Energetics of Load Carrying in Nepalese Porters

See allHide authors and affiliations

Science  17 Jun 2005:
Vol. 308, Issue 5729, pp. 1755
DOI: 10.1126/science.1111513


Nepalese porters routinely carry head-supported loads equal to 100 to 200% of their body weight (Mb) for many days up and down steep mountain footpaths at high altitudes. Previous studies have shown that African women carry head-supported loads of up to 60% of their Mb far more economically than army recruits carrying equivalent loads in backpacks. Here we show that Nepalese porters carry heavier loads even more economically than African women. Female Nepalese porters, for example, carry on average loads that are 10% of their Mb heavier than the maximum loads carried by the African women, yet do so at a 25% smaller metabolic cost.

Nepalese porters routinely carry head-supported loads exceeding their body weight (Mb) for many kilometers up and down steep mountain footpaths at high altitudes. Except in African women (13), virtually nothing is known about the energetic cost of carrying head-supported loads. We set out to determine the loads and distances carried by Nepalese porters, their metabolic cost for carrying the loads, and whether their optimal walking speed (as determined by the minimum cost for carrying a load) changed as a function of load.

The town of Namche (at an altitude of 3500 m) near Mount Everest hosts a weekly bazaar. Porters (Fig. 1A), predominantly ethnic Rai, Sherpa, or Tamang, typically take 7 to 9 days to travel to Namche from the Kathmandu valley. The route, no more than a dirt footpath, covers a horizontal distance of ∼100 km, with total ascents (river crossings to mountain passes) of ∼8000 m and total descents of ∼6300 m.

Fig. 1.

(A) Method of load carriage in Nepal. Porters use a head strap (namlo) to support a basket (doko) containing the load. The T-shaped stick (tokma) is used to support the basket during the frequent rest periods taken by the porters. (B) Metabolic cost of carrying 1 kg of load over a distance of 1 m. Cload is shown as a function of the load in Nepalese porters (symbols and continuous line). The dashed line represents the Cload of control subjects using backpacks (5). (C) Gross metabolic power increase (power loaded, PL, divided by power unloaded, PU) due to carrying a load as a function of the increase in load (total weight, Mtot, divided by body weight, Mb). Nepalese porters (symbols and continuous line), carrying head-supported loads, are the most economical, particularly at high loads. African women (dotted line) (1), also carrying head-supported loads, equal the Nepalese at low loads. Control subjects (dashed line) (5) are close to the proportionality line. The number of data points averaged is indicated near the symbols in (B) and is the same in (C); the bars indicate the standard deviation. All measurements were near the optimal walking speed (0.8 to 1.4 m s–1).

One day before the bazaar, we counted 545 male and 97 female porters (and 32 yaks) en route to Namche; others passed by earlier and later in the darkness. We weighed randomly selected porters and their loads (4). The men carried loads of 93 ± 36% of their Mb (mean ± SD, n = 96 male porters), whereas the women carried 66 ± 21% of their Mb (n = 17 female porters). The youngest porter was 11 years old, and the oldest 68; the greatest load measured was 183% of Mb, and 20% of the men carried >125% of their Mb. More than 30 tons of material were ported to Namche that day.

We asked eight Nepalese porters to walk around a 51-m flat track at five speeds, carrying six or seven different loads, according to their ability. The energy costs of loaded and unloaded walking were determined from their oxygen consumption and carbon dioxide production (4). Control experiments on European subjects carrying backpacks have been reported separately (5).

For all loads, the weight-specific cost of carriage Cload (i.e., the energetic cost loaded minus the cost unloaded at the same speed, divided by the load weight, given in J kg–1 m–1) as a function of speed presented a minimum at ∼1.1 m s–1, both in Nepalese porters and in control subjects. However, at all speeds and loads, the porters had a smaller Cload than the controls. The average Cload near the optimal walking speed can be shown as a function of load (Fig. 1B). Cload increased from 0 J kg–1 m–1 at 15% of Mb to ∼3 J kg–1 m–1 at 100% of Mb in the Nepalese porters, whereas it was always >3 J kg–1 m–1 in the controls (5).

The increase in metabolic power while carrying a load (the loaded oxygen consumption rate divided by the unloaded oxygen consumption rate at the same speed) as a function of the total weight Mtot (Mb plus the load) is shown in Fig. 1C for the Nepalese porters, control subjects (5), and African women (1) at about the optimal speed. The Nepalese porters were far more economical than the control subjects at all loads and more economical than the African women at all except the lightest loads. Loads lighter than ∼20% of Mb are carried “for free.” Above 20% of Mb, the Nepalese porters' advantage increases with increasing loads. The Nepalese porters' economy allows them to carry loads that are on average 30% of Mb heavier than the maximum loads carried by the African women, for the same increase in metabolic rate.

The load versus speed versus energy-cost trade-off chosen by these porters is to walk slowly for many hours each day, take frequent rests, and carry the greatest loads possible. We observed, for example, a group of heavily loaded porters making slow headway up a steep ascent out of a river gorge. Following whistled commands from their leader, they would take up their loads and labor uphill for no more than ∼15 s at a time, followed by a ∼45-s period of rest. Incredibly, this group of barefoot porters was headed for Tibet, across the Nangpa glacier (altitude 5716 m), about another week's travel beyond Namche.

So how do they do it? They might reduce the muscular work required to carry a load or increase their overall efficiency. The actual mechanism is unknown at this time.

Supporting Online Material

Materials and Methods

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article