
                                  MEMORANDUM

Tetra Tech, Inc.
10306 Eaton Place, Suite 340
Fairfax, VA 22030
phone	703-385-6000
fax	703-385-6007


DATE: 		December 8, 2008

TO:			Paul Shriner, EPA
	
FROM:		Kelly Meadows, Tetra Tech

SUBJECT:	Analysis of swim speed data

Tetra Tech was tasked with researching the origins of (and continued applicability of) the 0.5 feet per second (fps) threshold for fish swim speed in reducing impingement and impingement mortality in the 316(b) regulations.  The sections below discuss the treatment of swim speed in the previous rules, as well as provide discussion to consider in the development of the revised Phase II rule. 

Background

Impingement is generally correlated to three factors: intake flow, intake velocity, and fish swim speed.  The latter two factors are closely related, as the ability of fish to evade impingement depends on the swimming ability of the individual fish and the intake velocity against which it is attempting to escape.  EPA has consistently recognized that regulating the intake velocity at cooling water intake structures (CWIS) is an effective way to minimize impingement impacts.

Where to Measure Intake Velocity

Intake velocity can be measured at a number of locations, each of which can have a significant impact on the effectiveness of the velocity threshold:

   * Far-field: Measured at a large distance from the intake structure. Typically, the area of effect of an intake structure dissipates over a relatively small distance, as it become lost in ambient currents in the waterbody.  Far-field velocity is unlikely to be a useful metric for reducing impingement.
   * Approach velocity: Measured at some distance in front of the screen (often defined as a few inches [EPRI 2000]), just at or before the point where velocities increase before passing through the screen.
   * Through-screen: Measured in the open area of the screen, as water and organisms pass through.  Through-screen velocity has been EPA's preferred metric for the Phase I, II, and III rulemakings.

Intake velocity measurements should be measured at the "first point of contact," where an organisms would be presented with the first (and perhaps best) opportunity to detect the intake structure and avoid it.  For example, a facility that employs an intake canal should measure the intake velocity at the mouth of the canal, as organisms can detect the intake and avoid the canal.  If the intake velocity was measured at the intake screen (and not the mouth of the canal), organisms may be drawn into the canal but be unable to return to the primary waterbody and may eventually be impinged.

Adding numbers to this example, a facility with an intake canal has an intake velocity of 1.0 fps at the mouth of the canal and 0.5 fps at its traveling screens near the facility.  In this case, organisms are not provided with the optimal opportunity to avoid being impinged, as they may be drawn into the canal due to the higher intake velocity.  Once in the canal, escape may be difficult and the organism may eventually be impinged, despite the lower intake velocity at the screen itself.

Another example would be a facility with an offshore intake structure that supplies cooling water to an onshore forebay.  The intake velocity should be measured at the offshore intake structure, as an organism drawn into the CWIS and forebay would have no chance of escape (except possibly through a fish return; however, survival rates for organisms passing through a fish return is often low).

Similarly, measuring the velocity of flow in the conduits behind the screen is not appropriate, as any organism behind the screen will not have an opportunity to escape.

Other Biological Factors Affecting the Swim Speed Threshold

Two other biological factors can also affect the ability of fish to escape an intake structure.  EPA's approach to account for these factors is also discussed.

   * Horizontal versus vertical flow vectors: Fish are generally unable to detect changes in flow or velocity in a vertical direction.  Any intake that withdraws water vertically (e.g., an uncapped offshore intake) poses a significant impingement hazard.
         o All intake velocity measurements should be measured in the horizontal direction.

   * Categories of swim speed: Fish typically exhibit two general types of swimming -- prolonged continuous swimming or burst swimming (EPRI 2000).  The former is the most common and is slower.  Burst swimming is usually induced by an outside stimulus (e.g., chasing prey, escaping a disturbance) and is much higher, but typically the organisms only travel a short distance.
         o Burst swim speeds are higher and may provide fish with an improved ability to react to stimuli (including intake screens).  However, even with short periods of burst swimming, a fish may avoid initial contact with the screen but may be unable to eventually escape the area of the intake structure.  As such, EPA's conservative approach of using swim continuous/prolonged swimming speeds in its analysis is appropriate.

Phase I Rule

The Phase I rule established requirements for all new intakes to have a maximum design intake velocity of 0.5 fps.  The final rule describes the analysis EPA used: precedent from previous Federal studies, existing data, and an analysis of fish length versus swim speed. (See 66 FR 65274.)  The information collected indicated that a 0.5 fps threshold would be protective of 96% of the species and life stages found in the data.

In addition, the Electric Power Research Institute (EPRI) submitted a 2000 study with its comments on the Phase I Notice of Data Availability (NODA) titled "Technical evaluation of the utility of intake approach velocity as an indicator of potential adverse environmental impact under Clean Water Act Section 316(b)."  This comprehensive review of swim speed literature contained most of the data used by EPA in justifying the 0.5 fps threshold.  EPRI agreed that intake velocity was an appropriate regulatory criterion, and that while it preferred a site-specific approach to setting intake velocity limits, that a limit of 0.5 fps was a useful threshold for screening out significant impingement events at CWISs.

In the NODA and in a memo to the record, EPA discussed the intake velocities at existing facilities, noting that most facilities constructed in the last 15 years already meet the 0.5 fps threshold.

EPA's intake velocity threshold was challenged in litigation but was upheld by the Second Circuit.

Phase II and Phase III Rules

The 2004 Phase II rule used the same data, analyses and conclusions presented in Phase I to support a compliance alternative where an intake at a facility with a design through-screen velocity of 0.5 fps meets the impingement requirements.  Similarly, the proposed Phase III rule utilized the same regulatory framework as the Phase II rule, including the 0.5 fps intake velocity threshold.  In the final Phase III rule, EPA opted not to regulate land-based facilities, but continued to impose the intake velocity requirements on certain offshore facilities.

Revised Phase II Rule

In response to the Second Circuit decision, EPA is revising the Phase II rule.  As part of this rulemaking, EPA opted to briefly re-examine the basis for the 0.5 fps threshold, as well as conduct additional screening analyses.

In reviewing the swim speed data in the record, the previous conclusions continue to be supported by the data.  

First, many intakes already meet this standard, thereby reducing the burden of meeting the requirement.  In fact, many intake technologies installed today (e.g., cylindrical wedgewire screens) are specifically designed to meet the 0.5 fps threshold.

Second, as described above, a 0.5 ft/sec through-screen velocity would be protective of 96% of species.  It is possible that a lower intake threshold may be appropriate under certain conditions (e.g., Alaska's limits to protect salmonids); however, on a practical level, it is likely not appropriate to set a threshold lower than 0.5 fps on a national level.

Some have argued that the data on intake velocity and fish swim speed support a higher threshold (up to 1.0 fps), but it should be recognized that intake screens rarely operate at optimal conditions.  Given the potential for screen clogging and debris loading (which would reduce the open area of the screen and increase the through-screen velocity even further), the 0.5 fps threshold also provides for an appropriate safety margin for aquatic organisms.

Analyses were conducted to determine if the velocity threshold should vary by waterbody type.  The swim speed data from the EPRI report was plotted by fish assemblage, a categorization of fish species by waterbody type (e.g., Pacific Ocean, rivers in the Eastern U.S., etc.). (See attached.)  These plots did not show any clear differentiation of swimming ability between fish in the various waterbodies nor did any waterbody type appear to be any more vulnerable than another; it is therefore reasonable to conclude that the 0.5 fps national intake velocity limit is appropriate for all waterbody types.


References

Alaska Department of Fish and Game (ADF&G). 1998. Water Intake Structures: an Alternative to Traditional Screened-Box Enclosures for the Protection of Fish. Prepared by Robert F. "Mac" Mclean for the ADF&G Habitat and Restoration Division. March 20, 1998. Technical Report 97-8.

Electric Power Research Institute (EPRI). 2000. Technical Evaluation of the Utility of Intake Approach Velocity as an Indicator of Potential Adverse Environmental Impact under Clean Water Act Section 316(b). EPRI, Palo Alto, CA. 1000731.

Environmental Protection Agency (EPA). 2001. Preliminary Data Analyses Using Responses from the Detailed Industry Questionnaire: Phase II Cooling Water Intake Structures (Draft). DCN 4-4023C.





