Bibliography | Background | Hypotheses | Home

Water Quality: Setting Appropriate Temperature Standards 

Temperature Main Page

Temperature standards or objectives for most streams in the North Coast of California have not been officially defined beyond that stated in the Water Quality Control Plan for the North Coast Region (1996): 

       “Waters shall not be altered unless it can be demonstrated to the satisfaction of the Regional Water Board that such alterations in temperature does not adversely affect beneficial uses” and “at no time or place shall the temperature of any COLD water be increased by more than 5^oF above natural receiving water temperature.”  The establishment of water temperature standards and targets is challenged by complexity associated with different tolerances by species, different tolerances and times of exposure according to life stage, and different effects of temperature based on magnitude and time of duration. 

 Essig (1999) cautioned against setting temperature standards for too large a geographic area or for multiple species.  Setting one temperature threshold for large river basins may not be appropriate since higher order streams may warm naturally due to increased width and longer exposure to warm air temperatures (Lewis et al., 2000), yet all lower order streams should be evaluated with regard to tolerances of the single, most-sensitive species known to inhabit the basin (Essig 1999, Welsh et al 2001).  

While single-point maximas greater than 20° C and up to 26° C reflect increasing certainty of mortality for salmonids, much lower temperatures can cause impacts to growth and survival. Temperature indices that reflect some duration of high temperature can therefore be more relevant than single point maximas.   The most popular index is the maximum of the 7-day average temperature, also called the maximum weekly average temperature (MWAT) or the “7-day average”.  [ Note:  MWAT discussed here is different from the MWAT used by some researchers (e.g. Armour 1991) to establish laboratory-based indices for temperature tolerance]  An alternative is the maximum of the 7-day maximum temperature, also called the maximum weekly maximum temperature (MWMT) or “7-day maximum”. Comparison of MWAT and MWMT shows that MWMT better reflects  transient water temperature peaks (Welsh et al., 2001) and any acute effects of the single point maximum temperature. Also, see a comparison of three temperature variables for a better understanding of common water temperature metrics.

Many standards or targets have been proposed.  For a recent review, see Sullivan et al. (2000). The National Marine Fisheries Service (1996) characterized salmonid habitat as "at risk" when temperatures during spawning exceeded 15.5^o C and temperatures during rearing exceed 17.8 C.  The Washington State Department of Ecology (Hicks 2000) recommended that water temperatures remain below a 7-day maximum of 14-17° C to protect both returning adult salmon and rearing juvenile coho salmon and that maximum temperatures not exceed 20° C.   Such recommendations are based on a synthesis of available information.  

In northern California, both Welsh et al. (2001) and Hines and Ambrose (1998) found that coho salmon did not persist where the 7-day maximum temperature exceeded 18.3° C for any length of time.  Welsh et al. found similar results for a 7-day average of 16.8° C.  These simple investigations point to stream temperature levels that can be viewed as thresholds for the extirpation for endangered or threatened salmonids.  It is important, however, to remember that stream temperature effects rearing salmonids in interaction with many other factors, all of which vary with species and location.

Sullivan et al. (2000) conducted a risk assessment of temperature effects for coho salmon and steelhead trout based on growth potential. Although data from nineteen streams in Washington were used to develop the model, temperatures in those streams represented the range of temperatures found in the Pacific Northwest. The study is uniquely important for quantifying sublethal impacts of higher than optimal temperature. The principal result of the model is a chart showing percent reduction in maximum growth with reference to annual maximum temperature, MWAT, and MWMT.  Temperatures associated with the absence of coho salmon in north coastal California (18.3° C MWMT) correspond with somewhere between 10% and 20% reduced growth.