Total energy expenditure during arduous wildfire suppression : Medicine & Science in Sports & Exercise

Journal Logo

APPLIED SCIENCES: Physical Fitness and Performance

Total energy expenditure during arduous wildfire suppression


Author Information
Medicine & Science in Sports & Exercise 34(6):p 1048-1054, June 2002.
  • Free


Previous research has shown that the doubly labeled water (DLW) methodology serves as the gold standard for the measurement of total energy expenditure (TEE) of free-living individuals (18,27). However, the methodology requires rigorous quality control in regard to isotope preparation, sample analyses, and the application and timing of the isotopic dose relative to the sample collection period (27). These issues and costs make the use of the doubly labeled water methodology more difficult during actual field operations where the measurement period is unpredictable and conditions may be hazardous. Several studies have used the DLW technique in the field, including military operations (3,6,7,9,10,11,14), mountainous/arctic expedition (15,22,28), multiple-stage bicycle racing (24), extended training (20,23), and space travel (12,21). In the last decade, the use of doubly labeled water for measurement of TEE has received notable acceptance. However, the majority of prior “field” research has been conducted on a previously determined schedule (scheduled training/expedition period) with an anticipated start and finish time for the experiment.

Past research in field settings has used DLW for measurement periods of 5–8 d (7,10,24) and 10–12 d (3,9,11,14). Because the precision of the calculated TEE is dependent on the elimination rates of both isotopes (kH and kO), it is necessary for the measurement period to be long enough to accurately calculate the difference in the elimination of the 2H and 18O isotopes but not so long that the isotopic abundances return to near baseline. In the sedentary or recreationally active individual, this optimal period is between 4 and 16 d given the typical isotopic dose (approximately 0.12 g·kg−1 total body water for 2H2O and approximately 0.3 g·kg−1 total body water for H218O), anticipated water turnover, and low to moderate rates of TEE. However, when the work environment is likely to result in unusually high rates of water turnover, coupled with an elevated TEE, the measurement period should be shortened significantly, and adjustments may be needed to the initial isotopic dose to accommodate the more rapid rates of elimination.

During the summer months in the western part of the United States, land management agencies (United States Forest Service, Bureau of Land Management, and State Forestry) are involved in wildland fire suppression efforts. Wildland fire suppression is a seasonal occupation requiring long hours of heavy work under adverse conditions (extended work shifts up to 24 h, high ambient heat, compromised dietary intake, smoke inhalation, acute altitude exposure, and sleep deprivation). The common wildland-firefighting tasks often include hiking, fire-line construction, chain-saw work, and brush removal, typically averaging approximately 7.5 kcal·min−1 calculated from Douglas bag data (1). However, the 7.5 kcal·min−1 calculation represents steady state work and does not represent the common work:rest cycle each hour. Assuming a 3:1 work to rest ratio each hour and a 12- to 14-h workday, work-shift energy expenditure might range from about 16.8 to 19.7 MJ·12−1 and 14 h−1 MJ, respectively (4000–4700 kcal·12−1 and 14 h−1, respectively). Although accurate, the use of indirect calorimetry for extended measurement periods or on the actual fire line is not practical. Even if expired gas samples were periodically collected under these conditions, they may not be representative of the actual total daily energy expenditure. The use of DLW allows for an extended measure of TEE during an actual wildfire suppression assignment, provided issues related to dose delivery, initial isotope equilibration, and sample collection strategies can be accommodated.

The purpose of this study was to use the DLW methodology to determine the TEE in male and female wildland firefighters during short-term (3–5 d) arduous wildfire suppression. We hypothesized that the calculated rates of energy expenditure would be similar to previous research including military operations (3,6,7,9–11,14) and mountainous/arctic expedition (15,22,28).



Subjects included male (N = 8) and female (N = 9) wildland firefighters recruited from three Interagency Hot Shot Crews (Lolo, Bitterroot, and St. Joe crews) from Western Montana and Idaho. Subjects were recruited through an informative meeting arranged between the Principal Investigator and all Interagency Hot Shot Crew Supervisors before the 1997 and 1998 fire seasons. An informational meeting was then arranged between the Principal Investigator and the entire crew. At this time, the objectives of the study and the outline of data collection were discussed. Potential subjects were selected and subsequently tested upon deployment during five different fire assignments in Montana (N = 1), California (N = 4), Florida (N = 5), Washington (N = 4), and Idaho (N = 3). Each wildfire assignment in the west (MT, CA, WA, and ID) included extended hiking over rough terrain and fire-line construction on steep hillside slopes (20–40% grade). Individual load carriage was not measured for each subject. However, the typical line gear pack weighs approximately 12–20 kg. Considering the average weight of the subjects, load carriage is likely to range from 14 to 27 and 17 to 31% of BW (kg) for the men and women, respectively.

All wildland firefighters wear the same regulation personal protective gear, which includes mid-calf leather logger boots, Nomex pants, 100% cotton short sleeve undershirt, Nomex long sleeve shirt, leather gloves, and hard hat. This combination of protective gear was consistent for each subject in the present study. The only time firefighters may deviate from this equipment is during mop-up at which time the gloves are typically removed to test for warm spots in the soil. The average weight of the above mentioned protective gear is approximately 5.0–5.5 kg, depending on boot size and is considerably lighter compared with the protective gear of the structural firefighter, which may range from 20 to 30 kg.

Preliminary screening.

Before data collection, all subjects read and signed an Institutional Review Board (IRB)–approved human subjects consent form.

Isotope administration.

Upon arrival to the incident, subjects were provided with an oral dose of doubly labeled water (0.39 g·kg estimated TBW−1 H218O, 0.23 g·kg estimated TBW−12H2O, Cambridge Isotope Laboratories, Andover, MA; Isotec, Inc., Miamisburg, OH). The initial dose was given in the evening hours (approximately 2200 h) before initiating fire suppression efforts the following morning. After the initial isotope consumption, the dose vials were rinsed three times with approximately 30 mL of tap water. The protocol for isotopic dosing and the daily collection schedule is detailed in Figure 1. At each wildfire assignment, a non-DLW–dosed subject (1 placebo subject: 2 experimental subjects) provided urine samples at the same time points as the experimental subjects. The nondosed subjects were members of the same Hot Shot crew and performed the same work shift duties as the experimental subjects throughout the experimental period.

Dose and daily collection schedule for measures of TEE, using the doubly labeled water technique.

Although the initial dose is somewhat higher (for H218O) compared with estimates suggested by Schoeller (16), a larger dose was used to compensate for the anticipated rates of high water turnover, potential shifts in background enrichment, and an expected high TEE. All overnight voids were collected to correct the measures of total body water (TBW) before and after the TEE-measurement period.

Daily urine samples were collected and stored in 5-mL cryogenic vials on ice for the duration of the experimental period. TBW was calculated from the change in isotopic enrichment (background vs the second void urine). In addition to the initial dose of doubly labeled water, a second dose of 2H2O (approximately 2.0 g) was administered on the eve of day 5 after the collection of a second background sample. The following morning, first void and second urine samples were collected. First void samples were collected on mornings 1, 2, 3, and 5 for isotopic analyses and calculation of TEE after approximately 72 and 120 h of wildfire suppression activity.

Body composition was calculated before fire suppression from TBW (calculated from 2H2O dilution) in each subject during the 1997 and 1998 seasons. Body composition was estimated from the TBW values (calculated from 2H2O dilution, and corrected for overnight dose loss and baseline decay for the final TBW). Fat-free mass (FFM) was calculated as TBW/0.73. Fat mass (FM) was calculated from the difference in the nude body weight and the calculated FFM.

Experimental fire suppression period.

During deployment to five different wildfire incidents (MT, CA, FL, WA, and ID), subjects were studied during five consecutive days of wildfire suppression. Additional subjects (2–4 subjects for each fire) who did not receive the isotopic dose were also studied on each fire for adjustments in background abundances due to a change in the geographic location and water source. Wildfire suppression assignments commenced shortly after (<2 h) the collection of the second void sample on day 1 (see Fig. 1). Wildfire suppression included extensive hiking with a load (15–20 kg), fire-line construction with a Pulaski (modified axe for ground scraping), chain-saw work, and brush removal. Work shifts ranged from 12 to 18 h during the entire experimental period on all wildfire assignments. The collection of daily samples and the analysis of urine samples from the beginning, middle, and end of the experiment allowed TEE to be calculated for days 1–3 and days 1–5, and estimated for days 4–5. The rationale for this was to determine whether there was variation in the calculated average TEE during the initial attack portion of wildfire suppression as compared with the 5-d segment. Typically, the Interagency Hot Shot crews will work extended shifts (up to 24 h) during the initial stages of an assignment in an attempt to gain control of the wildfire. However, after the fire is controlled (a fire line has been constructed around the entire perimeter), work-shift energy expenditure may decline as firefighters “hold” the fire line (monitoring the interior of the fire with some hiking but less hand tool or saw work).

Isotopic analyses and TEE calculations.

The Nutritional Sciences Laboratory at the University of Wisconsin, Madison, performed isotopic analyses of all urine samples as described elsewhere (19). Briefly, urine was cleaned with 200 mg of dry carbon, black Deuterium analysis was performed by reducing 0.8 μL of cleaned fluid over chromium at 850°C (8), and the deuterium abundance measured using a Finnigan MAT Delta Plus isotope ratio mass spectrometer (Thermo Finnigan, San Jose, CA). All analyses were performed in duplicate and all specimens from the same participant were analyzed during the same batch.

The second aliquot was equilibrated with 1 mL (STP) of carbon dioxide at constant temperature (17). A Finnigan MAT Delta S isotope ratio mass spectrometer was used to measure the 18O/16O ratio under dynamic flow conditions (17). Analyses were performed in duplicate and all specimens from the same participant were analyzed during the same batch. Isotope dilution space was calculated as described by Coward and Cole (4). Total body water was calculated by averaging the deuterium dilution space/1.041. Repeat analysis indicated the analytical precision for TEE had a CV of 4% (19).

The components of TEE were estimated from the summary equation TEE = DIT + basal metabolic rate (BMR) + energy expenditure of physical activity (EEA), where DIT is the thermogenic effect of the dietary intake during the measurement period (estimated as TEE × 0.1). BMR was calculated based on total body water using the prediction of Cunningham et al. (5) ([21.6·(TBW/0.73)] + 370). Due to the nature of field collection, actual measures of BMR were impractical. Values of TEE (MJ·d−1 and as multiples of BMR) and EEA (MJ·d−1) were calculated for days 1–3 and 4–5.

Food intake.

Food intake of the subjects was monitored each day by using dietary recall and food records. Subjects were asked to complete detailed diaries each day and to update the diary after the completion of each meal and snack. Subjects also retained the packages of unique food items for subsequent analyses. All dietary intake records were updated each evening using The Food Processor (version 7.0, esha Research, Salem, OR) for Windows. Caterer recipes were analyzed to quantify foods consumed by the subjects in the fire camps.

Statistical methods.

All descriptive data are expressed as means ± SD. A one-way repeated measures ANOVA across time (pre vs post) was used to determine differences in body weight and composition (FFM and FM) from 2H2O dilution (N = 17). A 2 × 2 repeated measures ANOVA (intake vs expenditure and measurement period (short vs long)) was used to determine differences between the reported intake and expenditure and differences in the TEE across the two measurement periods (kcal·d−1). A 2 × 2 mixed design ANOVA (gender and short vs long measurement periods) was used to determine variations in measures of TEE and EEA.


Subject characteristics.

Descriptive data for all subjects are presented in Table 1. Although data were collected on eight men and nine women, one man from the 1997 season was not included in the D2O dilution results for body composition due to an error during the postexperimental-dosing period. The change in body weight, TBW, FFM, or FM was not statistically significant during the experimental periods.

Table 1:
Descriptive data for the sample of wildland firefighters studied during the 1997–98 fire season experimental periods; data are presented as means ± SD.

Data for the measure of TEE are shown in Table 2. Although the men had a significantly higher rate of absolute daily energy expenditure, there were no differences between the genders relative to estimated BMRs (TEE expressed as multiple of BMR). However, when TEE expenditure was expressed as a function of the energy expenditure associated with physical activity, the men demonstrated significantly higher values. Again, when this was expressed relative to total body weight, there were no differences between the men and women. Individual subject data are presented in Table 3 and demonstrate the variability between the various wildfire locations and assignments.

Table 2:
Average energy expenditure in the wildland firefighters studied during the 1997–98 fire seasons, calculated from total measurement period (5.1 ± 0.8 d); data are presented as means ± SD.
Table 3:
Individual values of TEE and EEA during wildfire suppression efforts calculated from the 1–5 d elimination rates; individual values are separated for each fire indicating location; average daily high and low temperatures and relative humidity have are shown in parentheses after the state.

To determine the relationships between TEE (kcal·d−1), EEA (kcal·d−1), and body mass (kg), simple correlations were calculated. Absolute TEE (kcal·d−1) was not significantly correlated to the preexperiment total body weight (r = 0.34, P > 0.05). Although the correlation was modest, TEE (kcal·d−1) was significantly correlated to the preexperimental measure of FFM (r = 0.53, P < 0.05). The energy expenditure from physical activity EE (EEA) was not significantly correlated to total body weight (r = 0.08) or FFM (r = 0.25). An additional objective of the study was to determine the energy cost of wildfire suppression during different stages of the fire (early, days 1–3 and later, days 4–5). Table 4 shows the comparison for TEE across the two measurement periods. Although there were some differences across gender, TEE was similar for each measurement period.

Table 4:
Comparison of the total energy expenditure across the two measurement periods (days 1–3 and days 4–5); the data from the original analyses (days 1–5 have been included for comparison; data are presented as means ± SD.

Self-reported energy intake data are presented in Table 5. Although the total intake was not significantly different across gender, the men tended to have a higher total intake. Expressed as a percent of the total intake, the men consumed less carbohydrate, and more fat and protein in comparison with the women. The self-report data indicated that daily carbohydrate consumption was 6.5 ± 1.7 and 7.3 ± 2.0 g·kg BW−1·d−1 for the men and women, respectively. Although there were no differences across gender, the total carbohydrate intake was slightly lower than recommendations associated with this type of exercise/work (approximately 10 g·kg−1 BW−1·d−1) (2). The reported protein intake was 2.2 ± 0.6 and 1.7 ± 0.5 g·kg BW−1·d−1 for the men and women, respectively, and are similar to the recommended amounts (approximately 1.8 g·kg−1BW−1·d−1) for this type of work/exercise (13).

Table 5:
Total energy intake and the percent contribution of each macronutrient during the experimental periods; data are presented as means ± SD.

In addition to the total intake patterns during the experimental period, the intake patterns during the “post shift” hours were evaluated to determine whether subjects were consuming adequate carbohydrate to ensure glycogen resynthesis. Over the post shift-time period (average estimate = 5 h), the carbohydrate intake averaged 0.62 ± 0.2 and 0.49 ± 0.2 g·kg−1·h−1 for the women and men, respectively. Although the differences between genders were not statistically significant (P = 0.23), these values are lower than what might be recommended for adequate glycogen resynthesis after exercise (approximately 1.0 g·kg BW−1·h−1).

Average data from local weather stations was collected to compute average daily high, low, and relative humidity. These data have been included in Table 3.


Although previous research has used the DLW methodology in the field during cycling (24), swimming (23), military operations (9,10), and mountaineering (15,25,26), previous protocols have been somewhat predictable in terms of start and finish times. However, the present study represents an attempt to study the application of the DLW technique in an unpredictable and hazardous occupational setting (wildland fire suppression).

Previous studies have documented rates of energy expenditure at levels similar to those noted in the present sample of wildland firefighters. Hoyt et al. (9) described the energy expenditure of Marines during simulated combat/training scenarios approaching 21 MJ·d−1. Similarly, Trappe et al. (23) documented rates of TEE as high as 23 MJ·d−1 in elite-level female swimmers. Westerterp et al. (24) has described the “energetic ceiling” for humans to be near 4–5 times BMR (29–37.7 MJ·d−1). However, Westerterp et al. (24) has also recognized the methodological concerns of the DLW methodology during measurement periods with extreme body water turnover (isotopic fractionation issues) and the need to account for these shortcomings. These issues include high rates of anticipated isotope turnover and alterations in background enrichment.

It is interesting to note that the inclusion of female subjects in previous field operations research has been minimal. Trappe et al. (23) measured the TEE of elite female swimmers while training at the U.S. Olympic Training Center. Westerterp et al. included two (25) and four female subjects (26) during high-altitude mountaineering expeditions on Mt. Sajama, Bolivia, and on Mt. Everest. Trappe et al. (23) noted TEE values of 23.4 ± 2.1 MJ·d−1 during 5 days of two swim sessions per day, lasting a total of 5–6 h·d−1. In the present investigation, there were minimal differences in TEE across gender expressed relative to estimated BMR, whereas EEA was significantly higher in the male subjects (by approximately 3.8 MJ·d−1). However, relative to total body weight (EEA·kg−1), the difference between male and female subjects was not statistically significant. Considering the low correlations between EEA and BW, the observed gender difference in EEA cannot be fully explained by variations in BW or FFM. Therefore, this difference may be attributed in part to variations in the test locations (geography of the fire), the fire assignment, and/or selected work rate.

Regardless of the harsh environment, the overall data suggest that subjects were able to maintain energy balance with the self-selected energy intake. However, this is not suggested by the comparison between self-reported intake and TEE. Although the correlation between TEE and intake was significant (r = 0.64, P < 0.05), total energy intake represented 88.6 ± 19.1% of the TEE (P < 0.05 between TEE and TEI). The disparity of the self-report intake data and the calculated TEE further emphasizes that the subjects underestimated total dietary intake (TEE = 17.5 ± 4.4 MJ·d−1 (1–5 d calculation) vs TEI = 15.2 ± 3.5 MJ·d−1). This is not necessarily surprising as the majority of previous research has demonstrated similar disparity and underreporting in the TEI of subjects during field-like measurement periods (7,23,27). Moreover, the subtle changes in body weight noted in the male subjects and the changes in FM noted in the female subjects may have resulted from an inconsistent net caloric deficit among the subjects. Therefore, it is difficult to conclude from these data that energy balance was maintained in all subjects. The TEI data coupled with the subtle changes in body weight and composition suggest the possibility of underreporting and/or an energy imbalance during the measurement period.

The most dramatic threat to the accuracy of the DLW in field conditions is associated with shifts in the background enrichment. This became apparent in the present investigation as all of our firefighter subjects were from western Montana and Idaho based crews. Corrections to TEE were not required for the measurement periods in the western states (MT, CA, WA, and ID) because background enrichments were minimally affected (changes in background enrichment for 18O were −9.21 ± 1.03, −8.75 ± 0.96, and −9.03 ± 1.00 for baseline, day 2, and day 5, respectively; changes in background enrichment for 2H were −83.90 ± 4.64, −81.90 ± 6.15, and −79.58 ± 8.71 for baseline, day 2, and day 5, respectively). However, the TEE calculations during wildfire suppression in Florida required correction due to changes in background enrichment (changes in background enrichment for 18O were −8.78 ± 0.96, −7.59 ± 1.76, and −6.28 ± 1.06 for baseline, day 2 and day 5, respectively; changes in background enrichment for 2H were −85.20 ± 4.69, −63.30 ± 2.64, and −53.80 ± 2.58 for baseline, day 2 and day 5, respectively). This variation is most likely a function of the change in geographic locations from the crew’s home base in Montana and Idaho to northern Florida.

The practical usefulness of the DLW methodology is limited because of obvious expense, methodological issues surrounding sample analyses, and calculation issues (isotopic fractionation). Regardless, the methodology appears robust and appropriate during unpredictable environmental conditions where water turnover (rH2O) may vary due to environmental stress (heat and humidity) and may exceed 8.5 L·day−1 so long as attempts are made to adjust for isotopic fractionation due to an exaggerated evaporative water loss. Rates of isotope elimination over the 5-d measurement period are presented in Table 6 and further demonstrate the accelerated rate of isotopic turnover.

Table 6:
Calculated rates of isotopic elimination during the 5-d measurement period; data are presented as means ± SD.

We used the assumption that the routes of fractionated evaporative water loss were transcutaneous water loss and breath water loss. The equation used assumes that breath water loss will be in proportion to CO2 production and that this assumption is very robust. The equation used assumes that transcutaneous water loss (nonsweat gland loss) is equal to 30% of breath water loss. This is based on average rates of transcutaneous water loss per m2 of body surface area and average 24-h ventilatory volumes in the typical adult population. In this population, the 24-h ventilatory volume will be proportionately elevated with increased CO2 production. Although total water loss at the skin surface increases with temperature, the transcutaneous loss does not increase. The increase is due to activation of the sweat glands and the resulting nonfractionated sweat loss. As such our calculation may have overestimated fractionated water loss. The overestimate of transcutaneous water loss would be proportional to the increased CO2 production relative to the average adult, which amounts to approximately 50%. Back calculating through the derivation of the TEE equation, this corresponds to an 8% overestimate of fractionated water loss. In these subjects in whom the ratio of ko/kd averaged 1.26, this would result in a 4% error in the calculated TEE. By comparison, had we used the frequently used assumption of a constant fraction percentage of 25% of water turnover, the error would have been in error by 20%.

Previous research has also noted that the accuracy of the DLW method is maintained even in situations of negative energy balance (9,10). In the current study, the overall coefficient of variation for TEE (MJ·d−1 and xBMR) for the 1–3 and 4–5 d measurement periods was 15.7%. However, the coefficient of variation for EEA (MJ·d−1) for the 1–3 and 4–5 d measurement periods was 28.6%. The magnitude of this variation is likely due to individual and day-to-day variations in physical activity due to geographic location, actual fire assignment, and the self-selected work rate of each firefighter. The coefficient of variation in TEE (MJ·d−1) comparing the 1–5 d calculation and the 1–3 d measurement periods was only 6.2% and further demonstrates the accuracy of this method during short measurement periods.

These data indicate a consistent day-to-day pattern of energy expenditure in the WLFF, and the robust qualities of the DLW technique during varying environmental conditions when the necessary corrections are done to compensate for any shift in the baseline abundance of the 2H and 18O isotopes. These data suggest that wildfire suppression in the United States requires an average daily energy expenditure equivalent to approximately 2.5–3.0 xBMR. Moreover, the energy expenditure associated with physical activity during multiple days of wildfire suppression may range from 4.2 to 12.6 MJ·d−1 and is mostly affected by work assignment, self-selected work intensity, and the location of the fire.

This study was funded by the Defense Women’s Health Research Grant Program, DAMD17-96-1-6329 and DAMD17-00-P-0429. The authors wish to recognize the cooperation from the research subjects, the various fire crews, fire management teams, and the Missoula Technology and Development Center during the planning and collection of the data.

Address for correspondence: Brent C. Ruby, Ph.D., Director, Human Performance Laboratory, Department of Health and Human Performance, The University of Montana, Missoula, MT 59812-1825; E mail: [email protected]


1. Budd, G. M., J. R. Brotherhood, A. L. Hendrie, et al. Project Aquarius 5: activity distribution, energy expenditure, and productivity of men suppressing free-running wildland fires with hand tools. Int. J. Wildlandfire 7: 105–118, 1997.
2. Burke, L. M., G. R. Cox, N. K. Culmmings, and B. Desbrow. Guidelines for daily carbohydrate intake: do athletes achieve them? Sports Med. 31: 267–299, 2001.
3. Burstein, R., A. W. Coward, W. E. Askew, et al. Energy expenditure variation in soldiers performing military activities under cold and hot climate conditions. Mil. Med. 161: 750–754, 1996.
4. Coward, A., and T. Cole. Precision and accuracy of the doubly labeled water energy expenditure by multipoint and two-point methods. Am. J. Physiol. 263: E965–E973, 1992.
5. Cunningham J. J. Body composition as a determinant of energy expenditure: a synthetic review and proposed general prediction equation. Am. J. Clin. Nutr. 54: 963–969, 1991.
6. Delany, J. P., D. A. Schoeller, R. W. Hoyt, E. W. Askew, and M. A. Sharp. Field use of D2 18O to measure energy expenditure of soldiers at different energy intakes. J. Appl. Physiol. 67: 1922–1929, 1989.
7. Forbes-Ewan, C. H., B. L. Morrissey, G. C. Gregg, and D. R. Waters. Use of doubly labeled water technique in soldiers training for jungle warfare. J. Appl. Physiol. 67: 14–18, 1989.
Gehre, M., R. Hoeflin, P. Kowski, and G. Stauch. Sample preparation device for quantitative hydrogen isotope analysis using chromium metal. Anal. Chem. 68: 4414–4417, 1996.
9. Hoyt, R. W., T. E. Jones, T. P. Stein, et al. Doubly labeled water measurement of human energy expenditure during strenuous exercise. J. Appl. Physiol. 71: 16–22, 1991.
10. Hoyt, R. W., T. E. Jones, C. J. Baker-Fulco, et al. Doubly labeled water measurement of human energy expenditure during exercise at high altitude. Am. J. Physiol. 35: R066–R971, 1994.
11. Jones, P. J., I. Jacobs, A. Morris, and M. B. Ducharme. Adequacy of food rations in soldiers during an arctic exercise measured by doubly labeled water. J. Appl. Physiol. 75: 1790–1797, 1993.
12. Lane, H. W., R. J. Gretebeck, D. A. Schoeller, J. Davis-Street, R. A. Socki, and E. E. Gibson. Comparison of ground-based and space flight energy expenditure and water turnover in middle-aged healthy male US astronauts. Am. J. Clin. Nutr. 65: 4–12, 1997.
13. Lemon, P. W. Beyond the zone: protein needs of active individuals. J. Am. Coll Nutr. 19 (Suppl. 5): 513S–521S, 2000.
14. Mudambo, K. S., C. M. Scrimgeour, and M. J. Rennie. Adequacy of food rations in soldiers during exercise in hot, day-time conditions assessed by doubly labeled water and energy balance methods. Eur. J. Appl. Physiol. 76: 346–351, 1997.
15. Pulfrey, S. M., and P. J. Jones. Energy expenditure and requirement while climbing above 6,000 m. J. Appl. Physiol. 81: 1306–1311, 1996.
16. Schoeller, D. A. Energy expenditure from doubly labeled water: some fundamental considerations in humans. Am. J. Clin. Nutr. 38: 999–1005, 1983.
17. Schoeller, D. A., and A. H. Luke. Rapid 18O analysis of CO2 samples by continuous flow isotope ratio mass spectrometry. J. Mass. Spectrom. 32: 1332–1336, 1997.
18. Schoeller, D. A. Recent advances of doubly labeled water to measurement of human energy expenditure. J. Nurtr. 129: 1765–1768, 1999.
19. Schoeller, D. A., A. S. Colligan, T. Shriver, H. Hairigh, and C. Bartok-Olson. Use of an automated reduction system for hydrogen isotope ratio analysis of physiologic fluids applied to doubly labeled water. J. Mass. Spectrom. 35: 1128–1132, 2000.
20. Sjodin, A. M., A. B. Andersson, J. M. Hogberg, and K. R. Westerterp. Energy balance in cross-country skiers: a study using doubly labeled water. Med. Sci. Sports Exerc. 26: 720–724, 1994.
21. Stein, T. P., M. J. Leskiw, M. D. Schluter, et al. Energy expenditure and balance during spaceflight on the space shuttle. Am. J. Physiol. 276: R1739–R1748, 1999.
22. Stroud, M. A., W. A. Coward, and M. B. Sawyer. Measurements of energy expenditure using isotope-labeled water (2H218O) during an arctic expedition. Eur. J. Appl. Physiol. 67: 375–379, 1993.
23. Trappe, T. A., A. Gasteldelli, A. C. Jozsi, J. P. Troup, and R. R. Wolfe. Energy expenditure of swimmers during high volume training. Med. Sci. Sports. Exerc. 29: 950–954, 1997.
24. Westerterp, K. R., W. H. Saris, M. van Es, and F. ten Hoor. Use of the doubly labeled water technique in humans during heavy sustained exercise. J. Appl. Physiol. 61: 2162–2167, 1986.
25. Westerterp, K. R., B. Kayser, F. Brouns, J. P. Herry, and W. H. Saris. Energy expenditure climbing Mt. Everest. J. Appl. Physiol. 73: 1815–1819, 1992.
26. Westerterp, K. R., B. Kayser, L. Wouters, J. L. Le Trong, and J. P. Richalet. Energy balance at high altitude of 6,542 m. J. Appl. Physiol. 77: 862–866, 1994.
27. Westerterp, K. R. Body composition, water turnover and energy turnover assessment with labelled water. Proc. Nutr. Soc. 58: 945–951, 1999.
28. Westerterp, K. R., E. P. Meijer, M. Rubbens, P. Robach, and J. P. Richalet. Operations Everest III: energy and water balance. Pflugers Arch. 439: 483–488, 2000.


©2002The American College of Sports Medicine