Eur J Appl Physiol (1985) 54:7--11 European Journal of

Applied Physiology and Occupational Physiology �9 Springer-Verlag 1985

The influence of clothing ensembles on the lower critical temperature

Masahiko Sato 1, Shigeki Watanuki 2, Koichi Iwanaga ~, and Fumiko Shinozaki 3

1 Department of Ergonomics, Kyushu University of Design Sciences, 4-9-1 Shiobaru, Minami-ku, Fukuoka, 815 2 Department of Science of Clothing System, Faculty of Science of Living, Osaka City University, Sugimoto, Sumiyoshi-ku, Osaka 3 Department of Ergonomics, Koran College, Yokote, Minami-ku, Fukuoka, Japan

Summary. This paper describes the effect of clothing insulation on the lower critical tempera- ture (LCT). Twelve young adult females were ex- posed to a temperature of 10~ for 2~ h. LCT was estimated at five different clothing conditions according to the intersect method. The total weights of clothing ensembles (CW) of the five conditions were 0.45, 0.52, 0.82, 1.34, and 2.56 kg, and their thermal insulating values (/) were 0.44, 0.60, 0.77, 1.21, and 2.14 clo, respectively. LCT of the five clothing ensembles were estimated to be 26.4, 25.4, 23.5, 21.5, and 17.5~ respectively. The regression equation of the logarithm of LCT on CW was calculated as logLCT=l.4469-- 0.08283 CW and that on I was logLCT= 1.4613-- 0.10526 I, respectively. The rate of changes in LCT is suggested to be dependent on the clothing conditions as for the following equations: dLCT/ d C W = - 5 . 3 4 exp(-0.1907 CW) or dLCT/ dI= - 7.01 e x p ( - 0.2424/).

Key words: Critical temperature -- Insulating ef- fect of clothing -- Clothing weight -- Clo value -- Cold tolerance

Introduction

The lower critical temperature (LCT) has been re- garded as a sensitive index for estimating cold ad- aptability of homeothermia, since Scholander et al. (1950b) suggested it to be near 27~ in most tropical species and with a range from - 5 0 ~ to 15~ in arctic species. They placed naked man

Offprint requests to: M. Sato at the above address Mailing address: M. Sato, Department of Ergonomics, Kyushu University of Design Sciences, 4-9-1 Shiobaru, Minami-ku, Fukuoka, 815, Japan

among the more temperature sensitive of tropic animals from such high LCT as around 27~ (Dubois 1936) to 29~ (Winslow and Herrington 1949). In addition, Scholander et al. (1950a) pointed out that thick fur insulation is the main reason for the low LCT of arctic species. There- after, LCT has been determined in several subject groups, mostly dressed in light clothing, and has been found to vary over a wide range from 21.7~ (Ishii 1976) to 27~ (Scholander et al. 1957). The investigation of clothing effect seems to be one of the necessary steps for more progress in research on LCT.

Methods

Twelve young female students, ranging in age from 18 to 20 years, all in good physical condition, volunteered for this study. Their mean and standard deviation of stature were 157.8+4.3 cm and the corresponding values of body weight were 51.1 + 3.9 kg, respectively.

The LCT of each subject was estimated for five different clothing conditions according to the intersect method (Erikson et al. 1956). The subjects quietly assumed a reclining position for 1.5 h at 10~ in a climatic chamber. The relative humidity and air velocity were kept at about 50% and 25 m - rain -1, re- spectively. They reclined with their legs fully extended to max- imize the exposure. Then, they moved to a bicycle ergometer in the same ambient conditions, and sat and pedalled for about 1 h to maintain the rectal temperature (Tr) constant (Erikson et al. 1956; Yoshimura and Yoshimura 1969). Before entering the chamber, all the subjects rested at 28~ and 50% humidity in a recumbent position for more than thirty minutes. The total weights of clothing except shoes (CW) in the five conditions were 0.45, 0.52, 0.82, 1.34, and 2.56 kg, respectively. The thermal insulating values (/) were estimated to be 0.44, 0.60, 0.77, 1.21, and 2.14 clo respectively, according to the for- mula of Sprague and Munson (1974) and the standard of Ne- vins et al. (1974). Every subject was measured in each clothing condition on five separate days in July or August. The order of clothing conditions was randomized between the subjects.

Each exposure was performed with the subject in the postabsorptive state. Expired gas was collected with Douglas

8 M. Sato et al.: Clothing effect on lower critical temperature

bags through low-resistance valves for 3 min at intervals of 15 min during the resting and pedalling exposures. The ventila- tory volume was measured with a wet gas meter and aliquots of mixed expired gas were analyzed for oxygen and carbon dioxide by a mass spectrometer. Tr at a depth of 10 cm was continuously recorded with a thermistor. The mean skin tem- perature ('I's) was calculated from continuous recordings of the skin temperature at the chest, upper arm, thigh, and lower leg, according to the formula of Ramanathan (1964). An addi- tional exposure to 28~ at 0.60 clo was conducted for each subject following the other exposures to obtain control data in a thermoneutral condition.

R e s u l t s

Figure 1 shows the means and standard devia- tions of metabolic rate (MR), Tr, and Ts of the twelve subjects at the final stage of resting expo- sure. The analysis of variance confirmed that the cold exposure significantly influenced MR, Tr, and I"s (p < 0.005) and the influences of CW and I on MR and "Fs during the cold exposure were sig- nificant (p < 0.005).

~'100- I

4o]

40] ~ 3 5 "-~- . . . .

,~ 30

25

...2

20

15

TN 0 .5 1.0 1.5 2,0 2.5 3.0 CW(kg) .44".77 1.21 2,i4 I(r

.60 Fig. 1. Metabolic rate, rectal temperature, and mean skin tem- perature at the final stage of resting exposure to 10~ and lower critical temperature at each clothing condition. TN on the abscissa marks the thermoneutral exposure

The correlation coefficient between MR and Ts at the final stage of resting exposure was calcu- lated at -0.57 (p < 0.01). The regression equation of MR (W. m - 2 ) o n i"s (~ was MR=226.5-5.7 Ts. The standard deviation of the regression coefficient was 1.09, hence the lower and upper borders of 95% confidence inter- vals of the regression coefficient were estimated

to be -7 .9 and -3.6, respectively. The correla- tion coefficient between the changes in metabolic rate (AMR) and those of mean skin temperature (AI"s) was calculated at -0.63 (p <0.01). The re- gression equation of the former on the latter was AMR= - 3.7- 7.7 A~'s. Since the standard devia- tion of the regression coefficient was 1.25, the lower and upper borders of the 95% confidence intervals of the regression coefficient were esti- mated to be -10.2 and -5.2, respectively.

LCT was estimated by using MR during ped- alling the ergometer. Fig. 1 also shows the means and standard deviations of LCT for each clothing condition. The analysis of variance confirmed that the clothing factor significantly influenced LCT (p <0.005). The regression equation of the logarithm of LCT on CW was calculated as logLCT=l.4469-O.08283 CW. The correlation coefficient between logLCT and CW was calcu- lated at 0.845 and was greater than that between LCT and CW. The standard deviation of the re- gression coefficient was 0.00688. The above equa- tion can be rewritten as the following exponential equation

LCT---27.98 exp(-0.1907 CW). (1)

And the equations obtained from the lower and upper borders of the 95% confidence intervals of the regression coefficient were written as

(.9 ..3

40

30

20

10 r

0

\

X \

\

1

1 2 3

CW(kg) Fig. 2. The regression curve of L C T on total weights of cloth- ing ensembles and its 95% confidence intervals for the mean and individual values

M. Sato et al. : Clothing effect on lower critical temperature

LCT=29.01 exp(-0.2225 CW) (2)

and

LCT=26.99 exp(-0.1589 CW) (3),

respectively. Fig. 2 shows the regression curve of Eq. (1) and its 95% confidence limits for the mean and individual values.

As for the rate of change in L C T with increase in CW, it follows from Eq. (1) that

dLCT

dCW - - - 5.34 exp(-0.1907 CW).

From Eq. (2) and (3), the following derived func- dLCT

tions were obtained: d C W - - 6.45 exp(-0.2225

dLCT CW) and d C ~ - 4.29 exp(-0.1589 CW), re-

spectively. Thus the rate depends on CW (Fig. 3).

0 L-

-2 'l:J1

d - 3

5 - 4 (.9 "O

- 5 ._1 "O

-6

-7

CW(kg) 1 2 3 ! ! I

/ /

/

Fig. 3. Changing rate of L C T with changes in CW. Two thin lines show the rates calculated from the lower and upper bord- ers of 95% confidence intervals of the regression coefficient of L C T on CW

The regression equation of the logarithm of L C T on I was calculated as l ogLCT= 1.4613- 0.10562 I. The correlation coefficient between l ogLC T and I was calculated at 0.846 and larger than that between L C T and I. The standard devia- tion of the regression coefficient was 0.00870. The above equation can be rewritten as the following exponential equation

LCT = 28.93 e x p ( - 0.2424 I). (4)

The equations obtained from the lower and upper limits of the 95% confidence levels of the regres- sion coefficient were written as

L C T = 30.15 �9 exp(-0.2825 I) (5)

and

LCT = 27.76 e x p ( - 0.2023 I) (6),

respectively. Fig. 4 shows the regression curve of Eq. (4) and its 95% confidence intervals for the mean and individual values.

? v

_.1

40

30

20

\ \

\ \

\

\ \

10 |

0 I 2 3 I (clo)

Fig. 4. The regression curve of L C T on clo values and its 95% confidence intervals for the mean and individual values

As for the rate of changes in L C T with in- creases in I, it follows from Eq. (4) that

dLCT

dI - - - -7 .01 exp(-0 .2424 I).

From Eq. (5) and (6), the following derived func- dLCT

tions were obtained: dI - 8.52 e x p ( - 0.2825

dLCT I) and dI - 5.61 exp(-0.2023 I), respec-

tively (Fig. 5).

10

-3

'O -~ -4 U

.--- - 5 "O F- U -6 ..J -o

-7

-8

-9

I (clo) 0 1 2 3 I I I ,~

j / " f

////// //I/

/ /

/ /

Fig. 5. Changing rate of LCT with changes in I. Two thin lines show the rates calculated from the lower and upper borders of 95% confidence intervals of the regression coefficient of LCT on I

M. Sato et al.: Clothing effect on lower critical temperature

MR on Ts and that o f A M R onATs (Ogata 1973; Ishii 1976; Sato et al. 1984).

Seppanen et al. (1972) and Mihira (1977) have studied the relationship between C W and I. In the present study, the regression equation of I on C W was calculated to be 1=0.137+0.787 CW. Al- though this relationship is approximately interme- diate between the results of Seppanen et al. (1972) and of Mihira (1977), the indirect estimation of I might have brought about some error in the pres- ent study. Furthermore, the insulation of clothing while bicycling has been reported to be 0.06 clo different from that when sitting (Olesen et al. 1982). This kind of postural influence on I might have led to another error in the present measure- ment of LCT. The present results should be con- sidered with the restriction of the above approxi- mations of clo values.

Acknowledgements. The authors express their sincere thanks to our twelve subjects and to the members of the Department of Ergonomics, Kyushu University of Design Sciences for their important assistance in this study.

Discussion

Erikson et al. (1956) found the L C T of a male dressed in winter sports clothing to be 14 ~ C. The present authors observed a lowering in L C T with heavier ensembles of garments in a preliminary research (Sato et al. 1984). The present study may be one of the few studies on the relationship be- tween L C T and C W or I.

If LCT at 0 kg or 0 clo could be extrapolated along Eq. (1) or Eq. (4), the mean value and its lower and upper borders of 95% confidence inter- vals would be 28.0, 26.8, 29.2 (~ or 28.9, 27.6, 30.4 (~ respectively. These values correspond well with those previously estimated in naked man by Dubois (1936) or Winslow and Herring- ton (1949).

Erikson et al. (1956) tested five male Cauca- sians with shorts and shoes and concluded that LCT in white men could lie between 25~ and 27 o C, as in tropical mammals. As for the LCT of male Japanese in light dress, early studies have determined it as 24~ (Yoshimura and Yoshi- mura 1969), 21.7~ (Ishii 1976), or 26.2~ (Sato et al. 1979). Such large differences in LCT might derive from differences in C W or I. If similar dif- ferences in L C T are to be expected among female Japanese, the present subjects would pertain to a group with insufficient cold tolerance as Japa- nese, judging from the regression coefficient of

References

Dubois EF (1936) Basal metabolism in health and disease. Philadelphia. [quoted from Scholander et al. 1950b]

Erikson H, Krog J, Andersen KL, Scholander PF (1956) The critical temperature in naked man. Acta Physiol Scand 37:35--39

Ishii M (1976) Cold tolerance of Japanese assessed by the lower critical temperature, and so on. J Anthrop Soc Nippon 84:93--104

Mihira K (1977) Measurement of thermal resistance of gar- ments using thermal manikins. Proc Symp on Man-Ther- mal Envir System 1:15--18

Nevins RG, McNall Jr PE, Stolwijk JAJ (1974) How to be comfortable at 65 to 68 degrees. ASHRAE J 16:41--43

Ogata K (1973) A consideration on effect of training through the responses of cold exposure. Jpn J Phys Fit Sport Med 22:125--131

Olesen BW, Sliwinksa E, Madsen TL, Fanger PO (1982) Effect of body posture and activity on the thermal insulation of clothing: Measurements by a movable thermal manikin. ASHRAE Trans 88:791--799

Ramanathan NL (1964) A new weighting system for mean sur- face temperature of the human body. J Appl Physiol 19:531--533

Sato M, Katsuura K, Yasukouchi A (1979) The lower and up- per critical temperatures in male Japanese. J Hum Ergol 8:145--153

Sato M, Yamasaki K, Watanuki S, Shinozaki F (1984) The in- fluence of clothings on the lower critical temperature. J Anthrop Soc Nippon 92:163--169

Scholander PF, Andersen KL, Krog J, Lorentzen FVI Steen J (1957) Critical temperature in Lapps. J Appl Physiol 10:231--234

M. Sato et al.: Clothing effect on lower critical temperature 11

Scholander PF, Hock R, Walters V, Irving L (1950a) Adapta- tion to cold in arctic and tropical mammals and birds in re- lation to body temperature, insulation, and basal metabolic rate. Biol Bull 99:259--271

Scholander PF, Hock R, Walters V, Johnson F, Irving L (1950b) Heat regulation in some arctic and tropic mam- mals and birds. Biol Bull 99:237--258

Seppanen O, McNall PE, Munson DM, Sprague CH (1972) Thermal insulating values for typical indoor clothing en- sembles. ASHRAE Trans 78:120--130

Sprague CH, Munson DM (1974) A composite ensemble method for estimating thermal insulating values of cloth- ing. ASHRAE Trans 80:120-- 129

Winslow CEA, Herrington LP (1949) Temperature and human life. Princeton. [quoted from Scholander et al. 1950b]

Yoshimura M, Yoshimura H (1969) Cold tolerance and critical temperatures of the Japanese. Int J Biometerol 13:163--172

Accepted December 10, 1984