Table l. — Density of salmon eggs and alevins, and concentra- 

 tion of ammonia in intragravel waters of Sashin Creek, south- 

 eastern Alaska. Pink salmon eggs or alevins predominated, al- 

 though a few coho salmon, Oncorhynchus kisutch, eggs, <10%, 

 were occasionally present at the sample sites and are included in 

 the totals. Four surface samples were also measured, ranging 

 from 0.005 to 0.019 ppm total ammonia (average 0.013 ppm). 



upstream versus downstream locations) nor with 

 the density ofeggs and alevins (r = -0.17,P>0.10) 

 (Table 1). None of the five sample sites with the 

 highest ammonia concentrations was among the 

 five sites with the highest densities of eggs and 

 alevins. 



The lack of correlation of ammonia concentra- 

 tions with alevin densities may be due to the var- 

 iability of intragravel water flow. Intragravel 

 water flow varies considerably from site to site in 

 all streams and is affected by surface water veloc- 

 ity, volume of water flow, stream gradient, gravel 

 size, and obstructions such as trees or ice (Vaux 

 1968). Intragravel flow may differ in adjacent 

 redds (cross-stream or upstream-downstream). 

 Brickell and Goering (1972) found that ammonia 

 concentrations in surface waters of Sashin Creek 

 during the fall spawning were generally greater at 

 the downstream sites than at the upstream sites. 

 In contrast, we found no relation between concen- 

 trations of ammonia sampled in the spring from 

 intragravel waters at upstream and downstream 

 sites at Sashin Creek. Furthermore, the concen- 

 trations of ammonia in surface water in our study 

 were much lower and more uniform than the con- 

 centrations in the intragravel water. We conclude 

 that measurements of ammonia in surface water 

 are poor estimates of ammonia concentrations of 

 intragravel water. 



Twice each year, in early spring and in fall, am- 

 monia concentrations in salmon streams can be 



expected to reach levels potentially harmful to 

 salmon eggs and alevins. In the fall, a large mass 

 of decaying salmon carcasses may litter the 

 stream. Brickell and Goering (1972) measured 

 ammoniacal nitrogen in surface waters of Sashin 

 Creek during and after a heavy run of salmon and 

 found concentrations of ammonia to be greater 

 than concentrations of ammonia that we found in 

 surface waters in the spring. Brickell and Goering 

 concluded that the ammonia was from the decay- 

 ing carcasses and not from excretion by pre-eyed 

 salmon eggs. Pre-eyed salmon eggs have low 

 ammonia-excretion rates (Rice and Stokes 1975; 

 Bailey et al. 1980). Unfortunately, no samples of 

 intragravel water were measured in the study by 

 Brickell and Goering, but the potential for harm to 

 developing eggs is probably low because pink 

 salmon eggs are quite tolerant of ammonia at this 

 life stage (Rice and Bailey 1980). In early spring 

 when water flows are low and excretion rates of 

 developing alevins are maximum, high concentra- 

 tions of ammonia could result. We found concen- 

 trations of ammonia in some of the salmon redds to 

 be higher than the concentrations in concurrent 

 samples of surface water and even higher than the 

 concentrations reported by Brickell and Goering 

 in the surface water in the fall. 



Although the probability of exposure to high 

 levels of ammonia in spawning grounds is greatest 

 in the spring when alevins are most sensitive, the 

 highest level we observed was below dangerous 

 levels. The highest concentration of ammonia that 

 we found in the intragravel samples was 0.24 ppm 

 total ammonia, which is about 0.1 ppb of toxic 

 un-ionized ammonia at the pH and temperature of 

 Sashin Creek. This concentration is only about 

 one-tenth of the lowest concentration that affected 

 the size of fry resulting from alevins exposed to 

 ammonia for 61 days (Rice and Bailey 1980) and 

 about two-thirds of the maximum concentration 

 found in hatchery incubators containing unusu- 

 ally high densities of eggs (Bailey et al. 1980). Our 

 highest value for intragravel water exceeded the 

 highest concentrations found in surface water of 

 Sashin Creek (Brickell and Goering 1972) when 

 many decaying salmon carcasses were present. 



In subarctic and arctic streams where water 

 temperature and pH are low, it seems unlikely 

 that ammonia will accumulate in intragravel 

 waters to concentrations that will significantly 

 affect size or survival of salmon alevins. Ammonia 

 toxicity may be significant at higher tempera- 

 tures, especially in more alkaline streams. 



810 



