The Contribution of Nasal Countercurrent Heat Exchange to Water Balance in the Northern Elephant Seal, Mirounga Angustirostris

Northern elephant seals, Mirounga angustirostis (Gill), are exceptional among pinnipeds in the duration of their terrestrial breeding fast. During this time, they voluntarily forgo both food and water while remaining active on the rookery. The length of these fasts varies with age, sex and social status. Bulls may fast 3 months while vigorously defending a harem (Le Boeuf & Peterson, 1969), while females fast for 5 weeks prior to and during nursing (Le Boeuf, Whiting & Gantt, 1972). Following weaning, pups fast 8 to 12 weeks before departing from the rookery (Reiter, Stinson & Le Boeuf, 1981). During this time, total water loss does not exceed metabolic water production (MWP) (Ortiz, Costa & Le Boeuf, 1978). Ortiz et al. (1978) suggested that respiratory evaporation represents the predominant avenue of water loss and that ‘the complex turbinate processes in the nasal passages of pinnipeds may function as countercurrent heat exchanger, thereby reducing respiratory water loss’.

Cooling of expired air results in a significant reduction in respiratory evaporative water loss in several terrestrial vertebrates (Jackson & Schmidt-Nielsen, 1964; Collins, Pilkington & Schmidt-Nielsen, 1971; Schmid, 1976; Langman, Maloiy, Schmidt-Nielsen & Schroter, 1979; Schmidt-Nielsen, 1981). In temporal countercurrent heat exchange, the nasal lining is cooled by convection and evaporation upon inhalation, which results in heating and saturation of incoming air. During exhalation, air saturated with water at body temperature (T_b) passes over this cooled lining, losing heat and water. The relative decline in expired air temperature (T_e), and hence the quantity of water recovered, is a function of the geometry and temperature of the surfaces available for heat exchange. In animals where water recovery via temporal nasal countercurrent exchange is high, this area is large (Collins et al. 1971).

Few studies of nasal heat exchange in marine birds and mammals have been undertaken. In penguins, 82 % of the water and 83 % of the heat added to ambient air are recovered upon exhalation (Murrish, 1973) ; in two species of porpoise T_e falls below Tb (Coulombe, Ridgway & Evans, 1965).

In the present study we investigated temporal nasal countercurrent heat exchange in the northern elephant seal by analysis of nasal morphology, measurement of nasal mucosal temperature gradients and expired air temperatures. These data yield an estimate of the contribution of countercurrent exchange to overall water economy in this species.

Measurements of heat exchange surface area and the smallest and largest widths were made on 20 separate, sectioned turbinate structures taken from 10 skulls of weanlings and of an adult male skull. The total surface area of a single turbinate was estimated by photographing each end of 1-cm serial cross sections. Each photograph was then enlarged 10 ×. The length of the total exposed cross-sectional surfaces was estimated by tracing their outlines with a rolling map measure (Keuffel & Esser Co). Each sectional area was calculated by multiplying the mean of the total cross-sectional lengths of a section by the section length ( 1 cm). The total area of the turbinate was estimated as the sum of the section areas.

Eight weanling elephant seal pups were collected at Ano Nuevo State Reserve and transported to the Long Marine Laboratory at the University of California, Santa Cruz, where all laboratory measurements were made on resting, restrained animals.

The temperature gradient within the nasal capsule was measured on two animals at T_a = 15 °C and at T_a = 5 °C, near the low temperature normally encountered in the natural habitat. A 30-gauge copper-constantan thermocouple, attached to a 26-gauge stainless steel stiffening wire, was threaded into the nasal cavity and temperatures were recorded at 1-cm intervals for approximately 10 cm using a Bailey Model 3 telethermometer.

Field measurements of expired air temperatures were made on 27 animals at Ano Nuevo State Reserve, California, from February through May: four adult males, five subadult males, three adult females and 15 juvenile or weaned pups.

Exhaled air and body temperatures were measured with copper-constantan thermocouples calibrated against a mercury thermometer to ±0·1 °C. Ambient temperatures and relative humidity (RH), measured with a sling psychrometer or dial reading hygrometer, were recorded at the beginning and end of each observation period.

Weanlings and juveniles were restrained manually during measurements (Pernia, Hill & Ortiz, 1980). The nasal probe was inserted 1 cm into the nasal cavity, in the air stream, where the temperatures of ten or more expirations were recorded. Measurements obtained from restrained and unrestrained animals were not significantly different. Body temperature was measured by inserting the rectal probe approximately 20 cm into the anus.

Temperature of expired air in unrestrained adults was recorded using a thermocouple threaded into a 0⁻6m length of rigid polyethylene tube (1 mm diameter) with the distal 1 cm exposed. The temperature of 10 or more expirations was measured in each animal.

In the laboratory, a series of similar recordings was made on each of eight weanlings at ambient temperatures between 1·2 and 17 °C.

Mean respiratory flow was measured in three animals (average weight: 97-8kg). Each animal was restrained with its head in a plastic hood. Flexible neoprene provided an airtight seal around the neck. One-way valves were used on the incurrent and excurrent port. To test for consistency, gas volumes were measured in one of two ways, using a Singer volume meter. In the first experiments, 5-min samples from the excurrent port were trapped in an 80-1 Douglas bag equipped with a tight-fitting gas valve. This volume was subsequently drawn through the meter at approximately 301 min⁻¹ using a vacuum system. In the second method, the entire excurrent flow was routed directly through the Singer meter. Experiments were conducted on all three animals at ambient laboratory temperatures (15–17 °C) and on two animals at 5 °C. In each case a minimum sample of 1 h was obtained from each animal. For subsequent calculations, volumes were adjusted to expired air temperature, barometric pressure (P_b) of 760 mmHg, and assumed to be saturated. The volume at the point of actual respiratory water loss is most accurately reflected at the expired air temperature, ambient barometric pressure and fully saturated with water (ETPS).

The configuration of the elephant seal nasal passage is complex (Fig. 1). The nasal turbinates provide a very large surface area for exchange of heat and water. The total surface area of both turbinate structures in weanlings is approximately 720 ±23 cm² (±S.D., N =20) and in an adult male this area was estimated to be 3140 cm².

The temperature gradients within the nasal passage of a weanling at 15 °C and 5 °C are shown in Fig. 2. The distal nasal mucosa is well below core temperature. The temperature rises continuously through the turbinates until T_b is reached.

The expired air temperatures of 35 elephant seals of various ages at T_a values from 1-2 °C to 21 °C are summarized in Table 1. Expired air temperature increases linearly with T_a (Fig. 3). The cooling of the exhalant stream (Tb–T_e) and the percentage water recovery vary inversely with T_a. The efficiency of the system is increased at lower ambient temperatures. At a mean T_a of 13·72 ± 5·4°C (N – 35) the mean water recovery is 71·5 ± 5·5 %. At T_a = 21 °C, T_e is 7·2°C below T_b and water recovery is 50·7%, while at T_a = 1·2°C, T_e is 21·9°C below Tb and water recovery is 79·1 %. Percentage water recovered does not vary significantly with sex, age or restraint.

Mean respiratory flow for all animals under all conditions was 12·2±4·31_ETPS min⁻¹. There was no significant difference between respiratory flows at different temperatures.

The adaptive value of the proposed mechanism is associated with the terrestrial breeding behaviour of elephant seals. Since Ortiz et al. (1978) have clearly demonstrated long-term positive water balance in these animals, all avenues of water loss must be reduced to levels below or equal to metabolic water production (MWP). Costa & Ortiz ( 1980) have shown that respiratory EWL accounts for 65 % of the metabolic water production. The remaining 35 % is lost as urine, faeces and cutaneous evaporation. This low EWL/MWP ratio is comparable to that found among several small, desert endotherms (Bartholomew, 1972; MacMillen, 1972; MacMillen & Grubbs, 1976).

What is the contribution of the described mechanism to overall water balance in a typical weanling? Daily EWL for a 100-kg weanling at 19 °C and 68% RH may be estimated from the calculated water content of the expired air minus the water content of the ambient air and the measured respiratory flow at 730 1_ETPS h⁻¹. This calculation yields a loss of approximately 140 ml day⁻¹ as compared to a direct measurement of 209 ml day⁻¹ at similar T_a and RH (Costa & Ortiz, 1980). In the absence of nasal countercurrent exchange, air exhaled at Tb would contain 534 ml day⁻¹, and thus exceed the available MWP of 326 ml day⁻¹ (Costa & Ortiz, 1980). Under these conditions, maintenance of positive water balance during the prolonged terrestrial fast would be impossible (Fig. 4).

Nasal countercurrent heat exchange is a major factor in the reduction of EWL and maintenance of positive water economy in the northern elephant seal. This mechanism may contribute to water balance in other pinnipeds and perhaps marine mammals in general.

We thank Drs B. J. Le Boeuf, C. L. Ortiz and R. J. Berger for critical analysis and discussion of the manuscript, and J. Walker, L. Walker, E. Keith and J. Palea for assistance in the field, and the California Department of Parks and Recreation at Ano Nuevo State Reserve for coopertion and logistic support. This work supported in part by NIH DRR-SO6-RRo8132-04, NSF DEB-8117024 and NIH postdoctoral fellowship F-32-AM-060930-01. Skeletal material provided by B. J. Le Boeuf and UCSC Vertebrate Collections.