The Herbivore-Based Trophic System 397 



behavior and activity patterns of caribou. The model indicated that ac- 

 tivity significantly affected the average daily metabolic rate and that the 

 energy spent in grazing and evading insects needed to be determined. 

 Grazing involves almost continuous movement, so it was necessary to 

 estimate the energy cost of walking on tundra (White and Yousef 1978). 

 ACTIVE calculated that, compared to days with no harassment, the aver- 

 age daily metabolic rate increased 1 .06 times during mild harassment and 

 1.6 times during severe harassment, and averaged 2.8 times the standard 

 rate during summer. Energy expended on locomotion increases from 

 \1% of the average daily metabolic rate on insect-free days to approxi- 

 mately 60% during severe insect harassment. 



Because of their heavy insulation, thermoregulation is not a prob- 

 lem for caribou in winter (White 1975), but the energy required for win- 

 ter activities has not been determined. The assumption was made that 

 energy expended in digging through snow was no higher than that ex- 

 pended during mild insect harassment. The average daily metabolic rate 

 was then calculated to be 2.2 times the standard fasting rate. Thus, dur- 

 ing the year the average daily metabolic rate in non-lactating caribou var- 

 ied from 884 to 1244 kJ kg"" ". 



Estimates of the production efficiency of each cohort of a caribou 

 population were made from data of Krebs and Cowan (1962) and Kelsall 

 (1968). In calves 3.0 and 4.6*^0 of gross energy intake were used for pro- 

 duction by males and females, respectively. Efficiencies declined to ap- 

 proximately 2% in animals between two and three years old and to zero 

 in animals older than five years. Energy secreted in milk was taken as 

 production of calves rather than a component of female production. The 

 main reason for the low efficiency of production is the amount of energy 

 required to support metabolism during winter, particularly from Decem- 

 ber to June when productivity is negative, and the animals lose weight. 



The metabolic requirement for milk production during the first 

 three weeks of lactation is very high— 10.5 to 12.6 MJ day"' or 40 to 50% 

 of the average daily metabolic rate. This energy is required during May 

 and June when primary production is negligible, and the predicted rates 

 of energy intake would be low. Preliminary studies on reindeer grazing in 

 shrub tundra in central Alaska indicate that the peak rate of milk secre- 

 tion can vary from 0.8 to 2.2 liters day"' (5.4 to 14.7 MJ day"'), depend- 

 ing on food intake (White, unpubl. obs.). Thus, lactation can be less 

 than optimal, and the growth rate and survival of calves may be related 

 to the diet. Data for grazing reindeer show that the growth rate of calves 

 depends on milk production for at least 50 days. After 50 days milk pro- 

 duction declines rapidly (Holleman et al. 1974), and females begin to 

 rebuild the nutrient pools in their bodies (Cameron and Luick 1972). 



The general demographic pattern and data on production and gross 

 intake were used to calculate population energetics for caribou (Table 



