Marine Mammals; Environmental Assessment on Preventing California Sea Lion Foraging and Predation on Salmonids at the Willamette Falls, Oregon

DEPARTMENT OF COMMERCE

National Oceanic and Atmospheric Administration

[I.D. 120497B]

Marine Mammals; Environmental Assessment on Preventing California

Sea Lion Foraging and Predation on Salmonids at the Willamette Falls,

Oregon

AGENCY: National Marine Fisheries Service (NMFS), National Oceanic and

Atmospheric Administration (NOAA), Commerce.

ACTION: Notice of availability and finding of no significant impact.

-----------------------------------------------------------------------

SUMMARY: NMFS announces the availability of an Environmental Assessment

(EA) that examines the

[[Page 56]]

environmental consequences of preventing California sea lion foraging

and predation on salmonids at the Willamette Falls in Oregon. The

proposed action consists of non-lethal measures that are authorized

under the Marine Mammal Protection Act (MMPA). NMFS has evaluated the

environmental consequences of the proposed action and has concluded

that it is unlikely to result in any significant impacts on the human

environment and, therefore, has made a finding of no significant impact

(FONSI).

ADDRESSES: A copy of the final EA may be obtained by writing to William

Stelle, Jr., Regional Administrator, Northwest Region, NMFS, 7600 Sand

Point Way NE, Seattle, WA 98115.

FOR FURTHER INFORMATION CONTACT: Joe Scordino (206)526-6143, or Tom

Eagle (301)713-2322.

SUPPLEMENTARY INFORMATION: The National Marine Fisheries Service, in

cooperation with the Oregon State Department of Fish and Wildlife

(ODFW), prepared an EA that examines the environmental consequences of

three alternatives for preventing sea lion foraging and predation on

returning adult salmonids and outmigrating smolts at Willamette Falls:

(1) No action; (2) non-lethal removal of California sea lions (proposed

action); and (3) lethal removal of sea lions foraging at the Falls. The

proposed action is to implement a program of non-lethal measures to

prevent sea lion predation at the Willamette Falls while continuing to

monitor the resource conflict at this site. The proposed action is

authorized under section 109(h)(1)(C) of the MMPA, which allows the

non-lethal removal of nuisance marine mammals by local, state, and

Federal officials.

A draft EA was made available for a 30-day public comment period.

NMFS published a notice in the Federal Register on March 13, 1997 (62

FR 11845), that announced the availability of the draft EA and

requested public comments. Seven public comments were received, and the

EA was revised in response to the comments. A summary of the comments

received and responses to the comments are given here:

Comment 1: The situation at the Willamette Falls does not warrant

lethal removal.

Response: Lethal removal of sea lions at Willamette Falls is not

proposed because it has not been authorized under section 120 of the

MMPA. Section 120 provides a process for a state to obtain authority

for lethal removal, but Oregon has not applied for this authority.

Comment 2: The proposed action does not address all of the

potential factors causing depletion of salmonids in the system. One

commenter suggested that causes of salmonid population decline should

be investigated, and another recommended that NMFS and ODFW evaluate

and assess predation in comparison to other factors.

Response: The State is addressing other factors that may be

affecting the decline of salmonids in the Willamette River basin;

however, the principal cause for decline appears to be the reduced

ocean survival. The scope of the EA and the proposed action, which

complements State efforts to address other factors affecting salmonids,

is limited to addressing the increasing presence of California sea

lions foraging at the Falls and the prevention of predation from

escalating to a point where it may impact salmonids, especially if the

salmonid stocks remain low or decline further.

Comment 3: The proposed action is consistent with general state

fish and wildlife authorities.

Response: NMFS agrees.

Comment 4: The EA does not show that predation has caused the

decline of the runs or is likely to have caused a negative effect on

the run. Commenters noted that the decline of steelhead and spring

chinook salmon occurred before sea lions could have had any noticeable

effect, and, therefore, actions to reduce sea lion predation are

unwarranted. One commenter supported the no-action alternative because

sea lions are not the cause of the decline.

Response: NMFS agrees that sea lion predation is not the cause of

the decline; however, if action is not taken to address increasing

foraging by sea lions, predation may increase to a point where

predation is impacting salmonid stocks in the Willamette River,

especially if the number of returning adults remains low or declines

further.

Comment 5: An Environmental Impact Statement (EIS) should be

prepared in order to provide a more comprehensive appraisal of this

action.

Response: An EIS is not required for this action because the

environmental consequences of non-lethally removing a few sea lions

from the Willamette Falls area will not result in any significant

impact to the environment.

Comment 6: The removal (lethal or non-lethal) of sea lions could

result in increased predation. Commenters were concerned that the

removed sea lions will be quickly replaced by other animals. One

commenter also was opposing the use of underwater firecrackers or other

methods which may inadvertently result in an increase of predation in

the long term because these methods have not been shown to have lasting

effectiveness in other applications.

Response: Because sea lions are opportunistic predators, predation

patterns develop relative to animal presence, prey availability, and

vulnerability. Based on observations at the Ballard Locks in

Washington, different methods of sea lion removal may be more or less

effective in reducing sea lion presence or in reducing the

vulnerability of fish to predation, depending upon the number of

animals involved and the location or circumstances of the predation.

NMFS believes that the proposed action will prevent sea lion foraging

and predation on salmonids at the Willamette Falls because the number

of sea lions to be removed is still small, the patterns of predation do

not appear well established, and the area is geographically remote from

where sea lions normally occur; thus, inseason replacement is unlikely.

In contrast, the alternative of taking no action to prevent foraging

and predation will likely result in escalation of the problem because

animals already present will become more effective at catching

salmonids at the site, and new animals will learn these effective

strategies as they arrive.

Comment 7: An additional alternative should be added to investigate

the real and primary cause of the fish run declines (e.g., hatchery

fish competition, fish passage problems due to construction and

operation of the fishway and dam, water, and general habitat

degradation) and to implement solutions to mitigate them.

Response: The scope of the proposed action is limited to preventing

sea lion predation; measures to address other causes of salmonid

declines are underway by the State, and a separate alternative on such

actions is unnecessary and outside the scope of this action. Natural

production (wild spawning) of spring chinook is low, owing primarily to

lost spawning habitat. As mitigation for lost wild production, the

majority of the spring chinook are hatchery produced. Hatchery produced

spring chinook originate from native stocks and are virtually

indistinguishable from wild spawners. Hatchery release practices and

harvest regulations for hatchery steelhead are designed to minimize

competition for available wild spawning habitat. Ocean productivity

over the past several years has been influenced by a multi-year

climatic event (El Nino) that has impacted ocean survival of

[[Page 57]]

salmonid stocks, including those returning to the Willamette.

Nonetheless, if numbers remain low or decline further, the potential

for sea lion predation to have a significant impact remains real, and

non-lethal removal actions are warranted.

Comment 8: No actions should be taken with sea lions until the

proposed non-lethal deterrents are tested and an implementation plan is

developed. The commenter recommended that an independent group of

pinniped and fisheries biologists be established to oversee the

development of a monitoring and research plan for evaluating the

effectiveness of various non-lethal deterrents.

Response: NMFS has tested and implemented the non-lethal deterrence

measures in pinniped interactions elsewhere on the Pacific Coast, with

no discernable deleterious affects on California sea lions or serious

injuries to personnel. Implementation of the individual measures will

be dependent on available resources during a given season. NMFS will

continue to request assistance from independent experts when necessary;

however, the formation of an oversight committee is not necessary or

warranted for actions taken under section 109 of the MMPA.

Comment 9: Non-lethal removal should not be authorized under

section 109 (h)(1)(C) because the EA does not specify the numbers of

animals to be taken, specify the exact methods to be used, specify the

risk of injury or mortality to individual animals, provide evidence

that sea lion predation is adversely affecting fish passage, or provide

scientific data on the degree of impact of sea lion predation on the

affected stocks.

Response: Section 109(h)(1)(C) of the MMPA authorizes the taking of

marine mammals by public officials during the performance of their

official duties. This authorization does not require the specification

of the number of animals to be taken, exact methods, degree of risk, or

evidence that the animals to be taken have exceeded some pre-determined

behavioral threshold. However, some of these factors would need to be

considered for authorization for the lethal removal of individually

identifiable pinnipeds under section 120 of the MMPA.

Comment 10: The proposed action does not appear likely to

contribute to the enhancement of Willamette River fish runs. One

commenter stated that non-lethal removal of sea lions can only give a

false hope of salmonid recovery because sea lions have not been

determined to be negatively affecting the fish runs.

Response: The proposed action is to reduce or eliminate sea lion

predation on salmonids and to prevent it from escalating to a point

where it may negatively impact salmonid runs at this site. Predation is

one of the factors affecting survival of adult spawners, and reduction

or elimination of this mortality factor should, therefore, contribute

to the enhancement and recovery of the involved salmonid runs.

Comment 11: Neither the regulations nor the statute provides a

definition of what constitutes a ``nuisance animal,'' and, lacking a

definition, the commenter found it difficult to evaluate whether sea

lions at Willamette Falls are a nuisance animal.

Response: NMFS acknowledges that neither the statute nor the

implementing regulations provide a specific definition for ``nuisance''

marine mammal. However, the legislative history of the MMPA includes

removal of seals from a fish ladder as an appropriate interpretation of

the nuisance animal provision. Sea lions constitute a nuisance at the

Falls because their foraging and predatory behavior is contrary to the

purpose of the fishway to pass fish upstream, and uncontrolled

predation at freshwater sites outside the normal habitat of sea lions,

especially where fish are congregated and vulnerable to predation, is

contrary to conservation efforts for recovering depressed and declining

fish stocks.

Comment 12: The EA incorrectly states that Willamette Falls is

outside the normal range of California sea lions.

Response: As the California sea lion population has increased since

the early 1970s, reports of animals occurring in areas previously not

documented have also increased. NMFS is not aware of any documented

historical occurrence of California sea lions at the Willamette Falls

other than the sightings noted in the EA and, therefore, considers the

occurrence of sea lions far upriver at the Falls in a freshwater

environment to be beyond the normal range.

Comment 13: The nuisance determination is not appropriate because

the effect of sea lions on fish runs may be only negligible.

Response: Section 109 of the MMPA does not establish a threshold of

damage that must be exceeded in order for a determination to be made on

whether an animal is a nuisance. The non-lethal removal measures

proposed are to reduce or eliminate sea lion predation on salmonids and

to prevent it from escalating to a point where it may negatively impact

the fish runs. If lethal removal were to be used under section 120 of

the MMPA, then it would be necessary to show that individual pinnipeds

are having a significant negative impact on the status or recovery of

salmonid populations that are listed under the Endangered Species Act

(ESA) or approaching listing.

Comment 14: To effectively recover the salmonid populations,

additional restrictions should be placed on commercial and recreational

fisheries, barriers to passage should be removed, spawning habitat

should be restored, hatchery operations should be improved, and power

generating operations should be evaluated. The commenter recommended

that the burden to conserve fish stocks should be distributed

proportionately among all human causes before penalizing sea lions for

eating fish.

Response: The State is addressing factors affecting the status of

salmonid populations, including restricting commercial and recreational

fisheries. Reducing or eliminating sea lion predation will be

complementary to other State efforts to enhance and restore salmonid

runs. In regard to barriers to passage, the Willamette Falls is a

natural barrier to fish passage and the fishway was constructed to

enhance adult passage to spawning habitat.

Comment 15: The design and construction of existing fishways should

be re-evaluated to devise ways for salmonid species to avoid sea lion

predation.

Response: Plans are underway to modify the fishway to improve fish

passage. An engineering evaluation of the fishway was completed in

1992, and that report is now referenced in the EA. Fishway design and

alteration information were not included in the draft EA because

contract work and planning processes for fishway maintenance and

modification are proceeding separately and are outside the scope of the

EA. The area of focus for preventing sea lion foraging and predation on

salmonids is outside the fishway in adjacent areas including below the

Falls.

Comment 16: The monitoring program should have been implemented

before an EA was considered, rather than basing the proposed action on

undocumented observations.

Response: The proposed action is based on results of observations

by biologists in 1995 as well as on ODFW-conducted monitoring programs

in 1996 and 1997 (as described in the EA), which documented sea lion

predation on steelhead and spring chinook.

Comment 17: Introduced salmonid runs do not warrant the

conservation protection of native runs.

Response: Introduced salmonid runs in the Willamette basin, such as

summer

[[Page 58]]

steelhead, have been made possible by the improved fish passage

conditions afforded by the construction of the fish passage facility.

These fish have been added to increase and support fishing

opportunities in response to public demand. Sport fishing for salmonids

is a popular and economically significant industry in the Willamette

River basin. The introduced runs have been maintained over several

decades without detrimental effect to native Willamette River basin

salmonid runs because of hatchery release practices and harvest

regulations. Timing of the two steelhead stocks overlap below the

Falls, and sea lions are, therefore, likely to intercept both native

and non-native stocks when foraging.

Comment 18: The methods of capturing and relocating sea lions are

inadequately described.

Response: The EA has been revised to provide additional information

on capture and translocation of sea lions. More detailed information on

California sea lion captures and relocation is included in prior EAs

prepared by NMFS (referenced in the EA) for non-lethal measures

implemented at the Ballard Locks, and these EAs are available to the

public.

Comment 19: The non-lethal options should not be considered safe

because they have not been adequately tested.

Response: The non-lethal options included in the proposed action

have been used previously in other locations and will be implemented

under protocols to ensure safety to sea lions as well as personnel

involved. The possibility of a sea lion mortality resulting from the

proposed measures is very remote.

Comment 20: The use of underwater firecrackers may deafen sea

lions.

Response: Observations at the Ballard Locks show that individual

sea lions continue to respond to noise stimuli in spite of repeated

exposures to firecrackers. Nonetheless, it is possible that a close

exposure to an exploding firecracker may cause temporary or possibly

permanent deafness, so dispatch of firecrackers should be used with

caution.

Comment 21: Aversive conditioning should not be used because this

technique did not successfully deter sea lions at the Ballard Locks.

Response: Aversive conditioning was previously found to be

ineffective for use at the Ballard Locks because of difficulties in

administering repeat treatments, which are necessary to achieve lasting

effect. This method has been included in the proposed action because

repeat treatment opportunities may be available at Willamette Falls.

Comment 22: The EA incorrectly states that sea lions have

negatively affected steelhead at the Ballard Locks.

Response: Based on extensive studies since 1985, NMFS has

determined that predation by sea lions is a principal factor affecting

the spawning escapement of returning adult winter steelhead in the Lake

Washington basin (migrating through the Ballard Locks). The

determination is well documented in several EAs prepared by NMFS and by

the Washington Department of Fish and Wildlife.

Comment 23: If sea lions are deterred from the area, it should be

done in a minimally invasive and humane manner. One commenter

recommended that NMFS should limit the study and implementation of sea

lion deterrence measures to those that are humane and realistically

promising (e.g., alternative barrier designs, expanded acoustic

deterrence devices).

Response: Section 109 of the MMPA specifies that the taking of a

marine mammal by public officials during the performance of their

duties shall be accomplished in a humane manner. The non-lethal

measures included in the proposed alternative are not expected to cause

mortality or serious injury and are intended to have the desired effect

of removing foraging sea lions from the area. Additional use of barrier

gates in other entrances to the fish ladder will be considered if

observations indicate that sea lions are entering the fishway through

those entrances. The use of acoustic deterrent devices is included in

the proposed action.

Comment 24: The funds spent on sea lions should be used for such

other factors as fish passage, competition with hatchery fish, and

habitat concerns.

Response: The State is addressing other factors that may be

affecting salmonids in the Willamette River basin, and the removal of

sea lions will complement those efforts. Non-lethal removal measures

will be combined with the NMFS-funded sea lion monitoring program to

minimize costs. Efforts to improve and update the fishway are

proceeding under different funding.

Comment 25: The EA should provide more information on why fish use

fishway entrance 1 so much less than other ladder entrances.

Response: It is difficult to fully assess passage through entrance

1 in comparison with the other three fishway entrances because of

fishway configuration. The different entrances have been constructed to

provide passage opportunities for fish under a wide range of flow

conditions. Passage conditions during the spring result in greater

passage by spring chinook and steelhead through fishway entrance 2,

whereas fall chinook more frequently use fishway entrance 1. The EA has

been modified to provide this clarification.

Comment 26: The goal of resource managers should be the restoration

of native fish runs that have declined rather than reducing sea lion

predation.

Response: NMFS and ODFW agree that the restoration and maintenance

of native fish populations are important goals, and the State is active

in addressing these goals. Prevention of sea lion foraging in locations

where declining runs are concentrated and vulnerable does not conflict

with this goal.

Comment 27: The construction of dams is the single most likely

cause for salmonid declines, not sea lion predation.

Response: Dam construction in the Willamette River basin has been

completed for decades, and salmonid stocks have been maintained through

successful hatchery practices and fishery regulation. Low ocean

survival conditions over an extended period have affected returns in

recent years in spite of stable hatchery production.

Comment 28: The capture and relocation of sea lions are unlikely to

be successful and will not significantly benefit salmonids passing

through the Willamette Falls fishway. The commenters suggested new sea

lions would probably replace those that have been removed.

Response: NMFS agrees that previous translocation efforts with

California sea lions from the Ballard Locks have not been totally

successful. However, due to the distance inland to the Falls and the

small numbers of animals found far upriver, other sea lions may not

immediately replace animals that have been deterred or removed from the

area of the Falls.

Comment 29: Because experience with the use of the partially

submerged cage trap is inadequate, raising concerns for the safety of

personnel and the possible drowning of sea lions exist.

Response: The trap design maintains open air space above the

surface of the water to allow a captured animal to surface and breathe,

thereby negating a concern for animals drowning. The trap was

successfully used to capture and handle an adult harbor seal without

mishap or injury.

Comment 30: Active capture techniques will present high risk to sea

lions and humans.

Response: Techniques that involve an elevated level of risk for the

animals, such as tangle nets and anesthetizing

[[Page 59]]

drugs, are not proposed for use at Willamette Falls because protocols

for their implementation in the moving river environment have not been

developed. The final EA has been modified to clarify that active

capture using tangle nets in the river is not proposed.

Comment 31: Non-lethal removal of sea lions should not occur until

the salmonid stocks are threatened with extinction by predation.

Response: Section 109 of the MMPA does not require that salmonid

stocks must be approaching an endangered status before non-lethal

taking of sea lions can occur. The intent of the proposed action is to

be proactive and prevent predation from increasing to a point where it

may have a negative impact on the salmonid stocks.

Comment 32: The EA should provide more detail on the dams,

hatcheries, rivers and tributaries, river flows over time, fluctuations

in salmonid populations, numbers of salmonids using the locks, and

suitable conditions for passage. The commenters also stated that it

would be helpful if the document was expanded to explain the operation

of the locks, the paper mill and power generation, and the allocation

of water between fish passageways, and to provide more information on

genetic relationships of runs, limiting factors on salmonid

populations, water quality or industrial outflows, redd counts, habitat

considerations, harvest regulation, and hatchery surpluses.

Response: The EA has been modified to address additional background

information, and references that provide more details have been

incorporated in the EA.

Comment 33: The information on fish runs and passage should be

presented in a tabular format for clarification.

Response: The EA has been modified to include tables on spring

chinook and steelhead runs and passage.

Comment 34: The EA does not demonstrate that sea lions are having

a significant, deleterious effect on passage.

Response: Non-lethal removal of sea lions from the fish passage

facility are authorized under section 109(h), which does not require a

demonstration that a significant, deleterious effect is occuring;

however, NMFS and ODFW have investigated fish passage at the Willamette

Falls facility. Observations suggested that sea lions were adversely

affecting fish passage by foraging at the entrance to the fish ladder

and preventing access, and consuming and dispersing adult salmonids

that were attempting to enter the fishway to progress upstream. Until a

barrier was installed in entrance 1 to the fish ladder, sea lions were

foraging on salmonids inside the fish ladder, thereby preventing fish

passage.

Comment 35: The EA should describe the possibility that the

California sea lion population, with its population growth, may be

poised for a population crash.

Response: There is no evidence of density dependent signals to

indicate that the sea lion population is approaching carrying capacity.

When that occurs, the population will fluctuate in response to factors

that limit continued growth.

Comment 36: The EA should explain why Willamette River chinook

salmon are candidates for listing under the ESA.

Response: A coastwide status review of chinook salmon on the

Pacific coast is in progress to determine the status of chinook salmon

populations with respect to the ESA; therefore, until the status review

is completed, Willamette River spring chinook are considered candidate

species under the ESA. The EA has been modified to include this

clarification.

Comment 37: The EA does not specify which run of steelhead was

consumed by sea lions.

Response: Winter and summer steelhead are present below the Falls

concurrently, and observers are not able to differentiate steelhead

when predation is observed.

Comment 38: Summer steelhead are hatchery-produced fish with no

shortage of availability; management strategies can provide flexibility

for the time being.

Response: The focus of the proposed action is to prevent predation

on winter steelhead and spring chinook, and summer steelhead are

present during the same period. Nonetheless, the summer steelhead

population also has declined in spite of hatchery production due to

reduced ocean survival conditions that are also affecting winter

steelhead and spring chinook salmon. If ocean survival conditions do

not improve and run numbers continue to decline, management options

will continue to erode and hatchery operations could be jeopardized.

Comment 39: The EA incorrectly states that there is no controversy

or uncertainty on the effects of the proposed non-lethal removal

measures.

Response: The proposed action is to use non-lethal measures that

have been used and assessed at the Ballard Locks. These actions have

been demonstrated to have no adverse effect on California sea lions,

and, therefore, there is no scientific controversy or uncertainty on

the effects of the proposed non-lethal removal actions. The final EA

includes a complete description of the finding of no significant impact

of the proposed action.

Comment 40: The decline in winter steelhead from 1995 to 1996 was

reported as 72 percent, but it should be 62 percent.

Response: The steelhead run declined from 4,693 in 1995 to 1,801 in

1996, which is a 62 percent decline. The EA has been corrected.

Comment 41: The total time that sea lions were present in 1995 and

an estimate of total predation are not in the EA.

Response: Observations in 1995 were quite limited and no data were

collected on the total time spent foraging by sea lions that year;

therefore, no extrapolation of predation was attempted. An estimated

kill rate for the limited time observed in 1995 is included in the EA.

Comment 42: The EA mischaracterizes animal protection groups'

support for the no-action alternative because the benefit is that sea

lions would not be disturbed.

Response: NMFS has received comments favoring no action to prevent

sea lion foraging and predation, and the EA has been modified to

reflect this.

Comment 43: The EA incorrectly states that the no-action

alternative will likely result in a negative reaction by a large sector

of the public. The commenter suggested that this applies only to the

opinions of fishers.

Response: NMFS and ODFW have received numerous telephone calls from

members of the public requesting that the resource agencies take some

action to remove sea lions from Willamette Falls. The characterization

of total representation in comparison to general population has been

deleted from the EA.

Comment 44: The EA is not correct that many people would resent

their tax dollars being spent on hatchery production that results in

food only for sea lions. The commenter felt that many people would

resent tax dollars spent on non-lethal removal of sea lions.

Response: NMFS and ODFW have received numerous complaints from

members of the public regarding the past lack of action by resource

agencies to stop sea lions from feeding on salmonids at Willamette

Falls while fisheries are being restricted and fish numbers are low.

The EA has been modified to indicate that comments have been received

favoring no action as well.

Comment 45: The EA should provide more detailed information on the

barrier gate and its effectiveness. One

[[Page 60]]

commenter noted that observations made at fishway entrance 1 indicate

that sea lions commonly forage at the face of the barrier gate, and out

to about 10 feet (3.048 meters) below the barrier. One commenter

questioned whether the barrier gate could be expanded from riverbank to

riverbank to keep sea lions out of area.

Response: The EA has been modified to include additional

observations on the barrier gate. The barrier gate prevents predation

from occurring within the fish ladder at fishway entrance 1, but it has

not stopped sea lions from foraging at the face of the barrier and

areas adjacent to the fish ladder entrance. The installation of barrier

gates at other fish ladder entrances will be assessed if foraging

inside those entrances is noted. A physical barrier across the

Willamette River is not feasible or practical.

ACTION: The EA has been modified as described in the responses to

the comments. NMFS has evaluated the environmental consequences of the

alternatives and has concluded that the proposed action is unlikely to

result in any significant impacts on the human environment and,

therefore, has made a finding of no significant impact (FONSI). The EA

and FONSI have been prepared in accordance with National Environmental

Policy Act (NEPA) and with implementing regulations at 40 CFR parts

1500 through 1508 and NOAA guidelines concerning implementation of NEPA

found in NOAA Administrative Order 216-6.

Copies of the EA and FONSI are available (See ADDRESSES).

Dated: December 22, 1997.

Hilda Diaz-Soltero,

Director, Office of Protected Resources, National Marine Fisheries

Service.

[FR Doc. 97-34145 Filed 12-31-97; 8:45 am]

BILLING CODE 3510-22-F