PUBLISHED ONLINE: 3 AUGUST 2015 | DOI: 10.1038/NCLIMATE2741
Energy consumption in buildings and female
Boris Kingma*and Wouter van Marken Lichtenbelt
Energy consumption of residential buildings and oces adds
up to about 30% of total carbon dioxide emissions; and
occupant behaviour contributes to 80% of the variation in
energy consumption1. Indoor climate regulations are based
on an empirical thermal comfort model that was developed
in the 1960s (ref. 2). Standard values for one of its primary
variables—metabolic rate—are based on an average male,
and may overestimate female metabolic rate by up to 35%
(ref. 3). This may cause buildings to be intrinsically non-
energy-ecient in providing comfort to females. Therefore, we
make a case to use actual metabolic rates. Moreover, with a
biophysical analysis we illustrate the eect of miscalculating
metabolic rate on female thermal demand. The approach
is fundamentally dierent from current empirical thermal
comfort models and builds up predictions from the physical and
physiological constraints, rather than statistical association to
thermal comfort. It provides a substantiation of the thermal
comfort standard on the population level and adds ﬂexibility
to predict thermal demand of subpopulations and individuals.
Ultimately, an accuraterepresentation of thermal demand of all
occupants leads to actual energy consumption predictions and
real energy savings of buildings that are designed and operated
by the buildings services community.
As the built environment is focusing more on design of energy-
eﬃcient buildings (for example, near-zero-energy buildings), we
argue that indoor climate standards should accurately represent
the thermal demand of all occupants. Otherwise there is a great
risk that occupants will adapt their behaviour to optimize personal
comfort, which may in turn nullify the eﬀects of supposed
energy-eﬃcient designs. Furthermore, various fields in commerce,
science and policymaking depend on accurate predictions of
building energy consumption. For instance, commercial incentives
for building renovations premised on energy-saving predictions;
scientific climate change simulations require building energy
consumption predictions to account for warming eﬀects in winter4;
and policymaking for resource management requires integrated
resource assessments including energy consumption by buildings5.
The total variation in building energy consumption that is
explained by occupant behaviour includes operating the thermostat,
windows or air conditioning system1. In general, females prefer a
higher room temperature than males in home and oﬃce situations,
and mean values may diﬀer as much as 3 K (males: 22 ◦C versus
females: 25 ◦C; refs 6,7). Despite this discrepancy in preferred room
temperature, no significant gender eﬀect is found with respect to
the mean skin temperature range that is associated with thermal
comfort (males: 32.8–33.8 ◦C versus females: 32.4–33.6 ◦C; ref. 8).
Indoor thermal environment design is primarily based on
PMV/PPD (predicted mean vote/percentage people dissatisfied)
criteria. The PMV is expressed on the ASHRAE 7-point Thermal
Sensation Scale ranging from cold (−3) to hot (+3). This vote
is linked to thermal discomfort through the PPD (ref. 9). Two
main input variables for the model are metabolic rate and
clothing insulation; however, the accuracy of these variables is
in general poorly defined10,11. Nevertheless, standard reference
values for the metabolic rate and clothing are tabulated and
used worldwide2,12,13. With respect to the metabolic rate, the
metabolic equivalent (MET) is used to express the metabolic
cost of an activity relative to the resting metabolic rate, and its
value (1 MET =4.186 kJ kg−1h−1≈58 W m−2) is set by convention
based on the resting metabolic rate of only one 70kg, 40-year-old
male3. This may have significant consequences because 58 W m−2
may overestimate resting heat production of women up to
35% (ref. 3). Similarly, with increasing age, basal metabolic
rate decreases14. Thus, current indoor climate standards may
intrinsically misrepresent thermal demand of the female and senior
subpopulations10,15. The PMV/PPD model uses the metabolic rate to
calculate the environmental conditions that satisfy thermal balance
between the body and the environment (see Fig. 1, right part: skin
to environment). However, from a biophysical perspective, thermal
balance within the body has to be satisfied as well (see Fig. 1).
Thermal balance within the body is dictated by both metabolism
and the composite thermal insulation provided by tissues (that
is, body composition and skin blood flow). The influence of
thermal insulation is especially relevant in the case of lean versus
obese. The larger insulation provided by adipose tissue results in
greater core-to-skin temperature gradient and a lower mean skin
temperature for obese compared with lean16. Consequently, these
physiological characteristics co-determine the thermal demand
from the environment. The PMV/PPD model was published in the
1970s and at that time biophysical models that incorporate the
influence of tissue insulation were not widely used. However, since
that time several biophysical models of human thermal balance have
been developed17–19. Therefore, the knowledge gained from these
models could be used to enhance the PMV/PPD model.
It has been suggested that thermal balance within the ther-
moneutral zone is a prerequisite of steady-state thermal comfort20.
The thermoneutral zone is defined in physiological terms as the
range of operative temperatures where an organism can maintain its
body temperature without regulatory changes in metabolic rate (for
example, shivering or non-shivering thermogenesis) or sweating21
(see Fig. 2). In relation to thermal comfort this means that opera-
tive temperatures that are thermally comfortable (thermal comfort
zone) coincide with, or at least form a subset of, the temperatures
where the body requires no regulatory metabolic heat production or
sweating to maintain thermal balance (thermoneutral zone)20,22. The
exact positioning of the thermoneutral zone may thus change with
activity, body composition (tissue insulation) and clothing level.
In this study we investigate the thermal state of young adult
females performing light oﬃce work and we use a biophysical
modelling approach to test whether these thermal states fall within
Department of Human Biology, NUTRIM School of Nutrition and Translational Research in Metabolism of Maastricht University Medical Center+,
PO Box 616, 6200 MD Maastricht, The Netherlands. *e-mail: email@example.com
NATURE CLIMATE CHANGE | ADVANCE ONLINE PUBLICATION | www.nature.com/natureclimatechange 1
LETTERS NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE2741
Air temperatureSkin temperature
Tissue insulation Clothing insulation
Body fat mass, muscle
mass and skin blood ﬂow
inﬂuence tissue insulation
Conduction, convection Conduction, convection, radiation, evaporation
Figure 1 | Schematic view of human heat balance. Heat balance from core to skin (within body) and from skin to air (between body and the environment).
The open arrows denote heat ﬂow within the body via conduction through tissues and convection through blood ﬂow; and between the body and
environment via conduction, convection, radiation and evaporation. Body tissue insulation provides resistance to metabolic heat within the body and
determines the temperature gradient between core and skin. Likewise, clothing also provides resistance to body heat loss and co-determines the
temperature gradient between skin and air.
their thermoneutral zone (see Methods and Supplementary Table).
With this analysis we aim to point out the importance of using the
actual metabolic rate, instead of a standard one based only on a male.
Therefore, the biophysical model may provide a constructive way
forward from the empirical thermal comfort standard.
The measured group average metabolic rate for young adult
females while performing light oﬃce work is 48 ±2 W m−2, which is
significantly lower (p<0.01) than the ASHRAE standard values for
metabolic heat production associated with this activity (from resting
seated: 60 W m−2or 20% overestimation, to seated filing: 70 W m−2
or 32% overestimation). The biophysical analyses using both the
measured and range of reference values for metabolic rate is shown
in Fig. 3 (see Methods for details on how this Figure is constructed).
The grey areas in Fig. 3 indicate the thermoneutral zones; that is,
they depict the area where heat loss equals measured metabolic
heat production. Open circles indicate actually measured mean skin
temperature and operative temperature baseline recordings. When
using the measured metabolic rate (Fig. 3, right area), measurements
are located inside the thermoneutral zone. This is in great contrast to
where all model parameters are kept equal except for the metabolic
rate for which the standard reference values for light oﬃce work are
used (Fig. 3, left area).
This biophysical analysis shows that mean skin temperature and
thermal environment of young adult females performing light oﬃce
work falls within their thermoneutral zone, but only if the correct
(actual) metabolic rate is used. Furthermore, we confirm that the
metabolic rate of young adult females performing light oﬃce work
is significantly lower than the standard values for the same type
of activity. With these results we argue that the current metabolic
standards should be adjusted by including the actual values for
females to reduce gender-discriminating bias in thermal comfort
predictions, and consequently, to reduce prediction bias in building
The body senses its thermal state through temperature-sensitive
receptors in core and skin tissues20. Various studies have examined
skin temperatures that are associated with thermal comfort,
and estimates range from wide (31.5 ≤Ts≤35.5 ◦C) to more
conservative (32.4 ≤Ts≤33.6 ◦C) values8,23. The former seems to
coincide with the entire thermoneutral zone of young adult females
performing light oﬃce work (see 31.5≤Ts≤35.5 ◦C, Fig. 3), whereas
the latter comprises only a subset of the thermoneutral zone (see
32.4 ≤Ts≤33.6 ◦C, Fig. 3). This discrepancy between studies may
in part have been caused by diﬀering number of skin sites that
have been measured. In general, more skin sites yields more reliable
results. Using a standard that consists of less than 10 skin sites
leads to significantly lower reliability24. On top of that, within limits
imposed by physics and physiology, human psychological factors
(for example, thermal adaptation due to geography) may also play a
role in what skin temperatures are considered comfortable25.
Constraining the model results further to skin temperatures that
are associated with thermal comfort it is possible to identify a
biophysical thermal comfort zone22. For the given conditions, the
biophysical thermal comfort zone for females ranges from 23.2 to
26.1 ◦C (for mean skin temperature ranging 32.4≤Ts≤33.6 ◦C).
As introduced, we make a case to use more reliable values
for female metabolic rate in thermal comfort prediction. Future
technological advances may yield devices that accurately measure
individual metabolic rate (for example, via smart watches and
so on). Until that time, one way to go forward may be by using
resting metabolic rate equations that take into account the eﬀects
of age, sex and body size (for example, revised Harris and Benedict
equations for resting metabolic rate26). The resting metabolic rate
can be converted from watt to watt per square metre by using the
appropriate equation for body surface area (for example, ref. 27).
Furthermore, the resting metabolic rate can also be scaled to the
activity type using the MET scaling factors.
The use of more accurate metabolic rates implies that the
PMV/PPD model requires recalibration. The reason for this is in
the very nature of the PMV/PPD model: it is an empirical model
and it has been fitted against thermal sensation votes using the
standardized values for metabolic rate. In the long run, recalibration
to any unforeseen subpopulation is not a sustainable strategy to
maintain. It has been suggested that a better method to improve
the PMV/PPD model is to revise its physiological construction15,28.
This in turn makes it possible to fundamentally understand and
take into practice the eﬀects of group and individual diﬀerences
temperature Lower critical
Thermoneutral zone Thermoneutral zone
Figure 2 | Two methods to depict the thermoneutral zone. a, The classical
method indicating the dependence of metabolic rate versus operative
temperature. b, The method used in this paper describing the relation
between skin temperature versus operative temperature. The range to the
left of the thermoneutral zone depicted with ‘shivering’ indicates that more
heat production is required to maintain thermal balance, and the range to
the right depicted by ‘sweating’ indicates that more heat loss is required to
maintain thermal balance. Within the open bounds it is possible for the
body to maintain core temperature.
2NATURE CLIMATE CHANGE | ADVANCE ONLINE PUBLICATION | www.nature.com/natureclimatechange
NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE2741 LETTERS
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Operative temperature (°C)
Mean skin temperature (°C)
measured metabolic rate
Measured thermal state of
young adult females
performing light oce work
standard metabolic rate
Figure 3 | The steady-state thermoneutral zone (grey areas) and baseline
measurements (open circles). The bottom of the thermoneutral zone is
associated with maximal tissue insulation, and vice versa for the top. Left
area: thermoneutral zone using the standard value for metabolic rate
associated with light oce work (ranging from 60 to 70W m−2). Right
area: thermoneutral zone using measured metabolic rate of females
performing light oce work.
in physiological characteristics such as lean versus obese, or the
consequences on preferred thermal environment of changes in
physiology (for example, ageing, acclimatization or illness).
The biophysical approach in this study is an essential step towards
such a revision. In our opinion it allows for a significant improve-
ment of the empirical model because of the physiological construc-
tion that includes the constraint for thermal balance within the body.
To do so, it is crucial to take into account not only the metabolic
rate but also the physiological range of body tissue insulation, which
may vary with body composition, gender and age, but also between
individuals. Thus, the biophysical model provides insight into ther-
mal comfort boundaries for subpopulations such as females, males,
children and seniors. These should be addressed in future studies
that measure subjective thermal comfort as well as physiological
characteristics such as metabolic rate and tissue insulation. The
biophysical approach provides a fundamental substantiation of the
PMV/PPD model on the population level and adds flexibility to
predict thermal comfort of subpopulations or individuals. Another
issue that deserves attention is thermal behaviour. How do physio-
logical characteristics relate to behaviour, and to what extent does a
deviation from the thermoneutral zone trigger an action?
The main points here are that thermal comfort models need
to adjust the current metabolic standard by including the actual
values for females. Consequently, thermal comfort models need
either to be recalibrated or enhanced using a biophysical approach
as presented here. This in turn will allow for better predictions
of building energy consumption, by reducing the bias on thermal
comfort of subpopulations and individuals.
Methods and any associated references are available in the online
version of the paper.
Received 20 May 2015; accepted 30 June 2015;
published online 3 August 2015
1. IEA Energy Conservation in Buildings and Community Systems Programme
ECBCS Annual Report 2011 (ECBCS, 2011).
2. ASHRAE Thermal Environmental Conditions for Human Occupancy ASHRAE
standard 55-2010 (ASHRAE, 2010).
3. Byrne, N. M., Hills, A. P., Hunter, G. R., Weinsier, R. L. & Schutz, Y. Metabolic
equivalent: One size does not fit all. J. Appl. Physiol. 99, 1112–1119 (2005).
4. Zhang, G. J., Cai, M. & Hu, A. X. Energy consumption and the unexplained
winter warming over northern Asia and North America. Nature Clim. Change
3, 466–470 (2013).
5. Howells, M. et al. Integrated analysis of climate change, land-use, energy and
water strategies. Nature Clim. Change 3, 621–626 (2013).
6. Beshir, M. Y. & Ramsey, J. D. Comparison between male and female subjective
estimates of thermal eﬀects and sensations. Appl. Ergon. 12, 29–33 (1981).
7. Karjalainen, S. Gender diﬀerences in thermal comfort and use of thermostats
in everyday thermal environments. Build. Environ. 42, 1594–1603 (2007).
8. Liu, W. W., Lian, Z. W. & Deng, Q. H. Use of mean skin temperature in
evaluation of individual thermal comfort for a person in a sleeping posture
under steady thermal environment. Indoor Built Environ. 24, 489–499 (2015).
9. Fanger, P. O. Thermal Comfort (McGraw-Hill, 1970).
10. Havenith, G., Holmer, I. & Parsons, K. Personal factors in thermal comfort
assessment: Clothing properties and metabolic heat production. Energy Build.
34, 581–591 (2002).
11. Parsons, K. Human Thermal Environments 2nd edn (Taylor & Francis, 2003).
12. Ainsworth, B. E. et al. 2011 compendium of physical activities: A second
update of codes and MET values. Med. Sci. Sports Exerc. 43, 1575–1581 (2011).
13. McCullough, E. A., Jones, B. W. & Huck, J. A Comprehensive Data Base for
Estimating Clothing Insulation (ASHRAE, 1985).
14. Van Someren, E. J. W. Thermoregulation and aging. Am. J. Physiol. 292,
15. van Hoof, J. Forty years of Fanger’s model of thermal comfort: Comfort for all?
Indoor Air 18, 182–201 (2008).
16. Wijers, S. L., Saris, W. H. & van Marken Lichtenbelt, W. D. Cold-induced
adaptive thermogenesis in lean and obese. Obesity 18, 1092–1099 (2010).
17. Gagge, A. P. Rational temperature indices of man’s thermal environment and
their use with a 2-node model of his temperature regulation. Fed. Proc. 32,
18. Tanabe, S., Kobayashi, K., Nakano, J., Ozeki, Y. & Konishi, M. Evaluation of
thermal comfort using combined multi-node thermoregulation (65MN) and
radiation models and computational fluid dynamic (CFD). Energy Build. 34,
19. Fiala, D., Lomas, K. J. & Stohrer, M. Computer prediction of human
thermoregulatory and temperature responses to a wide range of environmental
conditions. Int. J. Biometeorol. 45, 143–159 (2001).
20. Hensel, H. Thermoreception and Temperature Regulation
(Academic Press, 1981).
21. IUPS Thermal Commission Glossary of terms for thermal physiology. Jpn. J.
Physiol. 51, 245–280 (2001).
22. Kingma, B. R. M., Frijns, A. J. H., Schellen, L. & van Marken Lichtenbelt, W. D.
Beyond the classic thermoneutral zone: Including thermal comfort.
Temperature 1, 142–149 (2014).
23. Gagge, A. P., Stolwijk, J. A. & Hardy, J. D. Comfort and thermal sensations and
associated physiological responses at various ambient temperatures. Environ.
Res. 1, 1–20 (1967).
24. Liu, W. W., Lian, Z. W., Deng, Q. H. & Liu, Y. M. Evaluation of calculation
methods of mean skin temperature for use in thermal comfort study. Build.
Environ. 46, 478–488 (2011).
25. Nicol, J. F. & Humphreys, M. A. Adaptive thermal comfort and sustainable
thermal standards for buildings. Energy Build. 34, 563–572 (2002).
26. Roza, A. M. & Shizgal, H. M. The Harris Benedict equation reevaluated:
Resting energy requirements and the body cell mass. Am. J. Clin. Nutr. 40,
27. Mosteller, R. D. Simplified calculation of body-surface area. N. Engl. J. Med.
317, 1098 (1987).
28. Humphreys, M. A. & Nicol, J. F. The validity of ISO-PMV for predicting
comfort votes in every-day thermal environments. Energy Build. 34,
The authors would like to express their gratitude to C. Jacquot and L. Schellen for
performing measurements, and A. Frijns for fruitful discussions. This study was
supported by grants from AgentschapNL (INTEWON: EOSLT10033) and TKI Energo
and TKI Solar Energy (TRECO: TEGB|13023).
B.K. contributed to experimental work, project planning, data analysis, biophysical
modelling, and manuscript writing. W.v.M.L. contributed to project planning, data
analysis, manuscript writing and project funding.
Supplementary information is available in the online version of the paper. Reprints and
permissions information is available online at www.nature.com/reprints.
Correspondence and requests for materials should be addressed to B.K.
Competing ﬁnancial interests
The authors declare no competing financial interests.
NATURE CLIMATE CHANGE | ADVANCE ONLINE PUBLICATION | www.nature.com/natureclimatechange 3
LETTERS NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE2741
The thermoneutral zone. The thermoneutral zone is classically depicted as
described in ref. 29: metabolic rate versus operative temperature and bound by the
lower critical temperature and the upper critical temperature (see Fig. 2a). From
the perspective of thermal balance, thermoneutral operative temperatures near the
lower critical temperature correspond to relatively low skin temperatures, and vice
versa for thermoneutral operative temperatures near the upper critical
temperature. This leads to a new method to depict the thermoneutral zone, which
is used in the remainder of this paper (see Fig. 2b). The open range to the left of the
thermoneutral zone depicted with ‘shivering’ indicates that more heat production is
required to maintain thermal balance, and the open range to the right depicted by
‘sweating’ indicates that more heat loss is required to maintain thermal balance.
Note that depending on mean skin temperature there is overlap between
thermoneutral operative temperatures, and operative temperatures that require
extra heat production or extra heat loss. Thermal balance in the thermoneutral
zone is always a constellation of skin temperature and operative temperature.
Therefore, it may be inadequate to define a thermal comfort zone only by
The exact skin temperature range of the thermoneutral zone is bounded by the
metabolic rate and the capacity of the body to regulate tissue insulation22. The body
regulates tissue insulation by constricting and dilating blood vessels in skin tissue.
To preserve body heat the body constricts blood vessels (that is, vasoconstriction),
which results in maximal tissue insulation and minimal skin temperatures. To
enhance body heat dissipation the body dilates blood vessels (that is, vasodilation),
which results in minimal tissue insulation and maximal skin temperatures30.
Biophysical analysis. For the biophysical analysis a model is used that describes
the relation between body core temperature, skin temperature and operative
temperature (see Fig. 1)22. The model is used to determine core temperature for a
range of mean skin temperatures (28–38◦C) and operative temperatures
(14–32 ◦C), whilst satisfying thermal balance. Overall thermal balance is achieved
when the following conditions are met: (1) Metabolic heat production equals
internal body heat transport. (2) Metabolic heat production equals external body
heat loss. These conditions can be rewritten to an equivalent such that internal
body heat transport equals external body heat loss. Note that metabolic heat
production is then eliminated from the condition of thermal balance, yet still
determines the actual body heat transport and heat loss. The model details are
described in ref. 22. For the purpose of this paper the steps for constructing the
young adult female thermoneutral zone as depicted in Fig. 3 are given below:
Define body surface area (A), clothing insulation (Icl), relative humidity (ϕ)
skin wettedness, (w), wind speed (vair) and Lewis relation (λ).
Icl =0.105 m2K W−1(∼0.68 clo)
vair =0.09 m s−1
λ=2.2 ◦C mmHg−1
Define minimal and maximal metabolic rate (Mmin and Mmax) based on 99%
confidence interval of energy expenditure measurements and correct for
respiratory heat loss (that is, consider only heat transfer to and from skin surface;
assumed 8% of metabolic rate31).
Mmin =(1−0.08)×84 W=77 W
Mmax =(1−0.08)×97 W=89 W
Define minimal and maximal body tissue insulation (Ibody,min and Ibody,max), (skin +
fat layer ∼4 mm)30,32.
Ibody,min =0.031 m2K W−1
Ibody,max =0.112 m2K W−1
Define minimal and maximal body core temperature (Tc,min and Tc,max).
Tc,min =36.5 ◦C
Tc,max =37.5 ◦C
Calculate minimal and maximal skin temperature that support internal heat
balance (Ts,min and Ts,max) according to ref. 22:
Ts,min =Tc,min −Mmax ×Ibody,max/A[◦C]
Ts,max =Tc,max −Mmin ×Ibody,min/A[◦C]
Define 500 ×500 point grid for Tsand Ta
Tsbetween 28 and 38 ◦C
Tabetween 14 and 32 ◦C
For each point in the grid (Tsand Ta): Calculate combined convective (hc) and
radiative (hr) contribution to insulation provided by air (Ia) according to ref. 30:
hc=0.19×(100×vair )0.5 ×(298/(Ta+273.15)) [clo−1]
Calculate evaporative heat loss according to ref. 17:
Qe=w×λ×hc×(Ps−ϕPair)×Fpcl [W m−2]
Ps=γ×100exp(18.965−4, 030/(Ts+235)) [Pa]
Pair =γ×100exp(18.965−4, 030/(Ta+235)) [Pa]
γ=0.00750061683 [mmHg Pa−1]
Fpcl =1/(1+0.143×(hc/0.155)×(Icl/0.155)) [−]
Calculate body tissue insulation according to ref. 22:
Ibody =Ibody,max +(Ts−Ts,min)
×(Ibody,max −Ibody,min)/(Ts,min −Ts,max )[m2K W−1]
where (Ibody,min ≤Ibody ≤Ibody,max)
Calculate Tcsatisfying internal and external heat balance according ref. 22:
Calculate external heat loss (Qout)
Qout =A×((Ts−Ta)/(Icl +Ia)+Qe)[W]
Keep points for which Tcbetween Tc,min and Tc,max. Keep points for which Qout
between Mmin and Mmax and plot remaining points in Figure (Tsversus Ta).
The exact position of the thermoneutral zone depends on the metabolic rate,
body insulation and clothing level. Higher metabolic rate and/or clothing level
shifts the thermoneutral zone to lower operative temperature and higher body
insulation shifts the neutral zone to lower operative and mean skin temperature,
and vice versa for lower metabolic rate and/or clothing level and lower
Accuracy statement. The model in this paper combines the main principles that are
important to describe internal and external human thermal balance. Nevertheless,
it is a simplification of reality and may not always well-predict temperatures,
especially for non-steady-state conditions. Likewise, the model parameters
are obtained from elaborate experimental studies, but may include measurement
errors. Therefore, to obtain a conservative view of how the thermoneutral
zone is situated we used parameter ranges instead of set values for metabolic
rate, core temperature and body tissue insulation. All steady-state, comfortable
(32.4≤Ts≤33.6◦C) measurements fall within the computed thermoneutral
zone for these conditions (see Fig. 3). When including non-steady-state baseline
measurements and skin temperatures outside comfortable skin temperature
range, 14% of measurements are situated just outside the model prediction
(mean absolute distance from thermoneutral zone for these points is 0.3 K).
Measured data. The data for the analysis were obtained in the context of a larger
study performed in our laboratory on thermal preference in young adult females33.
NATURE CLIMATE CHANGE |www.nature.com/natureclimatechange
NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE2741 LETTERS
During the study 16 young female participants were lightly clothed
(∼0.58 Clo +0.10 Clo provided by chair), sitting behind a desk and were
randomly exposed to room temperature protocols in a climate chamber. For the
purpose of this study only steady-state baseline data from these protocols are used.
Operative temperature and relative humidity are measured using
wireless sensors (iButton, DS1923, Maxim Integrated Products; accuracy
±0.1 ◦C). Skin temperature is also measured with iButtons (DS1922L,
accuracy ±0.1 ◦C) at the 14 positions as described by ISO 9886 standard33.
Energy expenditure of young females performing light oﬃce work is
measured by indirect calorimetry (Maastricht Instruments, accuracy ±5%).
Recordings of baseline CO2production and O2uptake are converted into their heat
equivalent using the Weir equation34. The ASHRAE listed values for seated light
oﬃce work range from ‘seated quiet’ (60 Wm−2) to ‘seated filing’ (70 Wm−2).
Measured metabolic rate is compared with the standard values using a one-sample
t-test with significance level α=0.05. Data are presented as mean ±s.e.m. Whole
body fat percentage is measured through dual X-ray absorptiometry (Hologic,
29. Scholander, P. F., Hock, R., Walters, V. & Irving, L. Adaptation to cold in arctic
and tropical mammals and birds in relation to body temperature, insulation,
and basal metabolic rate. Biol. Bull. 99, 259–271 (1950).
30. Burton, A. C. & Edholm, O. G. Man in a Cold Environment
(Edward Arnold, 1955).
31. deGroot, D. W. & Kenney, W. L. Impaired defense of core temperature
in aged humans during mild cold stress. Am. J. Physiol. 292,
32. Veicsteinas, A., Ferretti, G. & Rennie, D. W. Superficial shell insulation
in resting and exercising men in cold water. J. Appl. Physiol. 52,
33. Jacquot, C. M., Schellen, L., Kingma, B. R., van Baak, M. A. &
van Marken Lichtenbelt, W. D. Influence of thermophysiology on thermal
behavior: The essentials of categorization. Physiol. Behav. 128, 180–187 (2014).
34. Weir, J. B. New methods for calculating metabolic rate with special reference to
protein metabolism. J. Physiol. 109, 1–9 (1949).
NATURE CLIMATE CHANGE |www.nature.com/natureclimatechange