Assessing ambien empe a u e
measu emen s in oad cycling aces
Ai o Pinedo-Jau egi
1,2,3
and Ande Roma a e
4,5
Abs ac
The impac o hea s ess on spo s pe o mance is a highly ele an opic. En i onmen al ac o s, such as empe a u e
and humidi y, play a significan ole in de e mining a hle ic pe o mance. Cycling compu e s a e commonly used in oad
cycling o cap u e a ious da a, including empe a u e. Howe e , i is c ucial o emphasize he impo ance o accu a ely
measu ing ambien empe a u e. This s udy aimed o assess he eliabili y and alidi y o he Ga min Edge 830 cycling
compu e s in measu ing ambien empe a u e du ing oad cycling compe i ions. We analyzed empe a u e da a collec ed
by 12 di e en cycling compu e s. We ocused on assessing hese de ices’ eliabili y and alidi y compa ed o measu e-
men s ob ained om an o ficial me eo ological s a ion. The in e class co ela ion coe ficien be ween he de ices was
0.468 (95% CI: 0.254 o 0.626; p < 0.001). The empe a u e eadings exhibi ed a significan a iabili y ac oss he di e en
de ices. Fu he mo e, when compa ing he da a wi h he eadings om he me eo ological s a ion, he e was no ag ee-
men , as indica ed by a Lin’s Conco dance Coe ficien (LCC) o 0.398 and a Mean Absolu e Pe cen E o (MAPE) o
8.87%. This esea ch aises conce ns ega ding he eliabili y and alidi y o cycling compu e de ices o measu ing ambi-
en empe a u e.
Keywo ds
Hea , me eo ological s a ion, mic oclima e, echnological de ice, wea he
In oduc ion
The impac o hea s ess on spo s pe o mance has
ecei ed inc easing a en ion.
1
En i onmen al ac o s such
as empe a u e and humidi y ha e a subs an ial impac on a h-
le ic pe o mance, leading o a g owing body o esea ch in
his field.
2–5
Among hese ac o s, empe a u e is pa icula ly
impo an and can p esen significan challenges o a hle es.
6
In endu ance spo s like cycling, empe a u e egula ion is c i -
ical as i di ec ly a ec s me abolic e ficiency, ca dio ascula
s ain, and he body’s abili y o main ain op imal pe o mance
le els du ing p olonged e o s.
6–8
Cu en ly, o measu e he e ec s o empe a u e on pe -
o mance, he we bulb globe empe a u e (WBGT) is used
o quan i y he mal s ess.
6,9
Howe e , when WBGT mea-
su emen s a e una ailable, ambien empe a u e and humid-
i y a e used as al e na i e indica o s.
9
In many cases,
coaches only ha e empe a u e da a and use his alue as
a e e ence o hea s ain. Fo example, coaches can
p edic dehyd a ion based on he ambien empe a u e.
6
In oad cycling, cycling compu e s a e expec ed o
eco d da a du ing aces and aining sessions o quan i y
in e nal and ex e nal load. These cycling compu e s a e
equipped wi h empe a u e senso s ha allow he eco ding
o empe a u e da a h oughou he en i e ou e.
Consequen ly, some s udies ha e used hese da a o
assess he impac on spo s pe o mance.
10
The au ho s sug-
ges ed ha he powe alues o he cyclis s changed
Re iewe : And ew G unds ein (Uni e si y o Geo gia, USA)
1
Socie y, Spo s, and Exe cise Resea ch G oup (GIKAFIT), Depa men o
Physical Educa ion and Spo , Facul y o Educa ion and Spo , Uni e si y o
he Basque Coun y UPV-EHU, Vi o ia-Gas eiz, Spain
2
AKTIBOki: Resea ch G oup in Physical Ac i i y, Physical Exe cise and
Spo , Depa men o Physical Educa ion and Spo , Facul y o Educa ion
and Spo , Uni e si y o he Basque Coun y (UPV/EHU), Vi o ia-Gas eiz,
Spain
3
Bioa aba, Physical Ac i i y, Exe cise and Heal h G oup, Vi o ia-Gas eiz,
Spain
4
Depa men o Physical Educa ion and Spo , Facul y o Educa ion and
Spo , Uni e si y o he Basque Coun y UPV-EHU, Vi o ia-Gas eiz, Spain
5
Emen4spo , Pe o mance, Heal h, and Spo , Leioa, Spain
Co esponding au ho :
Ai o Pinedo-Jau egi, Depa men o Physical Educa ion and Spo , Facul y
o Educa ion and Spo , Uni e si y o he Basque Coun y (UPV/EHU),
Po al de Lasa e, 71 - 01007 Vi o ia-Gas eiz, Spain.
Email: ai o [email p o ec ed]
O iginal esea ch
In e na ional Jou nal o Spo s Science
& Coaching
2025, Vol. 20(2) 742–747
© The Au ho (s) 2024
A icle euse guidelines:
sagepub.com/jou nals-pe missions
DOI: 10.1177/17479541241308504
jou nals.sagepub.com/home/spo
depending on he ambien empe a u e and concluded ha
he op imal empe a u e ange is be ween 10°C and 25°C.
Howe e , i mus be aken in o accoun ha he cycle
compu e s a e placed on he handleba s o he bicycle.
This placemen is no he mos sui able o measu ing
ambien empe a u e because he senso is exposed o
di ec sola adia ion, which may a ec he measu emen .
In addi ion, o example, he echnical specifica ions p o-
ided o Ga min Edge de ices lack essen ial de ails such
as accu acy da a, sampling a e, and esponse ime o em-
pe a u e senso s. As a esul , i is di ficul o assess he eli-
abili y o he senso igh om he s a .
In p e ious s udies o ma a hon unning, esea che s
ha e disco e ed he p esence o mic oclima es along he
compe i ion cou se.
11
While no specific esea ch has been
conduc ed on mic oclima es in cycling, i is easonable o
specula e ha simila condi ions could exis in cycling
e en s due o hei simila i ies wi h unning. Accu a ely
measu ing he ambien empe a u e is essen ial o es ima -
ing he he mal s ess expe ienced by cyclis s. To ensu e
he eliabili y and alidi y o empe a u e measu emen s,
i is impo an o conside bo h concep s. Reliabili y
ensu es consis en measu emen s, while alidi y confi ms
he accu acy o he de ice’s eadings. Bo h ac o s a e
c ucial when elying on de ice measu emen s, pa icula ly
in he case o ambien empe a u e.
Mo eo e , i is impo an o conside ha a s anda dized
p o ocol mus be ollowed when measu ing ambien em-
pe a u e. I is ecommended o conduc empe a u e mea-
su emen s in shaded a eas, away om di ec sola
adia ion, as i has been obse ed ha sunligh can dis o
empe a u e eadings.
12,13
The e o e, accu a e empe a u e measu emen s a e
essen ial o ob aining unbiased and eliable da a. The eli-
abili y and alidi y o measu ing ambien empe a u e
using cycling compu e s moun ed on bike handleba s
ha e ye o be es ablished. Ou main hypo hesis is ha
cycling compu e s yield biased ambien empe a u e ead-
ings. Hence, he objec i e o his s udy was o analyze he
eliabili y and alidi y o he Ga min EDGE 830 cycling
compu e s in measu ing ambien empe a u e du ing oad
cycling compe i ions. I was hypo hesised ha he Ga min
EDGE 830 cycling compu e s a e use ul o measu ing
ambien empe a u e.
Me hods
The da a we e ob ained om 12 Ga min EDGE 830
(Ga min In e na ional Inc, Ola he, KS) de ices used in
he unde -23 (U23) oad cycling ace in Amu io (A aba,
Basque Coun y, Spain). The oad ace ook place on a
ci cui a ound he illage. The esea che s collec ed he
“.fi ”file, which was eco ded using a cycling compu e
Ga min EDGE 830. All he de ices we e moun ed in
hei s anda d loca ion on he bike handleba . All he da a
we e eco ded h oughou he ace. The empe a u e was
eco ded wi h an in eg a ed senso in a cycling compu e .
We analyzed he empe a u e da a o he 2-h du a ion o
he ace. This esea ch ollows he e hical p inciples es ab-
lished by he Decla a ion o Helsinki and i s subsequen
e isions. As he Uni e si y o he Basque Coun y e hics
commi ee indica ed, e hical app o al is unnecessa y o
using ambien empe a u e da a.
The o ficial me eo ological s a ion (Euskalme , Basque
Coun y) was used as a e e ence me hod. This s a ion p o-
ided empe a u e da a a 10-min in e als. The a hes
poin o he ci cui and he me eo ological s a ion we e
app oxima ely 5800 me e s apa . The al i ude o he
me eo ological s a ion was 230 m, and he ange al i ude
o he ci cui was be ween 182 m and 352 m. On he ace
day, he sky was clea , and he empe a u e anged
be ween 21.6°C and 24.6°C. In o de o synch onize bo h
da ase s, we ex ac ed he empe a u e om he de ice
e e y 10 min, esul ing in a o al o 12 ime poin s.
S a is ical analysis
Da a analysis was pe o med using he R p og amming lan-
guage. The no mali y o he da a was assessed using he
Shapi o-Wilks es . Means, s anda d de ia ions (SDs), and
coe ficien s o a ia ion (CVs) we e calcula ed using s and-
a d s a is ical me hods.
Rela i e eliabili y was assessed using in aclass co el-
a ion coe ficien s (ICCs) wi h 95% CIs. ICC ( wo-way
mixed model, single measu es, absolu e ag eemen ) was
selec ed based on guidelines published elsewhe e.
14
One-way ANOVA was conduc ed, ollowed by a
Bon e oni pos- es o examine pai wise di e ences. The
e ec size (ES) o one-way ANOVA e ec s was de e -
mined using pa ial e a squa ed (ηp2).
Cons uc alidi y was assessed using independen
S uden ’s T- es o compa e mean di e ences be ween he
wo me hods. The di e ences’magni ude was e alua ed
using a 95% Confidence In e al (CI) and Cohen’s e ec
size (ES). Ag eemen was assessed using Lin’s
Conco dance Coe ficien (LCC), Mean Absolu e Pe cen
E o (MAPE) and Bland–Al man plo s wi h limi s o
ag eemen (LoA). Homoscedas ici y was assessed by isu-
alizing Bland–Al man plo s. Significance was se a p <
0.05.
Resul s
Reliabili y
The cycling compu e de ices eco ded a empe a u e ange
be ween 18°C and 33°C. Figu e 1 shows he a e age o all
de ices a each ime poin . (Figu e 1).
ICC was 0.468 (95% CI: 0.254 o 0.626; p < 0.001).
Significan di e ences we e obse ed be ween de ices (p
Pinedo-Jau egi and Roma a e 743
< 0.05; ηp2 =0.464). All pai wise combina ions we e sig-
nifican excep o wo pai wise compa isons. The mean
SD o all eco ded ime poin s was 2.14 °C, anging
om 1.66 °C o 3.08 °C. The mean CV was 9.09%,
anging om 6.80% o 11.60%. The mean ange be ween
he minimum and maximum eco ded empe a u es a
each ime poin was 5.7 ±0.86°C (min =4°C; max =9°
C). Table 1 p esen s he desc ip i e da a o 12 ime poin s
eco ded by he de ice.
Validi y
We obse ed significan mean di e ences be ween he
cycling compu e s and he me eo ological s a ion (p =
0.006, ES =0.255). The LCC was 0.398 (95% CI: 0.32 o
0.472), and MAPE was 8.87%. The ag eemen can be
obse ed in Figu e 2 and Figu e 3. The mean bias was
0.55 ±2.36°C (Lowe LoA =-4.07; Uppe LoA =5.18).
On Figu e 3
Discussion
Accu a ely measu ing ambien empe a u e is essen ial o
es ima ing he a hle e’s pe o mance in hea si ua ions.
6,10
This s udy ep esen s he fi s analysis o he eliabili y and
alidi y o ambien empe a u e eco dingsusing cycling com-
pu e s. The findings e ealed ha he cu en eco ding
me hod using he Ga min EDGE 830 is inaccu a e.
Figu e 1. Tempe a u e da a om cycling compu e s du ing wo-hou ace ime.
No e: Mean: mean o all cycling compu e s pe second. Max: maximum empe a u e egis a ion om cycling compu e s. Min:
minimum empe a u e egis a ion om cycling compu e s. Range: Di e ence be ween maximum and minimum empe a u e
egis a ion. SD: s anda d de ia ion o all cycling compu e s.
744 In e na ional Jou nal o Spo s Science & Coaching 20(2)
Ou findings e ealed ha hese cycling compu e s
eco ded a wide ange o empe a u es a he same ime
poin , indica ing a low ICC and significan di e ences
be ween a iables. This lack o eliabili y p esen s a chal-
lenge when compa ing da a among cycle compu e s.
Gi en ha cycling pe o mance is highly sensi i e o en i -
onmen al condi ions, pa icula ly empe a u e, e ec i e
he mo egula ion s a egies a e essen ial o op imizing
endu ance and p e en ing hea - ela ed impai men s.
6
S udies ha e shown ha e en mode a e inc eases in
ambien empe a u e can exace ba e ca dio ascula s ain,
educe muscle blood flow, and accele a e he onse o
a igue.
7,15
Thus, wi h a alid de ice o measu ing
ambien empe a u e, coaches could make decisions based
on ambien empe a u e.
Despi e pa icipa ing in he same ace, i we look a he
empe a u e da a, i may seem ha each cyclis pa icipa ed
in a di e en ace. Fo ins ance, in a s udy conduc ed by
Table 1. Desc ip i e da a o 12 ime poin s om cycling compu e s.
Mean (°C) SD (°C) CV (%) MIN (°C) MAX (°C)
1 22.08 2.11 9.55 19.00 25.00
2 21.00 2.22 10.55 18.00 24.00
3 21.08 1.98 9.37 19.00 24.00
4 22.42 2.07 9.21 20.00 25.00
5 22.25 2.34 10.52 19.00 26.00
6 23.25 2.09 9.01 21.00 26.00
7 24.92 2.19 8.80 22.00 28.00
8 23.50 2.15 9.16 21.00 26.00
9 25.42 1.93 7.59 23.00 29.00
10 25.17 2.04 8.10 22.00 28.00
11 25.58 2.11 8.24 23.00 28.00
12 28.08 2.11 7.51 25.00 31.00
No e: Mean: mean o all cycling compu e s a each poin ; Max: maximum empe a u e egis a ion om cycling compu e s; Min: minimum empe a u e
egis a ion om cycling compu e s; SD: s anda d de ia ion; CV: coe ficien o a ia ion.
Figu e 2. Bland-Al man analysis o compa e he ag eemen be ween cycling compu e s and he o ficial me eo ological s a ion.
Pinedo-Jau egi and Roma a e 745
Valenzuela e al. 2022
10
cycling compu e s we e used o
examine he impac o ambien empe a u e on powe
ou pu . Howe e , based on ou s udy, he esul s seem o
indica e ha hese de ices may no be alid.
The esul s show lack o alidi y, as e idenced by low
LCC and a high MAPE. Rega ding ag eemen , al hough
he mean bias is low, he LoAs a e conside able. The em-
pe a u e accu acy is se a 0.1°C below ou mean bias.
12
Mo eo e , isual analysis e ealed he p esence o he e o-
scedas ici y, as he mean bias inc eases wi h highe
ambien empe a u es. The disc epancy be ween he
me eo ological s a ion and cycling compu e s may be a ib-
u ed o ex e nal ambien a iables ha could influence he
measu emen s. The lack o accu acy could be a ibu ed
o di ec sola adia ion, as obse ed in o he s udies.
12
A
solu ion o he adia ion issue could in ol e eloca ing he
empe a u e senso on he bike, o example, unde nea h
he saddle. Fu he mo e, he disc epancy be ween he
wea he s a ion and he EDGE de ices could be caused
by local mic oclima es. Simila o he ma a hon scena io,
11
he e may be mic oclima es along he cou se ha make i
challenging o compa e wi h he wea he s a ion.
This lack o ag eemen p esen s challenges in unde -
s anding he influence o empe a u e on pe o mance and
hinde s da a compa ison. None heless, using cycling com-
pu e s o empe a u e measu emen emains p e alen
among scien is s and aine s.
10
Howe e , hese cycling
compu e s seem o no be alid o quan i ying WBGT
du ing a ace. I is c ucial o conside ha empe a u e
accoun s only o 10% o he WBGT equa ion.
13
As high-
ligh ed by Lemke and Kjells om 2012,
13
a one-deg ee
change in ai empe a u e oughly co esponds o a one-
deg ee change in WBGT. Ano he cu en me hod o
assessing hea s ain in ol es conside ing empe a u e and
humidi y.
9
Ne e heless, hese cycling compu e s a e no
sui able ye o accu a ely assessing hea s ain.
Ou s udy ound ha empe a u e senso s on cycle
compu e s, o hei placemen , may encoun e issues
when calcula ing ambien empe a u e. One p oposed
solu ion o his p oblem is o a ious companies o
p oduce a small, po able wea he s a ion ha can be
placed in an op imal loca ion on he bicycle. This idea
has al eady been sugges ed o measu ing he mal
s ess using ex e nal body empe a u e senso s.
16
The
au ho s emphasize he impo ance o combining bo h
measu emen s o p edic an a hle e’s he mal s ess
based on ambien empe a u e.
The main limi a ion o his s udy is he “black box
e ec ”, whe e he e is a lack o knowledge ega ding he
specific mechanisms by which he Ga min EDGE 830
cycling compu e s eco d empe a u e. Unde s anding his
p ocess is c ucial o comp ehending he obse ed lack o
accu acy. Fo ins ance, while we obse ed a 1°C accu acy
in he de ice empe a u e ou pu , we do no ha e in o ma-
ion ega ding he sampling equency o he ype o senso
used.
Figu e 3. The cycle compu e s and he me eo ological s a ion empe a u e eco ded a 12 di e en ime poin s.
746 In e na ional Jou nal o Spo s Science & Coaching 20(2)
Ano he limi a ion is he ange o ac ual empe a u es
measu ed du ing he ace. The p esence o he e oscedas i-
ci y highligh s he impo ance o analyzing accu acy
unde lowe and highe empe a u e condi ions.
Addi ionally, since he o ficial me eo ological s a ion
epo ed da a e e y 10 min, inc easing he sampling e-
quency may lead o sligh ly di e en ou comes. Finally,
he esea ch was conduc ed only o one ace and 12
de ices. While his app oach has acili a ed he compa ison
wi h he same me eo ological s a ion, i is wo h no ing ha
he ci cui ’s cha ac e is ics could impac he measu emen s.
To ackle his issue, conside ing he po en ial in alidi y o
he senso s, i is ad isable o u u e esea ch o adop he
me hodology u ilized by Kosaka e al. 2018
11
which com-
pa ed he mobile wea he s a ion wi h he cyclo-swi ches.
Mo eo e , i would be beneficial o include a wide a ay
o de ices om di e en manu ac u e s o enhance he
sample size o de ice. Las ly, he sample size was
limi ed, and despi e e o s o each ou o mo e de ices,
i was no easible. As a esul , he an icipa ed s a is ical
powe was no a ained. None heless, he findings demon-
s a e he beha iou o hese de ices, which should be
checked in u u e esea ch.
Conclusions
In conclusion, his esea ch aises conce ns ega ding he
eliabili y and alidi y o he Ga min EDGE 830 cycling
compu e de ices o measu ing ambien empe a u e.
Coaches and scien is s should exe cise cau ion when u iliz-
ing da a om hese de ices, conside ing he obse ed dis-
c epancies wi h he o ficial me eo ological s a ion. Fu u e
esea ch should cla i y whe he he esul s ound in his
s udy a e gene alizable o o he models and b ands o
cycling compu e de ices. Also, u u e esea ch should p i-
o i ize iden i ying op imal loca ions o bicycle empe a u e
senso s o enhance da a accu acy and alidi y by compa ing
wi h mobile wea he s a ions.
Acknowledgmen s
The au ho s wan o exp ess hei g a i ude o he cycling eams o
sha ing he da a and o D Txus o his aluable ad ice.
Decla a ion o conflic ing in e es s
The au ho s decla ed no po en ial conflic s o in e es wi h espec
o he esea ch, au ho ship, and/o publica ion o his a icle.
E hical app o al
As he Uni e si y o he Basque Coun y e hics commi ee indi-
ca ed, e hical app o al is unnecessa y.
Funding
The au ho s ecei ed no financial suppo o he esea ch, au ho -
ship, and/o publica ion o his a icle.
ORCID iD
Ai o Pinedo-Jau egi h ps://o cid.o g/0000-0002-7939-297X
Re e ences
1. Kenney WL, Ha eni h G and Jay O. The mal physiology,
mo e ele an han e e be o e. J Appl Physiol (1985).
2022; 133: 676–678.
2. Cha lo K, Fau e C and An oine-Jon ille S. Influence o ho
and cold en i onmen s on he egula ion o ene gy balance
ollowing a single exe cise session: A mini- e iew. Nu ien s
2017; 9: 592.
3. Ruíz-Mo eno C, Gu ié ez-Hellín J, González-Ga cía J, e al.
E ec o ambien empe a u e on a oxida ion du ing an
inc emen al cycling exe cise es . Eu J Spo Sci 2021; 21:
1140–1147.
4. Gagnon DD, Pe ie L, Do man SC, e al. Ambien empe a-
u e influences me abolic subs a e oxida ion cu es du ing
unning and cycling in heal hy men. Eu J Spo Sci 2020;
20: 90–99.
5. Pé ia d JD, Racinais S and Sawka MN. Adap a ions and
mechanisms o human hea acclima ion: applica ions o com-
pe i i e a hle es and spo s. Scand J Med Sci Spo s 2015; 25:
20–38.
6. Pé ia d JD, Eijs ogels TMH and Daanen HAM. Exe cise
unde hea s ess: he mo egula ion, hyd a ion, pe o mance
implica ions, and mi iga ion s a egies. Physiol Re 2021;
101: 1873–1979.
7. Nybo L, Rasmussen P and Sawka MN. Pe o mance in he
hea -physiological ac o s o impo ance o hype he mia-
induced a igue. Comp Physiol 2014; 4: 657–689.
8. Cheung SS and Slei e GG. Mul iple igge s o hype he -
mic a igue and exhaus ion. Exe c Spo Sci Re 2004; 32:
100–106.
9. A ms ong LE, Casa DJ, Milla d-S a o d M, e al. Exe ional
Hea Illness du ing T aining and Compe i ion. Med Sci Spo s
Exe cise 2007; 39: 556–572.
10. Valenzuela PL, Ma eo-Ma ch M, Zabala M, e al. Ambien
empe a u e and field-based cycling pe o mance: insigh s
om male and emale p o essional cyclis s. In J Spo s
Physiol Pe o m 2022; 17: 1025–1029.
11. Kosaka E, Iida A, Vanos J, e al. Mic oclima e a ia ion and
es ima ed hea s ess o unne s in he 2020 Tokyo Olympic
ma a hon. A mosphe e (Basel) 2018; 9: 192.
12. Yang J, Liu Q, Chen G, e al. Op imized design and expe i-
men al alida ion o a empe a u e senso o su ace ai em-
pe a u e obse a ion. Sens Ac ua o s, A 2021; 323: 112646.
13. Lemke B and Kjells om T. Calcula ing wo kplace WBGT
om me eo ological da a: a ool o clima e change assess-
men . Ind Heal h 2012; 50: 267–278.
14. Koo TK and Li MY. A guideline o selec ing and epo ing
in aclass co ela ion coe ficien s o eliabili y esea ch. J
Chi op Med 2016; 15: 155–163.
15. Sawka MN, Leon LR, Mon ain SJ, e al. In eg a ed physio-
logical mechanisms o exe cise pe o mance, adap a ion,
and maladap a ion o hea s ess. Comp Physiol 2011; 1:
1883–1928.
16. Ve del N, Podloga T, Ciuha U, e al. Reliabili y and alidi y
o he CORE senso o assess co e body empe a u e du ing
cycling exe cise. Senso s 2021; 21: 5932.
Pinedo-Jau egi and Roma a e 747
