Select Page

A Review and Synthesis of Effects of Alterations to the Water Temperature Regime on Freshwater Life Stages of Salmonids, with Special Reference to Chinook Salmon

Feb 22, 1999


Despite the great significance of the Columbia River Thermal Effects Study (a joint publication of the USEPA, NMFS, and AEC) and a compilation of temperature criteria and methods by the National Academy of Sciences in the 1970s that has been key EPA guidance on water temperature, too little attention has been placed on the key role of thermal pollution of river systems in reducing fish survival and production. The current review and synthesis of effects of water temperature on salmonids is an attempt to update important aspects of these earlier works in light of current ecological understanding. This revision is in terms of numeric criteria by species and life stage but, more importantly, is an explanation of the concepts that must be considered to fully protect salmonids from thermal effects under the Endangered Species Act.

Regulation of water temperature in salmon-bearing streams of the Pacific Northwest involves selection of appropriate biologically-based temperature standards and then creation of implementation procedures that insure that the biological intent is effected on necessary spatial and temporal scales. Selection of standards involves a thorough review of the effects of temperature on key life stages of species, variation in response among species and stocks, and an ability to estimate immediate and delayed biological responses from temperature statistics. Implementation involves consideration of problems such as where, when, and how to monitor water temperature in a watershed and developing a process for responding to violations of standards. This report deals primarily with a review of biological aspects of temperature in the environment, but the best understanding of the influence of temperature will not be effective unless implementation procedures are also meaningful. The tight interconnection between these two elements necessitated a broader review that incorporated a wide variety of spatial and temporal issues in fish ecology. In this context, full protection was viewed in terms of entire life cycle effects, single and multiple species, variation among stocks of a species, multiple environmental gradients, watershed to reach scales, and multiple biological responses (survival, growth, preference, fitness, reproduction, migration, swimming, feeding, etc.).

In consideration of the entire life cycle of spring chinook, it would be most preferable for adults to enter the Columbia River with water temperatures of 3.3-13.3&deg;C. Temperatures of 21.0&deg;C must be avoided because they represent thermal blockages and also are near adult upper incipient lethal temperatures. Temperatures >15.5&deg;C greatly enhance incidence of disease and mortality rate. Adults must be able to proceed rapidly upstream to holding areas without encountering migration blockages, excessive delays, and without excessive metabolic expenditures negotiating falls or ladders. Holding areas must have adequate cover and low temperatures to reduce metabolic losses and pre-spawning mortality. Holding temperatures <16.0&deg;C (maximum) at the time of egg maturation in holding females are essential for maximum egg viability and initiation of spawning activities. Temperatures <12.8&deg;C and declining provide best spawning conditions and highest survival upon egg deposition. High egg mortality can be expected if temperatures are <1.7&deg;C at egg deposition, although this is uncommon. Proper incubation temperature requires accumulation of necessary thermal units so that development rate results in best emergence timing. Prolonged autumn high temperatures or early spring warming can result in early emergence if they prematurely fulfill thermal unit cumulation. Temperature at emergence must be limited to 14.0&deg;C (max.) for initiation of feeding. Optimum juvenile growth occurs between 10 and 15.6&deg;C. Growth rate declines to zero at approximately 21&deg;C, but considering mortality rate when juveniles are fed at higher temperatures, zero net growth of the population occurs at a lower temperature (19.1&deg;C). The upper incipient lethal level for juveniles is 25.1&deg;C. In the field the distributional limit is approximately 22-24°C. Smoltification and outmigration proceed best at temperatures <12.2°C. It is vital for smolts to begin their downstream migration early enough to avoid higher temperatures that discourage migration. In the mainstem, temperatures >18.3&deg;C can result in desmoltification and residualism.

Temperature regulation and monitoring must be done at a watershed scale in order to fully protect the species found along the river continuum. Survey data are very useful in establishing the downstream limits of distribution. Most salmonids are restricted to stream zones having maximum temperatures <22-24°C. Bull trout and redband trout represent the lower and upper extremes in thermal tolerance. The general uniformity in thermal response among stocks of a species tends to obviate the need for geographically different biological standards to protect endangered species. Densities of various salmonid species have been shown to decline from a maximum under near optimum conditions to zero at the downstream distribution limit. Distribution is a function of temperature, but species are also distributed along multiple environmental gradients. Optimum habitat for a species involves temperature regime, but also is strongly influenced by channel gradient, substrate composition, pool frequency, large woody debris and other cover elements, oxygen concentration, and competitors or predators. In the context of management of an entire stream system, constituting the fish production unit of an endangered population, the most difficult challenge in application of any single quantitative standard to a stream is to not have the standard applied immediately to headwater areas. The effect of such management is to elevate temperature in the headwaters where temperature may then become ‘optimal’ but numerous other environmental factors may not be optimal, thereby losing the productive capacity of habitats in the normal center of distribution further downstream.

Survival and fitness of a fish population should be evaluated on multiple time scales. A conventional scale is a weekly temperature regime during periods of extreme temperatures. It is assumed that regulating the maxima will result in appropriate regimes in other seasons. A bioenergetic analysis involves evaluation of the effect of temperature regime during the entire life cycle (or freshwater phase) on the energy balance (i.e., energy storage, diversion to growth and reproduction versus energy costs attributed to thermal stress). Within shorter time frames the capacity to cope with environmental stress is a function of power budgeting. An organism with a high thermal load has a reduced scope for activity when confronted with instantaneous power demands (e.g., peak swimming, predator avoidance). Under a reduced scope, the organism is at higher risk of exhaustion, predation, or disease.



McCullough, D.A. 1999. A Review and Synthesis of Effects of Alterations to the Water Temperature Regime on Freshwater Life Stages of Salmonids, with Special Reference to Chinook Salmon. Seattle, Wash, U.S. Environmental Protection Agency, Region 10.



Report No.


Media Type

Inter-Agency Report