Day- o-day changes in expe imen al elec on densi y p ofiles and
hei implica ions o IRI model
G. Mi o
´ Ama an e
a,*
, M. Cue o San ama ı
´a
b
, M. Mose de Gonza
´lez
c
,
S.M. Radicella
a
, R. Ezque
d,e
a
Ae onomy and Radiop opaga ion Labo a o y, Abdus Salam ICTP, S ada Cos ie a 11, T ies e 34014, I aly
b
Geophysics and Me eo ology Depa men , Physics Facul y, UCM Uni e si y, A da. Complu ense, S/N 28040 Mad id, Spain
c
CASLEO CONICET, A da. Espan˜a (Su ) 1512–5400, San Juan, A gen ina
d
e
Physics Depa men , Ionosphe e Lab., CONICET, Buenos A ies 296, 4000 S.M. de Tucuma
´ n, A gen ina
Regional Facul y o Tucuma
´ n, Technological Na ional Uni e si y, Ri ad ia 1050, 4000 S.M. de Tucuma
´
n, A gen ina
Abs ac
The elec on densi y a iabili y a fixed heigh s is s udied o use in he In e na ional Re e ence Ionosphe e IRI model. Mon hly
median, uppe and lowe qua ile alues we e ob ained o
o
F2, hmF2, B0 and B1 as deduced om ionog ams. The IRI elec on
densi y p ofiles es ablished wi h hese alues we e hen compa ed wi h he median and qua iles a fixed heigh s. Resul s a e shown
o he h ee ionosphe ic s a ions E1 A enosillo (37.1 N, 353.3 E), Tucuman (26.9 S, 294.6 E) and San Juan (31.5 S, 290.4 E) as a
unc ion o sola ac i i y, season and local ime.
As ound by o he au ho s he heigh o maximum a iabili y, h max, is loca ed below he peak elec on densi y heigh (hmF2) in
all he cases. Values o diffe ences be ween hmF2 and h max a e analized.
Va iabili y defined as he in e qua ile diffe ence and h max esul s a e calcula ed om expe imen al elec on densi y p ofiles
measu ed a he h ee s a ions.
Keywo ds: Va iabili y-Qua ile-Median-IRI; Elec on densi y p ofiles; Ionosphe e
1. In oduc ion
One o he aims o he In e na ional Re e ence Iono-
sphe e Task Fo ce Ac i i ies held a Abdus Salam ICTP
is o de elop a model o a iabili y o IRI use s. The o-
cus so a has been o s udy he peak (
o
F2, hmF2) and
p ofile (B0, B1) pa ame e s a iabili y which de e mines
he whole p ofile a iabili y. In he las yea s, se e al
au ho s (Alazo e al., 2003; Lazo e al., 2003; Mose
e al., 2003; Adeniyi and Radicella, 2003) ha e es ima ed
and analyzed he de ia ion om he mon hly mean o
specified condi ions o model he a iabili y o hese ion-
osphe ic pa ame e s.
Con inuing his ea lie wo k, his s udy ocuses on
wo main goals. Fi s o all, o find ou how he pa am-
e e s
o
F2, hmF2, B0 o B1 affec he al i udinal changes
in elec on densi y a iabili y. Median, uppe and lowe
qua iles o he elec on densi y alues a fixed heigh s
will be compa ed wi h he IRI elec on densi y p ofiles
ob ained using he co esponding median, uppe and
lowe qua iles o he peak and p ofile pa ame e s o
each hou , mon h, yea and loca ion.
The second aim is o analyze he elec on densi y a -
iabili y a fixed heigh s o h ee ionosphe ic s a ions
*
Co esponding au ho . Tel.: +39 040 224 0338; ax: +39 040 224
604.
E-mail add ess: [email p o ec ed] (G.M. Ama an e).
using expe imen al elec on densi y p ofiles co espond-
ing o diffe en ionosphe ic condi ions. As has been ec-
ommended by he IRI Task Fo ce Ac i i y, we use he
in e -qua ile ange as measu e o he day- o-day a i-
abili y which is defined as ollows:
IQC ¼ðUppe qua ile Lowe qua ileÞ=median:ð1Þ
Lowe , middle and uppe qua ile a e defined as
hose alues ha ma k he 25%, 50% and 75% poin s
o he da a dis ibu ion. Simila s udies ha e been
Fig. 1. Median (ci cle shaped poin s), uppe (squa e shaped poin s) and lowe qua iles ( iangle shaped poin s) o elec on densi y p ofiles measu ed
in Oc obe 1999 (high sola ac i i y) a 00 LT (le column) and 12 LT ( igh column) wi h El A enosillo digisonde. These a e compa ed wi h he IRI
p ofiles calcula ed using he median (solid line), uppe (do ed line) and lowe (dashed line) qua iles o peak and p ofiles pa ame e s ob ained by he
digisonde. The sepa a e ows om op downwa ds show p ofiles o median and qua ile alues espec i ely o
o
F2, hmF2, B0 and B1, wi h in each
case he o he p ofile pa ame e s being aken as he median alues.
p esen ed in he las yea s o de e mine he ionosphe ic
a iabili y a fixed heigh s. Mose de Gonzalez and
Radicella (1995) and Radicella e al. (1994) show he
ollowing esul s: (1) he ela i e a iabili y o elec on
densi y inc eases wi h heigh in he F2 egion wi h a
maximum below he F2 peak, (2) he e is a minimum
o a iabili y a abou 180–210 km al i ude a ound
noon, (3) he nigh - ime IQC is appa en ly la ge han
du ing day- ime, (4) he day- ime a iabili y is la ge
a lowe sola ac i i y han a high sola ac i i y and
he opposi e occu s abo e 220 km by nigh - ime and
(5) he ela i e a iabili y has a clea dependence on
sola zeni h angle.
Sole
´and Al adill (1997) de e mined he ionosphe ic
a iabili y a fixed heigh s using da a om a low sola
ac i i y yea o he Eb o Obse a o y s a ion (40.4 N,
Fig. 2. Same as in Fig. 1, a El A enosillo o Oc obe 1995 (low sola ac i i y).
0.3 E). In his case, h ee pa ame e s, which a e ob ained
wi h uppe and lowe deciles and i s s anda d de ia ion,
a e conside ed o define he absolu e and ela i e a ia-
bili y. The main esul s hey ha e ound abou he ela i e
a iabili y a e ha : (1) i inc eases wi h al i ude du ing
day ime, bu dec eases du ing nigh ime, (2) du ing day-
ime, i shows a ela i e minimum be ween 120 and 150
km and i minimizes nea o a F2 peak du ing nigh ime,
(3) i has a minimum a ound midday, (4) i is la ge in
win e and smalle in summe , du ing nigh ime.
Ezque e al. (2002) conside ed low magne ic ac i i y
da a se s om wo Sou h Ame ican ionosphe ic s a ions
and a ela i e a iabili y coefficien based on he
s anda d de ia ion alue. Thei esul s show ha : (1)
Fig. 3. Same as in Fig. 1, a San Juan o Oc obe 1971 (middle sola ac i i y).
a midnigh he a iabili y inc eases wi h heigh eaching
a maximum below he peak densi y; (2) he highes a -
iabili y was obse ed a 280 km o nigh equinox and
high sola ac i i y; (3) by noon, he a iabili y dec eases
wi h inc easing sola ac i i y; (4) he a iabili y is la ge
by nigh han by day abo e 220 km, below ha al i ude
his beha io is e e sed.
2. Da a sec ion
In o de o each he wo goals desc ibed in he p e-
ious sec ion, ionosphe ic pa ame e s (peak and p ofiles
pa ame e s) and bo omside elec on densi y p ofiles ob-
ained wi h digisondes a e analysed. The h ee digi-
sondes chosen in his s udy a e loca ed in San Juan
Fig. 4. Same as in Fig. 1, a Tucuman o Oc obe 1969 (high sola ac i i y).

(31.5 S, 290.4 E), El A enosillo (37.1 N, 353.3 E) and
Tucuman (26.9 S, 294.6 E). They co e mid-la i ude
and c es egions and No he n and Sou he n Hemi-
sphe es. Diffe en condi ions ha e been s udied aken
in o accoun diu nal (00, 06, 12, 18 LT), seasonal (Jan-
ua y, Ap il, July, Oc obe ) and sola ac i i y (high, low)
a iabili y.
Since he ionosphe ic da a base a ailable o Tucu-
man is smalle han hose om San Juan and El A eno-
sillo and conside ing ha hese wo loca ions ha e he
same geog aphic la i ude bu in diffe en Hemisphe e,
esul s shown in his pape co espond mainly o hese
wo s a ions. Howe e , a simila beha io is also p e-
sen ed a Tucuman.
3. Me hodology
3.1. Median, uppe and lowe qua iles o elec on densi y
p ofiles p ocessing
All he expe imen al elec on densi y p ofiles we e ob-
ained by applying he ionog am in e sion echnique
NHPC (Huang and Reinisch, 1995) o abou 1600 dig-
i al ionog ams. Be o e his p ocess, he ionog ams we e
manually co ec ed.
F om hese p ofiles, median, uppe and lowe qua -
iles o daily elec on concen a ion alues a fixed
heigh s we e ob ained o each yea , mon h and hou .
The e o e, we calcula ed h ee elec on densi y alues
o each heigh : median, uppe and lowe qua iles o
elec on uppe and lowe densi y. Taking in o accoun
hese h ee alues o all he heigh ange, median, qua -
iles p ofiles we e p ocessed a e aging alues a 10 km
in e al. These h ee p ofiles will be compa ed wi h he
IR1 p ofiles ha a e ob ained as desc ibed in he IRI
p ofile p ocessing sec ion.
3.2. Ionosphe ic pa ame e s calcula ion
Peak pa ame e s,
o
F2 and hmF2, and p ofile pa am-
e e s, B0 and B1 we e ex ac ed om he edi ed iono-
g ams and he indi idual expe imen al elec on
concen a ion p ofiles. Median, uppe and lowe qua -
iles o hese pa ame e s o each hou , mon h and yea
we e calcula ed. These peak and p ofile pa ame e s we e
used as inpu in he IRI p ofile p ocessing.
3.3. IRI p ofile p ocessing
Median, uppe and lowe IRI p ofiles o
o
F2 we e
calcula ed using he ollowing ionosphe ic pa ame e s
o each hou , mon h and yea :
IRI
o
F2 median p ofile: calcula ed using median
o
F2, hmF2, B0 and B1 alues.
IRI
o
F2 uppe p ofile: calcula ed using uppe
o
F2
and median hmF2, B0 and B1 alues.
IRI
o
F2 lowe p ofile: calcula ed using lowe
o
F2
and median hmF2, B0 and B1 alues.
The same p ocedu e is epea ed o he ionosphe ic
pa ame e s hmF2, B0 and B1. These h ee p ofiles will
be compa ed wi h he co esponding median, uppe
and lowe qua iles o he elec on densi y p ofiles de-
duced om he measu ed ionog ams.
The IRI e sion used o calcula e he elec on concen-
a ion p ofiles was di ec ly downloaded om he web
page: h p://nssdc p.gs c.nasa.go /models/ionosphe ic/
i i/. Due o he la ge numbe o cases p ocessed, some
changes we e made o simpli y he da a inpu o he p o-
g am. As a consequence, he expe imen al ionosphe ic
pa ame e s,
o
F2, hmF2, B0 and B1, we e in oduced
au oma ically om a file in o he IRI model.
3.4. Elec on densi y a iabili y a fixed heigh s
The second aim o his pape , he analysis o he al -
i udinal changes in elec on densi y a iabili y, was s ud-
ied using he measu ed bo omside elec on densi y
p ofiles. The a iabili y pa ame e , IQC, defined in he
p e ious session, was calcula ed e e y 10 km. Resul s
o diffe en condi ions will be discussed in he ollowing
sec ion.
4. Resul s
Examples o median, uppe and lowe qua iles o
elec on densi y p ofiles (ci cle, squa e and iangle
poin s in he plo s) and median, uppe and lowe qua -
iles IRI model p ofiles (solid, do ed and dashed lines
in he plo s) co esponding o
o
F2, hmF2, B0 and B1 ob-
ained by he me hod explained be o e a e plo ed in
Figs. 1–4 o El A enosillo (high and low sola ac i i y),
San Juan (medium sola ac i i y) and Tucuman (high so-
la ac i i y). In each figu e he igh column shows day-
ime esul s and he le column nigh ime esul s. The
ou panels in each column show he effec s o
o
F2
(uppe mos panel), hmF2, B0 and B1 (lowes panel),
espec i ely. The IRI cu es in he uppe mos panel,
o example, we e ob ained by using he median, uppe
and lowe qua ile o
o
F2 and he median o he es
o he pa ame e s. The plo s show ha he pa ame e
wi h he g ea es influence on he elec on densi y is he
c i ical equency
o
F2. This is well illus a ed in Fig. 1.
The h ee IRI p ofiles calcula ed wi h he median, uppe
and lowe qua iles
o
F2 alues (solid, do ed and dashed
lines in he plo s o he fi s ow) ep oduces well he elec-
on densi y a iabili y a fixed heigh s (ci cle, squa e
and iangle poin s). I is also impo an o poin ou ha
he pa ame e s B0 and B1 (3 d and 4 h panels in each
ow) ha e a negligen ly small effec on he a iabili y o
he elec on densi y p ofile (median, uppe and lowe
qua iles IRI cu es a e undis inguishable).
These examples indica e ha he a iabili y o he
elec on concen a ion p ofile could be ep oduced ea-
sonably well by a model o he
o
F2 a iabili y based
on median and uppe and lowe qua ile condi ions.
Howe e , he combina ion o
o
F2 and hmF2 a iabil-
i y seems o be a be e app oxima ion. Fu he s udies
will include he effec o combining hese ionosphe ic
pa ame e s a iabili y in he elec on densi y p ofile
a iabili y.
Wi h ega d o he s udy o he elec on densi y a -
iabili y a fixed heigh s, he pa ame e IQC is calcula ed
om he expe imen al bo omside p ofiles o he diffe -
en condi ions. Figs. 5 and 6 show he IQC alues a El
A enosillo and San Juan o diffe en sola ac i i ies,
seasons and hou s.
Fig. 5. IQC alues a El A enosillo o 1995 (low sola ac i i y) and 1999 (high sola ac i i y), diffe en season and hou .
As can be seen, he ela i e a iabili y exp essed by
IQC gene ally eaches he highes alue below he peak
densi y hmF2. The alue o hmF2 is ex ac ed om he
median o elec on densi y p ofile. The median o he
dis ance be ween he peak densi y hmF2 and he heigh
we e IQC is maximum (he ea e h max) is 105 km o
ou condi ions.
Mo eo e , he diu nal a ia ion o he maximum o
IQC (IQCmax) gene ally p esen s a maximum a nigh
ime (00 and 06 LT) and a minimum a midday and
he seasonal a ia ion eaches he smalles alues du ing
summe . The dis ance be ween hmF2 and he h max in-
c eases wi h he sola ac i i y.
In gene al, i was ound ha IQC eaches i s la ges
alues a high sola ac i i y (1969, 1981, and 1999) a
heigh s be ween 120 and 220 km. In ela ion o he sea-
sonal a iabili y a fixed local ime, highes alues co e-
spond o au umn and win e o 00 and 18 LT.
Fig. 6. IQC alues a San Juan o 1971 (middle sola ac i i y) and 1981 (high sola ac i i y), diffe en season and hou .
5. Conclusions
The esul s o ou s udy could be summa ized as
ollows:

o
F2 a iabili y is he dominan ac o ha de e -
mines he a iabili y o he whole elec on densi y
p ofile.
The highes ela i e a iabili y, exp essed by he
in e qua ile diffe ence di ided by median alue, is
loca ed a ound 105 km below hmF2, inc easing when
he sola ac i i y inc eases.
The diu nal a ia ion o he highes ela i e a iabil-
i y gene ally exhibi s a minimum a ound noon wi h
peaks a midnigh (00 LT) and dawn (06 LT).
The seasonal a ia ion o he ela i e a iabili y p e-
sen s a maximum a au umn and win e o 00 and 18
LT.
The ela i e a iabili y is maximum a high sola
ac i i y and i is loca ed be ween 120 and 220 km.
Acknowledgemen s
The au ho s a e g a e ul o he ionosphe ic g oup a
El A enosillo INTA (Huel a, Spain) o p o iding da a
used in his s udy. One o he au ho s (G.M.A.) also
wan s o hank o he Ma ie Cu ie Fellowship o he
Eu opean Communi y Fi h F amewo k P og amme
unde Con ac Numbe HPMF-CI-2002-01867 o sup-
po ing his esea ch.
Re e ences
Adeniyi, J.O., Radicella, S.M. Va iabili y in
o
F2 a an Equa o ial
S a ion and he Influence o Magne ic Ac i i y, in: P oceedings o
he IRI Task Fo ce Ac i i y 2002, Abdus Salam In e na ional
Cen e o Theo e ical Physics, Repo IC/IR/2003/3, T ies e, I aly,
pp. 27–38, 2003.
Alazo, K., Lazo, B., Rod ı
´guez, M., Calzadilla, A.
o
F2 Va iabili y
o e Ha ana, in: P oceedings o he IRI Task Fo ce Ac i i y 2002,
Abdus Salam In e na ional Cen e o Theo e ical Physics, Repo
IC/IR/2003/3, T ies e, I aly, pp. 1–8, 2003.
Ezque , R.G., Mose , M., Radicella, S.M., Jadu , C.A. S udy o he
elec on densi y a iabili y a fixed heigh s o e San Juan and
Tucuman. Ad . Space Res. 29 (6), 993–997, 2002.
Huang, X., Reinisch, B.W. Ve ical elec on densi y p ofiles om
digisonde ionog ams. The a e aged ep esen a i e p ofile-ARP,
COST 238/PRIME, in: B adley, P.A., de la Mo ena, B.A., (Eds.),
Wo kshop on De elopmen an Tes ing o an Elec on Densi y
Heigh P ofile Model o PRIME, pp. 73–86, 1995.
Lazo, B., Alazo, K., Rod ı
´guez, M., Calzadilla, A. hmF2 Va iabili y
o e Ha ana, in: P oceedings o he IRI Task Fo ce Ac i i y 2002,
Abdus Salam In e na ional Cen e o Theo e ical Physics, Repo
IC/IR/2003/3, T ies e, I aly, pp. 9–16, 2003.
Mose de Gonzalez, M., Radicella, S.M. S udy o ionosphe ic
a iabili y a fixed heigh s using da a om sou h Ame ica. Ad .
Space Res. 15, 61–65, 1995.
Mose , M., Ezque , R., Mi o, G., Co bella, R., De la Ze da, L. A
P elimina y Analysis o he Va iabili y o Ionosphe ic Cha ac e -
is ics om lonosonde Da a, in: P oceedings o he IRI Task Fo ce
Ac i i y 2002, Abdus Salam In e na ional Cen e o Theo e ical
Physics, Repo IC/IR/2003/3, T ies e, I aly, pp. 17–26, 2003.
Radicella, S.M., Zhang, M.L., Jodogne, J.C. Va iabili y o elec on
densi y a fixed heigh s using digisonde da a, Scien ific Repo
COST Documen COST238TD(94)010, pp. 121–128, 1994.
Sole
´, J.G., Al adill, D. Ionosphe ic a iabili y a fixed heigh s, in:
P oceedings o he Join COST 251/IRI Wo kshop and Wo king
G oup Sessions, Ins i u e o A mosphe ic Physics, Kuhlungsbo n,
Ge many, 27–30 May, pp. 109–116, 1997.