p oblem add essed by he
SP, a comp ehensi e desc ip ion o he me hodology employed o a ain
he solu ion, and a de ailed analysis o he esul ing ou comes and
de i ed conclusions.
3.1. Case s udy 1: All condi ions – 24/7 ield spec al measu emen s
To e alua e he eliabili y o he SP’s pe o mance in measu ing
su ace e lec ance in he ield while a oiding he a o emen ioned
p oblems such as poo ou doo adia ion, i e soil samples om Is ael
Fig. 1. Le : Ske ch o he SoilPRO de ice (p o ile iew). Righ : SoilPRO sys em se up con igu a ion ( op iew). (1) main body, (2) illumina ion sou ce, (3) illu-
mina ion beam, (4) op ic ibe , (5) ibe holde , (6) ibe FOV, (7) handle, (8) powe adap e , (9) powe co d, (10) po able ba e y, (11) ASD Field Spec P o
spec ome e and ope a ing lap op.
Fig. 2. Use o SoilPRO in he ield. (a, b) Ease o ca ying he po able spec ome e o e di e en a eas, (c) whi e e e ence calib a ion p ocess, (d) su ace
measu emen .
Table 1
Soil samples selec ed o he SP exe cise.
Sample Is aeli name De ini ion USDA G ea G oup soil
axonomy
A Ham a B ownish ed sandy
soil
Haploxe al s
B Rendzina Rendzina moun ain
soil
Rendolls
C Ve isol B own allu ial soil Ch omoxe e s
D Loess Loess aw soil Cambo hids
E Coas al beach
sand
Coas al beach sand To ipsammen s
E.B. Do e al.
Geode ma 438 (2023) 116636
4
ep esen ing i e di e en USDA o de s (see Table 1 and Fig. 3) we e
selec ed. All samples we e collec ed om he uppe 5 cm, b ough o he
labo a o y, ai -d ied, and sie ed o pass a 2 mm sie e. Each soil was
placed on a shallow aluminum pla e pain ed ma e black wi h a diame e
o 30 cm and hickness o 2 cm o comple ely ill he SP’s a ea diame e
o 25 cm. The aim was o examine he s abili y o he SP appa a us
ela i e o BF p ac ices o e se e al soil ypes and wi h changing sun
ele a ion and a mosphe ic condi ions.
The e lec ance o each soil was inc emen ally acqui ed on 23 May
2020 using a BF, ou side he labo a o y unde a clea sky, and hen using
he SP o 8 h (08:30–16:30 h). A 12:30 h, clouds appea ed empo a ily
du ing he measu emen s. The e o e, wo measu emen s we e acqui ed
a ha ime, wi h and wi hou clouds. The BF was moun ed on a ipod
acing nadi and co e ing he soil a ge wi h a 30% ma gin (a ea o
1000 cm
2
). Fig. 4 shows he spec a h oughou he day as measu ed
wi h he SP and BF.
Rega ding he SP measu emen s, i can be no iced ha a s able and
cons an e lec ance was deli e ed h oughou he day (Fig. 4), wi h no
a iance in albedo alue o spec al shape o he spec a aken a
di e en imes. Only a mino a ia ion in albedo was seen in sample (d)
a 12:30 h, bu he spec um s ill sa is ied ou expec a ion o s abili y. In
con as , he BF measu emen s showed a ia ions in albedo a each ime
poin due o he ins abili y o he a mosphe ic condi ions and changes in
sun angle. Ma ginal changes appea ed du ing he clouds’ appea ance a
12:30 h wi h he BF (Fig. 4). We assume ha hose changes we e he
esul o bidi ec ional e lec ance dis ibu ion unc ion (BRDF) e ec s
due o he changing sola ele a ion and s able nadi iewing angle, as
well as he inconsis en illumina ion due o he passing clouds. In Fig. 4
(ba e ibe ), wo ex eme whi e e e ence alues a e shown: in he one
ha is lowe han 100% e lec ance, he whi e e e ence was measu ed
when he clouds we e passing o e , and is ela i e o he whi e e e ence
s anda d which was measu ed unde a clea sky; he one ha is highe
han 100% was measu ed unde a clea sky, and is ela i e o he whi e
e e ence measu ed unde a cloudy sky. The s able esul s o he SP
measu emen s sugges ha i can p o ide accu a e and s able eadings
despi e he unce ain ies a ising om a mosphe e, illumina ion, mea-
su emen geome y, and use expe ience.
This examina ion con i med he abili y o he SP assembly o acqui e
s able ield measu emen s, while lea ing he soil su ace undis u bed (i.
e., p ese ing i in i s na u al s a e unde cons an illumina ion and
geome y). I can se e as a apid measu emen ool o de elop a ield
SSL in pa allel o a labo a o y SSL, which will enable ob aining ield
in o ma ion and eliable g ound- u h measu emen s (Mil on e al.,
2009). F ancos and Ben-Do (2022) ecen ly es ablished a ans e
unc ion o es ima e ield e lec ance om a labo a o y SSL using mu ual
measu emen s wi h he ield (SP) and labo a o y (CP) SSL p o ocols. In
addi ion, his de ice can close he gap, when soil spec al in o ma ion is
needed unde ex eme condi ions such as du ing he nigh , a oiding
complica ed measu emen schemes as conduc ed, o example, by Wise
and Ma s (2022).
3.2. Case s udy 2: Vica ious calib a ion o hype spec al senso s om he
g ound
Hype spec al (HSR) senso s equi e pe iodic adiome ic and spec-
al calib a ion. Aside om onboa d calib a ion ou ines (e.g., pa ial-
ape u e calib a o s, s anda d lamps, and sola di use panels) (Then-
kabail, 2016), ica ious calib a ion (VC) p ac ices a e he mos use ul
me hod o examining and ec i ying he senso ’s adiome ic and
spec al esponse, and checking i s hema ic pe o mance (Xu e al.,
2020). The VC me hod uses na u al o a i icial si es on he g ound o
he pos -launch calib a ion o senso s and spec al measu emen s o
hese si es ha p ecisely ep esen he e lec ance o he a ea wi h high
ideli y (Mo ain and Budge, 2004). Acco dingly, VC is a signi ican pa
o many g oups’ ac i i y wo ldwide, and hose who ope a e and use HSR
da a in es a g ea deal o e o in he calib a ion/ alida ion (CAL/VAL)
o he mission a hand, such as he IEEE P4001 (h ps://s anda ds.ieee.
o g/ieee/4001/7314/) wo king g oup, he GEO wo king g oup, he
CAL/VAL Wo king G oups o Su ace Biology and Geology (SBG)
(Kokaly and Tu pie, 2019), and he RadCalNe Wo king g oup (Bou e
e al., 2019).
The mos common way o ca y ou VC is o use na u al and homo-
geneous g ound a ge s on ea h ha a e s able in space and ime. The
measu emen is conduc ed du ing o close o he senso ’s o e pass ime,
es ima ing op o he a mosphe e (TOA) adiance a -senso . The c i ical
issue in VC is he measu emen o he g ound e lec ance in a e y
p ecise, eliable, and ep esen a i e way, e y close o o du ing he
o e pass. This is also ue o any o he g ound- u h mission and hence
equi es a ca e ully execu ed p o ocol and me hod.
In gene al, o achie e ep esen a i e e lec ance on he g ound o
he mission, g ound spec al measu emen s a e well-planned, wi h
se e al known p o ocols. In addi ion, he VC a ea should be s able wi h
minimum changes o e ime, and i should allow o equen and
epea ed g ound measu emen s. Due o oday’s a ailabili y o po able
spec ome e s, he e a e many p o ocols o acqui ing ield e lec ance
measu emen s, om jus poin ing he BF a a nadi -looking angle
manually o using a goniome e ; bo h ely on sola adia ion and a e
s ongly a ec ed by he a mosphe e’s a enua ion and he sun’s ele a-
ion (e.g. S one e al. 1980, Peddle e al. 2001,Schop e , 2008, Kodai a
and Shibusawa, 2013, Laba e e al 2019). Ano he common p ac ice is
o use a CP u nished wi h a i icial illumina ion while ouching he
su ace o measu emen s. In each o hese me hods, samples ep e-
sen ing he selec ed and s able a ea a e needed whe e se e al limi a-
ions, such as sensi i i y o he measu emen geome y and e en o he
colo o he clo hes ha he use s a e wea ing, ma e (SVC, 2019). The
goniome e appa a us is a good me hod o a single measu emen bu
canno co e la ge a eas, whe eas he CP assembly, which can do so, is
p oblema ic as i has a e y small a ea. In addi ion, i may des oy he
su ace due o he need o con ac he agile g ound. To a oid some o
hese p oblems, Ray Kokaly (pe sonal communica ion, 2022) de eloped
a po able assembly ha simul aneously measu es he upwelling and
downwelling adia ion a a cons an geome y, he eby minimizing
some o he main unce ain ies ( ecen ly, a new concep e med ASD
Field Spec Dual has also been adop ed o o e come hese obs acles; h
Fig. 3. Fi e soil samples used o case s udy 1 (see Table 1). Acco ding o he USDA G ea G oup Soil Taxonomy, hese soils a e: (a) Haploxe al s, (b) Rendolls, (c)
Ch omoxe e s, (d) Cambo hids, (e) To ipsammen s.
E.B. Do e al.
Geode ma 438 (2023) 116636
5
ps://www.mal e npanaly ical.com/en/p oduc s/p oduc - ange/asd- a
nge/ ieldspec- ange/so wa e/ ieldspec-dual).
In his case s udy, ou aim was o minimize, using he SP appa a us,
he e ec s o measu emen geome y, a mosphe ic a enua ion, and sun
ele a ion changes du ing he sho ime o he senso ’s o e pass.
Acco dingly, we examined he SP’s pe o mance du ing a eal calib a-
ion mission o e a well-known VC si e selec ed o HSR pe o m-
ance—Amiaz plain in sou he n Is ael (31◦04
′
20.98
″
N, 35◦22
′
14.83
″
E).
The a ea is in he Judean Dese (Fig. 5) and is a homogeneous b igh -
su ace playa (5 km ×5 km) wi h high e lec ance alues (>0.4).
Because ege a ion is sca ce and i is s able in space and ime, i was
selec ed o he adiome ic e alua ion and c oss-calib a ion o PRISMA,
DESIS, EnMAP and EMIT o bi al senso s (Helle -Pea lsh ien e al., 2021,
2023; Helle -Pea lsh ien and Ben-Do , 2022).
Du ing Augus 2021 and Augus 2022, and close o he o e pass o
EnMAP and EMIT senso s, we conduc ed VC measu emen s o e he
a ea. We used he EnMAP measu emen p o ocol (Helle -Pea lsh ien
e al., 2023) ha cap u es measu emen s o a 90 m ×90 m a ea and he
SP measu emen p o ocol o he VC mission. Fig. 5 p o ides he scheme
o he EnMAP measu emen s om Augus 2022. An ASD BF wi h 25◦
o e op ics was also used om a heigh o 1.5 m co e ing an a ea o
abou 1800 cm
2
. The measu emen s we e conduc ed on a clea mo ning,
and he use s main ained a s able geome y o back o he sun while
measu ing he su ace e lec ance e e y 3 m. Each 90 m line cap u ed 30
measu emen s, and ano he 48 poin s we e aken andomly be ween he
lines. A e each line, a whi e e e ence measu emen was aken o
compensa e o any changes in sun ele a ion o possible changes in he
op ical dep h du ing he measu emen s. All measu emen s in he a ea
ook abou 45 min. In o al, he BF measu emen s consis ed o 138 poin s
(Fig. 5).
A e he BF session, we emeasu ed he a ea wi h he SP using he
same 90 m ×90 m es a ea, hough wi hou any geome y limi a ions.
Du ing he measu emen s, cumulus and ci us clouds appea ed, bu we
con inued measu ing. In o al, we measu ed 70 poin s wi h he SP, and i
ook app oxima ely 20 min. No in e calib a ion wi h a whi e e e ence
was conduc ed du ing he measu emen session. The whi e e e ence a
Fig. 4. Case s udy 1 spec a om ba e ibe ( op) s. SoilPRO (bo om) o soils a–e (see Table 1) a di e en imes and unde a clea and cloudy sky a 12:30 h. Whi e
Re e ence (WR).
E.B. Do e al.
Geode ma 438 (2023) 116636
6
he end o he session was measu ed and compa ed o he i s whi e
e e ence, demons a ing main enance o s abili y wi h simila uni
alues o bo h.
Fig. 6 shows he spec a o he BF unde he EnMAP p o ocol and he
SP measu emen s, along wi h he s anda d de ia ion (SD) calcula ed
om all o he spec a. The SP p o ided mo e s able spec a wi h SD =
0.38 compa ed o BF, wi h SD =0.46. In addi ion, he SP p o ided in-
o ma ion ac oss he en i e spec al ange, including he well-de ined
spec al egion o he a mosphe ic wa e apo a enua ion ange
a ound 1400 and 1900 nm.
Fig. 7 shows a compa ison o he measu emen s ob ained using he
BF and SP in Augus 2022 e sus 2021 (1 yea apa ) a he exac same
loca ion in he Amiaz plain. The a e age spec a showed signi ican
a ia ion o he BF compa ed o he SP measu emen s. I should be
no ed ha he wo se s o da a we e acqui ed by di e en ASD spec-
ome e s (FieldSpec3 in 2021 and FieldP o4 in 2022). The ASDS alues
be ween he a e age spec um in each yea o bo h SP and BF we e
calcula ed, (0.14 and 0.83, espec i ely). As a lowe ASDS alue sugges s
a close simila i y be ween wo spec a, his indica es ha he SP p o-
ided be e esul s han he adi ional BF measu emen . Mo eo e , he
SP enabled compensa ing no only o he a mosphe ic a ia ion bu also
o di e en spec ome e models. I should be no ed ha in e nal
calib a ion o c oss-calib a ion be ween wo o mo e senso s is an
impo an app oach ha has no ye been consolida ed; hese esul s
hus call o mo e s udies on c oss-calib a ion be ween bo h g ound and
o bi al senso s. Fig. 8 shows pho os om he BF ield measu emen s
compa ed o SP and whi e e e ence measu emen s.
3.3. Case s udy 3: P oximal sensing o soil wa e epellency in a ield
Wa e - epellen soils e e o soils ha a e no eadily we ed (also
e med hyd ophobic). Soil hyd ophobici y a ec s he soil’s physical and
hyd ological p ope ies. I inc eases hys e esis o he wa e - e en ion
cu e (Ri sema e al., 1998; Bau e s e al., 2000), gene a es uns able
we ing on s due o inge ed low (Hend ickx e al., 1993), educes
in il a ion capaci y ela i e o we able soils (e.g., Wang e al., 2003;
Doe e al., 2006), and induces g ea e su ace uno and e osion
(Bu ch e al., 1989; Bena ides-Solo io and MacDonald, 2001). Soil wa e
epellency (SWR) depends, among o he ac o s, on wa e con en and
quali y, and soil su ace cha ac e is ics.
The deg ee o SWR is commonly quan i ied by he wa e d op
pene a ion ime (WDPT) es (Doe , 1998). This me hod consis s o
placing a d op o dis illed wa e on he soil su ace and eco ding he
ime aken o he d op o pene a e comple ely in o he soil. The in-
ensi y o he epellency is usually de e mined by app oxima ing he
su ace ension equi ed o ini ia e in il a ion ins an ly (Le ey e al.,
2000), a measu e o how s ongly a soil epels a wa e d op a he ime o
applica ion (i.e., how much o he wa e d op “balls up” on he su ace).
Whe eas bo h me hods can be ca ied ou in he labo a o y and in he
ield, hey ha e he disad an age o being semi-objec i e, ime-
consuming, and equi ing expe use s and sophis ica ed echniques (e.
g., a well-aligned high-speed ideo came a). SWR can be s ongly al e ed
by a ia ions in en i onmen al condi ions, such as humidi y (Jex e al.,
1985), empe a u e (Diehl and Schaumann, 2007), mois u e con en
(T¨
aume e al., 2005), o ganic ma e con en and species, and he hin
soil laye ’s s uc u al cha ac e is ics; whe e he measu emen mus be
applied on undis u bed samples unde ield condi ions. Wallach e al.
(2005) epo ed s ong de elopmen o wa e epellency in sandy soil
unde o cha d ees ha we e i iga ed wi h seconda y- ea ed was e-
wa e o 30 yea s, esul ing in he accumula ion o hyd ophobic o ganic
compounds on he soil su ace.
Recen s udies ha e shown ha soil spec oscopy in he
VIS–NIR–SWIR egion (350–2500 nm) has he po en ial o p edic SWR
indices. Such s udies ha e been pe o med in he labo a o y wi h
dis u bed soil samples o di e en o de s om New Zealand (Kim e al.,
2014), and o homogeneous soils om Denma k (Knadel e al., 2016).
To he bes o ou knowledge, assessing hyd ophobici y indices, such as
WDPT, by VIS–NIR–SWIR e lec ance spec oscopy in he ield unde
eal condi ions on an undis u bed soil su ace has ne e been ca ied
ou . The aim o his case s udy was o use he SP o be e measu e he
e lec ance o he undis u bed soil su ace, a labo a o y quali y, in o de
o ex ac mo e eliable ield SWR alues han is adi ionally done.
We he e o e conduc ed he ollowing exe cise using he SP assembly
o measu e he undis u bed soil su ace wi h a pa allel measu emen o
WDPT. Two se s o samples we e aken om a 50-yea -old o cha d plo
ha had been i iga ed wi h seconda y- ea ed e luen wa e o 30
yea s (Wallach e al., 2005): 146 samples we e aken o he labo a o y
unde “undis u bed” su ace sampling (Pe sson and Be gs ¨
om, 1991)
and subjec ed o spec al measu emen s using he ASD CP assembly in
he labo a o y and WDPT measu emen s acco ding o a labo a o y
p o ocol (see la e ); 68 samples we e spec ally measu ed in he ield
using he SP be o e he “undis u bed” soil-sampling p ocedu e. These
samples we e also subjec ed o WDPT measu emen s ollowing he same
labo a o y p o ocol.
Fig. 5. Amiaz plain ield es si e, 90 m ×90 m. Th ee s aigh 90 m lines we e measu ed wi h an addi ional 48 andomly measu ed poin s be ween hese lines (sun
ele a ion 70.54
◦, sun azimu h 152.13◦).
E.B. Do e al.
Geode ma 438 (2023) 116636
7
The labo a o y p o ocol o he WDPT measu emen consis s o
placing h ee 50 mL d ops o dis illed wa e on he su ace o a soil
sample, and de e mining he ime elapsed o he d op’s comple e ab-
so p ion (Doe , 1998). The ime was measu ed using a high-speed
came a ha ollowed he wa e d op angle. The spec al measu emen
was pe o med o bo h se s o samples wi h he same ASD spec ome e ,
which was hooked up o he CP (in he labo a o y se ) and he SP (in he
ield se ). The o cha d ee a ea and he sampling loca ions o all soils o
Fig. 7. Compa ison o ba e ibe and SoilPRO® e lec ance measu emen s (a e age o all measu emen s) o e he 90 m ×90 m es a ea in Amiaz plain in 2021 (solid
line) s. 2022 (dashed line).
Fig. 6. Compa ison o ep esen a i e e lec ance and a ia ion o e he Amiaz plain a ea using he ba e ibe and he SoilPRO measu emen s o a 90 m ×90 m
es a ea.
E.B. Do e al.
Geode ma 438 (2023) 116636
8
WDPT measu emen s a e shown in Fig. 9. Mos o he sampling poin s
we e in shaded a eas, wi hou he di ec sola adia ion needed o o -
dina y ield spec al measu emen s.
Fig. 10 shows he spec a o some selec ed soil samples acqui ed in
he ield by he SP. The WDPT alues o each o he soils a e also gi en.
The spec a a e e y s able, simila o labo a o y quali y. Especially
no iceable is he conca e slope ac oss he VIS–NIR–SWIR egion ha
sugges s he p esence o soil o ganic ca bon (SOC) in bo h samples (#69
and #64) wi h he highes WDPT alues (502 and 2215, espec i ely).
This is in good ag eemen wi h Wallach e al.’s (2005) inding ha in
his a ea, he hyd ophobic o ganic subs ances in he e luen wa e ha
ha e accumula ed o e he long i iga ion pe iod a e he d i ing o ce
Fig. 8. Field measu emen s using ba e ibe (a) compa ed o SoilPRO (b), and whi e e e ence measu emen s using he SoilPRO (c).
Fig. 9. The exac loca ions and he WDPT alues (in second) ha measu ed in
he o cha d a ea. Fig. 10. Spec a o soil samples om he o cha d as aken by he SP.
E.B. Do e al.
Geode ma 438 (2023) 116636
9
o he wa e epellency cha ac e is ics o he soil su ace.
The p oxima e spec al-based model was un on bo h da a se s: 146
samples (calib a ion n =108, alida ion n =38) wi h he CP measu e-
men in he labo a o y and 68 samples (calib a ion n =51, alida ion n
=17) wi h he SP measu emen in he ield. To quan i a i ely p edic he
WDPT alues by spec oscopy, we an a pa ial leas squa es eg ession
(PLSR) p ocedu e using se e al p ep ocessing s ages on he e lec ance
alues and hei combina ions: i s de i a i e, Log (1/R), s anda d
no mal a ia e (SNV) (Dhanoa e al., 1989) and con inuum emo al
(CR) using he PARACUDA-II engine (Gholizadeh e al., 2018) (Table 2).
The SP da a se (in-si u SP measu emen s wi h e lec ance SNV p e-
p ocessing) ga e he bes alida ion esul s o he PLSR analyses,
p o iding highe accu acy han he CP da a se (wi h e lec ance p e-
p ocessing) om he labo a o y (R
2
=0.86 and 0.52, and RPIQ =4.43
and 2.16, espec i ely). Al hough he samples ha we e b ough o he
labo a o y we e kep “undis u bed”, using he CP o he measu emen
p ocess may ha e des oyed he agile soil seal and hus, could no
mimic he eal in-si u ield condi ion as achie ed by he SP measu e-
men . The esul s demons a ed he po en ial o he SP assembly o
measu ing a hin undis u bed soil su ace in he ield by co e ing a la ge
ep esen a i e a ea wi h cons an geome y and no con ac ing he soil
seal.
In his case s udy, he SP assembly enabled la ge-scale, p ecise and
ep esen a i e spec al measu emen s o he eal undis u bed soil su -
ace wi h no need o b ing he soil o he labo a o y. In addi ion, he SP
makes i possible o measu e he soil in he ield in shaded a eas unde
he ee plan canopy. The SP p ese es he soil su ace’s condi ion as i
is in he ield and he e o e p o ides be e p oximal spec al modeling
o es ima e he SWR s a us. Easy ope a ion o he SP, which does no
equi e skilled pe sonnel, and he abili y o measu e a la ge numbe o
samples in a sho ime (100 samples’ spec a measu ed in abou 60 min)
and in shaded a eas, place his assembly as a game-change in assessing
he hyd ophobici y s a us o soils benea h ee canopies and close o he
i iga ion pi o o he d ip i iga ion p ac ice.
3.4. Case s udy 4: Assessing he a e o wa e in il a ion in o soil p o ile
This sec ion is a summa y o comp ehensi e esea ch ha has been
published in he li e a u e be ween 2021 and 2022. This wo k has
adop ed he SP o measu e he soil su ace condi ion and i s e ec on he
wa e egime. In he p e ious sec ions, i was demons a ed ha he SP
gi es a be e ep esen a ion o he uppe ( hin) soil su ace laye
(“seal”) condi ion in he ield and enables be e p oximal sensing o soil
su ace- ela ed p ope ies (e.g., WIR). As he uppe laye may be agile
and sensi i e o small spec al changes in he ield such as i e (A gen-
ie o e al., 2021; Yang e al., 2021) o ain ene gy (Goldshlege e al.,
2009; Li e al., 2020), we u he examined he SP’s pe o mance wi h
ano he su ace- ela ed p ope y—wa e -in il a ion a e (WIR).
WIR is a e y impo an hyd ological pa ame e ha con ols uno ,
su ace leaching, soil e osion, wa e -s o age capaci y in he soil p o ile,
and wa e a ailabili y o c ops (F anzluebbe s, 2002). Mo eo e , WIR is
in ol ed in se e al soil p ocesses and has been co ela ed wi h ege a-
ion co e (Walke e al., 1981; Thu ow e al., 1986, 1988; Van de
Koppel e al., 1997, Van de Koppel e al., 2002). Acco dingly, WIR may
be a c i ical soil p ope y in comba ing he dese i ica ion phenomenon
in he cu en e a o global wa ming, concomi an wi h high ain ene gy
and inc eases in p ecipi a ion olume.
WIR in o he soil is s ongly ela ed o he soil’s su ace p ope ies,
such as clay mine als, which a e no necessa ily d i en by SOC, as
demons a ed wi h he wa e epellency issue (SOC may a y om one
sou ce o he nex and may change he WIR in opposi e di ec ions). The
WIR is s ongly dependen on he in e ace be ween he soil and he
a mosphe e and is mainly con olled by he hin seal ha e ol es be-
ween hese wo phases (also e med “physical c us ” by Agassi e al.,
1985, Agassi e al., 1994). Lado e al. (2004) s a ed ha an inc ease in
SOC limi s soil seal o ma ion and he e o e may induce in il a ion. On
he o he hand, in some cases (e.g., case s udy 3 in his pape and sup-
po ed by Cap iel, 1997), some species o SOC may cause hyd opho-
bici y and hus dec ease he WIR alues. Ben-Hu and Le ey (1989)
ound ha clay dispe sion a he soil su ace de e mines WIR and a ec s
seal o ma ion. S e n e al. (1991) examined he ela ionship be ween
clay mine als and WIR s a us and ound ha soils ich in smec i e
mine als a e mo e ulne able o sealing and e osion. They sugges ed
ha kaolini e and illi e ha e opposi e e ec s on soil sealing and he
esul ing e osion. Goldshlege e al. (2009) p edic ed WIR ia spec-
oscopy and PLSR analyses in dis u bed soils unde a con olled ain
simula o using a ying aind op ene gy. As he soil unde he ain
simula o did no mimic he eal in-si u ield condi ion ( he soil su ace
was dis u bed when i was b ough o he labo a o y), he ain simula o
showed only ends, and no he ac ual ield condi ion. As a esul , he
eal ield WIR mus be examined wi h a di e en p o ocol.
The aim o his case s udy was o minimize he des uc i e e ec o
he adi ional CP measu emen on he soil c us , and a oid indi ec
assessmen o he WIR in he labo a o y. Acco dingly, we used he SP
assembly o measu e he e lec ance alues o he soil su ace in he ield
while simul aneously measu ing WIR wi h a mini disk in il ome e
(Decagon De ices, Inc). Se e al samples om di e en loca ions ac oss
Medi e anean coun ies ha p esen a ia ions in soil ex u e we e
selec ed: one place in I aly, one place in Is ael and h ee in G eece
(Fig. 11). A e spec al and WIR measu emen s in he ield, he soil
samples we e b ough o he labo a o y and he spec al measu emen s
we e pe o med on a dis u bed mixed soil sample, using Ben Do e al.’s
(2015) p o ocol o labo a o y ASD spec ome e measu emen s wi h
he CP assembly.
To gene a e he spec al-based models o he WIR p edic ion, he
spec al da a we e subjec ed o se e al p ep ocessing echniques. The
spec al eadings in he ield (using he SP) and he labo a o y (using he
CP) we e compa ed o he WIR measu ed in he ield. Table 3 p o ides
he esul s o he alida ion s age o bo h spec al measu emen p ac-
ices a e Sa i zky–Golay and i s -de i a i e p ep ocessing app oaches
we e execu ed, and he PLSR models we e gene a ed. The mixed da a se
(gene ic) showed excellen esul s in he ield domain using he SP (R
2
Val
=0.70). Al hough he alida ion o he labo a o y-based model also
showed an accep able pe o mance (R
2
Val
=0.57). These esul s
Table 2
The bes esul s o he PLSR model o he alida ion se using he CP and SP
appa a us and he p ep ocessing me hod o he e lec ance alue.
CP da a se (n ¼146 o al: calib a ion; n ¼108, alida ion n ¼38)
Spec a R
2
RMSE RPIQ Bias Slope Fac o s
Re lec ance 0.52 0.73 2.16 −0.24 0.46 7
SP da a se (n ¼68 o al, calib a ion; n ¼51, alida ion; alida ion n ¼17)
Spec a R
2
RMSE RPIQ Bias Slope Fac o s
Re lec ance *SNV 0.86 0.51 4.43 −1.39 0.91 1
RMSE, oo mean squa e e o o p edic ion; RPIQ, a io o pe o mance o
in e qua ile dis ance; SNV, s anda d no mal a ia e.
E.B. Do e al.
Geode ma 438 (2023) 116636
16
Re e ences
Agassi, M., Mo in, J., Shainbe g, I., 1985. E ec o aind op impac ene gy and wa e
salini y on in il a ion a es o sodic soils. Soil Sci. Soc. Am. J. 49, 186–190. h ps://
doi.o g/10.2136/SSSAJ1985.03615995004900010037X.
Agassi, M., Bloem, D., Ben-Hu , M., 1994. E ec o d op ene gy and soil and wa e
chemis y on in il a ion and e osion. Wa e Resou . Res. 30, 1187–1193. h ps://
doi.o g/10.1029/93WR02880.
A gen ie o, I., Ricci, G.F., Elia, M., D’es e, M., Giannico, V., Ronco, F.V., Gen ile, F.,
Sanesi, G., 2021. Combining me hods o es ima e pos - i e soil e osion using emo e
sensing da a. Fo es s 12. h ps://doi.o g/10.3390/ 12081105.
Ba˜
n´
on, M., Jus el, A., Vel´
azquez, D., Quesada, A., 2013. Regional wea he su ey on
Bye s Peninsula, Li ings on Island, Sou h She land Islands, An a c ica. An a c ic Sci.
25, 146–156. h ps://doi.o g/10.1017/S0954102012001046.
Bau e s, T.W.J., Dica lo, D.A., S eenhuis, T.S., Pa lange, J.Y., 2000. Soil wa e con en
dependen we ing on cha ac e is ics in sands. J. Hyd ol. 231–232, 244–254.
h ps://doi.o g/10.1016/S0022-1694(00)00198-0.
Behe a, S.K., Adamchuk, V.I., Shukla, A.K., Pandey, P.S., Kuma , P., Shukla, V.,
Thiyaga ajan, C., Rai, H.K., Hadole, S., Sachan, A.K., Singh, P., T i edi, V.,
Mish a, A., Bu ail, N.P., Kuma , P., P ajapa i, R., Tiwa i, K., Su i, D., Sha ma, M.,
2022. The scope o using p oximal soil sensing by he a me s o India.
Sus ainabili y 14. h ps://doi.o g/10.3390/su14148561.
Ben Do , E., Ong, C., Lau, I.C., 2015. Re lec ance measu emen s o soils in he labo a o y:
s anda ds and p o ocols. Geode ma 245–246, 112–124. h ps://doi.o g/10.1016/j.
geode ma.2015.01.002.
Bena ides-Solo io, J., MacDonald, L.H., 2001. Pos - i e uno and e osion om
simula ed ain all on small plo s, Colo ado F on Range. Hyd ol. P ocess. 15,
2931–2952. h ps://doi.o g/10.1002/hyp.383.
Ben-Do , E., Banin, A., 1995. Nea -in a ed analysis as a apid me hod o simul aneously
e alua e se e al soil p ope ies. Soil Sci.Soc. Am. J. 59, 364–372. h ps://doi.o g/
10.2136/sssaj1995.03615995005900020014x.
Ben-Do , E., G ano , A., No esco, G., 2017. A simple appa a us o measu e soil spec al
in o ma ion in he ield unde s able condi ions. Geode ma 306, 73–80. h ps://doi.
o g/10.1016/j.geode ma.2017.06.025.
Ben-Hu , M., Le ey, J., 1989. E ec o polysaccha ides, clay dispe sion, and impac
ene gy on wa e in il a ion. Soil Sci. Soc. Am. J. 53, 233–238. h ps://doi.o g/
10.2136/SSSAJ1989.03615995005300010041X.
Bou e , M., Thome, K., Be helo , B., Bialek, A., Czapla-Mye s, J., Fox, N.P., Go yl, P.,
Hen y, P., Ma, L., Ma cq, S., Meyg e , A., Wenny, B.N., Woolliams, E.R., 2019.
RadCalNe : a adiome ic calib a ion ne wo k o ea h obse ing image s ope a ing
in he isible o sho wa e in a ed spec al ange. Remo e Sens. 11 h ps://doi.o g/
10.3390/ s11202401.
Bu ch, G.J., Moo e, I.D., Bu ns, J., 1989. Soil hyd ophobic e ec s on in il a ion and
ca chmen uno . Hyd ol. P ocess. 3, 211–222. h ps://doi.o g/10.1002/
hyp.3360030302.
Cap iel, P., 1997. Hyd ophobici y o o ganic ma e in a able soils: in luence o
managemen . Eu . J. Soil Sci. 48, 457–462. h ps://doi.o g/10.1111/j.1365-
2389.1997. b00211.x.
Ch is y, C.D., 2008. Real- ime measu emen o soil a ibu es using on- he-go nea
in a ed e lec ance spec oscopy. Compu . Elec on. Ag ic. 61, 10–19. h ps://doi.
o g/10.1016/j.compag.2007.02.010.
Dhanoa, M.S., Lis e , S.J., Ba nes, R.J., 1989. S anda d no mal a ia e ans o ma ion
and de- ending o nea -in a ed di use e lec ance spec a. Appl. Spec osc. 43,
772–777.
Diehl, D., Schaumann, G.E., 2007. The na u e o we ing on u ban soil samples: we ing
kine ics and e apo a ion assessed om sessile d op shape. Hyd ol. P ocess. 21,
2255–2265. h ps://doi.o g/10.1002/HYP.6745.
Doe , S.H., 1998. On s anda dizing he “wa e d op pene a ion ime” and he “mola i y
o an e hanol d ople ” echniques o classi y soil hyd ophobici y: a case s udy using
medium ex u ed soils. Ea h Su . P ocess. Land o ms 23, 663–668. h ps://doi.o g/
10.1002/(SICI)1096-9837(199807)23:7<663::AID-ESP909>3.0.CO;2-6.
Doe , S.H., Shakesby, R.A., Dekke , L.W., Ri sema, C.J., 2006. Occu ence, p edic ion
and hyd ological e ec s o wa e epellency amongs majo soil and land-use ypes in
a humid empe a e clima e. Eu . J. Soil Sci. 57, 741–754. h ps://doi.o g/10.1111/
j.1365-2389.2006.00818.x.
F ancos, N., Romano, N., Nas a, P., Zeng, Y., Szab´
o, B., Man eda, S., Ci aolo, G.,
M´
esz´
a os, J., Zhuang, R., Su, B., Ben-do , E., 2021. Mapping wa e in il a ion a e
using g ound and ua hype spec al da a: A case s udy o alen o, i aly. Remo e Sens.
13, 1–30. h ps://doi.o g/10.3390/ s13132606.
F anzluebbe s, A.J., 2002. Wa e in il a ion and soil s uc u e ela ed o o ganic ma e
and i s s a i ica ion wi h dep h. Soil Tillage Res. 66, 197–205. h ps://doi.o g/
10.1016/S0167-1987(02)00027-2.
Gholizadeh, A., Sabe ioon, M., Ca mon, N., Bo u ka, L., Ben-Do , E., 2018. Examining
he pe o mance o PARACUDA-II da a-mining engine e sus selec ed echniques o
model soil ca bon om e lec ance spec a. Remo e Sens. 10 h ps://doi.o g/
10.3390/ s10081172.
Goldshlege , N., Ben-Do , E., Chudno sky, A., Agassi, M., 2009. Soil e lec ance as a
gene ic ool o assessing in il a ion a e induced by s uc u al c us o
he e ogeneous soils. Eu . J. Soil Sci. 60, 1038–1051. h ps://doi.o g/10.1111/
j.1365-2389.2009.01162.x.
Gomez, C., Visca a Rossel, R.A., McB a ney, A.B., 2008. Soil o ganic ca bon p edic ion
by hype spec al emo e sensing and ield is-NIR spec oscopy: An Aus alian case
s udy. Geode ma 146, 403–411. h ps://doi.o g/10.1016/j.geode ma.2008.06.011.
Helle -Pea lsh ien, D., Ben-Do , E., 2022. CalVal E alua ion o DESIS p oduc s in Amiaz
Plain and Makh esh Ramon Tes si es, in: 1s DESIS Use Wo kshop – Imaging
Spec ome e Space Mission, Calib a ion and Valida ion, Applica ions, Me hods, 28
Sep .–1 Oc . 2021, Vi ual. The In e na ional A chi es o he Pho og amme y,
Remo e Sensing and Spa ial In o ma ion Sciences, pp. 13–21.
Helle -Pea lsh ien, D., Pigna i, S., G eisman-Ran, U., Ben-Do , E., 2021. PRISMA senso
e alua ion: a case s udy o mine al mapping pe o mance o e Makh esh Ramon.
Is ael. In . J. Remo e Sens. 42, 5882–5914.
Helle -Pea lsh ien, D., Pigna i, S., Ben-Do , E., 2023. Vica ious CAL/VAL app oach o
o bi al hype spec al senso s using mul iple si es. Remo e Sens. 15, 771. h ps://doi.
o g/10.3390/ s15030771.
Hend ickx, J.M.H., Dekke , L.W., Boe sma, O.H., 1993. Uns able we ing on s in wa e -
epellen ield soils. J. En i on. Qual. 22, 109–118. h ps://doi.o g/10.2134/
jeq1993.00472425002200010014x.
Jex, G.W., Bleakley, B.H., Hubbell, D.H., Mun o, L.L., 1985. High humidi y-induced
inc ease in wa e epellency in some sandy soils. Soil Sci. Soc. Am. J. 49, 1177–1182.
h ps://doi.o g/10.2136/SSSAJ1985.03615995004900050021X.
Kim, I., Pullanaga i, R.R., Deu e , M., Singh, R., Huh, K.Y., Clo hie , B.E., 2014. The use
o isible and nea -in a ed spec oscopy o he analysis o soil wa e epellency.
Eu . J. Soil Sci. 65, 360–368. h ps://doi.o g/10.1111/ejss.12138.
Knadel, M., Masís-Mel´
endez, F., De Jonge, L.W., Mold up, P., A hu , E., G e e, M.H.,
2016. Assessing soil wa e epellency o a sandy ield wi h isible nea in a ed
spec oscopy. J. Nea In a ed Spec osc. 24, 215–224. h ps://doi.o g/10.1255/
jni s.1188.
Kodai a, M., Shibusawa, S., 2013. Using a mobile eal- ime soil isible-nea in a ed
senso o high esolu ion soil p ope y mapping. Geode ma 199, 64–79. h ps://doi.
o g/10.1016/j.geode ma.2012.09.007.
Kokaly, R.F., Tu pie, K.R., 2019. Calib a ion and alida ion wo king g oup o su ace
biology and geology (SBG). In: IGARSS 2019-2019 IEEE In e na ional Geoscience
and Remo e Sensing Symposium. IEEE, pp. 8517–8518.
Kusumo, B.H., 2018. In si u measu emen o soil ca bon wi h dep h using nea in a ed
(NIR) spec oscopy. IOP Con . Se . Ma e . Sci. Eng. 434 h ps://doi.o g/10.1088/
1757-899X/434/1/012235.
Laba e, S., Jacquemoud, S., Fe a i, C., Delo me, A., De ien, A., G andin, R.,
Tanguy, B., 2019. Re ie ing soil su ace oughness wi h he Hapke pho ome ic
model: Con on a ion wi h he g ound u h. Remo e Sens. En i on. 225, 1–15.
h ps://doi.o g/10.1016/j. se.2019.02.014.
Lado, M., Paz, A., Ben-Hu , M., 2004. O ganic ma e and agg ega e-size in e ac ions in
sa u a ed hyd aulic conduc i i y. Soil Sci. Soc. Am. J. 68, 234–242. h ps://doi.o g/
10.2136/sssaj2004.2340.
Le ey, J., Ca illo, M.L.K., Pang, X.P., 2000. App oaches o cha ac e ize he deg ee o
wa e epellency. J. Hyd ol. 231–232, 61–65. h ps://doi.o g/10.1016/S0022-1694
(00)00183-9.
Li, J., Wang, W., Guo, M., Kang, H., Wang, Z., Huang, J., Sun, B., Wang, K., Zhang, G.,
Bai, Y., 2020. E ec s o soil ex u e and g a el con en on he in il a ion and soil
loss o spoil heaps unde simula ed ain all. J. Soil. Sedimen . 20, 3896–3908.
h ps://doi.o g/10.1007/s11368-020-02729-6.
L´
opez-Ma ínez, J., Ma ínez de Pis´
on, E., Se ano, E., A che, A., 1996.
Geomo phological map o Bye s Peninsula, Li ings on Island. BAS Geomap Se ies.
Wi h Supplemen a y Tex . B i ish An a c ic Su ey, Camb idge.
Mil on, E.J., 2010. Re iew a icle: p inciples o ield spec oscopy. In . J. Remo e Sens. 8,
1807–1827. h ps://doi.o g/10.1080/01431168708954818.
Mil on, E.J., Schaepman, M.E., Ande son, K., Kneubühle , M., Fox, N., 2009. P og ess in
ield spec oscopy. Remo e Sens. En i on. 113, S92–S109. h ps://doi.o g/10.1016/
j. se.2007.08.001.
Mo ain, S.A., Budge, A.M., 2004. Pos -Launch Calib a ion o Sa elli e Senso s. Pos -
Launch Calib a ion Sa ell. Senso s d. 10.1201/9780203026830.
Peddle, D.R., Pe e Whi e, H., So e , R.J., Mille , J.R., LeD ew, E.F., 2001. Re lec ance
p ocessing o emo e sensing spec o adiome e da a. Compu . Geosci. 27, 203–213.
h ps://doi.o g/10.1016/S0098-3004(00)00096-0.
Pe sson, L., Be gs ¨
om, L., 1991. D illing me hod o collec ion o undis u bed soil
monoli hs. Soil Sci. Soc. Am. J. 55, 285–287. h ps://doi.o g/10.2136/
SSSAJ1991.03615995005500010050X.
P i zne , K., Ba olo, R., Ca , G., Espa on, A., Bollh¨
o e , A., 2011. S anda ds o
e lec ance spec al measu emen o empo al ege a ion plo s. Supe ising Scien is
Repo 195. Supe ising Scien is , Da win NT. ISBN: 978-1-921069-16-1.
RadCalNe Po al, n.d. h ps://www. adcalne .o g/#!/ (accessed 22 July 2021).
Ri sema, C.J., Niebe , J.L., Dekke , L.W., S eenhuis, T.S., 1998. S able o uns able
we ing on s in wa e epellen soils – e ec o an eceden soil mois u e con en .
Soil Tillage Res. 47, 111–123. h ps://doi.o g/10.1016/S0167-1987(98)00082-8.
Rodiono , A., Welp, G., Dame ow, L., Be g, T., Amelung, W., P¨
a zold, S., 2015. Towa ds
on- he-go ield assessmen o soil o ganic ca bon using Vis-NIR di use e lec ance
spec oscopy: de eloping and es ing a no el ac o -d i en measu ing chambe . Soil
Tillage Res. 145, 93–102. h ps://doi.o g/10.1016/j.s ill.2014.08.007.
Rollin, E., Mil on, E., Eme y, D., 2000. Re e ence panel aniso opy and di use adia ion –
some implica ions o ield spec oscopy. In . J. Remo e Sens. 21, 2799–2810.
h ps://doi.o g/10.1080/01431160050121258.
Schmid, T., Nie o, A., L´
opez-Ma ínez, J., Guillaso, S., Koch, M., Oli a-U cia, B.,
Lamb´
an, L.J., 2021. Cha ac e izing he ice- ee a ea o Cie a Poin (An a c ic
Peninsula) using e lec ance spec oscopy. IGARSS IEEE In e na ional 6178–6181.
h ps://doi.o g/10.1109/IGARSS47720.2021.9554835.
Schop e , J.T., 2008. Spec odi ec ional g ound-based emo e sensing using dual- iew
goniome y. Field BRF e ie al and assessmen o he di use i adiance dis ibu ion
in spec odi ec ional ield measu emen s. Remo e Sens. Se . 53, 93.
Schweige , A.K., 2020. Spec al ield campaigns: planning and da a collec ion, in:
Ca ende -Ba es, J., Gamon, J.A., Townsend, P.A. (Eds.), Remo e Sensing o Plan
Biodi e si y. Sp inge , Cham, pp. 385–423. 10.1007/978-3-030-33157-3_15.
Simas, F.N.B., Schae e , C.E.G.R., Michel, R.F.M., F ancelino, M.R., Bockheim, J.G.,
2015. Soils o he Sou h O kney and Sou h She land, in: Bockheim, J.G. (Ed.), The
E.B. Do e al.
Geode ma 438 (2023) 116636
17
Soils o An a c ica. Wo ld Soils Book Se ies. Sp inge , Cham, pp. 227–273. 10.1007/
978-3-319-05497-1_13.
Smellie, J.L., Pankhu s , R.J., Thomson, M.R.A., Da ies, R.E.S., 1984. The Geology o he
Sou h She land Islands: VI. S a ig aphy, Geochemis y and E olu ion. B i ish
An a c ic Su ey, Camb idge (B i ish An a c ic Su ey Scien i ic Repo s 87).
S enbe g, B., Visca a Rossel, R.A., Mouazen, A.M., We e lind, J., 2010. Visible and nea
in a ed spec oscopy in soil science. Ad . Ag on. 107, 163–215. h ps://doi.o g/
10.1016/S0065-2113(10)07005-7.
S e n, R., Ben-Hu , M., Shainbe g, I., 1991. Clay mine alogy e ec on ain in il a ion,
seal o ma ion and soil losses. Soil Sci. 152, 455–462. h ps://ag is. ao.o g/ag
is-sea ch/sea ch.do? eco dID=US9311511.
S one , E.R., Baumga dne , M.F., Weismille , R.A., Biehl, L.L., Robinson, B.F., 1980.
Ex ension o labo a o y-measu ed soil spec a o ield condi ions. Soil Sci. Soc. Am.
J. 44 (3), 572–574. h ps://doi.o g/10.2136/
SSSAJ1980.03615995004400030028X.
SVC, 2019. Field Spec oscopy Guide wi h SVC i-se ies Spec o adiome e s 1–76.
T¨
aume , K., S o egen, H., Wessolek, G., 2005. De e mina ion o epellency dis ibu ion
using soil o ganic ma e and wa e con en . Geode ma 125, 107–115. h ps://doi.
o g/10.1016/j.geode ma.2004.07.004.
Thenkabail, P.S., 2016. Remo ely Sensed Da a Cha ac e ized, Classi ica ion and
Accu acies. Remo e Sensing Handbook, Volume 1. Taylo & F ancis G oup,
Ox o dshi e.
Thu ow, T.L., Blackbu n, W.H., Taylo , C.A., 1986. Hyd ologic cha ac e is ics o
ege a ion ypes as a ec ed by li es ock g azing sys ems, Edwa ds Pla eau, Texas.
J. Range Manag. 39, 505. h ps://doi.o g/10.2307/3898758.
Thu ow, T.L., Blackbu n, W.H., Taylo , C.A., 1988. In il a ion and in e ill e osion
esponses o selec ed li es ock g azing s a egies, Edwa ds Pla eau, Texas. J. Range
Manag. 41, 296. h ps://doi.o g/10.2307/3899382.
Van de Koppel, J., Rie ke k, M., Weissing, F.J., 1997. Ca as ophic ege a ion shi s and
soil deg ada ion in e es ial g azing sys ems. T ends Ecol. E ol. 12, 352–356.
h ps://doi.o g/10.1016/S0169-5347(97)01133-6.
Van de Koppel, J., Rie ke k, M., Van Lange elde, F., Kuma , L., Klausmeie , C.A.,
F yxell, J.M., Hea ne, J.W., Van Andel, J., De Ridde , N., Skidmo e, A.,
S oosnijde , L., P ins, H.H.T., 2002. Spa ial he e ogenei y and i e e sible
ege a ion change in semia id g azing sys ems. Am. Na . 159, 209–218. h ps://doi.
o g/10.1086/324791.
Visca a Rossel, R.A., Wal oo , D.J.J., McB a ney, A.B., Janik, L.J., Skjems ad, J.O.,
2006. Visible, nea in a ed, mid in a ed o combined di use e lec ance
spec oscopy o simul aneous assessmen o a ious soil p ope ies. Geode ma 131,
59–75. h ps://doi.o g/10.1016/j.geode ma.2005.03.007.
Visca a Rossel, R.A., Beh ens, T., Ben-Do , E., B own, D.J., Dema ˆ
e, J.A.M.,
Shephe d, K.D., Shi, Z., S enbe g, B., S e ens, A., Adamchuk, V., Aïchi, H.,
Ba h`
es, B.G., Ba holomeus, H.M., Baye , A.D., Be noux, M., B¨
o che , K.,
B odský, L., Du, C.W., Chappell, A., Fouad, Y., Geno , V., Gomez, C., G unwald, S.,
Guble , A., Gue e o, C., Hedley, C.B., Knadel, M., Mo ´
as, H.J.M., Noci a, M.,
Rami ez-Lopez, L., Roudie , P., Campos, E.M.R., Sanbo n, P., Selli o, V.M.,
Suddu h, K.A., Rawlins, B.G., Wal e , C., Winowiecki, L.A., Hong, S.Y., Ji, W., 2016.
A global spec al lib a y o cha ac e ize he wo ld’s soil. Ea h-Science Re . 155,
198–230. h ps://doi.o g/10.1016/j.ea sci e .2016.01.012.
Walke , B.H., Ludwig, D., Holling, C.S., Pe e man, R.M., 1981. S abili y o semi-a id
sa anna g azing sys ems. J. Ecol. 69, 473–498. h ps://doi.o g/10.2307/2259679.
Wallach, R., Ben-A ie, O., G abe , E.R., 2005. Soil wa e epellency induced by long- e m
i iga ion wi h ea ed sewage e luen . J. En i on. Qual. 34, 1910–1920. h ps://
doi.o g/10.2134/jeq2005.0073.
Wang, Z., Wu, Q.J., Wu, L., Ri sema, C.J., Dekke , L.W., Feyen, J., 2003. E ec s o wa e
epellency on in il a ion a e and low ins abili y, in: Ri sema, C.J., Dekke , L.W.
(Eds.), Soil Wa e Repellency – Occu ence, Consequences, and Amelio a ion.
Else ie , Ams e dam, pp. 235–244. 10.1016/B978-0-444-51269-7.50024-2.
Wenjun, J., Zhou, S., Jingyi, H., Shuo, L., 2014. In si u measu emen o some soil
p ope ies in paddy soil using isible and nea -in a ed spec oscopy. PLoS One 9 (8),
e105708.
Wise, J.E., Ma s, J.C., 2022. Field e lec ance measu emen s a nigh o beach and dese
sands wi hin a pa icula e BRDF model. Remo e Sens. 14, 5020. h ps://doi.o g/
10.3390/ s14195020.
Xu, H., Zhang, L., Huang, W., Li, X., Si, X., Xu, W., Song, Q., 2020. On-boa d absolu e
adiome ic calib a ion and alida ion based on sola di use o HY-1C SCS.
Guangxue Xuebao/Ac a Op . Sin. 40, 30015–30034. h ps://doi.o g/10.3788/
AOS202040.0928002.
Yang, Y., Hu, X., Cao, X., Jin, T., Wang, Y., 2021. Medium- e m e ec s o di e en
wild i e se e i ies on soil p ope ies: a case s udy o Hengduan Moun ains,
sou hwes e n China. IOP Con . Se .: Ea h En i on. Sci. 861 h ps://doi.o g/
10.1088/1755-1315/861/6/062021.
E.B. Do e al.
