Salmon Findings of Fact Judge Oliver Wanger

Case 1:09-cv-01053-OWW-DLB Document 347 Filed 05/18/2010 Page 1 of 134
1 UNITED STATES DISTRI CT CO URT

2 FOR THE EASTERN DISTRICT OF C ALIFORNIA
3
4
5 The Cons olidated Sal monid 1:09-cv- 1053 OWW DLB
Cases
6
FINDINGS OF FA CT AND
7 CONCLUSIONS OF LAW R E:
PLAINTIFFS’ RE QUEST FOR
8 PRELIMINARY IN JUNCTI ON
(Docs. 161 & 2 30)
9
10
I. INTRODUCTION
11
Plaintif fs San Luis & Delta Mendota Water Authori ty
12
(the “Au thority”) an d Westlands Water District
13
(“Westla nds”) (colle ctively “San Luis Plaintiffs” ) seek a
14
15 Temporar y Restrainin g Order (“TRO”) 1 and a P relimi nary
16 Injuncti on (“PI”) ag ainst the implementation of
17 Reasonab le and Prude nt Alternative (“RPA”) Action IV.2.1

18 set fort h in the Nat ional Marine Fisheries Servic e’s
19
(“NMFS”) June 4, 200 9 Biological Opinion (“2009 S almonid
20
BiOp”), which addres ses the impacts of the coordi nated
21
operatio ns of the fe deral Central Val ley Pr oject (“CVP ”)
22
and Stat e Water Proj ect (“SWP”) on the Central Va lley
23
24 winter-r un and sprin g-run Chi nook salmon, Central Valley
25 1
Pl ai nt if fs ’ re qu es t for a T RO a ga i nst t he i mm in en t
imp le me nt at io n of A ct ion I V. 2. 1, w h ich t oo k ef fe ct a s of Ap ri l 1,
26 Doc . 23 3, w as d en ie d for t he r ea so n s s ta te d in o pe n co ur t o n th e
rec or d on M ar ch 3 1, 2 010 . D oc . 30 6 . Th e de ni al o f a TR O m ot io n is
27 not d is po si ti ve o f th e m er it s of a rel at ed m ot io n fo r pr eli mi na ry
inj un ct io n. Se e Of fi ce of P er so nn e l M an ag em en t v. A m. F ed’ n of
28 Gov ’t E mp lo ye es , 47 3 U.S . 13 01 , 13 0 5 ( 19 85 ).
1
1 steelhea d, Southern Distinct Population Segment o f Green

2 Sturgeon , and Southe rn Resident Killer Whales (“L isted
3
Species” ). Both mot ions were filed F ebruary 22, 2010.
4
Docs. 23 0, 233.
5
Plaintif fs State Wat er Contractors, Stockton East
6
Water Di strict, Oakd ale Irrigation District, and South
7
8 San Joaq uin Irrigati on District, and Plaintiff-Intervenor
9 Californ ia Departmen t of Water Resources (“ DWR”) filed
10 statemen ts of non-opposit ion. Docs. 247, 2 48 & 2 51.

11 Federal Defendants a nd Defendant-Intervenor s opposed.
12
Docs. 27 3 & 274.
13
Addition ally, San Lu is Plaintiffs seek a PI again st
14
the impl ementation o f Action IV.2.3 in the 2009 S almonid
15
BiOp. D oc. 164 (fil ed Jan. 27, 2010). Plaintiff s Kern
16
17 County W ater Agency and Coalition for a Sustainab le Delta
18 joined. Doc. 181. DWR filed a partial joinder i n and

19 statemen t of non-oppositi on to the motion. Doc. 249.
20
Federal Defendants a nd Defendant-Intervenor s opposed.
21
Docs. 27 3 & 274.
22
The PI m otions came on for evidentiary hearing an d
23
argument , in Courtro om 3 of the above-capti oned Court
24
25 from Mar ch 30 throug h April 2, 2010. The parties were
26 represen ted by couns el, as noted on the record in open
27 court.
28
2
1 After co nsideration of the testimony of the

2 witnesse s, the exhib its received in evidence, the written
3
briefs o f the partie s, oral arguments, and the pa rties’
4
proposed findings of fact and conclusions of law, Docs.
5
316 & 31 4, and disap provals thereto, Docs. 320, 3 21 &
6
336, the following f indings of fact and conclusio ns of
7
8 law conc erning the m otion for interim relief/prel iminary
9 injuncti on are enter ed.
10 To the e xtent any fi nding of fact may be interpre ted

11 as a con clusion of l aw or any conclusion of law may be
12
interpre ted as a fin ding of fact, it is so intend ed.
13
14 II. BACKGROUND
15 The 2009 Salmonid Bi Op found that planned coordin ated
16 Project operations w ould jeopardize the continued

17 existenc e of and/or adversely modify the critical habitat
18
of sever al of the Listed Species. 2 BiOp at 1-2. As
19
required by law, NMF S proposed a Reasonable and P rudent
20
Alternat ive (“RPA”) that imposes a number of oper ating
21
22 restrict ions and oth er measures on the Projects. The RPA
23 included numerous el ements for §each of the vario us
24 project divisions an d associated stressors, which NMFS

25 conclude d “must be i mplemented in its entir ety to avoi d
26
2
Je op ar dy w as f ou nd a s t o al l of t h e c ov er ed s pe ci es ; ad ver se
27 hab it at m od if ic at io n was f ou nd a s t o t he d es ig na te d cr it ica l ha bi ta t
of wi nt er -r un , sp ri ng -ru n, s te el he a d, an d gr ee n st ur ge on . Bi Op a t
28 1-2 .
3
1 jeopardy and adverse modification.” Id. at 578 ( empha sis

2 added). The descrip tion of the RPA comprises
3
approxim ately 90 pag es of the 2009 Salmonid BiOp. See
4
id. at 581-671 .
5
The RPA includes fiv e principle components, with
6
numerous sub-p arts, but Plaintiffs currently seek to
7
8 restrain only:
9 • Action I V.2.1, which will limit pumping based on San
10 Joaquin River inflow , measured at Vernalis, from

11 April 1 through May 31; and
12
• Action I V.2.3, which imposes restrictions on nega tive
13
flows in Old and Mid dle Rivers (“OMR”) between
14
January 1 and June 1 5, or until average daily wat er
15
temps at Mossdale (a location on the San Joaquin
16
17 River we st of Mantec a, California) are greater th an
18 72°F, wh icheve r is earlier.

19
III. SUMMARY OF MOT ION
20
Plaintif fs seek prel iminary injunctive relief aga inst
21
22 implemen tation of Ac tions IV.2.1 and IV.2.3 on th e
23 grounds that:
24 1) the d istrict cour t already found that the Unit ed

25 States B ureau of Reclamation (“Reclamation”) fail ed
26
to compl y with the N ational Environmental Policy Act
27
(“NEPA”) in implemen ting the 2009 Salmonid BiOp; and
28
4
1 2) the 2 009 Salmonid BiOp is arbitrary, capriciou s,

2 and cont rary to law because:
3
a) NMFS allegedly co nducted an effects analysis
4
that imp roperly over states impacts attributable
5
to the c oordinated o perations of the CVP and
6
SWP;
7
8 b) NMFS failed to cl early define or consistently
9 apply a relevant env ironmental baseline;
10 c) NMFS failed to di stinguish between

11 discreti onary and no n-dis cret ionary CVP and SWP
12
activiti es, which ov erstated the effects of
13
coordina ted operatio ns of the Projects; and
14
d) RPA A ctions IV.2. 1 and IV.2.3 are arbitrary
15
and capr icious, beca use they are without factual
16
17 or scien tific justif ication and/or not supp orted
18 by the b est availabl e science.

19 Plaintif fs further c laim that the implementation of
20
Actions IV.2.1 and I V.2.3 will cause them continu ing
21
irrepara ble harm and that the public interest and balance
22
of hards hips favor i njunctive relief.
23
24 IV. STANDARD OF DE CISION
25 Injuncti ve relief, w hether temporary or permanent , is
26
an “extr aordinary re medy, never awarded as of rig ht.”
27
Winter v . Natural Re sources Defense Council, 129 S. Ct .
28
5
1 365, 376 (2008); Weinberg er v. Romero-Barce lo, 456 U.S.

2 305, 312 (1982). Four factors must be esta blishe d by a
3
preponde rance of the evidence to qualify for temp orary
4
injuncti ve relief:
5
1. Likeliho od of succes s on the merits;
6
2. Likeliho od the movin g party will suffer
7
8 irrepara ble harm abs ent injunctive relief;
9 3. The bala nce of equit ies t ips in the moving
10 parties’ favor; and

11 4. An injun ction is in the public interest.
12
Winter, 129 S. Ct. a t 374; Am. Trucking Ass ’n v. City of
13
Los Ange les, 5 59 F.3d 1046, 1052 (9th Cir. 2009).
14
15 V. FINDINGS OF FA CT
16 A. The Agen cy Action.
17 1. The agen cy action is the coordinated operation
18 of the C VP and SWP, pursuant to an Agreement for the
19
Coordina ted Operatio n of the two projects (“COA”) .
20
2. Accordin g to the Riv ers and Harbors Act of 1937,
21
the dams and reservo irs of the CVP “shall be used, fir st,
22
23 for rive r regulation , improvement of navigation a nd flood
24 control; second, for irrigation and domestic uses ; and,
25 third, f or power.” 50 St at. 844, 850.

26 3. The CVP was reauthor ized in 1992 through the
27
Central Valley Improvemen t Act (“CVPIA”), w hich m odifi ed
28
6
1 the 1937 Act and add ed mitigation, protection, an d

2 restorat ion of fish and wildlife as project purpo ses.
3
Pub. L. 102-57 5 § 3402, 106 Stat. 4600, 4706 (1992). One
4
of the s tated purpos es of the CVPIA is to address impacts
5
of the C VP on fish a nd wildlife. § 3406(a). The CVPIA
6
made env ironmental p rotection and water deliverie s co-
7
8 purposes .
9 4. This cas e presents a critical conflict between
10 these du al legislati ve purposes, providing water service

11 for agri cultur al, domestic, and industrial use ve rsus
12
enhancin g environmen tal protection for fish speci es whose
13
habitat is maintaine d in rivers, estuaries, canal s, and
14
other wa terways that comprise the Sacramento-San Joaquin
15
Delta.
16
17 5. It is of manifest si gnificance to the public
18 interest that DWR, a co-opera tor and the State

19 contract ual partner of Reclamation, disagrees wit h at
20
least so me portions of the RPA and seeks limited
21
injuncti ve relief ag ainst RPA Action IV.2.3.
22
23 B. Facts Re levant to NE PA Claims.
24 6. It is un disputed tha t neither NMFS nor
25 Reclamat ion engaged in any NEPA analysis in conne ction
26
with pre paration or implementation of the 2009 Sa lmonid
27
BiOp.
28
7
1 7. It is un disputed tha t a March 17, 2010 Order

2 granted San Luis Pla intiffs’ motion for summary j udgment
3
on their claim that Federal Defendants violated N EPA when
4
they ado pted and imp lemented the 2009 NMFS BiOp R PA
5
without conducting t he required NEPA analysis. D oc. 288.
6
8. NMFS ass erts that it did consider a range o f
7
8 alternat ive RPA acti ons, including those proposed by
9 Reclamat ion and DWR, and “carefully avoided presc ribing
10 measures that are no t necessary to meet section 7

11 requirem ents.” BiOp at 578, 580 & 720 (NMFS ende avored
12
“through the iterati ve consultation process to avoid
13
developi ng RPA actio ns that would result in high water
14
costs, w hile still p roviding for the survival and
15
recovery of listed s pecies.”). However, this pro cess did
16
17 not full y or suffici ently evaluate, explain or an alyze
18 the exte nt and gravity of the harms to huma ns and the

19 environm ent visited upon Plaintiffs by Project wa ter
20
service reductions a nd pumping restrictions.
21
9. The 2009 Salmonid Bi Op phases in some elements
22
of the R PA over time , provides a health and safet y
23
exceptio n to ensure a minimum level o f water exports,
24
25 uses mon itoring prog rams and adaptive management to
26 initiate RPA actions when species are present and
27 protecti ons are most needed, and includes specifi c

28
8
1 scientif ic studies a nd engineering programs to re fine RPA

2 elements . Id. at 579-80, 719-23. In addition, the
3
challeng ed RPA Actio ns were modified between the draft
4
and fina l RPA to les sen water supply impacts, inc luding
5
shorteni ng the durat ion of Action IV.2.1 from 90 to 60
6
days. I d. at 723; N MFS A R 104419.
7
8 10. A legall y sufficient NEPA analysis should
9 identify and analyze alternatives that minimize h arm to
10 humans a nd the human environment. Federal Defend ants do

11 not clai m that they engaged in a systematic consi deration
12
of impac ts to humans and the human en vironment an d/or the
13
alternat ives that wo uld have minimized harm to hu man and
14
the huma n environmen t while still protecting the species.
15
11. Federa l Defe ndants did not take the hard look
16
17 required to achieve, to the maximum extent possible, the
18 co-equal Recla mation Law objective of providing water

19 service.
20
21 C. Facts Re levant to ES A Challenges.
22 (1) Current Status of th e Species.
23 a. Sacramen to River Win ter-Run C hinook Salmon.
24 12. Sacramen to River win ter-run C hinook salmon
25 (Oncorhy nchus tshawytscha) (“winter-run”) a re lis ted a s
26 “endange red” under t he ESA. 70 Fed. Reg. 37,160 (June

27 28, 2005 ). Winter-run critical habit at inc ludes porti ons
28
9
1 of the S acramento Ri ver and other waters. Histor ical

2 winter-r un populatio n estimates were as high as
3
approxim ately 100,00 0 fish in the 1960s, but decl ined to
4
under 20 0 fish in th e 1990s. Gov’t Salmon Ex. 4 (Second
5
Stuart D ecl., Doc. 2 73-3), ¶45. In recent years,
6
populati on surveys o f winter-run estimated a high of
7
8 17,344 f ish in 2006, followed by a decline in 200 7 (2, 542
9 fish) th at persisted into 2008 (2,830 fish). Id. In
10 2009, th ere was a mo dest increase in adult escape ment

11 (4,658 f ish). Id. Winte r-run are “current ly not
12
viable.” BiOp at 88 ; see als o 4/1/10 Tr. 175: 11 -12.
13
13. Winter-r un juvenile rearing and migration
14
typicall y occurs bet ween July and February in the upper
15
Sacramen to River, wi th juvenile emigration downst ream
16
17 through the Delta ta king place between November t hrough
18 May or J une. BiOp a t 81, 94; Pac. Co ast Fed’n of

19 Fishermans’ As s’ns. v. Gutierrez (“Gutierrez II”), 606 F.
20
Supp. 2d 1195, 1216- 17 (E .D. Cal. 2008); 4/ 1/10 T r.
21
167:5-19 ; Gov’t Salm on Ex. 1 (First S tuart Decl., Doc.
22
190-4) a t (int ernal) Exhibit 1a. Historically, the peak
23
emigrati on period fo r winter-run occurs during March.
24
25 Gov’t Sa lmon Ex. 4, ¶47.
26 14. During t he current e migration year (2009-20 10),
27 juvenile winter-run began entering the Delt a in O ctobe r

28
10
1 2009. I d. at ¶46. On Ap ril 1, 2010, Mr. S tuart, an N MFS

2 biologis t, testified that “about 1,600 winter-run ”
3
juvenile s have been salvaged at the pumping facil ities
4
for the season. 4/1 /10 Tr. 174:11.
5
15. The esti mate of juve nile winter-run p roduct ion
6
(known a s the Juveni le Production Estimate (“JPE” )) for
7
8 2009 is 1,144,860. Gov’t Salmon Ex. 1, at ¶3. T he Bi Op
9 sets an incidental t ake limit of two percent of t he JPE
10 of winte r-run salmon , or 22,897. BiOp at 7 75; 3/31/10

11 Tr. 112: 16-25 – 113:1.
12
16. In addit ion, althoug h winter-run are currently
13
at the “ tail end” of their emigration throu gh the Delt a
14
(90% mov ed through t he Delta by the end of March) ,
15
3/31/10 Tr. 172:3-6, Mr. Stuart opined that the “ tail
16
17 end” of the winter-run migration peri od is “significan t”
18 because it “represen ts fish that would probably s how a

19 differen t life history than fish that occur during the
20
other pa rts” and, “p rotecting the tail end would be
21
importan t to maintai n the diversity of that winte r-run
22
populati on,” 4/1/10 Tr. 174:19 - 175: 8.
23
17. The em igrati on per iod f or winter-run is all but
24
25 concluded for this water year.
26 18. Designat ed critical habitat for winter-run
27 includes the Sacrame nto River, the Delta, and dow nstream
28
11
1 bays to the Golden G ate Bridge. Gutierrez II, 606 F.

2 Supp. 2d at 1217. C urrently, the value of winter -run
3
critical habit at is “degraded.” BiOp at 93.
4
5 b. Central Valley Sprin g-Run Chi nook Salmon.
6 19. Central Valley sprin g-run Chi nook salmon (O.
7 tshawyts cha) ( “spring-run ”) a re listed as “threat ened”

8 under th e ESA. 71 F ed. Reg. 834 (June 5, 2005); 70 Fed.
9
Reg. 371 60 (June 28, 2005) (critical habitat desi gnated).
10
Spring-r un are not c urrently viable. 4/1/10 Tr. 179:12-
11
15. Spr ing-ru n Chinook have been declining over recen t
12
years; t his past yea r was one of the lowest adult
13
14 escapeme nts ever see n. 3/31/10 Tr. 1 37:22-138:2.
15 20. It is estima ted th at th e entire Evolution arily
16 Signific ant Unit (“E SU”) consists of 3,800 adults .

17 4/1/10 T r. 180:9-11; Gov’ t Salmon Ex. 4 at (internal) Ex.
18
7 (March 2010 popula tion estimates).
19
21. The emig ration perio d for spring-run extend s
20
from Nov ember to May , see Gov’t Salmon Ex. 4, ¶50 ,
21
22 although spring-run may occur in the Delta in low
23 abundanc e in June, see Go v’t Salmon Ex. 1 at (int ernal)
24 Exhibit 1a. Histori cally, April is the peak peri od for

25 spring-r un salvage a t the CVP and SWP. 3/31/10 T r.
26
125:14; see also Gov’t Salmon Ex. 4, ¶52.
27
22. Emigra tion f or spr ing-run f or 2009-2010 is
28
12
1 substant ially comple te.

2 23. During t he current e migration year (2009-20 10),
3
spring-r un began entering the Delta i n Octo ber 2009.
4
Gov’t Sa lmon Ex. 4, ¶52. Under the 2009 Salmonid BiOp,
5
NMFS use s the releas e of specially-marked late fall-ru n
6
Chinook as a surroga te for determining take of sp ring- run
7
8 Chinook at the expor t pumps. BiOp at 776, 782; G ov’t
9 Salmon E x. 4, ¶52; id. at (in ternal) Exhibit 10 ( graph
10 showing peak of spri ng-run salvage in April). Fo r

11 spring-r un, the inci dental take limit is one perc ent of
12
the mark ed fall-run surrogates. 3/31/10 Tr . 113: 1-2; see
13
also BiO p at 7 76. Take of the tagged late-fall surrog ate
14
releases exceeded th e caution level of 0.5% this year,
15
which wo uld have tri ggered a reduction in negativ e OMR
16
17 flows un der RPA Acti on IV.2.3. See 3/31/10 Tr. 113:1- 4;
18 Gov’t Sa lmon Ex. 4, ¶52; BiOp at 649. However, b ecause

19 Action I V.2.3 was en joined, NMFS coul d not implement
20
Action I V.2.3 for se veral days. See Gov’t Salmon Ex. 4,
21
¶52.
22
24. Designat ed critical habitat for spring-run
23
includes the Sacrame nto River, tributaries suppor ting
24
25 spring-r un, the Delt a, and downstream bays to the Gold en
26 Gate Bri dge. Gutierrez II, 6 06 F. Supp. 2d at 1217. The
27 value of spring-run critical habitat curren tly is

28
13
1 “degrade d.” BiOp at 101, 104.

2
c. Central Valley Steel head.
3
25. Central Valley steel head (O. mykiss) (“CV
4
5 steelhea d”) ar e listed as “threatened” under the ESA. 71
6 Fed. Reg . 834 (Jan. 5, 2006). Wild CV steelhead are
7 confined mostly to t he upper Sacramento River and its

8 tributar ies. BiOp a t 107. Recent surveys also h ave
9
detected small, self-sustaining popul ations on the
10
Stanisla us, Mokelumn e, and Calaveras rivers, as w ell as
11
observat ions of juve nile steelhead on the Tuolumn e and
12
Merced r ivers. Id. These small popu lation s make up t he
13
14 remainin g representa tives of the Southern Sierra Nevada
15 Diversit y Group (“SS NDG”) of CV Steelhead. Id. at 198.
16 26. Approxim ately 90% of historical CV Steelhead

17 range is blocked by dams. 3/31/10 Tr. 99:25-100:2.
18
Mortalit y rates for CV steelhead, estimated by us ing
19
fall-run Chino ok as surrogates, are approxi mately 70 t o
20
90%. Id . at 102:21- 23.
21
22 27. While th ere is limit ed information on population
23 size, on e population estimate in 2005 calculated that
24 there we re approxima tely 3,600 female CV steelhea d

25 spawning in the enti re Central Valley, compared w ith
26
40,000 s pawners in the 1960s. BiOp at 106.
27
28. All avai lable data i ndicate that the CV
28
14
1 steelhea d population continues to decline. Id. at 108-

2 09; see also i d. at 100:8.
3
29. The SSND G is one of the population groups of the
4
CV steel head. 3/31/10 Tr . 98:2-3. U nder t he Viable
5
Salmonid Population (“VSP”) concept and the Lindl ey
6
(2007) p aper applyin g the VSP concept to Central Valley
7
8 salmonid s, NMFS must maintain all extant populati ons
9 within t he Central V alley, in order to maintain t he
10 viabilit y of t he Dis tinct Population Segmen t (“DPS”) a s a

11 whole. Id. at 98:3- 7.
12
30. The SSND G is a very small population,
13
represen ted by very few adult fish moving back in to the
14
system a nd potential ly only a few hundred to a fe w
15
thousand juveniles m oving out each year. Id. at 98:9- 12;
16
17 100:12-2 3. These nu mbers are an “assumption” bec ause of
18 the limi ted monitori ng data available. Id. at 98:12-15.

19 31. The risk of extirpat ing the SSNDG is very high
20
because 100% of this very small population must travel
21
through the Delta, w here it is exposed to numerou s risks.
22
Id. at 1 03:2-1 1. Mr. Stuart opined that this diversity
23
group ha s a “very te nuous hold on survival” and t hat
24
25 “[i]t wo uldn’t take much to extirpate it.” Id. a t Tr.
26 104:11-1 3. Extirpat ion of this diver sity g roup would
27 further decrease the viability of the CV steelhea d DPS as

28
15
1 a whole. Id. at 103:24-104:3 .

2 32. The CV s teelhead DPS as a whole is not currently
3
viable. Id. a t 99:8-11.
4
33. Juvenile CV steelhea d typically emigrate through
5
the Delt a from late September through June. Gov’ t Salmon
6
Ex. 1, a t (internal) Exhibit 1a. “Peak entrainme nt
7
8 typicall y occurs bet ween mid-February and m id-Mar ch wi th
9 a prolon ged tail int o June.” Gov’t Salmon Ex. 4, ¶57.
10 CV steel head are cur rently migrating through the Delta,

11 includin g the Sacram ento and San Joaquin Rivers a nd their
12
associat ed tributari es. See 3/31/10 Tr. 118:8-10. As of
13
March 15 , 2010, appr oximately 420 wild CV steelhe ad had
14
been tak en at the CV P since October 2009, and 204 wild
15
steelhea d had been t aken at the SWP. Gov’t Salmo n Ex. 4,
16
17 ¶57. Th e “highest r ates of fish collection did o verlap
18 with the period in w hich the TRO [issued in this case

19 against the implemen tation of Action IV.2.3] allo wed
20
increase d exports (Februa ry 5 through Febru ary 10,
21
2010).” Id.
22
34. The 20 09 inc identa l tak e for CV steelhead is
23
3,000 fi sh based on “fairly old data.” 3/31/10 T r.
24
25 135:19-2 0.
26 35. CV steel head critica l habitat is severely
27 degraded . 3/31/10 T r. 67:21-68:8.

28
16
1 36. Despit e over five (5) y ears of active

2 controve rsy over rel evant ESU designation and
3
preserva tion of CV s teelhead, Federal Defendants have no
4
credible population figures, nor a reliable life cycle
5
model fo r this speci es.
6
7 d. Southern DPS of North America n Green
Sturgeon .
8
37. The sout hern distinc t population segment of the
9
10 North Am erican green sturgeon (“green sturgeon”)
11 (Acipens er medirostr is) i s li sted as “threatened” under
12 the ESA. 71 Fed. Re g. 17757 (Apr. 7, 2006); 73 Fed.

13 Reg. 52, 084 (critica l habitat designated).
14
38. Green sturg eon ar e ana dromous fish that spawn
15
and rear in freshwat er rivers and estuaries but s pend
16
most of their lives in the ocean. Gov’t Salmon E x. 4,
17
¶58. Ju venile green sturgeon are present i n the Delta
18
19 year rou nd. I d. at ¶59. The green sturgeo n “is at
20 substant ial risk of future population declines” d ue to,
21 among ot her things, “loss of juvenile green sturg eon due

22
to entra inment at th e project fish collection fac ilities
23
in the S outh D elta.. ..” BiOp at 126.
24
39. There ar e no populat ion counts or figures for
25
the Sout hern DPS gre en sturgeon. 3/31/10 Tr. 73: 1. Mr.
26
27 Stuart w as unable to provide an estimate of the a ctual
28 populati on of green sturgeon because relevant dat a is

17
1 sparse. 4/1/10 Tr. 177:7-8, 183:17-18. Th e BiOp

2 estimate s salvage of green sturgeon to be highly
3
variable , with a 10- year historical average of 74 adul ts
4
and 106 juveniles pe r year. BiOp at 777. Howeve r, Mr.
5
Stuart n oted that gr een sturgeon have not been de tecte d
6
in salva ge this year . 4/1/10 Tr. 177:10-11.
7
8 40. Green sturge on are anot her species for wh ich no
9 reliable population estimates and/or life cycle m odels
10 have bee n developed, preventing the formulation o f more

11 precise protective m easures.
12
13 e. Southern Resident Ki ller Whale.
14 41. The Sout hern Residen t killer whale DPS (“Sothern
15 Resident s”) of Orcinus orca was listed as a n “end anger ed”
16 species under the ES A on November 18, 2005. 70 F ed. Reg.

17 69,903 ( Nov. 18, 200 5).
18
42. S ou t h er n Re s i de n t s a r e f o un d th r ou g h ou t th e
19
c oa s t al w at e r s o f f W a sh i n gt o n , O re g o n, a nd V an c o uv e r I s l an d
20
a nd a re k no w n t o tr a v el a s f a r s ou t h a s ce n t ra l Ca l i fo r n ia
21
a nd a s f a r n o rt h as t he Q ue e n C h ar l o tt e Is l a nd s , B r i ti s h
22
23 C ol u m bi a . B i Op a t 1 5 9. The Southern Residents we re
24 formerly thought to range southward along the coa st to
25 about Gr ays Harbor o r the mouth of the Columbia R iver.

26 However, recent sigh tings of members of K and L p ods in
27
Oregon ( in 1999 and 2000) and California (in 2000 , 2003,
28
18
1 2005, 20 06 and 2008) have extended th e southern limit of

2 the Sout hern Residen t range. Id. at 160.
3
43. The Sout hern Residen ts have fewer than 90
4
members and loss of even a single individual, or the
5
decrease in reproduc tive capacity of a single ind ividual,
6
is likel y to reduce the likelihood of survival an d
7
8 recovery of the DPS. BiOp at 573. NMFS conclude d that
9 Southern Residents p refer Chinook salmon as prey. Id. at
10 163 (sal mon constitu te up to 96% of Southern Resi dent

11 prey, wi th Chinook s almon constituting 72% of that pre y);
12
id. at 573. I n addition, genetic and chemical evidence
13
indicate that Southe rn Residents consume Chinook salmon
14
from the Central Val ley. Id. at 164. Orca sight ings off
15
the Coas t of Califor nia coincide with large runs of
16
17 Central Valley salmo n. Id. at 159-62, 573.
18 44. NMFS con cluded that extinction of winter-ru n and

19 spring-r un Chinook s almon, as well as reductions in fall-
20
run Chin ook salmon p opulations 3 , “woul d reduce prey
21
availabi lity and inc rease the likelihood for local
22
depletio ns of prey i n particular locations and ti mes,”
23
which wo uld, in turn increase the risk of extinct ion of
24
25 the Sout hern Residen ts. Id. at 573-74.
26 45. There is no direct evid ence of orca morta lity
27
3
Fa ll -r un C hi no ok s al mon a re n ot l i ste d as t hr ea te ne d or
28 end an ge re d un de r th e ESA . 3 /3 1/ 10 Tr. 1 26 :1 9- 21
19
1 attribut able to the Projects.

2
(2) Effects of Ocean Con ditions on Salmon Declines.
3
46. Mr. Cram er testified that poor fall-run Chinook
4
5 adult re turns during 2007 and 2008 could be attri buted to
6 a change in ocean co nditions and very poor surviv al in
7 the ocea n. 3/30/10 Tr. 111:10-112:2; 117:17-118:2.

8 47. The BiOp cites the L indley (2009) analysis of
9
this fis hery collaps e for the proposition that “t he rapid
10
and like ly temporary deterioration in ocean condi tions
11
acted on top of a lo ng-term steady degradation of the
12
freshwat er and estuarine environment.” BiOp at 149. The
13
14 BiOp als o concludes:
15 Because the potentia l for poor ocean conditions

exists i n any given year, and there is no way
16 for salm on managers to control these factors,
any dele terious effe cts endured by salmonids in
17 the fres hwater envir onment can only e xacerb ate
18 the prob lem of an in hospitable marine
environm ent. Therefo re, in order to ensure
19 viable p opulations, it is important that any
impacts that can be avoided prior to the period
20 when sal monids enter the ocean must be carefully
consider ed and reduc ed to the greatest extent
21 possible .
22
Id. at 152-53
23
48. Mr. Cram er clarified that the fish of concern
24
were alr eady at low abundance and that, over the course
25
of decad es, there we re other factors operating on their
26
27 populati on trajector ies besides ocean conditions.
28 3/31/10 Tr. 2:18-3:2. Mr . Stuart testified that the

20
1 collapse of fall-run Chin ook was not exclus ively caused

2 by ocean conditions, but also was brought about b y
3
freshwat er environme ntal conditions, including reduced
4
flows, w ater tempera tures, predators, and non-nat ive
5
species. 3/31/10 Tr . 127:22-25; id. at 128 :1-11.
6
49. Other causes of fr eshwa ter degradation,
7
8 includin g, but not l imited to, toxics, increased
9 salinity , alien and invasive species, predators, riparian
10 pumping and in-Delta diversions are unaddressed by any

11 alternat ives. These other causes have not been
12
systemat ically addre ssed by Federal Defendants or any
13
other po tentially in terested agency or entity.
14
15 (3) Action I V.2.1.
16 a. Operatio n and Purpos e(s) of A ction IV.2.1.
17 50. The stat ed objective s of Action IV.2.1 are to:
18 (a) redu ce vulnerabi lity of emigrating CV Steelhe ad in
19
the San Joaquin Rive r (i.e., the SSNDG) to condit ions in
20
the Sout h Delta and at the pumps; and (b) enhance
21
likeliho od of salmon ids successfully exitin g the Delta by
22
23 creating more suitab le hydraulic conditions in th e
24 mainstem of the San Joaquin. BiOp at 641; 3/31/10 Tr.
25 65:10-18 .
26 51. NMFS’s a nalysis of t he scientific basis for
27
Action I V.2.1 is fou nd in Appendix 5 to the BiOP. Gov’t
28
21
1 Salmon E x. 20 (“BiOp App. 5”).

2 52. While sp ring flow in creases and export
3
reductio ns have been provided as part of the Vern alis
4
Adaptive Management Plan (“VAMP”) since 2000, the
5
proposed operation d id not carry VAMP forwa rd, as fund ing
6
for such flows was s et to expire in 2009, and the San
7
8 Joaquin River Agreem ent, a key to implementing VA MP,
9 expires in 2012. Id. at 2. Based on uncer tainty that
10 VAMP wou ld continue, NMFS determined it necessary to

11 develop an RPA which ensured the flows necessary for
12
successf ul juvenile outmigration and maintenance of
13
critical habitat. Id. at 3.
14
53. Action I V.2.1 is in effect from April 1 through
15
May 31 a nd has two r equirements. First, the Acti on
16
17 requires a minimum f low, as measured at Ver nalis, base d
18 on an in dex of stora ge at New Melones (“New Melon des

19 Index”). BiOp at 64 2. Based on this Index, the minimum
20
flow req uired at Ver nalis from April 1, 2010 to M ay 31,
21
2010 und er Action IV .2.1 is the greater of 3,000 cubic
22
feet per second (“cf s”) or the flow needed to mee t the
23
requirem ents of Stat e Water Resources Control Boa rd
24
25 Decision 1641 (“D-1641”). Gov’t Salmon Ex. 5 (Th ird
26 Milligan Decl.), ¶5. The Vernalis flow requireme nt is
27 not chal lenged here.

28
22
1 54. The seco nd requirement of Action IV.2.1

2 restrict s combined C VP and SWP export pumping bas ed on
3
the flow s at Vernali s, with the permissible expor ts
4
rising i n relation t o increased flows at Vernalis . BiOp
5
at 642. The baselin e export rate is set at 1,500 cfs, as
6
this has been deemed an operational minimum requi red to
7
8 address health and h uman safety needs. 3/31/10 T r. 64:9-
9 11. As of a March 1 5, 2010 estimate provided by the day-
10 to-day manager of the CVP, Ronald Milligan, Vernalis

11 flows ar e likely to be less than 6,000 cfs, meani ng th at
12
Action I V.2.1 likely will limit export pumping to 1,500
13
cfs. Bi Op at 642; G ov’t Salmon Ex. 5, ¶5.
14
55. Action I V.2.1 will n ot control exports for the
15
entire 6 0-day period , as VAMP will limit co mbined expo rts
16
17 to 1,500 cfs f or 30 days in April and May. Gov’t Salm on
18 Ex.5, ¶2 3. This yea r, VAMP likely will be initia ted

19 April 22 , 2010. Id.
20
56. Action I V.2.1 is des igned primarily to benefit
21
the SSND G (i.e. stee lhead that originate in the S an
22
Joaquin basin from the St anislaus, Tuolumne , and Merce d
23
Rivers). 3/31/10 Tr . 65:10-13. Acti on IV.2.1 wi ll also
24
25 benefit those salmon ids that emigrate out of the
26 Calavera s and Mokelu mne Rivers and those salmonid s that
27 come fro m the Sacram ento River basin but enter in to the
28
23
1 central and southern Delta through Georgiana Slou gh or

2 the Delt a Cross Chan nel (“DCC”) and the Mokelumne River
3
system w hen the DCC gates are open. Id. at 65:13 -18.
4
57. Increase d flows from Action IV.2.1 will also
5
benefit designated c ritical habitat f or the CV steelhe ad
6
within t his region b y enhancing riparian habitat, flow,
7
8 and decr easing ambie nt temperature, as well as in creasing
9 turbidit y and juveni le migration time, both of wh ich
10 lessen t he risk of p redation. 3/31/10 Tr. 67:2-17.

11 However, habitat pro tection is not one of the rat ionales
12
for Acti on IV.2.1 ar ticulated in the BiOp or Appe ndix 5.
13
14 b. Viable S almonid Popu lation Methodology.
15 58. There is considerabl e dispute about whether NMFS
16 went far enough in i ts use of the Viable Salmonid

17 Populati on (“VSP”) c oncept to evaluate the effect s of
18
Project operations o n the Listed Species.
19
59. It is un disputed tha t VSP can serve as a
20
“concept ual framewor k” around which the analysis of a
21
22 project can be struc tured. BiOp at 51-53. The B iOp
23 describe s VSP as fol lows:
24 The VSP concept prov ides specific guidance for

estimati ng the viabi lity of populations and
25 larger-s cale groups of Pacific salmonids such as
ESU or D PS. Four VS P parameters form the key to
26 evaluati ng populatio n and ESU/DPS via bility : (1)
abundanc e; (2) produ ctivity (i.e., population
27 growth r ate); (3) po pulation spatial structure;
and (4) diversity.
28
24
1 Id.
2 60. Under th e VSP concep t, abundance is just one of
3
several criteria tha t must be met for a populatio n to be
4
consider ed viable. BiOp at 84. ESU viability al so
5
depends on the numbe r of populations and subunits within
6
the ESU, their indiv idual status, their spatial
7
8 arrangem ent with res pect to each other and source s of
9 catastro phic disturb ance, and diversity of the
10 populati ons and thei r habitat. Id.; see also NMFS AR

11 00123481 (Lindley (2 007)).
12
61. The BiOp explains th at under the VSP framework,
13
viabilit y requires m ore than attaining a particul ar level
14
of popul ation abunda nce. “Rather, for an ESU to persist,
15
populati ons within t he ESU must be able to spread risk
16
17 and maxi mize future potential for adaptation.” B iOp at
18 84. Lin dley (2007) further found that an importa nt risk

19 facing s almonid ESUs is “that much of the diversi ty
20
historic ally present in these ESUs has been lost. ” NM FS
21
AR 00123 489. Lindle y (2007) thus recommends that “every
22
extant p opulation” o f the listed salmonids “be vi ewed as
23
necessar y for the re covery of the ESU,” because a ll three
24
25 ESUs “ar e far short of being viable, and extant
26 populati ons, even if not presently vi able, may be needed
27 for reco very.” NMFS AR 00123494. Based on this

28
25
1 recommen dation, the BiOp “assumed that if appreci able

2 reductio ns in any po pulation’s viability are expe cted to
3
result f rom implemen tation of the proposed action, then
4
this wou ld be expect ed to appreciably reduce the
5
likeliho od of both t he survival and recovery of t he
6
diversit y group the population belongs to as well as the
7
8 listed E SU/DPS.” Bi Op at 50.
9 62. The BiOp used the VS P concept, extensively
10 discusse d it, and ad dressed the various VSP facto rs in

11 consider ing the curr ent status of and the impacts of
12
proposed Project ope rations on the Listed Species . Se e
13
BiOp at 105 at 43; see also, id. at 5 0-53, 68, 84 -88, 93-
14
101, 108 -111, 124, 1 73, 309, 443, 451, 472. However,
15
NMFS use d VSP as a qualitative framew ork.
16
17 63. There is a dispute o ver whether NMFS should have
18 used the VSP as a st arting point for a quantitative

19 analysis . Mr. Crame r opines that the VSP concept
20
describe d in Lindely (2006) ( “NMFS Science Center
21
Evaluati on of the Pe er Reviews of the Long-Term Centra l
22
Valley P roject and S tate Water Project Operations Section
23
7 Consul tation”), id entifies attributes of a popu lation
24
25 that are useful in d etermining a population’s abi lity to
26 persist, but i s not a quantitative framewor k. 3/ 30/10
27 Tr. 105: 5-13.

28
26
1 64. Lindley 2006 states that the VSP framework was

2 designed to be a con ceptual framework. SLDMWA Ex . 379 at
3
5. Howe ver, Lindley 2006 also stated: “while VSP would
4
provide a conc eptual framework, an analytical framework
5
will sti ll need to b e assembled to assess the imp acts of
6
specific projects on VSP parameters.” Id.
7
8 65. Mr. Cram er opines th at there was data cited in
9 the 2009 Salmonid Bi Op that would have permitted
10 quantita tive analyse s within the VSP framew ork. 3/30/ 10

11 Tr. 123: 1-12.
12
66. However, the NMFS Sc ience Center’s 2006 peer
13
evaluati on of the pr evious salmonid biological op inion,
14
for whic h Lindley wa s the lead author, disagrees: “While
15
new info rmatio n or models,” beyond the VSP criter ia, “may
16
17 help mak e the analys is more transparent and rigor ous, it
18 is not r equired and many times is not realistic g iven the

19 limitati ons on time and resources.” SLDMWA Ex. 3 79 at 5.
20
67. Although the analysis in the BiOp could hav e
21
benefite d from the a pplication of quantitative
22
methodol ogies within the VSP framework, there is a
23
scientif ic dispute w hether the failure to do so
24
25 represen ts a breach of accepted scientific practi ce.
26
c. Populati on Modeling/ Life Cycle Analys is.
27
68. Mr. Cram er opines th at the BiOp should have
28
27
1 performe d population modeling and/or life cycle m odeling.

2 See 3/30/10 Tr . 94:8 – 96 :1. In the context of
3
anadromo us salmonids , the application of such a
4
methodol ogy involves evaluation of th e life history of
5
the popu lation, from adults spawning in fresh wat er, to
6
fry emer gence from g ravel, to downstream migratio n as
7
8 smolts r ear, and the n to the species’ salt-water life
9 history. At each st age, population modeling woul d be
10 used to evalua te the factors that affect survival. Id .

11 at 94:8 – 96:1. Mr. Cramer opined that proper us e of a
12
life cyc le model inv olves testing of a hypothesis against
13
availabl e data to de termine whether predicted out comes
14
match up with observ ed values. Id. a t 97:13 – 98 :8.
15
69. NMFS did not explici tly evaluate the impact of
16
17 project operations i n a life cycle model. This f ailure
18 has been criticized as not complying with accepte d

19 scientif ic principle s for population analysis.
20
Plaintif fs presented no evidence regarding the ex isten ce
21
or avail ability of s uch a life cycle model for th e
22
species in question. Plaintiffs did not present evidence
23
that the y, or anyone else developed or made avail able to
24
25 NMFS an appropriate life cycle model or the resul ts of an
26 appropri ate li fe cyc le analysis prior to the issu ance of
27 the BiOp .
28
28
1 70. The prim ary purpose of Action IV.2.1 is to

2 protect outmigrating juvenile members of the SSND G of CV
3
steelhea d, for which no population indices (wheth er
4
absolute or relative ) exist.
5
71. Despit e year s of c ontro versy and litigati on ove r
6
CV steel head, the ab sence of reliable population data
7
8 complica tes the anal ysis.
9
d. Lack of Statisticall y Significant
10 Correlat ion Between Exports and Effects on
Salmonid Survival.
11
72. The crux of Plaintif fs’ critique of Action
12
IV.2.1 i s that it is unsupported by the various s tudies
13
and anal yses actuall y relied upon in the BiOp. T he
14
15 rational e for Action IV.2.1, provided in Appendix 5 to
16 the BiOp , relies on a number of sources.
17
(1) Treatmen t of VAMP Da ta in the BiOp.
18
73. VAMP is a multi-agency collaborative effort
19
designed to test the hypothesis that exports and flow in
20
21 the San Joaquin Rive r influence survival of smolt s
22 emigrati ng down the San Joaquin River. 3/30/10 T r.

23 126:21 – 127:4. Ann ual reports prese nting the results of
24
the VAMP experiment have been produced since 2000 . Id .
25
at 127:5 -7.
26
74. Analyses of the evid ence gathered during VAMP
27
have bee n equivocal regarding the impact of expor ts on
28
29
1 survival . The BiOp recognized that “ r e ce n t p a p er s

2 e xa m i ni n g t h e e f f ec t s o f ex p o rt s o n sa l m on s ur v i va l ha v e
3 b ee n un a b le t o p r ov e a s t at i s ti c al l y s i g ni f i ca n t r e d uc t i on
4
i n s u rv i v al r el a t ed t o e x po r t s ( Ne w m an 2 00 8 ) .” Bi O p a t
5
4 26 .
6
75. Newman’s 2008 statis tical ana lyses of the V AMP
7
data con cludes that environmental variables could obscure
8
9 any rela tionship bet ween exports and survival. 3 /31/10
10 Tr. 88:1 1-14. This cavea t was recognized i n the BiOp.
11 BiOp at 426.
12 76. The VAMP experimenta l design has not been
13
implemen ted in full, in that not all of the plann ed
14
relation ships have b een tested. 3/31/10 Tr. 83:1 1-15.
15
Over the ten years V AMP data was collected, there have
16
been six replication s of conditions at 3,200 cfs Vernalis
17
18 flow and 1,500 cfs e xports. Id. at 84:2-4. Newman no ted
19 that the small numbe r of variables tested in the existing
20 VAMP dat a did not pr ovide the ability to discrimi nate

21 between survival eff ects. Id. at 88:19-22. Plai ntiff s’
22
expert, Mr. Cramer, and DWR’s expert, Mr. Cavallo ,
23
recogniz e thes e limitations in the VAMP data. Id. at
24
191:6-12 ; 4/1/10 Tr. 100:4-11.
25
77. The BiOp also recogn izes these limitations.
26
27 BiOp at 426. To bui ld a more robust data set, NM FS is
28 implemen ting a six-year acoustic tag study prescribed by

30
1 RPA Acti on IV. 2.2. 3/31/10 Tr. 87:11-15.

2 78. The BiOp considered the VAMP evidence and its
3
limitati ons and did not disregard any important
4
conclusi ons generate d from the VAMP data.
5
6 (a) Figure 10.
7 79. Notwiths tanding the lack of statistical
8 signific ance, eviden ce contained in the VAMP repo rts
9
demonstr ates that, d uring times when the Head of Old
10
River Ba rrier (“HORB ”) 4 was in place, as the ratio
11
between Vernalis flo w and exports increased, surv ival
12
increase d. 3/31/10 Tr. 86:6-9; BiOp App. 5 at 20 . 5
13
14 Figure 1 0 in Appendi x 5 of the BiOp demonstrates a
15 positive relationshi p between the Vernalis flow/e xport
16 ratio an d survival. BiOp App. 5 at 20. The rela tionship

17 was not statisticall y significant, but the BiOp s tates
18
that thi s may have b een due t o the narrow range of exp ort
19
rates te sted. Id.
20
80. RPA Acti on IV.2.1 as sumes a physical or non-
21
22 physical barrier wil l be installed at the head of Old
23 4
HO RB i s a re mo va bl e roc k ba rr ie r t hat “ wh en i ns ta ll ed , dir ec ts
flo ws o n th e Sa n Jo aq uin R iv er a wa y fr om t he O ld R iv er i nto t he
24 Cen tr al D el ta .” Fi nd ing o f Fa ct # 4 7 R e: I nt er im R em ed ie s R e: D el ta
Sme lt E SA R em an d an d Rec on su lt at io n , N RD C v. K em pt ho rn e, 20 07 W L
25 446 23 95 ( De c. 1 4, 2 00 7).
5
It i s un di sp ut ed t ha t w he n HO RB i s in p la ce , th er e is a
26 sta ti st ic al ly s ig ni fi can t re la ti on s hip b et we en V er na li s flo ws a nd
sur vi va l. Se e Bi Op A pp. 5 a t 20 ; T r. 3/ 30 /1 0 12 8: 3 – 13 0:1 1
27 (Cr am er ); S LD MW A Ex . 128 . T hi s is not e qu iv al en t to a s tat is ti ca ll y
sig ni fi ca nt e ff ec t of ex po rt s or t h e V er na li s fl ow /e xp or t r at io o n
28 sur vi va l.
31
1 River in order to pr event the fish from following the

2 flow spl it at the ju ncture of the maintstem San J oaqui n
3
and Old Rivers. 3/3 1/10 Tr. 92:4-8. However, because
4
the HORB negatively impacts the Delta smelt, NMFS worked
5
with Rec lamation, DW R, and other parties to devel op
6
alternat ive engineer ing solutions, which resulted in an
7
8 addition al RPA Action to study ways to sepa rate f ish f rom
9 the flow . Id. at 95:22-96:3.
10 81. A non-ph ysical barri er, or “bubble ba rrier,”

11 which us es bubbles, LED strobe lights, and acoust ic noise
12
to deter the fish fr om entering Old River is plan ned to
13
be insta lled this ye ar. Id. at 96:10-14. Based on a
14
2009 stu dy, the bubb le barrier was 83% successful in
15
blocking fish from m oving through the barrier. Id. at
16
17 96:19-21 . NMFS has determined that t he bubble barrier
18 will ser ve as an eff ective substitute for the physical

19 barrier at the head of Old River required by RPA Action
20
IV.2.1. Id. a t 96:22-25. As of March 31, the
21
installa tion of the bubble barrier was scheduled to
22
commence on April 6, 2010. Id. at 18 0:19.
23
82. Mr. Cram er opined th at without HORB in plac e,
24
25 studies of survival with HORB in place should not be
26 used. S ee id. at 132:13- 24; SLDMWA Ex. 129. Mr. Cramer
27 did not address whet her the alternative bubble ba rrier

28
32
1 will pro duce conditi ons similar enough to those p resent

2 with HOR B in place t o permit the reliance on survival
3
data fro m when HORB was in place.
4
83. The reco rd suggests that an effective barrier
5
will be in place at the head of Old River. It wa s not
6
unreason able for NMF S to consider data with HORB in
7
8 place.
9
(2) Escapeme nt Data.
10
84. In Figur e 11 of Appe ndix 5, the BiOp relied on
11
an analy sis presente d in the 2006 VAMP annual rep ort that
12
showed a positive re lationship between the spring
13
14 Vernalis flow/export ratio and adult escapement ( i.e.
15 return f rom the ocea n to freshwater) two and a half ye ars
16 later, b ased on data from 1951 through 2003. 3/3 1/10 Tr.
17 70:12-14 , 74:7-20; BiOp App. 5 at 21.
18
85. The anal ysis in Figu re 11 did not attempt to
19
account for variable ocean conditions or commerci al
20
harvest of salmonids. See ge nerally 3/31/10 Tr. 142-43
21
22 (Cramer) . Elsewhere in the BiOp, NMFS acknowledg es that
23 escapeme nt survival may be significantly impacted by
24 ocean co nditions. See, e.g., BiOp 96, 144-45, 148-53,

25 166-68, 218. There is a conceptual model in the
26
administrative record that suggests even though ocean
27
conditio ns and harve st may vary from year to year , the
28
33
1 species’ long- term declines may be attributed to other

2 factors affecting su rvival during the freshwater life
3
stages o f the specie s in question. DI 1002 (Lawson
4
conceptu al model).
5
86. Although Figure 11 d id not account for variable
6
ocean co nditions and /or commercial harvest, Plain tiffs’
7
8 expert, Mr. Cramer, testified that a reasonable b iologist
9 would us e this data. 3/30/10 Tr. 192:21-193:3. This
10 suggests that it was not unreasonable for NMFS to

11 consider the analysi s depicted in Figure 11.
12
13 e. Delta Ac tion 8 Studi es.
14 87. The BiOp also consid ered data from the so-called
15 “Delta A ction 8 stud ies,” which compared the rela tive

16 survival rates of coded-wire tagged salmon released at
17
(a) Ryde on the Sacr amento River and (b) Georgian a
18
Slough, a channel th at splits off of the Sacramen to River
19
at Walnu t Grove and leads to the interior Delta, joining
20
the Sout h Fork of th e Mokelumne River just before it
21
22 meets th e San Joaqui n River.
23 88. Evaluati ng the data from the Delta Action 8
24 studies, Newman (200 8) first explained that there was a

25 high lev el of enviro nmental variation in the data . Id .
26
at 78:18 -23. Dr. Ne wman performed further analys is to
27
reduce t he amount of environmental variation and
28
34
1 subseque ntly found a 98% probability that a negat ive

2 relation ship between exports and survival is pres ent.
3
Id. at 7 9:5-7. Mr. Stuart stated the significance of
4
Newman’s finding is that as exports increas ed, su rviva l
5
decrease s for those salmonid smolts that are movi ng down
6
into the San Joaquin River, where they would be e xposed
7
8 to the i nfluences of the export pumps. 4/2/10 Tr . 32:8-
9 34:12. For those fi sh released into Georgiana Sl ough,
10 survival was b etter when exports were lower.

11 89. This stu dy is releva nt to assessing the impacts
12
of expor t pumping on fish migrating through the S an
13
Joaquin River, becau se fish released into Georgia na
14
Slough m ust exit int o the San Joaquin River, wher e they
15
are subj ect to the i nfluence of the pumps. 3/31/ 10 Tr.
16
17 76:20-23 . The Georg iana Slough fish share a common
18 migrator y pathway wi th fish that exit the San Joa quin

19 River ba sin. Id. at 76:24-77:6. Regardles s of t heir
20
origin, once the fis h are in this common migratory
21
pathway, they are su bject to the same hydraulic
22
conditio ns. I d. at 78:1- 17.
23
90. Mr. Cava llo stated t hat his interpretation of
24
25 the Newm an (2008) st udy is that there is a weak
26 relation ship between exports and survival in the interior
27 Delta, b ut conceded that there was some relations hip.

28
35
1 4/1/10 T r. 98:24-99:4. M r. Stuart testifie d that

2 Newman’s studies are the best available and the f act that
3
Newman c ould find a relationship given the consid erable
4
amount o f “environme ntal noise” and t he very low signal
5
to noise ratio “show s that the relationship is pr obably
6
very rea l.” I d. at 159:6-10. Whether this opinion is
7
8 entitled to weight i s disputed by Plaintiffs.
9 91. A Septem ber 26, 2008 paper prepared by Dr.
10 Newman w ith Patricia L. B randes entitled “H ierarc hical

11 Modeling of Juvenile Chinook Salmon Survival as A
12
Function of Sacramen to-San Joaquin Delta Water Ex ports”
13
(“Newman and Brandes 2008”) examined the Delta Ac tion 8
14
data con cerning the relative survival rates for R yde and
15
Georgian a Slough rel eases and declared: what “we canno t
16
17 conclude is that exp orts are the cause of this lo wer
18 relative survival.” 4/1/ 10 Tr. 67:20-23 (e mphasis

19 added); DWR Ex. 507 at 22. Newman and Brandes 20 08
20
reached this conclus ion because “the evidence for an
21
associat ion between exports and survival is somew hat
22
weak” an d because of the study’s inability to ran domize
23
export l evels within a given outmigration season. 4/1/10
24
25 Tr. 68:1 -12; DWR Ex. 507 at 22-23. A later versi on of
26 this stu dy, dated 20 09, omitted this language fro m the
27
28
36
1 conclusi on. 4/2/10 Tr. 28:2-13. 6

2 92. The Delt a Action 8 s tudies seek to relate to
3
exports survival of juvenile salmonids and steelh ead
4
passing through the interior Delta from the San J oaquin
5
River ba sin. These studi es s how a negative relat ionship,
6
although admittedly weak, between export levels a nd
7
8 survival for fish pa ssing through this area of th e Delta.
9
f. Limited Amount of Wa ter Available in Storage
10 to Incre ase Flows at Vernalis.
11 93. Figure 1 1 and other studi es c ited in Append ix 5
12 suggest that maximiz ing the difference between Ve rnalis

13 flows an d export lev els (or maximizing the Vernal is
14
flow/exp ort ratio) i mproves survivial. BiOp App. 5 at 8,
15
20-21.
16
94. NMFS det ermined that , because there was a
17
limited amount of wa ter available to increase flo ws at
18
19 Vernalis , capping ex port levels would provide the
20 greatest differentia l between flows at Vernalis a nd
21 export l evels. 3/31 /10 Tr. 71:12-17; 97:14-21.

22
95. This rea son for cont rolling e xports is unre lated
23
to any d irect scient ific evidence connecting expo rt
24
levels t o fish survi val, making the reason arbitr ary,
25
26 6
Mr . St ua rt e xp la in ed th at a lt ho ug h th e Bi Op c it ed t he 2 008
ver si on o f th e Ne wm an an d Br an de s s tud y, h e ac tu al ly u se d t he 2 00 9
27 ver si on t o pr ep ar e th e B iO p an d th e 20 09 p ap er w as i n hi s r ef er en ce
lis t. He d oe s no t kn ow wh y th e Bi O p u se d th e 20 08 c it at ion . 4 /2 /1 0
28 Tr. 2 8: 2- 13 .
37
1 capricio us, unsuppor ted by reasonable explanation , and

2 not base d on best av ailable science.
3
4 g. Justific ation for Ra tios Used in Action
IV.2.1.
5
96. Although not the sub ject of extensive testimony
6
during t he evidentia ry hearing, there is little t o no
7
justific ation in the record for the exact flow ra tios
8
chosen f or RPA Actio n IV.2.1.
9
10 97. NMFS loo ked at the V AMP data to develop the
11 ratio.
12 Current VAMP studies have ratios of flow to

exports clustered ar ound 2:1, which have
13 provided low surviva l indices for upstream
14 releases compared to downstream releases,
particul arly in rece nt years. Studie s which
15 would ha ve had highe r flo ws ( i.e., 7,000 cf s) to
export ( 1,500 cfs) r atios were not conducted,
16 since th e necessary environmental conditions to
implemen t this part of the study protocol never
17 occurred . Recent con ditions in which high flows
did occu r in the San Joaquin River ba sin and
18
which wo uld have giv en flow to export ratios
19 greater than 3:1 in 2005 and 10:1 in 2006 were
confound ed by poor o cean conditions during the
20 smolts e ntry into th e marine environment, and
returnin g adult fall-run Chinook salm on
21 escapeme nt numbers from these brood years were
very low (brood year s 2004, 2005 which returned
22
in 2007 and 2008). From the available data,
23 includin g the inform ation contained in figures
10 7 and 1 1 8 , flow to export rat ios shou ld be at
24 least 2: 1 and prefer ably higher to increase
survival and abundan ce. In li ght of these
25
7
Fi gu re 1 0 su gg es ts t her e is a p os i tiv e re la ti on sh ip b et wee n
26 the r at io o f Ve rn al is fl ow t o ex po r ts an d su rv iv al o f sa lmo ni ds i n
the i nt er io r De lt a.
27 8
Fi gu re 1 1 re li ed o n an an al ys is p r ese nt ed i n th e 20 06 V AMP
ann ua l re po rt t ha t sh owe d a po si ti v e r el at io ns hi p be tw ee n t he s pr in g
28 Ver na li s fl ow /e xp or t rat io a nd a du l t e sc ap em en t.
38
1 factors, NMFS initia lly developed flow to export

ratios o f 4:1 for we t, above normal, below
2 normal, and dry year s, based on the m inimum
export l evel of 1,50 0 cfs and a targeted minimum
3
Vernalis flow of 6,0 00 cfs. Flows in critically
4 dry year s were targe ted to be a minimum 3,000
cfs, whi ch gives a f low to export ratio of 2:1
5 when exp orts are tar geted to be 1,500 cfs.
6 BiOp App . 5 at 22-23 (emp hasis added). The feasi bilit y

7 and wate r supply imp lications of implementing such flo w
8
versus e xport ratios were then examined through c omputer
9
modeling . Id. at 24-68. The BiOp re asoned that a 2:1
10
ratio wa s insufficie nt because the VAMP studies
11
12 demonstr ated low sur vival rates at that ratio, an d that
13 higher r atios would be “prefera[ble]” to in crease
14 survival and abundan ce. Yet, without any biologi cal

15 explanat ion, the BiO p chose to impose a 1,500 cfs limit
16
when flo ws at Vernal is are lower than 6,000 cfs, 9 and a
17
ratio of 4:1 ( as opposed to 2.5:1, or 3:1, or even 5:1 or
18
higher) when V ernali s flows are between 6,000 cfs and
19
21,750 c fs. I d. at 71-72.
20
21 98. The abse nce of expla nation and analysis for
22 adoption of these li mits uses no science, let alo ne the
23 best ava ilable and i s simply indefensible.

24
25
26
9
Th is 1 ,5 00 c fs l im it is t he m in im u m e xp or t le ve l th at w oul d
27 mai nt ai n he al th a nd s afe ty c ri te ri a . Bi Op A pp . 5 at 2 2. A t fl ow s
of 5, 00 0 cf s, f or e xa mpl e, t he r at i o w ou ld t he re fo re b e 5,0 00 /1 ,5 00
28 or ap pr ox im at el y 3. 33 :1.
39
1 h. Will Enj oining Actio n IV.2.1 Appreciably

Diminish The Likelih ood Of Survival Or
2 Recovery Of The List ed Species Or Adversely
Modify T heir Critica l Habitat?
3
99. The evid ence support s NMFS’s general finding
4
that som e form of re striction on the Vernalis flo w/export
5
6 ratio is needed to p revent jeopardy to the SSNDG of CV
7 Steelhea d. Enjoinin g any flow/export ratio restr iction
8 will app reciably dim inish the likelihood of the S SNDG’s

9 survival or recovery and/or adversely modify its critical
10
habitat.
11
a. Mr. Stua rt testified that enjoining Action
12
IV.2.1 w ould “jeopar dize” the SSNDG of CV steelhe ad,
13
3/31/10 Tr. 122:9, 1 21:3- 5, which in turn would “ further
14
15 decrease the viabili ty of the Central Valley” ste elhead
16 DPS, id. at 104:2-3. Pla intiffs’ expert, M r. Cra mer, did
17 not provide an opinion on the impact of enjoining Action

18 IV.2.1 o n the SSNDG of CV steelhead. Id. a t 24:2 3-25:1.
19
b. For crit ical habitat , Mr. Stuart opined that
20
Action I V.2.1 provid es benefits by enhancing migr atory
21
corridor s, increasin g riparian zones and re aring areas
22
23 which ca n be used by migrating juveniles, and sho rtening
24 migratio n time and i ncreasing turbidity, both of which
25 can decr ease vulnera bility to predation. Id. at 110:2 4-

26 111:14. Mr. Stuart testified that enjoining Acti on
27
IV.2.1 w ould remove these beneficial effect s. Id. at
28
40
1 111:1-2, 121:13-19; see a lso Gov’t Sa lmon Ex., ¶4

2 (enjoini ng Action IV .2.1 would “negate” the benef its
3
provided by Action I V.2.1). Mr. Cramer did not o pine
4
what eff ect enjoinin g Action IV.2.1 would have on CV
5
steelhea d crit ical habitat. 3/31/10 Tr. 25:7-11, 110: 24-
6
25, 111: 1-2 (S tuart testi mony that Mr. Cram er “didn’t
7
8 look at the effects of the flow on enhancing crit ical
9 habitat in migratory corridors in the Delta”).
10 100. The low levels of in cidental take of steelh ead

11 in this water year d o not undermine this conclusi on.
12
a. Mr. Cram er opined th at the current estimated
13
take of salmon and s teelhead is below the inciden tal take
14
limits i n the BiOp. See SLMW A Ex. 122, Doc. 244, Cram er
15
Decl., ¶ ¶ 41-4 3. 10
16
17 b. The purpose of the i ncide ntal take limit is
18 to ident ify a point at which reinitiation of cons ultation

19 should o ccur. 3/31/ 10 Tr. 113:20-22. It is not the
20
default level at whi ch the facilities should be o perated.
21
If the R PA works as designed, the incidental take limit
22
should n ever be reac hed. Id. at 113:25-114 :7, 13 3:15- 24.
23
24 10
Mr . Cr am er a ls o su gg est s in h is d e cla ra ti on t ha t “o nc e fis h
hav e en te re d th e so ut h D el ta , th ei r be st c ha nc e fo r su rv iva l is t o
25 be sa lv ag ed a t th e fi sh fa ci li ti es . ” SL DM WA E x. 1 22 , ¶2 6. H ow ev er ,
Mr. S tu ar t di sa gr ee d wit h th is p os i tio n an d po in te d ou t tha t, i n
26 add it io n to t he m or ta lit y at t he s a lva ge f ac il it y, t he re is a h ig h
cha nc e of p re da ti on f or th e fi sh r e lea se d ba ck i nt o th e wes te rn
27 Del ta a ft er s al va ge . 3/ 31 /1 0 Tr . 1 32: 16 -2 4. Th e be st o pti on i s to
kee p th e fi sh o ut o f Old R iv er . I d . a t 13 2: 24 -1 33 :1 . T his i s a
28 mat te r of s ci en ti fi c dis pu te a mo ng exp er ts .
41
1 Mr. Stua rt opines th at the take limits alone are not

2 sufficie ntly protect ive without implementation of the RPA
3
Actions. See, e.g., id. at 148:20-149:1; BiOp 105 at 729
4
(“If les s take occur s from the propos ed act ion than is
5
anticipa ted, this do es not indicate that the acti ons
6
compromi sing the RPA are not necessary to avoid
7
8 jeopardi zing listed species.”).
9 b. Take of salmon and s teelhead at the pumps is
10 only a “ small fracti on” of their overall mortality,

11 3/31/10 Tr. 126:5-7, and does not account f or ind irect
12
impacts of export pu mping. 3/31/10 Tr. 114:10-15. Mr.
13
Cramer, expressed no opinion whether enjoining Ac tion
14
IV.2.1 w ould increas e indirect mortality. 3/31/1 0 Tr.
15
36:22-37 :25.
16
17 101. Action I V.2.1 also h elps spring-run C hinook
18 salmon, because “the reduced export rates [caused by

19 Action I V.2.1] creat e a more positive OMR flow wi thin the
20
southern central Del ta,” resulting in less fish e ntrained
21
when ent ering the Sa n Joaquin River a t Mokelumne.
22
3/31/10 Tr. 124:9-15.
23
102. However, the record does not support a finding
24
25 that the specific Ve rnalis flow to export ratios imposed
26 by Actio n IV.2.1 (as opposed to lesser or greater ratios)
27 are nece ssary to avo id jeopardy and/or adverse

28
42
1 modifica tion to any of the Listed Species. The t otal

2 absence of explanati on for the exact flow limits chosen
3
makes Ac tion IV.2.1 arbitrary and capricious.
4
5 (4) Action I V.2.3.
6 103. Action I V.2.3 operat es from January 1 through
7 June 15 or until the average daily water temperature a t

8 Mossdale is greater than 72º F, and limits OMR fl ows to
9
no more negative tha n -2,500 to -5,000 cfs, depending on
10
juvenile entrainment levels. BiOp at 648-52. At the
11
first le vel of incre ased juvenile loss, exports must be
12
reduced to achieve a n average net flow of -3,500 cfs f or
13
14 a minimu m of five da ys, and at the second level, a more
15 positive OMR average of - 2,500 cfs must be achiev ed for
16 at least five days. Id. For each trigger, OMR averag es

17 can retu rn to
18
-5,000 c fs only afte r three consecuti ve days of not
19
meeting the higher-density juvenile l oss tr igger. Id.
20
104. Action I V.2.3 is mea nt to:
21
22 [r]educe the vulnera bility of emigrating
juvenile winter-run, yearling spring-run, a nd CV
23 steelhea d within the lowe r Sacramento and S an
Joaquin rivers to en trainment into the channels
24 of the S outh Delta a nd at the pumps due to the
diversio n of water b y the export facilities in
25 the Sout h Delta. Enh ance the likelihood of
salmonid s successful ly exiting the Delta at
26 Chipps I sland by cre ating more suitab le
hydrauli c conditions in the mainstem of the San
27 Joaquin River for em igrating fish, including
greater net downstre am flows.
28
43
1 BiOp at 648. RPA Ac tion IV.2.3 is intended to be nefit

2 fish com ing from bot h the Sacramento and Sa n Joaquin
3
River ba sins. 4/1/1 0 Tr. 101:18-102:2.
4
105. NMFS uti lized severa l sources of data to
5
determin e that expor t flow limitations would achi eve the
6
objectiv es of RPA Ac tion IV.2.3, including the
7
8 relation ship between OMR flows and salvage, parti cle
9 tracking model simul ations, and other studies eva luating
10 survival of fish wit hin the central and southern Delta.

11 4/1/10 T r. 134:5-17.
12
13 a. Reliance on Particle Tracking Model
Simulati ons.
14
106. Plaintif fs’ seminal challenge to Action IV.2.3
15
is that NMFS imprope rly based its rationale for t he
16
Action o n outputs fr om computer model runs utiliz ing the
17
so-called Part icle Tracking Model (“PTM”), which models
18
19 the flow of inert pa rticles as they move within a flowing
20 body of water.
21 107. PTM is a hydrodynami c simulation used to assess

22
the fate of particle s, as a function of flow, tid es,
23
exports, and other f actors. 4/1/10 Tr. 18:12-15; see
24
also id. at 143:9-25. NM FS used PTM to ass ess th e
25
effects of different OMR flows on the movement of
26
27 neutrall y buoyant pa rticles injected at nine diff erent
28 location s in the Del ta. Gov’t Salmon Ex. 23 at 2 ; BiOp

44
1 at 364-6 6. The 2009 Salmonid BiOp states that “N MFS uses

2 the find ings of PTM simulations to look at the ev entual
3
fate of objects in t he river over a d efined perio d of
4
time fro m a given po int of origin in the system.” BiOp
5
at 366. According t o the BiOp, “PTM data can be useful
6
to indic ate the magn itude of the net movement of water
7
8 through the channel after the junction split (and the
9 route se lected by th e fish), and thus can be used to
10 infer th e probable f ate of salmonids that are adv ected

11 into the se channels during their migration.” Id. at 3 67.
12
108. Mr. Cava llo opined t hat PTM data are not useful
13
to infer the probabl e fate of salmonids because, in
14
contrast to PTM part icles, which have no behavior
15
characte ristics, fis h have behavior, swim quickly , and
16
17 have a d estination i n mind. 4/1/10 Tr. 20:14 – 21:5.
18 Mr. Cram er explained that “[j]uvenile salmonids a re

19 strong s wimmers whos e movements are determi ned by a wi de
20
variety of factors v arying with species, size,
21
developm ental state, season, time of day, and wat er
22
temperat ure, as well as relative hydraulic condit ions in
23
a channe l. Unlike p assive particles, juveniles c an and
24
25 do swim against significant currents.” SLDMWA Ex. 120 at
26 ¶6. To illustrate t he problems with PTM, Mr. Stu art
27 compared PTM simulat ions to actual data from mark -

28
45
1 recaptur e studies of Chinook salmon. This compar ison

2 demonstr ated that sa lmon move approximately 3.5 t imes
3
faster t hough the wa ter than neutrall y buoy ant particl es
4
and woul d arrive at Chipps Island in a considerab ly
5
shorter time frame. 4/1/10 Tr. 37:13 – 38:4.
6
109. This was a concern e xpressed in other studies by
7
8 other ex perts. For example, the BiOp relied upon Wim J.
9 Kimmerer and Matthew Nobriga’s report entitled
10 “Investi gating Parti cle Transport and Fate in the

11 Sacramen to-San Joaquin Delta Using a Particle Tra cking
12
Model” ( “Kimmerer an d Nobriga 2008”). BiOp 105 a t 380-
13
381; Gov ’t Salmon Ex . 1 at ¶4; Gov’t Salmon Ex. 4 at ¶ 8.
14
Kimmerer and Nobriga 2008 disclaims: “[w]e do not claim
15
that the specific re sults presented here represen t actual
16
17 movement s of salmon; rather, these results indica te what
18 factors may or may n ot be important in determinin g how

19 salmon s molts may mo ve through the De lta.” DWR Ex. 50 1
20
at 18.
21
110. DWR expr essed simila r concerns in an email to
22
NMFS dat ed April 20, 2009 regarding the draft 200 9
23
Salmonid BiOp, asser ting that NMFS improperly app lied the
24
25 PTM resu lts in deter minin g th e eventual fate of
26 salmonid s. Attachme nt 1 to DWR’s comments is a
27 comparis on of the re sults of an experimental rele ase of

28
46
1 coded wi re tagged sa lmon in the San Joaquin River under

2 known hy drodynamic c onditions with a PTM simulati on under
3
identica l cond itions. 4/1/10 Tr. 32:19-33: 8. Th ese
4
results indicate tha t under low flow conditions, the
5
coded wi re tag salmo n reached the end location of Chipps
6
Island l ong before t he arrival of most of the PTM
7
8 particle s. The PTM results only partially corres ponded
9 with the coded wire tag results under high flow
10 conditio ns. I d. at 34:3- 35:18; DWR Ex. 502 at AR

11 00086765 , AR 0008676 7.
12
111. NMFS rec ognized the limitations of applying the
13
PTM mode l simulation to salmonids. 4/1/10 Tr. 14 4:2-8.
14
There we re dis cussions with DWR concerning this issue
15
during t he consultat ion process. Id. at 144:9-11. In
16
17 discussi ons between DWR and NMFS, NMFS indicated it was
18 using th e PTM to eva luate water movement and the

19 potentia l vulnerabil ity to particle entrainment f rom
20
various locations in the Delta. Id. at 144 :13-19. NMFS
21
was expl icit that it was not using PTM to predict exactly
22
how fish were moving within these same channels, but that
23
the info rmation glea ned from PTM about water move ment
24
25 through the Delta co uld provide infor mation on
26 vulnerab ility to ent rainment. Id. at 144:19-25.
27 112. DWR’s ex pert, Mr. Ca vallo, agrees with the BiOp

28
47
1 that PTM data can be useful to indicate the magni tude of

2 the net movement of water through a channel after a
3
junction split. Id. at 20:21 -23; BiO p at 367.
4
113. Mr. Cava llo also agr ees that PTM results may be
5
informat ive with reg ard to salmon movement. 4/1/ 10 Tr.
6
28:21-25 . Mr. Cavall o stated that und er the appropriate
7
8 conditio ns, PTM simu lations would be an appropria te tool
9 to descr ibe fish mov ement in discharge-driven portions of
10 the Delt a watershed. Id. at 86:8-10. Mr. Cavallo stated

11 that the Kimmerer an d Nobriga PTM study shows tha t “flow
12
has a bi g effect on the path that water takes thr ough the
13
Delta,” and that fish in a riverine system will t end t o
14
go with the flow. Id. at 30:11-15.
15
114. Mr. Cava llo’s time-step critique of t he PTM
16
17 simulati ons used in the BiOp is unsupported.
18 a. Mr. Cava llo opines t hat the correct approach

19 to PTM s imulations i s be to ensure that the time horizon
20
used in the model wa s consistent with the time ho rizon of
21
the fish being studi ed. Id. at 25:6-11. Mr. Cavallo
22
interpre ted particul ar graphs in the biological o pinion
23
to indic ate that NMF S used a 31-day t ime ho rizon in it s
24
25 PTM simu lations, id. at 26:6-16, and opined that this
26 time hor izon was too long and would skew the resu lts of
27 the simu lation, id. at 27:7-11.

28
48
1 b. The PTM simulations NMFS used were run by

2 DWR. Id . at 86:14-15; 146:9-10. These sim ulatio ns
3
included four model runs for the months of February
4
through June, using both wet year, a dry year, an d varied
5
whether HORB was ins talled during the April/May p eriod.
6
Id. at 1 46:14- 24, 147:4-6. T hree different OMR flows
7
8 were exa mined: -3,000 cfs, -2,500 cfs , and -1,250 cfs.
9 Id. at 147:15- 18. During that simulation, the particl es
10 actually were tracke d every five days for the fir st 30

11 days. I d. at 147:1- 4; Go v’t Salmon Ex. 23 at 2. Mr.
12
Cavallo was unsure t hat the particles were tracke d every
13
five day s, nor did h e review Mr. Stuart’s memorandum
14
explaini ng the PTM s imulation results. 4/1/10 Tr . 87:11-
15
13.
16
17 115. Mr. Cava llo’s critiq ue of the choice of
18 injectio n sites is w eakened by his agreement that at

19 least tw o of the par ticle injection sites modeled by DWR,
20
at NMFS’ request, we re useful in eval uating the moveme nt
21
of water particles a t channel junctions. Id. at 90:17 -
22
91:16. NMFS selecte d the particular injection si tes in
23
order to model the v ulnerability of particles wit hin the
24
25 waterway s of the sou th Delta. Id. at 147:2 2-149:13.
26 116. NMFS’ PT M simulation also showed that, as export
27 levels i ncrease, OMR levels became more negative. 4/1/10

28
49
1 Tr. 150: 21-21. Mr. Cavallo stated that exports a re

2 highly c orrelated wi th OMR flows. 4/1/10 Tr. 40: 25-41 :2.
3
117. NMFS’ PT M simulation showed that, as exports
4
increase d, the perce ntage of particles entrained at the
5
export f acilities in creased, particularly from th e
6
Mossdale and Union I sland sites and stations 912, 815,
7
8 902, and 915. 4/1/1 0 Tr. 150:22-25; see Gov’t Salmon Ex.
9 18 (map of injection sites). The proximity of th e
10 injectio n point to t he export facilities led to a much

11 higher l evel of part icle entrainment. 4/1/10 Tr. 151:1-
12
3. As e xports incre ased, the rate at which the p articles
13
arrived at the export facilities increased. Id. at
14
151:3-5; see also BiOp at 365-66; 4/1 /10 Tr. 151:21-15 3:9
15
(explain ing graphs i n biological opinion).
16
17 118. Despite the statemen t in the Kimmerer and
18 Nobriga study that t hey could not establish a “zo ne of

19 influenc e” of export s, Mr. Stuart tes tified that the
20
shorter time horizon used in NMFS’ PTM simulation s
21
distingu ished it fro m the Kimmerer and Nobriga
22
simulati ons, which u tilized a 90-day period . 4/2/10 T r.
23
23:21-24 :2.
24
25 119. Mr. Stua rt testified that there is no preci sely
26 defined boundary for the influence of the exports , and
27 that the boundary of influence depends on river f low,

28
50
1 tides, a nd the magni tude of the exports. Id. at 29:4- 9.

2 If there are extreme ly low-flow conditions and hi gh
3
exports, the e xtent of th e exports could tr avel
4
consider ably farther downstream, even towards the
5
junction of the Sacr amento and San Joaquin Rivers . Id .
6
at 29:9- 13. Typical ly, according to Mr. Stuart, the
7
8 boundary would be cl ose to station 815 at the con fluence
9 of Georg iana S lough and the Mokelumne River or sl ightly
10 farther downstream. Id. at 2 9:13-15. As the BiOp

11 explains :
12
The data output for the PTM simulation of
13 particle s injected a t the confluence of the
Mokelumn e River and the San Joaquin River
14 (Station 815) indica te th at a s net OMR flow
increase s southwards from -2,500 to -3,500 cfs,
15 the risk of particle entrainment nearly doubles
from 10 percent to 2 0 percent, and quadruples to
16
40 perce nt at -5,000 cfs. At flows more neg ative
17 than -5, 000 cfs, the risk of entrainment
increase s at an even greater rate, reaching
18 approxim ately 90 per cent at -7,000 cfs. Even if
salmonid s do not beh ave exactly as neutrally
19 buoyant particles, t he risk of entrainment
escalate s considerab ly with increasing exports,
20
as repre sented by the net OMR flows. The lo gical
21 conclusi on is that a s OMR reverse flows
increase , risk of en trainment into the channels
22 of the S outh Delta i s increased. Conversely, the
risk of entrainment into the channels of the
23 South de lta is reduc ed when exports are lower
and the net flow in the OMR channels is more
24 positive -- th at is, in t he d irection of th e
25 natural flow toward the ocean.
26 BiOp at 652.
27 120. This is a dispute am ong scientists. While DWR

28
51
1 criticiz es PTM model ing, Stuart and NMFS recogniz ed its

2 limitati ons and foun d PTM studies helpful to supp ort its
3
conclusi ons that: (a ) as exports increase, negati ve OMR
4
flows al so increase; and (b) that at Station 815 (the
5
confluen ce of the Mo kelumne River and the San Joa quin
6
River), particle ent rainment increases from 10% at -2, 500
7
8 cfs, to 20% at -3,500 cfs, to 40% at -5,000 cfs, and 90%
9 at -7,00 0 cfs. NMFS, through Mr. Stuart, took into
10 account inherent dif ferences in the movement of n eutrally

11 buoyant particles an d their speed and direction o f
12
travel. Admin istrative law requires deference to the
13
Agency. Additional record analysis is necessary to
14
determin e the extent of support for NMFS’s additi onal
15
opinion that exports affect salmonid survival.
16
17
b. Addition al Data Reli ed Upon by NMFS.
18
(1) Salvage Data.
19
121. NMFS als o relied on salvage data provided by
20
Plaintif f-Inte rvenor DWR. 4/1/10 Tr. 134:2 1; see Gov’t
21
Salmon E x. 1 at (int ernal) Exhibit 3. This data
22
23 collecte d monthly av erage OMR flows for the month s of
24 December to April 19 95-2007 and the monthly older
25 juvenile loss number s for both the state and the federal
26 faciliti es. I d. at 135:18-136:8.
27
122. This dat a was presen ted in Figures 6-65 and 6-66
28
52
1 of the B iOp:
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
BiOp at 361-62 .
21
123. Based on this data, NMFS determined that there
22
was a th reshold leve l of pumping, as reflected by OMR
23
flows, b elow which e ntrainment was low, but above which
24
25 entrainm ent at the P roject facilities markedly in creases.
26 4/1/10 T r. 139:11-16. Th e threshold level identi fied by
27 NMFS is -5,000 cfs. Id. at 1 39:18-21.

28
53
1 124. There is evidentiary support for the conclusions

2 that: (1 ) entrainmen t data show that as exports i ncrease,
3
so does juvenile sal vage; and (2) that at flows m ore
4
negative than -5,000 cfs, OMR salvage increases more
5
rapidly than at lower flow levels.
6
125. However, The compari sons of salvage to negative
7
8 OMR flow s relied upo n in the BiOp utilize raw sal vage
9 numbers, rather than scaling salvage to populatio n size.
10 See Doc. 179, Declaration of Richard B. Deriso at ¶¶ 3-5.

11 Scaling salvage to p opulation size is standard fi sheries
12
science practice and could have been accomplished for
13
several of the Liste d Species based on existing
14
populati on data. See id. at ¶¶ 5-6. This failure is a
15
fundamen tal and inex plicable error. Salvage may have
16
17 been hig her in some years simply because the popu lation
18 was high er, not beca use of any differences in neg ative
19 OMR flow s. Salvage may have been lower in other years
20
because the populati on was lower. Dr. Deriso
21
demonstr ated the pot ential significance of this f ailur e
22
by plott ing the popu lation adjusted Juvenile Chin ook
23
Incident al take rate against OMR flow. Based upo n this
24
25 revised analysis for spring-run and w inter-run, Dr.
26 Deriso c oncluded tha t there is no statistically
27 signific ant relation ship betw een the take index a nd OMR
28
54
1 flows. Id. at ¶6.

2 126. The BiOp ’s conclusio ns reached about the spring-
3
run and winter-run Chinook failed to utilize the best
4
availabl e scientific methodology, because populat ion data
5
was avai lable at the time the BiOp was issued that wou ld
6
have per mitted NMFS to perform the straightforwar d
7
8 populati on adjustmen t required to conform to stan dard,
9 generall y accepted p ractices for fisheries popula tion
10 measurem ents utilize d in their field of expertise . If,

11 in those years when salvage was greatest, populat ion
12
sizes ov erall were 1 0 or 100 times larger than ot her
13
years, t he effects m ight not be jeopardizing. Wi thout
14
adjustme nt for popul ation size, NMFS’s reliance o n that
15
figure w as arbitrary and capricious.
16
17 127. As to th e CV steelhe ad, for which no population
18 numbers are availabl e, it is less clear whether t he use

19 of raw s alvage numbe rs is always inappropriate. Figures
20
6-65 and 6-66 ambiguously reference monthly CVP a nd SWP
21
“Older J uvenile Loss ” on the y axis. Were most of the
22
salvaged fish repres ented on these charts Chinook salmon?
23
No reaso n is offered why NMFS did not segregate t he
24
25 steelhea d figures fr om those of Chinook salmon. If the
26 species had been eva luated separately, would it h ave been
27 reasonab le for NMFS to fa il to adjust the s teelhead

28
55
1 figures for populati on size? Separate analysis w as not

2 done.
3
4 (2) Delta Ac tion 8 Studi es.
5 128. NMFS rel ied upon New man’s 2008 analysis of the
6 Delta Ac tion 8 studi es discussed above. See also BiOp at
7 373 (Gen eral D iscussion of Relationship of Exports to

8 Salvage) . These res ults demonstrate that as expo rts
9
increase there is de creased survival for salmonid s
10
passing through the south and central Delta. Geo rgianna
11
Slough e nters the De lta at Station 815.
12
129. Newman’s and Brandes ’ (2009) Delta Action 8
13
14 studies found that d etermining the proportion of all
15 Sacramen to River smo lts volitionally migrating th rough
16 Georgian a Slough is essential to evaluating the

17 populati on level or biological significance of any exp ort
18
effects, at least on those populations that spawn in the
19
upper Sa cramento bas in (e.g., winter-run Chinook salmo n).
20
DWR Ex. 507 at 24. NMFS did not address relative
21
22 populati on impacts i n developing or explaining RP A Action
23 IV.2.3. 11
24 130. Even ass uming all sm olts traveled through

25 Georgian a Slough, Mr . Cavallo testified that unde r
26
11
Al th ou gh t he s am e fa ilu re a pp li es to NM FS ’s u se o f th e Del ta
27 Act io n 8 da ta i n IV .2 .1, t ha t Ac ti o n w as d es ig ne d to h el p t he S SN DG
of CV S te el he ad , al l of wh om m us t p ass t hr ou gh t he c en tr al De lt a on
28 the ir w ay t o th e oc ea n.
56
1 Newman’s weak export-mortality relati onship , a 2,000 c fs

2 increase (from 4,000 to 6,000 cfs) in exports wou ld
3
increase total morta lity by five percent. 4/1/10 Tr.
4
63:8-25. Howe ver, based on his review of a vailab le da ta,
5
Mr. Cava llo estimate d that no more than 22% of sm olts
6
originat ing in the S acramento River would pass th rough
7
8 Georgian a Slough, lo wering the impact on these
9 populati ons of a 2,0 00 cfs increase t o one percent. Id.
10 131. NMFS’s f ailure to ev aluate the population level

11 impacts of exports i s inexplicable. A population level
12
evaluati on would she d light on the relative impac t of
13
exports on the winte r-run , fo r which no populatio n spawns
14
in the S an Joaquin b asin. This failure is less c ritical
15
to the a nalysis of i mpacts on spring-run and CV
16
17 steelhea d, as both s pecies have important populat ions
18 that spa wn in tribut aries of the San Joaquin and

19 necessar ily must pas s through the interior Delta on th eir
20
way to t he ocean.
21
22 c. Perry & Skalsi.
23 132. The BiOp utilized th e Perry and Skalski (2008)
24 study th at concluded survival of fish moving into

25 Georgian a Slough and nearby channels was reduced compared
26
to those in the main stem of the Sacramento River. 4/1/10
27
Tr. 161: 20-162 :1. These fish enter a portion of the San
28
57
1 Joaquin River that N MFS found to be impacted by e xports

2 in its P TM simulatio n. Id. a t 162:5-17; 4/2/10 Tr.
3
18:12-20 , 19:22-20:11.
4
133. However, Perry and S kalski 2008 noted that
5
“there i s limited un derstanding of how water mana gement
6
actions in the Delta affect population distributi on and
7
8 route-sp ecific survi val of juvenile s almon.” SDLMWA Ex.
9 227 at 3 . Mr. Caval lo testified that Perry and S kalski
10 2008 doe s not provide sci entific support fo r the view

11 that sal monids are l ost due to water project-induced
12
alterati ons to Delta hydrologic conditions. 4/1/ 10 Tr.
13
66:5-9.
14
134. Mr. Stua rt admitted that Perry and Skalski 2008
15
did not address wate r project impacts on Delta hydrolo gy,
16
17 fish beh avior, or th e indirect mortality of fish in the
18 central and southern channels of the Delta. Mr. Stuart

19 further admitted tha t he reached his conclusions
20
regardin g water proj ect impacts on Delta hydrolog y, fish
21
behavior , and indirect salmonid morta lity b ased upon h is
22
personal extrapolati on from the data contained in Perry
23
and Skal ski 2008, an d not from any conclusions re ached by
24
25 Perry an d Skalski. 4/2/10 Tr. 19:2 – 21:24. However,
26 these pe rsonal extra polations are not documented or
27 otherwis e explained in the BiOp or elsewhere in t he

28
58
1 record.
2
d. Vogel.
3
135. The BiOp also relied upon Vogel (2004), which
4
5 reviewed telemetry-tagging data to in vestig ate fish ro ute
6 selectio n in the cha nnels leading to the south De lta.
7 See BiOp at 38 0-81. Based on Vogel’s work, the B iOp

8 found th at when expo rt levels were reduced and Sa n
9
Joaquin River flows were increased, more fish sta yed in
10
the main channel of the San Joaquin River, headin g
11
downstre am toward th e San Francisco Bay. Id.
12
136. Mr. Cava llo maintain s that Vogel (2004) does not
13
14 support the conclusi on that a reduction in export pumping
15 resulted in the redu ction of salmon leaving the m ainstem
16 of the S an Joaquin R iver and entering the souther n Delta.

17 4/1/10 T r. 47:20-24, 49:8 -13, 49:25 – 50:4, 50:17 -23; DWR
18
Ex. 505. The Vogel (2004) study concluded that t he
19
experime nts it condu cted “could not explain why some fish
20
move off the mainste m of the San Joaquin River in to the
21
22 south De lta channels ,” noting that “[d]ue to the wide
23 variatio n in h ydrologic conditions” during the course of
24 the expe riments, “it was difficult to determine t he

25 principa l factors af fecting fish migration. Bas ed on
26
the limi ted data fro m these studies, it may be th at a
27
combinat ion of a nea p tide, reduced exports, and
28
59
1 increase d San Joaqui n River flows is beneficial f or

2 outmigra ting smolts, but more research is necessa ry.”
3
DWR Ex. 505 at 37.
4
137. When ask ed about Vog el’s inconclusive results,
5
not disc ussed in the BiOp, Mr. Stuart admitted th at the
6
BiOp’s f ailure to di sclose the conclu sion w as “an
7
8 oversigh t on my part ,” for which he had no explan ation.
9 4/2/10 T r. 15:4-9.
10 138. It was not rational nor scientifically

11 justifie d for the Bi Op to rely on Vogel (2004) fo r
12
findings the authors themselv es refused to make.
13
14 e. Justific ation for Sp ecific Flow Levels.
15 139. The only discernable and scientifically
16 justifia ble support provided in the BiOp for the negative

17 5,000 cf s ceiling on OMR flows under Action IV.2. 3 is the
18
salvage data, represented in Figures 6-65 and 6-66 of the
19
BiOp. S ee Gov ’t Sal mon E x. 1 at (internal) Exhibit 3.
20
Based on this data, NMFS concluded that
21
22 -5,000 c fs represent ed a “threshold l evel” of pumping,
23 reflecte d by OMR flo ws, below which species entra inment
24 was low, but above w hich entrainment at the Proje ct

25 faciliti es markedly increases. 4/1/10 Tr. 139:11 -16.
26
The BiOp discusses F igures 6-65 and 6-66:
27
Loss of older juveni les at the CVP and SWP fish
28 collecti on facilitie s increase sharply at Old
60
1 and Midd le River flows of approximate ly -5, 000

cfs and depart from the initial slope at flows
2 below th is.
3
The reco rd does not explain whether NMFS utilized a
4
statisti cal analysis to choose -5,000 cfs a s the break
5
point, o r whether th at figure was based on a visu al
6
inspecti on of Figure s 6-65 an d 6-66.
7
8 140. NMFS con sidered sett ing more positive OMR flow
9 requirem ents, which would have been more benefici al for
10 the list ed salmonids , but would place more restri ctions

11 on expor ts. 4/1/10 Tr. 178:17-22. Mr. Stu art te stifi ed
12
that he “tried to fi nd a point that would be equitable ”
13
to balan ce species p rotection and burdens on the exports.
14
Id. at 178:24- 179:6 (emphasis added).
15
141. Mr. Stua rt testified that:
16
17 [T]he mi nus 5,000 wa s sufficiently [ ]
restrict ive to prote ct th e fi sh from
18 entrainm ent. To go more positive than that
would ha ve been bett er, but I don’t think that I
19 would ha ve gained th at much. And, you know, I
did, you know, consi der that to go more positive
20
you’d ha ve to put mo re restrictions on the
21 exports. And I trie d to balance that
relation ship. You k now, more negative would
22 have tak en more [ ] fish, which was less
protecti ve of our sp ecies. To go more positive
23 would ha ve been more protective, but it would
have bee n a very one rous burden on the exports.
24 [¶] So, you know, I tried to find a point that
25 would be equitable. I didn’t run a full
detailed hydraulic a nalysis and water analysis
26 on that, but, you kn ow, to balance those two was
in my mi nd as I was looking at the minus 5,000
27 as the t rigger point .
28
61
1 4/1/10 T r. 178:17 – 179:6. This effo rt to choose a

2 “balance point,” is not supported by any scientif ic
3
analysis .
4
142. Mr. Stu art testifie d that he “looked at ... the
5
level wh ere we saw i ncreasing take and use[d]
6
precauti onary ... princip les to protect the fish. ” Ye t,
7
8 nowhere in the BiOp (or any other document in the
9 administ rative recor d cited by the parties) does NMFS
10 disclose its intent to use a “precautionary princ iple” to

11 design t he RPA Actio ns, nor is that “level” speci fically
12
defined or justified .
13
143. The -5, 000 cf s OMR ceiling is based,
14
predomin antly on spe culation.
15
144. Moreove r, Figures 6-65 and 6-66, do not scale
16
17 salvage to populatio n size. This further undermi nes
18 NMFS’s e xtrapolation of the -5,000 cfs “bre ak poi nt,” and

19 affects the credibil ity of Mr. Stuart’s testimony .
20
21 f. Will Enj oining Actio n IV.2.3 Appreciably

Diminish The Likelih ood Of Survival Or
22 Recovery Of The List ed Species Or Adversely
Modify T heir Critica l Habitat?
23
145. Although the m oving paper s se ek an unlimite d
24
injuncti on of Action IV.2.3, at the evidentiary h earing,
25
Plaintif f-Inte rvenor DWR clarified that an injunction was
26
sought o nly against the so-called “calendar-based
27
28 triggers ” of Action IV.2.3, and that it does not oppose
62
1 the salv age-ba sed triggers of Action IV.2.3. 4/1/10 T r.

2 9:7-10:1 7. DW R accepts the underlying scientific
3
principl e that when significant salvage occurs at project
4
pumps, t he projects operations must be altered. Id. at
5
10:11-13 . In prior remedial proceedings, s ome Pl ainti ffs
6
have ack nowledged th at at flows more negative tha n -7, 000
7
8 cfs, Del ta smelt and the continued existence of t wo
9 Chinook salmon speci es are jeopardized. See, e.g., PC FFA
10 v. Gutie rrez, 2008 WL 4657785, *6 (Oct. 21, 2008). Th e

11 proposed injun ction appli es only to the “ca lendar -based
12
triggers ” of RPA Act ion IV.2.3.
13
146. There ar e serious qu estions whether there is
14
support in the recor d for the general proposition that
15
exports reduce survi val of salmonids in the inter ior
16
17 Delta.
18 a. The PTM studies do s tand for the proposition

19 that neu trally buoya nt particles injected at Stat ion 815
20
have a h igher chance of entrainment as negative O MR flows
21
increase . But, part icles are not a reasonably ac curate
22
prototyp e for the be havior of strong-swimming Chinook
23
salmon, steelhead, a nd sturgeon.
24
25 b. The salv age data was not scaled for
26 populati on size, whi ch any prudent and competent fish
27 biologis t and statis tician would have done, makin g NMFS’

28
63
1 reliance on the salv age data scientifically erron eous for

2 those sp ecies for wh ich abundance data are availa ble.
3
The effe ct of this e rror on NMFS’s evaluation of export
4
impacts on CV steelh ead is less clear.
5
c. NMFS’s r eliance on t he Perry & Skalski and
6
Vogel st udies is unj ustified and unreasonable, gi ven that
7
8 NMFS rel ied upon tho se studies to support conclus ions the
9 authors refused to r each without explanation.
10 d. The Delt a Action 8 s tudies, at the very

11 least, s upport the p roposition that, for those sa lmonid
12
populati ons spawning entirely within the San Joaq uin
13
basin, i ncreasing ex ports can negatively impact s almonid
14
smolt su rvival. Thi s data, coupled with the high ly
15
criticiz ed PTM studi es, are the questionable foun dation
16
17 underlyi ng NMFS’s ra tionale for Action IV.2.
18 e. Mr. Stua rt testified that if the calendar-

19 based po rtion of the Action were enjoined, jeopar dy to
20
the spec ies would no t be avoided because it would “affect
21
a large proportion o f the spring-run popula tion, a
22
portion of the steel head population, and that por tion of
23
the gree n sturgeon p opulation that’s currently within the
24
25 Delta.” Id. a nd 186:2-5. (Although, not one stu rgeon
26 has been taken as of April 4, 2010.) As further
27 explaine d in Mr. Stu art’s declaration:

28
64
1 Without the protecti on of RPA action IV.2.3, OMR

flows wi ll increase in relation to the increase
2 in expor ts, and more fish will be lost to the
export a ctions over current conditions. In
3 addition to the loss [of] salmonids during the
salvage process, it is expected that a greater
4 number o f listed fis h will be exposed to
stressor s in the delta as they are ad vected into
5 the chan nels of the central and southern delta
by the a ltered hydra ulic conditions. Loss to
6 predatio n, as well a s other stressors such as
contamin ants, is exp ected to occur as a result
7 of this increased ex posure.
8 Gov’t Sa lmon E x. 4, ¶62. Action IV.2.3 is design ed to
9 protect the fish fro m being pulled south towards the
10
faciliti es; a purely salvage-based operatio n is
11
reaction ary and refl ects the pre-biological opinion
12
status q uo, which NM FS determined was not suffici ently
13
protecti ve. 4 /1/10 Tr. 170:9-171:7.
14
15 147. Plaintif fs’ offer to use the species’ incidental
16 take lim its to avoid jeopardy is not sufficiently
17 protecti ve. The ITL is not meant to be a ceiling on

18 mortalit y, in part b ecause it “doesn’t address al l of the
19
differen t forms of t ake that can occur throughout the
20
whole Ce ntral Valley .” 4/1/10 Tr. 172:21-73:1.
21
148. NMFS’s c hoice of -5,000 c fs as the calendar
22
based ce iling for Ac tion IV.2.1 is not scientific ally
23
24 justifie d and is not based on best available science.
25
(5) Indirect Mortality.
26
149. Indirect mortality i s that mortality that does
27
not occu r directly a s a result of the entrainment process
28
65
1 at the P roject pumps . 3/31/10 Tr. 104:22-24. St ated

2 another way, it is t he sum of mortality that occu rs to
3
fish tha t are under the influence of the changed
4
hydrauli c field with in the Delta. Id. at 105:1-3.
5
150. Indirect mortality i s observed within the
6
channels and waterwa ys of the northern, central, and
7
8 southern Dela. Id. at 10 9:23-24.
9 151. DWR’s ex pert, Mr. Ca vallo, does not contend that
10 there is no indirect loss, 4/1/10 Tr. 94:10-12, n or th at

11 indirect mortality i s not a stressor on fish as t hey move
12
through the system, id. at 94:13-15. Mr. C avallo agre es
13
that a r easonable biologi st addressing the impacts of the
14
Projects should not have ignored indirect mortali ty. Id.
15
at 94:16 -19.
16
17 152. This bel ies DWR’s pr esent contention that
18 indirect mortality i s not related to Project oper ations,

19 as does information submitted by DWR in the prior
20
litigati on estimatin g indirect mortality attribut able to
21
exports. 4/1/10 Tr. 190:7-191:10; see D-I Ex. 1003 at
22
(interna l) Exhibit 2 . NMFS relied on this inform ation in
23
preparat ion of the c urrent biological opinion. 4 /1/10
24
25 Tr. 191: 13-18; see D-I 1011. The inf ormati on provided by
26 DWR sugg ests that, b ased on certain water year ty pes and
27 export t o inflow (“E /I”) ratios, there could be

28
66
1 substant ial export-related mortality in the interior

2 Delta. 4/1/10 Tr. 1 92:9- 14. Such mortality may be
3
substant ially greate r than direct take at the CVP and
4
SWP. Se e id. at 190:17-190:1 0; see a lso D-I Ex. 1011.
5
153. Plaintif fs’ expert, Mr. Cramer, did not deny the
6
existenc e of indirec t mortality, but stated that it had
7
8 not been adequately tested. 3/31/10 Tr. 19:2-15.
9 154. Acoustic tag studies are beginning to provide
10 estimate s of indirec t mortality in the Delta. Id. at

11 105:9-10 . The Perry and Skalski (200 8) paper showed a
12
survival rate of abo ut 30 to 35% for interior Del ta
13
waters. Id. a t 105:15-17, 108:15-18; see SLDMWA Ex. 2 27
14
(Perry & Skalski (20 08)). Perry and Skalski did not
15
attribut e any partic ular portion of this to the p rojects.
16
17
(6) Other St ressors.
18
155. It is un disputed tha t there are numerous
19
stressor s unrelated to project operat ions t hat adverse ly
20
affect a nd jeopardiz e the viability of the Listed Species
21
22 and the quality of t heir critical habitat. The B iOp
23 dedicate s a lengthy section to “Factors Responsib le for
24 the Curr ent Status o f Winter-Run, Spring-Run, CV

25 Steelhea d, and the S outhern DPS of Green Sturgeon .” BiOp
26
at 134-1 57. Among o ther causes, this section dis cusses
27
the foll owing factor s adverse to survival and hab itat
28
67
1 quality:
2 • Habitat blockage by dams of the CVP SWP and other
municipa l and privat e entities;
3
• Water di version and storage;
4
• Anderson -Cottonwood Irrigation District (“ACID”)
5 Dam and Red Bluff Di version Dam (“RBDD”);
6 • Water co nveyance and flood control facilities;
7 • Land use activities throughout the Central Valley ;
8 • Water qu ality degrad ation;
9 • Hatchery opera tions and p ractices;
10 • Over uti lization thr ough commercial and/or sport
harvest;
11
• Disease and predatio n;
12
• Environm ental variat ion (including natural
13 environm ental cycles , ocean productivity, and
global c limate chang e); and
14
• Non-Nati ve Invasive Species.
15
156. Whether and to what extent these factors are
16
17 exacerba ted by proje ct operations has been the su bject of
18 continui ng debate in this and the Consolidated Sm elt
19 Cases. It was not t he subject of briefing in the PI

20 motion i n this case.
21
157. Plaintif fs have argu ed that Federal Defendants
22
have wro ngfully igno red these other causes and ha ve put
23
24 the burd en of remedi ation wholly on the water sup ply and
25 Project operations. Plaintiffs contend that the
26 overwhel ming causes of jeopardy to the species and the ir

27 habitats are these o ther stressors.
28
68
1 158. Federal Defendants h ave not quantified relative

2 harms, nor has any party suggested what remedies will
3
effectiv ely address these other causes.
4
5 D. Irrepara ble Harm.
6 159. The evidence has established a variety of
7 adverse impacts to h umans and the human environme nt from

8 reduced CVP and SWP deliveries, including “irretr ievable
9
resource losses (per manent crops, fallowed lands,
10
destruct ion of famil y and entity farming business es);
11
social d isruption an d dislocation; as well as
12
environm ental harms caused by, among other things ,
13
14 increase d groundwate r consumption and overdraft, and
15 possible air quality reduction.” Doc. 202, 2/5/1 0 TRO
16 Decision , at 15:24-24 - 16:1-4.

17 160. At the s ame time, th e declining health of the
18
salmonid population is harming other interests, i ncluding
19
those of commercial fishermen and Native American s with
20
cultural and spiritu al interests in salmon.
21
22
(1) Water Su pply Impacts .
23
161. It has p reviously be en re cogn ized that “any lost
24
pumping capacity dir ectly attributable to the 200 9
25
Salmonid BiOp will c ontribute to and exacerbate t he
26
currentl y catastroph ic situation faced by Plainti ffs,
27
28 whose fa rms, busines ses, water service areas, and
69
1 impacted citie s and counties, are dependent, some

2 exclusiv ely, upon CV P and/or SWP water deliveries .” Doc
3
202, TRO Decision, a t 15:17-24.
4
162. Every ac re-foo t of p umpin g fo regone during
5
critical time period s is an acre-foot that does not re ach
6
the San Luis R eservoir where it can be stored for future
7
8 delivery to users du ring times of peak demand in the
9 water ye ar.
10 163. It is un disputed tha t, in the three water years

11 prior to the 2009-2010 wa ter year, Californ ia has
12
experien ced three co nsecutive years of drou ght
13
conditio ns. Gov’t S almon Exh. 5 at (internal) Ex hibit 1
14
at 18. This influen ces the amount of run-off for ecast ed
15
for 2010 and is indi cative of why reservoir stora ges were
16
17 at a low state enter ing the 2009-2010 water year. 4/1 /10
18 Tr. 208: 7-15. Hydro logic conditions are no t within th e

19 control of the parti es and have materially contri buted to
20
water se rvice reduct ions to contractors.
21
164. It is al so undispute d that other, non-project
22
factors, such as tid es, wind events, storm surges , San
23
Joaquin River flows, Contra Costa Water District
24
25 operatio ns, and dive rsions by in-Delta wate r users imp ose
26 limitati ons on how R eclamation must operate the p roject
27 to meet flow targets . See id. at 202:12-204:1.

28
70
1 165. The proj ects are sub ject to e xport reductions

2 required to protect species listed under the Cali fornia
3
Endanger ed Species A ct, including longfin smelt, delta
4
smelt, w inter- run Chinook salmon, and spring-run Chinook
5
salmon, which subjec t the water project operators to
6
controls under state law that are similar, and, i n some
7
8 cases, i dentical to those contained in the 2009 S almonid
9 BiOp and the United States Fish and Wildlife Serv ice’s
10 (“FWS”) December 15, 2008 Biological Opinion (“20 08 Delta

11 Smelt Bi Op”). See id. at Tr. 212:4-213:8; 4//10 Tr.
12
20:18-21 :20. In the absence of the B iOps’ RPAs, those
13
protecti ons are argu ed to have likely limited exp ort
14
pumping to levels be low those allowable under D-1641,
15
which al so limits Pr oject pumping at certain time s of the
16
17 year. S ee, e.g., SWC Ex. 938 (DWR’s 3/30/10 allocatio n
18 announce ment conside red several “SWP operational

19 constrai nts” includi ng “the incidental take permi t for
20
longfin smelt”).
21
166. Plaintif fs’ estimate s of water losses do not
22
account for or other wise offset losses attributab le to
23
proposed remedies in the consolidated Delta Smelt and
24
25 Salmon c ases. See 4/7/10 Tr. 17:10-20:14.
26
a. Water Su pply Impacts of Action IV.2.1.
27
167. Action IV.2.1 lasts from April 1, 2010 through
28
71
1 May 31, 2010. SLDMW A Ex. 105 at 641-643. The flow

2 requirem ents in Acti on IV.2.1 vary depending on t he
3
February New Melones Index. SLDMWA Ex. 105 at 64 2.
4
Based on the Februar y 2010 New Melones Index of 1 ,779
5
thousand acre- feet (“TAF”) under the 50% exceedance
6
forecast , 12 the minimum flo ws a t Vernalis und er Action
7
8 IV.2.1 w ill be those required to meet the D-1641
9 requirem ents or 3,00 0 cfs, whichever is greatest. Gov’t
10 Salmon E x. 55 at ¶5. Additionally, flows at Vern alis are

11 anticipa ted to be le ss than 6,000 cfs in April an d May
12
2010, wh ich means that co mbined exports wil l likely be
13
limited to 1,500 cfs in April and May when Action IV.2.1
14
controls . Gov’t Sal mon Ex. 55 at ¶5; SLDMWA Ex. 105 at
15
642.
16
17 168. Action I V.2.1 began affecting pumping and water
18 supply a llocations b eginning April 1. 4/6/10 Tr. 188:11-

19 14. Ter ry Erlewine, General Manager of the State Water
20
Contract ors, estimat ed that from April 1 through April 5,
21
2010 SWP and CVP exp erienced a loss of exports of
22
approxim ately 50,000 acre feet. 4/6/10 Tr. 188:1 8-19.
23
He also estima ted that the two Projects would incur
24
25 12
Re cl am at io n on ly c an es ti ma te w ha t wi ll b e co nt ro ll in g CVP
ope ra ti on s in t he f ut ure . 4 /1 /1 0 T r. 20 4: 5- 7. Th e de gr ee of
26 cer ta in ty i n pr ed ic ti ng wh at w il l c ont ro l Pr oj ec t op er at ion s,
par ti cu la rl y in t he w int er a nd s pr i ng, d ec li ne s ra pi dl y pas t tw o or
27 thr ee d ay s. Id . at 2 04: 7- 9. Re cl a mat io n us es D WR ’s m on thl y ru n- of f
for ec as ts t o de ve lo p mon th ly 5 0% a n d 9 0% e xc ee da nc e fo re cas ts o f CV P
28 ope ra ti on s. Id . at 2 06: 13 -2 07 :1 5.
72
1 addition al losses of approximately 50,000 acre fe et, or

2 more, du ring the mon ths of April and May 2010, as a
3
result o f the 2009 S almonid and 2008 Delta Smelt BiOps.
4
4/6/10 T r. 196:19-21; 199 :10-16, 23; SWC Ex . 939.
5
169. The 2009 Salmonid Bi Op estimates that, on
6
average, Action IV.2 .1 could reduce monthly expor ts by 73
7
8 percent in April and 67 percent in May. SLDMWA E x. 105,
9 App. 5 a t 44. NMFS has acknowledged that these
10 reductio ns are in ad dition to the reductions mandated

11 under th e 2008 Delta Smelt BiOp. Id. at 60. If Actio n
12
IV.2.1, Action IV.2. 3, or the 2008 Delta Smelt Bi Op RPA
13
are enjo ined, Reclam ation expects to increase CVP water
14
supply a llocations i n May and June. 4/1/10 Tr. 2 13:14-
15
20.
16
17
b. Water Supply Impacts of Actio n IV.2.3.
18
170. Action I V.2.3 began controlling Reclamation’s
19
and DWR’ s operation of the CVP and SWP, respectiv ely, on
20
January 20, 2010. 4 /1/10 Tr. 199:8-9; Gov’ t Salm on Ex . 5
21
22 at ¶6. This restric tion lasted until January 27, 2010 .
23 Id. at 1 99:11- 13; Gov’t Salmon Ex. 5 at ¶6. From Janu ary
24 27, 2010 through Feb ruary 5, 2010, Action IV.2.3 required

25 OMR flow reductions which, in turn, required Recl amation
26
to restr ict its pump ing at the CVP’s Jones Pumpin g Plant
27
to appro ximately 3,300 cf s. Gov’t Salmon E x. 5 at ¶6.
28
73
1 On Febru ary 6, 2010, Reclamation increased pumpin g at the

2 Jones Pu mping Plant to approximately 4,200 cfs in order
3
to compl y with the t emporary restraining order gr anted on
4
February 5, 2010. G ov’t Salmon Ex. 5 at ¶6. On February
5
10, 2010 , the OMR re quirement for the 2008 Smelt BiOp
6
began co ntrolling op eration of the pumping facili ties.
7
8 4/1/10 T r. 200:6-10.
9 171. From Feb ruary 19 thr ough March 15, 2010, NMFS
10 and FWS independentl y made flow recommendations of -5, 000

11 cfs for OMR flow tar gets, in order to comply with Action
12
IV.2.3 a nd the 2008 Delta Smelt BiOp, respectivel y.
13
4/1/10 T r. 200:5-7; Gov’t Salmon Ex. 5 at ¶ 8.
14
172. San Luis Plaintiffs estimate that for every day
15
that Act ion IV.2.3 c ontrols under a -5,000 cfs limit,
16
17 Reclamat ion’s pumpin g output is reduced by 500 cf s per
18 day. TR O Decision a t 14:8-15. Mr. Erlewin e esti mates

19 that los ses to the c ombined projects between Janu ary 20
20
and Janu ary 26, 2010 exceeded 90,000 acre-feet (“ AF”),
21
and comb ined losses from January 27 t hrough February 5,
22
2010 wer e approximat ely another 100,000 AF. TRO Decision
23
at 14:19 -22; TR 4/6/ 10 183:14-15; SWC Ex. 903. It has
24
25 been reo cognized tha t even if estimates of loss b y Thomas
26 Boardman and Erlewin e “are so excessive that they doub le
27 actual l oss, the fig ures are still significant.” TRO

28
74
1 Decision at 15:1-4.
2
c. Other Fa cts Relevant to Water Supply
3 Impacts.
4 173. It is un disputed tha t even in the absence of the
5
RPAs, th e quantity o f exportable water is still subjec t
6
to regul ation, e.g. under Decision 1641. 4/6/10 Tr. 184-
7
185. Ho wever, the q uantity of exportable water h as been
8
reduced by the imple mentation of the salmonid and smelt
9
10 RPAs. I d. Fr om Jan uary 20 through March 2 4, 2010, Mr .
11 Erlewine testified t hat p oten tial and actual expo rts were
12 diminish ed by 522,56 1 acre feet, of which a 433,0 00 AF

13 loss was attributabl e to the SWP and a 89,000 AF loss was
14
attribut able to the CVP. 4/6/10 Tr. 185:16-19; S WC
15
Demonstr ative Ex. 90 3.
16
174. DWR made its initial wate r su pply allocatio n
17
announce ment on Nove mber 30, 2009, allocating fiv e
18
19 percent of Table A c ontracted amounts for SWP wat er
20 contract ors. 4/6/10 Tr. 240:16-22; S WC Ex. 923, Ex. B .
21 As of Ma rch 30, 2010 , DWR increased the SWP alloc ation

22
for 2010 to a 20% al location. 4/6/10 Tr. 189:15- 17; SWC
23
Ex. 938; 4/1/10 Tr. 249:22-25.
24
175. Reclamat ion announce d its initial allocation of
25
CVP wate r on Februar y 26, 2010. Fed. Salmon Ex. 55 at
26
27 ¶1. Und er the 90% e xceedance forecast, Reclamati on
28 allocate d CVP agricu ltural users 5% o f their contract

75
1 amounts, and CVP mun icipal and industrial (“M&I”)

2 contract ors 55% of t heir contract amounts. Fed. Salmon
3
Ex. 55 a t ¶12. Unde r the 50% exceedance forecast , north-
4
of-Delta agric ultural and M&I contractors would receive
5
100% of their contra ct amounts, while south-of-Delta
6
agricult ural contrac tors would receive 30% and M& I
7
8 contract ors 75%. Id.
9 176. CVP wate r users face d similar reductions to
10 their in dividual all ocations. Farmers on the west side of

11 the San Joaquin Vall ey have received reduced CVP water
12
supply a llocations i n the 2007-2008, 2008-2009, a nd 20 09-
13
2010 wat er years, an d face similar reductions in 2010-
14
2011. S LDMWA Ex. 15 3 at ¶3; SLDMWA Ex. 154 at ¶4 ; SLDMWA
15
Ex. 156 at ¶4. In 2 007-2008, Reclama tion allocated to
16
17 Westland s 40% of its contract supply. In 2008-2009, that
18 allocati on was 10%. SLDMWA Ex. 155 at ¶8. For t he 2009-

19 2010 wat er year, Wes tlands was advised the initia l
20
allocati on was zero percent. SLDMWA Ex. 155 at ¶ 9.
21
177. On March 16, 2010, R eclamation announced an
22
increase in allocati ons, raising the allocation f or
23
south-of -Delta agric ultural users to 25% under a 90%
24
25 forecast and 30% und er a 50% forecast. 4/1/10 Tr .
26 210:14-2 2; Gov’t Sal mon Exh. 13.
27 178. Judicial notice is t aken of the fact that as of

28
76
1 April 1, 2010, CVP w ater supply allocations to so uth-of

2 Delta ag ricultural c ontractors were increased fro m 25% to
3
30%. Se e Doc. 318-2 (U.S. Department of th e Interior
4
Press Re lease). On April 23, 2010, DWR increased its
5
allocati on of SWP de liveries to 30%. See Doc. 32 3-2 (DWR
6
Press Re lease). Thi s does not alter the fact tha t water
7
8 deliveri es will like ly increase if the two RPAs a re
9 enjoined . 4/1/10 Tr . 213:14-20 (acknowledg ing th at
10 deliveri es would inc rease by 5% - 10% if th e RPAs were

11 enjoined ).
12
179. The quan tity of wate r lost through pumping
13
reductio ns translate s directly into water losses for
14
urban an d agricultur al water users. In the SWP s ervice
15
area, on e acre-foot of water serves about five to seven
16
17 people f or one year. 4/6/10 Tr. 186:25 - 187:1-3. Th e
18 SWP loss of 433,000 AF, if available to urban use rs,

19 would ha ve supplied approximately 2.6 million peo ple for
20
one year . 4/6/10 Tr . 187:8-11. Seve nty-five to eight y-
21
five per cent of SWP supply is provided for urban uses,
22
with the remainder p rovided to agricultural users .
23
4/6/10 T r. 187:15-17. Th e Metropolitan Wat er Dis trict of
24
25 Southern California alone serves approximately 20 million
26 urban us ers.
27 180. Water lo ss for agric ultural users results in

28
77
1 reductio n in the num ber of acres that may be sust ained

2 with act ual water su pply. Water duty is the amou nt of
3
water th at a crop ne eds per acre for a growing se ason.
4
4/6/10 T r. 187:21-22. DW R information indi cates that for
5
the SWP service area, the water duty is approximately
6
three AF per acre. 4/6/10 Tr. 187:22-25. If the 433,000
7
8 AF were withheld fro m almond crops, for example, almond
9 producti on would be reduced by approximately 140, 000
10 acres. 4/6/10 Tr. 1 88:1- 4.

11 181. Reduced CVP and SWP water supply allocations
12
have inc reased the c ost of supplemental water. F armers
13
have bee n forced to purchase supplemental water a t
14
drastica lly increase d cost. SLDMWA Ex. 154 at ¶7 , SLDMWA
15
Ex. 155 at ¶17, SLDM WA Ex. 156 at ¶6. Since 2007 , the
16
17 cost of securing sup plemental water has more than
18 tripled. SLDMWA Ex. 156 at ¶6; SLDMWA Ex. 154 at ¶7. As

19 of Janua ry 2010, the cost for buying replacement water
20
for tran sfer in a dr y year is at least $300 per a cre
21
foot, pl us transport ation costs. SLDMWA Ex. 157 at ¶1 2.
22
182. Increase d water allo cations may lessen this
23
increase d cost, and will mitigate anticipated har ms from
24
25 reduced water alloca tions. Farmers anticipate th at
26 increase d water allo cations would mitigate antici pated
27 damage t o crop s in p ropor tion to the amount of water

28
78
1 received and prevent further layoffs of farm empl oyees.

2 SLDMWA E x. 156 at ¶1 0.
3
183. In 2009, the Departm ent of the Interior
4
accounte d for action s taken under the Delta smelt
5
biologic al opinion, including federal export redu ctions,
6
as (b)(2 ) actions, p ursuant to section 3406(b)(2) of the
7
8 CVPIA. 4/1/10 Tr. 2 13:24-214 :2. In 2010, the Departm ent
9 of the I nterior inte nds to follow the same accoun ting
10 allocati on for feder al export reductions related to both

11 biologic al opinions, to the extent that (b)(2) as sets are
12
availabl e at the tim e the action is taken. Id. at
13
214:3-7.
14
15 (2) Other Re source Impac ts Caused or Exacerbated by
the 2009 Salmonid Bi Op RPA Actions.
16
184. Plaintif fs attribute a number of other huma n
17
impacts to reduction s in the water supply. There is
18
19 consider able dispute among the parties regarding the
20 extent t o which the 2009 Salmonid BiOp RPA Action s are
21 responsi ble for a nu mber of other impacts. It is

22
undisput ed that thes e RPA Actions are, at the very lea st,
23
exacerba ting the fol lowing impacts.
24
25 (1) Permanen t Crops.
26 185. Reductio ns in the qu antity of water supply
27 deliveri es have resu lted in changes to farming pr actices,
28
79
1 includin g an increas ed reliance on permanent crop s.

2 SLDMWA E x. 154 at ¶6 ; SLD MWA Ex. 155 at ¶¶ 18, 22; SLD MWA
3
Ex. 157 at ¶11.
4
186. Permanen t crops plac e farmers at greater risk
5
than row crops, as f armers cannot cut back on the water
6
to perma nent crops w ithout destroying them. SLDM WA Ex.
7
8 154 at ¶ 6; SLDMWA Ex. 155 at ¶¶ 18, 22; SLD MWA Ex . 157 at
9 ¶11.
10
(2) Fallowed Lands.
11
187. Because of reduced w ater forecasts and
12
13 uncertai nty regardin g future water supply, farmer s have
14 fallowed hundreds an d thousands of acres of field s.
15 SLDMWA E x. 155 at ¶1 0; SLDMWA Ex. 153 at ¶3 ; SLDM WA Ex .

16 156 at ¶ 5.
17
188. Fallowed lands and r educed water supply has
18
caused t he loss of t housands of acres of crops. Todd
19
Allen, a third-generation farmer in Fresno County, was
20
able to salvage and harvest only 40 acres of a wh eat c rop
21
22 out of a total arabl e 616 acres on his farm in 20 09.
23 SLDMWA E x. 153 at ¶3 .
24 189. For ever y 1,000 AF o f water lost by the San Luis

25 Plaintif fs’ member a gencies, approximately 400 ac res of
26
land may remain out of production. SLDMWA Ex. 157 at
27
¶13.
28
80
1 190. Fallowin g fields als o negatively impacts the air

2 quality of the San J oaquin Valley by increasing d ust and
3
particul ate matter. SLDMWA Ex. 155 at ¶20. Redu ced air
4
quality in turn impa irs major transportation rout es
5
through the va lley. SLDMWA Ex. 155 at ¶20.
6
191. The co mmander of L emoore Naval Air Station
7
8 describe d increased bird- on-aircraft strikes attr ibutable
9 to land fallowing. 4/7/10 Tr. 213:20 - 214:6.
10 Reclamat ion responde d by allocating an emergency water

11 supply t o farm s adja cent to Lemoore. See i d. at 213.
12
13 (3) Lack of Access to Cr edit.
14 192. The more unreliable the water supply, the more
15 difficul t it is for farmers to secure necessary f inancing

16 for thei r farming op erations. SLDMWA Ex. 153 at ¶4;
17
SLDMWA Ex. 154 at ¶13, SLDMWA Ex. 155 at ¶26, SLDMWA Ex.
18
156 at ¶ 7, SLDMWA Ex . 157 at ¶15. In some cases, lenders
19
deny loa n applicatio ns because of a lack of relia ble
20
water su pply. SLDMW A Ex. 153 at ¶4; SLDMWA Ex. 1 54 at
21
22 ¶13, SLD MWA Ex. 155 at ¶26, SLDMWA Ex. 156 at ¶7, SLDM WA
23 Ex. 157 at ¶15. In others, lenders’ concerns abo ut
24 availabi lity to land s irrigated by federally-supplied

25 water ha s required f armers to make a 50 percent d own
26
payment to secure an y loans. SLDMWA Ex. 156 at ¶ 7.
27
28
81
1 (4) Social D isruption an d Dislocation.

2 193. It is un disputed tha t farm employees and their
3 families have faced devastating losses due to red uctions
4
in the a vailable wat er supply. The impact on the farm
5
economy from the com bination of a three-yea r drou ght a nd
6
diversio n limi tation s rel ating to the delta smelt has
7
already been severe. SLDMWA Ex. 157 at ¶14.
8
9 194. Lost wat er supply ha s decreased the number of
10 producti ve agricultu ral acres, which has resulted in
11 reductio ns in employ ee hours, salaries, and posit ions,

12 devastat ing farm emp loyees and their families. S LDMWA
13
Ex. 154 at ¶11, SLDM WA Ex. 156 at ¶8.
14
195. The remo val of 250,0 00 acres from production
15
translat es to a loss of approximately 4,200 perma nent
16
agricult ural worker positions. SLDMWA Ex. 155 at ¶19.
17
18 Water sh ortages also cause jobs to be lost in
19 agricult ure-re lated businesses, such as packing sheds,

20 processi ng plants, a nd other related services. Id. The
21
projecte d agricultur e-rel ated wage loss for the S an
22
Joaquin Valley stand s at $1.6 billion. Id.
23
196. Dr. Mich ael, Defenda nt Intervenors’ economist
24
with exp ertise in re gional and environmental econ omics,
25
26 counters that “[a]lt hough water impacts have affe cted
27 parts of the west si de, there is no evidence that reduced
28
82
1 water de liveries have had a severe effect on farm or non-

2 farm emp loyment in t he Central Valley as a whole. ” D- I
3
Exh. 100 6 (Michael D ecl.) ¶10. Instead, it is a
4
combinat ion of facto rs, including the three-year droug ht,
5
the glob al economic recession, the foreclosure cr isis,
6
and the collapse of the real estate market and
7
8 construc tion industr y, that are mainly driving cr op and
9 job loss es, food ban k needs, and credit problems in the
10 Central Valley—not RPA Action IV.2.1. Id. at ¶¶ 6-10.

11 Dr. Mich ael estimate s that ESA-related pumping
12
restrict ions have re sulted in the loss of less th an 2,000
13
jobs. S ee id. at ¶4.
14
197. Unemploy ment has led to hunger on the west side
15
of the S an Joaquin V alley. SLDMWA Ex. 158 at ¶8. The
16
17 Communit y Food Bank, serving Fresno, Madera and K ings
18 Counties , estimates 435,000 people in the area it serves

19 do not h ave a reliab le source of food. SLDMWA Ex . 158 at
20
¶4. The Chief Execu tive Officer of the Community Food
21
Bank, Da na Wilkie, b elieves that hunger in the
22
communit ies served b y the Food Bank in the wester n San
23
Joaquin Valley will continue to increase in 2010 because
24
25 of ongoi ng water sho rtages. SLDMWA Ex. 158 at ¶5 . Ms.
26 Wilkie u nderstands t hat at least 42,000 people se rved by
27 the Food Bank in Oct ober 2009 were employed by fa rm-

28
83
1 related busine sses b efore losing their jobs . SLDMWA E x.

2 158 at ¶ 8.
3
4 (5) Groundwa ter Consumpt ion and Overdraft.
5 198. Reductio ns in the av ailable water supply have
6 caused w ater users t o increase groundwater pumpin g in
7 attempts to make up the difference between irrigation

8 need and allocated w ater supplies. SLDMWA Ex. 15 5 at ¶¶
9
4, 7; SL DMWA Ex. 157 at ¶10; 4/6/10 Tr. 216:6-7.
10
199. However, groundwater pumping is not always
11
availabl e, and canno t be used in all areas or for all
12
crops. SLDMWA Ex. 1 55 at ¶11 . Increased groundwater
13
14 pumping reduces the quality of water applied to t he soil
15 by incre asing soil s alinity. SLDMWA Id. at ¶15. Not all
16 fields a nd crops can be irrigated with groundwate r. Id.

17 at ¶¶ 11 , 15.
18
200. Increase d reliance o n and ove ruse of ground water
19
has caus ed groundwat er overdraft, which occurs wh en
20
pumping exceeds the safe yield of an aquifer. Id. at
21
22 ¶12. Ov erdraft caus es increased land subsidence and
23 potentia l damage to CVP conveyance facilities, id. at ¶¶
24 12-13, althoug h it is not clear that any su bsidence of

25 CVP faci lities has o ccurred as a result of the
26
implemen tation of th e 2009 Salmonid BiOp RPA Acti ons, as
27
the only reported in cident of subsidence at a SWP
28
84
1 conveyan ce facility predates current implementati on,

2 4/7/10 T r. 16: 1-13.
3
201. Increase d groundwate r pumping also increases
4
demand f or energy. SLDMWA Ex. 155 at ¶16. Due t o the
5
falling water table, wells require increased amou nts of
6
energy. Id. Westlands estimates that pumping of
7
8 groundwa ter in 2009 required approximately 425,00 0,000
9 kWh. Id . Adverse e nvironmental impacts are asso ciated
10 with suc h increased demand for and use of energy. Id.

11 202. Increase d groundwate r pumping has depleted
12
groundwa ter reserves . Groundwater reserves that were at
13
2 millio n acre feet in the beginning of 2007 are now less
14
than 900 ,000 AF. 4/ 6/10 Tr. 216:21-24. Wi thin M WD’s
15
service area, storag e levels are at 1.3 million A F, about
16
17 half of normal stora ge levels. 4/6/10 Tr. 217:4- 8.
18
b. Impacts of Decreased Salmonid Populations.
19
203. It is un disputed tha t declines in salmon
20
populati ons have cau sed harm to other residents o f
21
22 Californ ia, predomin antly the salmon fishing indu stry,
23 although the extent to which the Projects should be
24 assigned the blame f or such harms and the e xtent to wh ich

25 the RPA Actions will alleviate these harms is a m atter of
26
consider able dispute .
27
28
85
1 (1) Impacts on the Comme rcial and

Recreati onal Salmon Fishing Industries
2
204. Mr. Zeke Grader, Exe cutive Director of
3
Defendan t-Inte rvenor Pacific Coast Fe derati on of
4
5 Fisherme n’s Associat ions (“PCFFA”), testified tha t the
6 commerci al fishing i ndustry has suffered tremendo us
7 losses a s a result o f the near total collapse of

8 Californ ia’s salmon fishery, which precipitated a
9
shutdown of the salm on fi shin g seasons in 2008 an d 2009
10
and thre atens anothe r shutdown in the future. D- I Ex.
11
1007 (Su pp. Declarat ion of William F. “Zeke” Grad er) ¶¶
12
5, 8. T he fall-run (a non-listed species) collap se is
13
14 believed to have bee n brought about by a combinat ion of
15 environm ental stress ors in the Delta, including r educed
16 flows, w ater tempera ture, predation, and non-native

17 species, as well as declining ocean conditions. Id. a t
18
¶5; see also 3 /31/10 Tr. 127:22-128:1 0.
19
205. The evi dence establ ishes that the co sts of these
20
closures are substan tial: the 2008-2009 cl osures cost
21
22 the stat es of Califo rnia, Oregon, and Washington
23 approxim ately 4,200 jobs and well over $500 milli on. See
24 id. ¶7, Att. 3 ; see also D-I Ex. 1006 at ¶1 4.

25 206. Accordi ng to Mr. Stuart, fall-run Chinook
26
emigrate through the Delta during the same time p eriod as
27
Central Valley steel head (April and May). 3/31/1 0 Tr.
28
86
1 128:17-1 8. The BiOp notes, “[m]any RPA actions i ntended

2 to avoid jeopardy to listed winter-run and spring -run, or
3
adverse modifi cation of their critical habi tat, a re al so
4
expected to reduce a dverse effects of the action on the
5
short- a nd long-term abundance and the long-term
6
viabilit y of non-listed f all-run and late-fall ru n.”
7
8 BiOp at 715. RPA Ac tions IV.2.1 and IV.2.3 are also
9 designed to “reduce exposure of fall-run and late fall -
10 run juve niles to exp ort facilities and increase s urvival

11 for fall -run leaving the San Joaquin River.” Id. at 716,
12
717.
13
207. Reduced fall-run populations could lead to
14
further closures in future seasons, which, according t o
15
Mr. Grad er, “would h ave devastating effects on th e
16
17 commerci al fishermen of PCFFA and likely would le ad to
18 addition al job and i ncome losses. Continued fishe ry

19 closures threaten th e long term viability of the salmon
20
fishery, as the infr astructure and expertise that
21
sustains the fishery is lost.” D-I E x. 1007 (Supp.
22
Grader D ecl.) ¶8.
23
208. Dr. Mich ael compared the economic impacts to the
24
25 agricult ural and sal mon fishing industries and co ncluded
26 that the “short-run economic impacts of the endan gered
27 species pumping rest rictions and salmon fishery c losure

28
87
1 are of a similar sca le.” D-I Exh. 1006 at ¶ 16.

2
c. Impacts On the Winne mem Wintu Tribe’s
3 Cultural Interests i n Salmon
4 209. The Winn emem Wintu, a Native American tribe,
5
also hav e significan t interests in Sacramento Riv er
6
Chinook salmon that could be affected by injuncti ve
7
relief a gainst Actio ns IV.2.1 or IV.2.3. See D-I Ex.
8
1008 (De claration of Gary Hayward Slaughter Mulca hy
9
10 (“Mulcah y Decl.”)) ¶ ¶ 2-3. T he decla ration of Gary
11 Mulcahy demonstrates that, for centuries, salmon have
12 sustaine d the Winnem em Wintu and have formed the

13 foundati on of the Tr ibe’s cultural and spiritual
14
ceremoni es and belie fs. Id. at ¶3. However, lik e the
15
salmon, the Tribe is “struggling to survive ,” in part due
16
to the d ecline of na tive wild salmon and the diet ary and
17
health e ffects this has had on Tribal members. Id. at
18
19 ¶5. In addition, th e loss of native salmon runs has
20 transfor med the Winn emem Wintu’s way of life, whi ch once
21 involved commu nity celebrations, salmon bakes, and

22
festival s, all cente red around the salmon. Id. at ¶¶ 3,
23
6. The Winnemem Tri be’s connection to salmon is so
24
strong t hat they bel ieve “that if the salmon go, the
25
Winnemem Wintu will also disappear.” Id. a t ¶3.
26
27 210. To the extent that an injunction of either
28 Action I V.2.1 or Act ion IV.2.3 would harm Sacrame nto

88
1 River Ch inook salmon , as discussed above, it will

2 threaten the signifi cant cultural and spiritual i nterests
3
of the W innemem Wint u.
4
5 (3) Harm to Species.
6 211. The pot ential harms to the species of enjoining
7 Action I V.2.1 and/or IV.2.3 are discussed above.

8 212. The NMF S’s and rela ted fish agencies continuing
9
failure, after more than ten (10) years of disput es, to
10
acquire credib le and reliable species population figures,
11
perform impact analy ses in light of population le vels,
12
and deve lop appropri ate population life-cyc le mod els,
13
14 with exp licit knowle dge that such data and modeli ng are
15 generall y accepted s cientific methods in the fiel d, is
16 still un explained, e xcept that it is difficult to

17 accompli sh.
18
19 VI. CONCLUSI ONS OF LAW 1.
20 A. Jurisdic tion.
21 1. Jurisdic tion over cl aims brought under NEPA
22 exists u nder 28 U.S. C. § 1331 (Federal Question) and the

23 Administ rative Proce dure Act (“APA”), 5 U.S.C. § 702 e t
24
seq. Ju risdiction o ver the ESA claims exists und er the
25
ESA citi zen-su it provision, 16 U.S.C. § 1540(g)(1)(A).
26
Personal jurisdictio n over all the parties exists by
27
virtue o f their part icipation in the lawsuit as
28
89
1 Plaintif fs, De fendants, and Intervenors.

2
B. Likeliho od of Succes s on the Merits: NEPA Claim.
3
2. Plaintif fs have alre ady succeeded on their NEPA
4
5 claim. S ee Mem orandu m Dec ision Re Cross-Mot ions f or
6 Summary Judgment on NEPA Issues. Doc. 266.
7 3. NEPA ins ures that fe deral agencies “m ake

8 informed decisions a nd ‘contemplate the environme ntal
9
impacts of [their] a ctions.’” Ocean Mammal Inst. v.
10
Gates, 5 46 F. Supp. 2d 960, 971 (D. Hi. 200 8) (qu oting
11
Idaho Sp orting Cong. v. Thomas, 137 F .3d 1146, 1149 (9 th
12
Cir. 199 8).
13
14 4. “NEPA em phasizes the importance of coherent and
15 comprehe nsive up-front en vironmental analys is to insure
16 informed decision-making to the end that th e agen cy wi ll

17 not act on incomplet e information, only to regret its
18
decision after it is too late to correct.” Ctr. for
19
Biologic al Diversity v. U.S. Forest Serv., 349 F.3d 11 57,
20
1166 (9t h Cir. 2003) .
21
22 5. The agen cies’ violat ions of NEPA prevented the
23 required reasonable evaluation, analysis, “hard l ook at,”
24 and disc losure of th e harms and damag e of i mplementing

25 the 2009 Salmonid Bi Op RPA Actions to human healt h and
26
safety, the human en vironment and other environme nts not
27
inhabite d by the Lis ted Species.
28
90
1 6. Harms th at have been caused by RPA water supply

2 reductio ns include but are not limited to: destruction of
3
permanen t crops; fal lowed lands; increased ground water
4
consumpt ion; land su bsidence; reduction of air qu ality;
5
destruct ion of famil y and entity farming business es; and
6
social d isruption an d dislocation, such as increa sed
7
8 property crimes and intra-fam ily crimes of violence,
9 adverse effects on s chools, and increased unemplo yment
10 leading to hunger an d homelessness.

11 7. Where a federal agen cy takes action in violation
12
of NEPA, “that actio n will be set aside.” High Sierra
13
Hikers A ss’n v. Blac kwell, 390 F.3d 630, 640 (9th Cir.
14
2004).
15
8. However, a court may not issue an injunction
16
17 under NE PA that woul d cause a violation of other
18 statutor y requiremen ts, such as those found in se ction 7

19 of the E SA. S ee United States v. Oakland C annabi s
20
Buyers’ Coop., 532 U.S. 483, 497 (2001) (“A district
21
court ca nnot, for ex ample, override Congress’ pol icy
22
choice, articulated in a statute, as to what beha vior
23
should b e prohibited ”). Nor should an injunction issue
24
25 under NE PA whe n enjoining government action would resu lt
26 in more harm to the environment than denying inju nctive
27 relief. Save Our Ec osystems v. Clarke, 747 F.2d 1240,

28
91
1 1250 (9t h Cir. 1984) ; Am. Motorcyclis t Ass’ n v. Watt, 714

2 F.2d 962 , 966 (9th C ir. 1983) (holding public interest
3
does not favor grant ing an injunction where “gove rnment
4
action a llegedly in violation of NEPA might actua lly
5
jeopardi ze natural r esources”); Alpin e Lake s Prot. Soc ’y
6
v. Schla pfer, 518 F.2d 1089, 1090 (9th Cir. 1975)
7
8 (denying injunctive relie f in NEPA case whe re mor e har m
9 could oc cur to fores t from disease if injunction was
10 granted) .
11
C. Likeliho od of Succes s on ESA Claims.
12
13 (1) Legal St andards.
14 9. The Admi nistrative P rocedure Act (“APA”)
15 requires Plaintiffs to show that NMFS’s action was
16 “arbitra ry, capricio us, an abuse of discretion, o r

17 otherwis e not in acc ordance with law.” 5 U.S.C. §
18
706(2)(A ).
19
20 a. Record R eview.
21 10. A court reviews a bi ological opinion “based upon

22 the evid ence contain ed in the administrative reco rd.”
23
Arizona Cattle Growe rs’ Ass’n v. FWS, 273 F .3d 12 29, 1 245
24
(9th Cir . 2001). Ju dicial review under the APA m ust
25
focus on the adminis trative record already in exi stence,
26
not some new record made initially in a reviewing court.
27
28 Parties may not use “post-dec ision informat ion as a ne w
92
1 rational ization eith er for sustaining or attackin g the

2 agency’s decision.” Ass’n of Pac. Fi sherie s v. EPA, 615
3
F.2d 794 , 811- 12 (9th Cir. 1980).
4
11. Exceptio ns to admini strative record review for
5
technica l informatio n or expert expla nation make such
6
evidence admissible only for limited purposes, an d those
7
8 exceptio ns are narro wly construed and applied. Lands
9 Council v. Powell, 395 F. 3d 1019, 1030 (9th Cir. 2005).
10 12. Here, th e Court has considered expert testimony

11 only for explanation of technical terms and compl ex
12
subject matter beyon d the Court’s knowledge; to
13
understa nd the agenc y’s explanations, or lack the reof,
14
underlyi ng the RPA A ctions; and to determine if a ny bad
15
faith ex isted.
16
17
b. Deferenc e to Agency Expertise.
18
13. The Cour t must defer to the agency on matters
19
within t he agency’s expertise, unless the agency
20
complete ly failed to address some factor, conside ration
21
22 of which was essenti al to making an informed deci sion.
23 Nat’l Wi ldlife Fed’n v. NMFS, 422 F.3d 782, 798 (9th C ir.
24 2005). The court “m ay not substitute its judgmen t for

25 that of the agency c oncerning the wisdom or prude nce of
26
the agen cy’s action. ” River Runners for Wilderne ss v.
27
Martin, 539 F.3d 106 4, 1070 (9th Cir. 2009).
28
93
1 In condu cting an APA review, the cour t must

determin e whether th e agency’s decision is
2 “founded on a ration al connection between the
facts fo und and the choices made ... and whether
3 [the age ncy] has com mitted a clear error of
judgment .” Ariz. Cattle Growers’ Ass’n v. U.S.
4 Fish & W ildlife, 273 F.3d 122 9, 1243 (9th C ir.
2001). “The [agency ’s] action ... need be only
5 a reason able, not th e best or most reasonable,
decision .” Nat’l Wildlife Fed. v. Burford, 871
6 F.2d 849 , 855 (9th C ir. 1989).
7 Id.
8 14. Although deferential , judicial review under the
9
APA “is designed to ensure that the agency consid ered all
10
of the r elevant fact ors and that its decision con tained
11
no clear error of ju dgment.” Arizona v. Thomas, 824 F.2d
12
745, 748 (9th Cir. 1 987) (internal citations omit ted).
13
14 “The def erence accor ded an agency’s scientific or
15 technica l expertise is not unlimited.” Bro wer v. Evan s,
16 257 F.3d 1058, 1067 (9th Cir. 2001) (internal cit ations
17 omitted) . Deference is not owed when “the agency has
18
complete ly failed to address some factor co nsider ation of
19
which wa s essential to making an informed decisio n.” Id.
20
(interna l citations and quotations omitted).
21
[An agen cy’s decisio n is] arbitrary and
22 capricio us if it has relied on factors which
Congress has not int ended it to consider,
23 entirely failed to consider an important aspect
of the p roblem, offe red an explanation for its
24 decision that runs c ounter to the evidence
before t he agency, o r is so implausible that it
25 could no t be ascribe d to a difference in view or
the prod uct of agenc y expertise.
26
Motor Ve hicle Mfrs. Ass’n of U.S. v. State Farm M ut.
27
28 Auto. In s. Co., 463 U.S. 29, 43 (1983); see also Citiz ens
94
1 to Prese rve Overton Park, Inc. v. Volpe, 401 U.S. 402,

2 416 (197 1) (“A revie wing court may overturn an ag ency’s
3
action a s arbitrary and capricious if the agency failed
4
to consi der relevant factors, failed to base its decision
5
on those factors, an d/or made a clear error of
6
judgment .”).
7
8
c. General Obligations Under the ESA.
9
15. ESA Sect ion 7(a)(2) prohibits agency action that
10
is “like ly to jeopar dize the continue d existence” of any
11
endanger ed or threat ened species or “result in th e
12
destruct ion or adver se modification” of its criti cal
13
14 habitat. 16 U.S.C. § 1536(a)(2).
15 16. To “jeop ardize the c ontinued existence of” means
16 “to enga ge in an act ion that reasonably would be

17 expected , directly o r indirectly, to reduce appre ciably
18
the like lihood of bo th the survival and recovery of a
19
listed s pecies in th e wild by reducing the reprod uction,
20
numbers, or distribu tion of that species.” 50 C.F.R. §
21
22 402.02; see also Nat ’l Wildlife Fed’n v. NMFS, 524 F.3d
23 917 (9th Cir. 2008) (“NWF v. NMFS II”) (rej ecting agen cy
24 interpre tation of 50 C.F.R. § 402.02 that in effe ct

25 limited jeopardy ana lysis to survival and did not
26
realisti cally evalua te recovery, ther eby av oiding an
27
interpre tation that reads the provision “and reco very”
28
95
1 entirely out of the text). An action is “jeopard izing”

2 if it ke eps recovery “far out of reach,” even if the
3
species is able to c ling to survival. Id. at 931 .
4
17. “[A]n a gency may no t take action that will tip
5
a specie s from a sta te of precarious survival int o a
6
state of likely exti nction. Likewise, even where
7
8 baseline conditions already jeopardize a species, an
9 agency m ay not take action that deepens the jeopa rdy by
10 causing additi onal harm.” Id. at 930 .

11 18. To satis fy this obli gation, the federal agency
12
undertak ing the acti on (the “action agency”) must prepare
13
a “biolo gical assess ment” that evaluates the acti on’s
14
potentia l impacts on species and species’ habitat. 16
15
U.S.C. § 1536(c); 50 C.F.R. § 402.12(a).
16
17 19. If the p roposed acti on “is likely to adversely
18 affect” a threatened or endangered species or adv ersely

19 modify i ts designate d critical habitat, the actio n agency
20
must eng age in “form al consultation” with NMFS to obtain
21
its biol ogical opini on as to the impacts of the p roposed
22
action o n the Listed Species. 16 U.S.C. § 1536(a)(2),
23
(b)(3); see also 50 C.F.R. § 402.14(a), (g) . Onc e the
24
25 consulta tion process has been completed, NMFS mus t give
26 the acti on age ncy a writt en biological opin ion “settin g
27 forth [N MFS’s] opini on, and a summary of the info rmation

28
96
1 on which the opinion is based, detailing how the agency

2 action a ffects the s pecies or its critical habita t.” 16
3
U.S.C. § 1536(b)(3)( A); see a lso 50 C .F.R. § 402. 14(h) .
4
20. If NMFS determines t hat jeopardy or destruction
5
or adver se modificat ion of critical habitat is li kely,
6
NMFS “sh all suggest those reasonable and prudent
7
8 alternat ives which [ it] believes would not violat e
9 subsecti on (a)(2) of this sec tion and can be take n by the
10 Federal agency or ap plicant in implementing the a gency

11 action.” 16 U.S.C. § 153 6(b) (3)(A). “Following the
12
issuance of a ‘jeopa rdy’ opinion, the agency must either
13
terminat e the action , implement the proposed alte rnative,
14
or seek an exemption from the Cabinet-level Endangered
15
Species Committee pu rsuant to 16 U.S.C. § 1536(e) .”
16
17 National Ass’n of Ho me Builders v. Defenders of W ildlife,
18 551 U.S. 644, 652 (2 008).

19
d. Best Ava ilable Scien ce.
20
21. Under th e ESA, an ag ency’s ac tions must be based
21
22 on “the best scienti fic and commercial data avail able.”
23 16 U.S.C . § 1536(a)( 2); 50 C.F.R. § 402.14(g)(8) (“In
24 formulat ing its Biol ogical Opinion, any reasonabl e and

25 prudent alternatives , and any reasonable and prud ent
26
measures , the Service will use the best scientific and
27
commerci al data avai lable.”). “The obvious purpo se of
28
97
1 the [bes t available science requirement] is to en sure

2 that the ESA not be implemented haphazardly, on t he basis
3
of specu lation or su rmise.” Bennett v. Spe ar, 52 0 U.S .
4
154, 176 (1997). A failure by the agency to util ize the
5
best ava ilable scien ce is arbitrary and capriciou s. See
6
Gutierre z II, 606 F. Supp. 2d at 1144.
7
8 22. A decisi on about jeo pardy must be made based on
9 the best science ava ilable at the time of t he dec ision ;
10 the agen cy cannot wa it for or promise future stud ies.

11 See Ctr. for B iological Diversity v. Rumsfeld, 198 F.
12
Supp. 2d 1139, 1156 (D. Ariz. 2002).
13
23. The “bes t available science” mandate of the ESA
14
sets a b asic standar d tha t “p rohibits the [agency ] from
15
disregar ding availab le scientific evidence that i s in
16
17 some way better than the evidence [it] relies on. ” Am.
18 Wildland s v. Kemptho rne, 530 F.3d 991, 998 (D.C. Cir.

19 2008) (c itation omit ted).
20
24. What con stitutes the “best” available scien ce
21
implicat es core agen cy judgment and expertise to which
22
Congress requires th e courts to defer; a court sh ould be
23
especial ly wary of o verturning such a determinati on on
24
25 review. Baltimore G as & Elec. Co. v. Natural Res .
26 Defense Counci l, 462 U.S. 87, 103 (1983) (a court must be
27 “at its most deferen tial” when an agency is “maki ng

28
98
1 predicti ons within i ts area of special expertise, at the

2 frontier s of science ”). As explained by the en b anc
3
panel of the Ninth C ircuit in Lands C ouncil, 537 F.3d at
4
993, cou rts may not “impose on the agency their o wn
5
notion o f which proc edures are best or most likel y to
6
further some vague, undefined public good.” Id. In
7
8 particul ar, an agenc y’s “scientific methodology i s owed
9 substant ial deferenc e.” Gifford Pinchot Ta sk For ce v.
10 U.S. Fis h & Wildlife Serv., 3 78 F.3d 1059, 1066 (9th C ir.
11 2004).
12
25. This def erence exten ds to the use and
13
interpre tation of st atistical methodologies. As
14
explaine d by the D.C . Circuit in Appalachia n Power Co. v.
15
EPA, 135 F.3d 791 (D .C. Cir. 1998), in reviewing a
16
17 challeng e to a decis ion of the Environmental Prot ection
18 Agency ( “EPA”) under the “arbitrary and capriciou s”

19 standard of review:
20
Statisti cal analysis is perhaps the prime
21 example of those are as of technical wilderness
into whi ch judicial expeditions are best limited
22 to ascer taining the lay of the land. Although
computer models are “a useful and often
23 essentia l tool for p erforming the Herculean
labors C ongress impo sed on EPA in the Clean Air
24 Act,” [c itation] the ir scient ific nature do es
not easi ly lend itse lf to judicial review. Our
25 consider ation of EPA ’s use of a regression
analysis in this cas e must therefore comport
26 with the deference t raditionally given to an
agency w hen reviewin g a scientific analysis
27 within i ts are a of e xpert ise without abdica ting
our duty to ensure t hat the application of this
28 model wa s not arbitr ary.
99
1
Id. at 802.
2
26. More gen erally, “[w] hen specialists express
3
4 conflict ing views, a n agency must have discretion to rely
5 on the r easonable opinions of its own quali fied e xpert s
6 even if, as an origi nal matter, a court might fin d

7 contrary views more persuasive.” Lan ds Council, 537 F.3d
8
at 1000 (quoting Marsh v. Oregon Natural Re s. Cou ncil,
9
490 U.S. 360, 378 (1 989)).
10
27. Mere unc ertainty, or the fact that evidence may
11
be “weak ,” is not fa tal to an agency decision.
12
13 Greenpea ce Action v. Franklin, 14 F.3d 1324 , 1337 (9th
14 Cir. 199 2) (upholdin g biological opinion, despite
15 uncertai nty about th e effectiveness of management

16 measures , because decision was based on a reasonable
17
evaluati on of all av ailable data); Nat’l Wildlife Fed' n
18
v. Babbi tt, 12 8 F. S upp. 2d 1274, 1300 (E.D . Cal. 2000 )
19
(holding that the "m ost reasonable" reading of th e best
20
21 scientif ic data avai lable standard is that it “pe rmits
22 the [FWS ] to take ac tion based on imperfect data, so long
23 as the d ata is the b est available”).

24 28. The defe rence afford ed under the best available
25
science standard is not unlimited. For example, Tucson
26
Herpetol ogical Socie ty v. Salazar, 566 F.3d 870, 879 ( 9th
27
Cir. 200 9), held tha t an agency may not rely on
28
100
1 “ambiguo us studies a s evidence” to support findin gs made

2 under th e ESA. Beca use the studies did not lead to the
3
conclusi on reached b y FWS, the Ninth Circuit held that
4
these st udies provid ed inadequate sup port i n the
5
administ rative recor d for the determination made by FWS.
6
Id.; see also Rock Creek Alliance v. U.S. F ish & Wildl ife
7
8 Service, 390 F. Supp . 2d 993 (D. Mont. 2005) (rej ecting
9 FWS’s re liance on a disputed scientific report, w hich
10 explicit ly sta ted it s analysis was not applicable to the

11 small po pulations ad dressed in the challenged opi nion);
12
Greenpea ce v. NMFS, 80 F. Supp. 2d 1137, 11 49-50 (W.D.
13
Wash. 20 00) (where a gency totally failed to devel op any
14
projecti ons regardin g population viab ility, it could not
15
use as a n excuse the fact that relevant data had not been
16
17 analyzed ).
18 29. The pres umption of a gency expertise may be

19 rebutted if the agen cy’s decisions, although base d on
20
scientif ic expertise , are not reasoned. Greenpeace, 80
21
F. Supp. 2d at 1147. Agencies cannot disregard a vailable
22
scientif ic evidence better than the evidence on w hich it
23
relies. Kern County Farm Bureau v. Allen, 450 F.3d 10 72,
24
25 1080 (9t h Cir. 2006) ; S.W . Ct r. for Biological Di versity
26 v. Babbi tt, 21 5 F.3d 58, 60 ( D.C. Cir. 2000).
27 30. Courts r outinely per form substantive reviews of

28
101
1 record e vidence to e valuate the agency's treatmen t of

2 best ava ilable scien ce. The judicial review proc ess is
3
not one of blind acc eptance. See, e. g., Kern County, 450
4
F.3d 1072 (tho roughly reviewing three post-comment
5
studies and FWS’s tr eatment of those studies to d etermine
6
whether they “provid e[d] the sole, essential supp ort for”
7
8 or “mer ely suppleme nted” the data used to suppor t a
9 listing decision); Home Builders Ass’ n of N. Cal. v. U .S.
10 Fish and Wildlife Se rv., 529 F. Supp. 2d 1110, 1120 (N .D.

11 Cal. 200 7) (examinin g substance of challenge to F WS’s
12
determin ation that c ertain data should be disrega rded);
13
Trout Un limited v. L ohn, 645 F. Supp. 2d 929 (D. Or.
14
2007) (f inding best available science stand ard ha d bee n
15
violated after thoro ugh examination of rationale for
16
17 NMFS’s d ecision to w ithdraw its proposal to list Oregon
18 Coast Co ho salmon); Ocean a, Inc. v. Evans, 384 F. Supp .

19 2d 203, 217-18 (D.D.C. 2005) (carefully considering
20
scientif ic underpinn ings of challenge to Service’ s use of
21
a partic ular model, including post decision evide nce
22
presente d by an expe rt, to help the court underst and a
23
complex model, apply ing one of several record rev iew
24
25 exceptio ns articulat ed in Esc h v. Yeutter, 876 F.2d 97 6,
26 991 (D.C . Cir. 1989) , which are similar to those
27 articula ted by the N inth Circuit).

28
102
1 31. Courts a re not requi red to defer to an agency

2 conclusi on that runs counter to that of other age ncies or
3
individu als with spe cialized expertis e in a particular
4
technica l area. See, e.g ., Am. Turnb oat Ass’n v.
5
Baldrige , 738 F.2d 1 013, 1016-17 (9th Cir. 1984) (NMFS ’s
6
decision under the M arine Mammal Protection Act w as not
7
8 supporte d by substan tial evidence because agency ignored
9 data tha t was product of “many years’ effort by trained
10 research personnel”) ; Sie rra Club v. U.S. Army Co rps of

11 Eng’rs, 701 F.2d 101 1, 1030 (2d Cir. 1983) (“court may
12
properly be skeptica l as to whether an EIS’s conc lusions
13
have a s ubstantial b asis in fact if the responsible
14
agency h as apparentl y ignored the conflicting vie ws of
15
other ag encies havin g pertinent experience[]”) (i nternal
16
17 citation s omitted). Here, DWR has a scientifical ly-
18 based, c ontrary view of the science, has consider ed the

19 economic consequence s of the RPA Actions, a nd has
20
interven ed to protec t humans and the human enviro nment.
21
A court should “reje ct conclusory assertions of a gency
22
‘experti se’ where th e agency spurns unrebutted ex pert
23
opinions without its elf offering a credible alter native
24
25 explanat ion.” N. Spotted Owl v. Hodel, 716 F. Supp. 479,
26 483 (W.D . Wash. 1988 ) (citing Am. Tur nboat Ass’n, 738
27 F.2d at 1016).
28
103
1 32. In Conner v. Burford, 848 F.2d 1441, 1453-54

2 (9th Cir . 1988), the agency attempted to defend i ts
3
biologic al opinions by arguing that there w as a lack o f
4
sufficie nt informati on. In rejecting this defens e, the
5
court he ld that “inc omplete information about pos t-
6
leasing activities d oes not excuse the failure to comply
7
8 with the statutory r equirement of a comprehensive
9 biologic al opinion using the best inf ormati on availabl e,”
10 and it n oted that FW S could have completed more a nalysis

11 with the information that was available. Id. at 1454
12
(emphasi s added). T he Ninth Circuit stated:
13
In light of the ESA requirement that the
14 agencies use the best scientific and commer cial
data ava ilable ... t he FWS cannot ignore
15 availabl e biological info or fail to develop
projecti ons of oil a nd gas activities which may
16 indicate potential c onflicts between development
and the preservation of protected species. We
17 hold tha t the FWS vi olated the ESA by failing to
use the best informa tion available to prepare
18 comprehe nsive biolog ical opinions.
19 848 F.2d at 1454 (em phasis added).
20
(2) Environm ental Baseli ne.
21
33. Plaintif fs arg ue tha t the BiO p is flawed be cause
22
NMFS imp roperly attr ibuted negative effects to th e
23
24 Project that should have been included in the
25 environm ental baseli ne. Doc. 164 at 10-16.

26 34. The rele vant regulat ory definition of the
27
“environ mental baseline” is provided within the
28
104
1 definiti on of the “e ffects of the action”:

2 the dire ct and indir ect effects of an action on
the spec ies or criti cal habitat, together with
3 the effe cts of other activities that are
interrel ated or inte rdependent with that action,
4 that wil l be added t o the environmental
baseline . The envir onmental baseline includes
5 the past and present impacts of all Federal,
State, o r private ac tions and other human
6 activiti es in the ac tion area, the anticipated
impacts of all propo sed Federal projects in the
7 action a rea that hav e already undergone formal
or early section 7 c onsultation, and the impact
8 of State or private actions which are
contempo raneous with the consultation in
9 process.
10 50 C.F.R . § 402.02.
11
35. When det ermining the “effects of the action ,”
12
the agen cy first mus t evaluate the status of the species
13
or criti cal habitat, which will involve “consider ation of
14
the pres ent environm ent” in which the species or habitat
15
16 exists a s well as “t he environment that will exis t when
17 the acti on is comple ted, in terms of the totality of
18 factors affecting th e species or critical habitat .” 51

19 Fed. Reg . 19,926, 19 ,932 (June 3, 1986). This ev aluation
20
is to se rve as the “ baseline” for determining the effects
21
of the a ction on the species or critical habitat. Id.
22
However, it is all e valuated together as the “eff ects of
23
the acti on.”
24
25 36. If addit ional data w ould provide a better
26 informat ion base fro m which to formulate a biolog ical
27 opinion, the Directo r may request an extension of formal

28
105
1 consulta tion and tha t the action agen cy obt ain additio nal
2 data to determine ho w or to what extent the actio n may
3
affect l isted specie s or critical habitat. 50 C. F.R. §
4
402.14(f ); U.S. Fish and Wildlife Service and Nat ional
5
Marine F isheries Ser vice, Endangered Specie s Consultat ion
6
Handbook (March 1998 ) at 4-6. 13
7
8 37. The Nint h Circuit di rected NMFS to consider the
9 effects of its actio ns “within the context of oth er
10 existing human activ ities that impact the listed

11 species. ” NWF v. NMFS II, 524 F.3d at 930. “[T]he
12
proper b aseline anal ysis is not the proportional share of
13
responsi bility the f ederal agency bears for the d ecline
14
in the s pecies, but what jeopardy might result fr om the
15
agency’s proposed ac tions in the present and futu re human
16
17 and natu ral contexts.” Id. The relevant jeopardy
18 analysis is whether this Project will tip a speci es into

19 a state of “likely e xtinction.” 524 F.3d at 930.
20
Even und er the so-called aggregation approa ch
21 NMFS cha llenges, the n, an agency only
“jeopard ize[s]” a sp ecies if it cause s some new
22 jeopardy . An agency may still take action that
removes a species fr om jeopardy entirely, or
23 that les sens the deg ree of jeopardy. However, an
agency m ay not take action that will tip a
24 species from a state of precarious survival into
a state of likely extinction. Likewise, even
25 where ba seline condi tions already jeopardize a
species, an agency m ay not take action that
26 deepens the jeopardy by causing additional harm.
27 13
Ju di ci al n ot ic e ma y be ta ke n of t h is Ha nd bo ok , wh ic h is
ava il ab le a t:
28 htt p: // ww w. fw s. go v/ en dan ge re d/ co ns u lta ti on s/ s7 hn db k/ s7 hn dbk .h tm .
106
1
Our appr oach does no t require NMFS to include
2 the enti re environme ntal baseline in the “agency
action” subject to r eview. It simply requires
3 that NMF S appropriat ely consider the effects of
its acti ons “within the context of other
4 existing human activ ities that impact the listed
species. ” [citation] . This approach is
5 consiste nt wit h our instr ucti on (which NMFS does
not chal lenge) that “[t]he proper baseline
6 analysis is not the proportional share of
responsi bility the f ederal agency bears for the
7 decline in the speci es, but what jeopardy might
result f rom the agen cy's proposed actions i n the
8 present and future h uman and natural contexts.”
[citatio n].
9
Id. (footnote omitted).
10
38. The agen cy is not re quired to quantify and/or
11
12 parcel o ut the “prop ortional share” of harms amon g the
13 baseline and the pro posed action. See Pacific Coast
14 Fed’n of Fishermen's Ass'ns v. U.S. Bureau of

15 Reclamat ion, 4 26 F.3 d 108 2, 1093 (9th Cir. 2005); see
16
also Pac ific Coast F ed’n of Fishermen's Ass'ns v. U.S.
17
Bureau o f Reclamation, 22 6 Fed. Appx. 715, 718 (9 th Ci r.
18
2007) (r ejecting wat er users’ argument that agenc y act ion
19
must be the “histori cal cause” of the jeopardy to
20
21 salmon). However, t he record must reasonably dem onstrate
22 that the agency’s pr oposed actions, when viewed i n the
23 present and future h uman and natural contexts, wi ll cause

24 jeopardy or ad verse modification. 14
25
26 14
Pl ai nt if fs ’ mo ti on f or pr el im in ar y in ju nc ti on s pe ci fi ca lly
add re ss es t he t re at me nt of h at ch er i es an d gr av el l os s be low
27 Whi sk ey to wn D am . D oc . 1 64 a t 11 -1 2 . Ho we ve r, t hi s is su e w as n ot
pre se nt ed o r di sc us se d a t th e ev id e nti ar y he ar in g or i n Pla in ti ff s’
28 pro po se d fi nd in gs . I t a pp ea rs t ha t th es e sp ec if ic a rg um ent s ha ve
107
1 39. Here, Pl aintiffs ide ntify only two potential

2 flaws in the environ mental baseline in their Prop osed
3
Findings of Fact and Conclusions of Law, namely N MFS’s
4
general failure to s egregate discretionary from n on-
5
discreti onary actions, Doc. 316, Pltf’s Proposed Findi ngs
6
of Fact ## 65- 66, 80, and, more specifically, NMFS’s
7
8 failure to treat cer tain obligations arising unde r the
9 Coordina ted Operatio ns Agreement (“COA”) as “mand atory,”
10 id. at P ropose d Find ings of Fact ## 67-80. 15

11
a. Treatment of Discretionary v. Non-
12 Discreti onary Operat ions.
13 40. Plaintif fs complain that the BiOp does not
14
distingu ish between discretionary and non-discret ionar y
15
actions. Home Build ers, 551 U.S. 644, held that ESA §
16
7’s cons ultation req uirements do not apply to non -
17
discreti onary action s. Where an agency is required by
18
19 law to p erform an ac tion, it lacks the power to i nsure
20 that the action will not jeopardize the species. Id. at
21 667.
22
41. However, Home Builde rs sa ys n othing about
23
whether, once sectio n 7 c onsultation is tri ggered, the
24
jeopardy analysis sh ould segregate discretionary and non-
25
26 bee n ab an do ne d.
15
It i s un cl ea r wh et he r P la in ti ff s c ont en d th at a ll o th er
27 str es so rs n ow j eo pa rd izi ng t he S an Joa qu in a nd S ac ra me nt o R iv er s an d
the D el ta a re p ar t of th e Ba se li ne and m us t no t be c on si der ed
28 cum ul at iv el y wi th t he ef fe ct s of c o ord in at ed P ro je ct o pe rat io ns .
108
1 discreti onary action s, relegating the non-discret ionar y

2 actions to the envir onmental baseline. Hom e Buil ders
3
fundamen tally concer ns whether the section 7 cons ultat ion
4
obligati on attaches to a particular agency action at all.
5
See Home Builders, 551 U. S. at 679-80 (“duty does not
6
attach t o actions... that an agency i s required by
7
8 statute to undertake ....”) (emphasis added).
9
b. Reclamat ion’s Treatm ent of the Coordinated
10 Operatio ns Agreement .
11 The same reasoning a pplies to Plaintiffs’ related
12 argument that Federa l Defendants acted unlawfully by

13 attribut ing to the p roject the effects of “mandat ory”
14
complian ce with the Coordinated Operations Agreem ent
15
(“COA”). Even assuming, arguendo, th at any mandatory
16
obligati on exists un der the COA, a proposition th at is
17
question able given t he open-ended wor ding of the COA and
18
19 language in the CVPI A subjecting project operatio ns to
20 the ESA, Home Builde rs do es n ot require the agency to
21 segregat e discretion ary from non-discretion ary activit ies

22
during a n ESA § 7 co nsultation. 16 Moreover, this argument
23
was not presented in Plaintiffs’ opening brief. See
24
Alaska C tr. for Envt . v. U.S. Forest Serv., 189 F .3d 8 51,
25
858 n. 4 (9th Cir. 1 999) (“Ar guments not raised in
26
27 16
T o th e ex te nt t ha t Pla in ti ff s su g ges t th at s ec ti on 7 d oes
not a pp ly t o th e pr oj ect s at a ll u n der H om e Bu il de rs , th is pa ra di gm -
28 shi ft in g ar gu me nt h as no t pr op er ly bee n ra is ed o r br ie fe d.
109
1 opening brief are wa ived”).

2
(3) Southern Resident In direct Effects Analysis.
3
42. Plaintif fs raise ano ther argument based on an
4
5 alleged error in the effects analysis pertaining to the
6 impacts of the proje cts on Southern Resident Killer
7 whales. Doc. 164 at 16-19. While the parties briefed

8 the issu e, engaging in considerable debate over b oth the
9
appropri ate standard to be applied to indirect ef fects
10
analyses and the suf ficiency of the evidence cite d in the
11
record t o support NMFS’s conclusions, this issue was n ot
12
a focus of the evide ntiary hearing.
13
14 43. It is un necessary to reach this issue because,
15 even if, arguendo, Plaint iffs demonstrated a likelihood
16 of succe ss on this c laim, the alleged deficiencie s in the

17 BiOp’s a nalysis of i mpacts to orcas d o not justify
18
enjoinin g either RPA Action IV.2.1 or IV.2.3. An
19
injuncti on must be “ narrowly tailored” to give on ly the
20
relief t o which plai ntiffs are entitled. See Ora ntes-
21
22 Hernande z v. Thornbu rgh, 919 F.2d 549, 558 (9th C ir.
23 1990). Here, NMFS a dopted Actions IV.2.1 and IV. 2.3
24 primaril y for the be nefit of salmon, steelhead, a nd green

25 sturgeon that migrat e through the Delta and are h armed by
26
export p umping that interferes with their migrati ons, not
27
orcas wh ich reside i n the ocean. See 4/1/10 Tr. 184:4 -17
28
110
1 (Action IV.2.3 was n ot designed with the objectiv e to

2 protect orcas or fal l-run Chi nook salmon). The i ndirect
3
effect o f alleged re ductions of orca prey is not
4
mentione d as a direc t justification for either ch allenged
5
RPA.
6
7 (4) Challenges to Action IV.2.1.
8 a. Viable S almonid Popu lation Methodology/
Populati on Modeling/ Life Cycle Analysis.
9
44. Plaintif fs’ argument that NMFS failed to apply
10
11 the VSP methodology in a sufficiently rigorous ma nner is
12 unpersua sive. The B iOp did not ignor e the VSP
13 methodol ogy. Rather , it chose to use VSP in a

14 qualitat ive manner a s a conceptual framework, as
15
recommen ded by Lindl ey (2006). Although the anal ysis in
16
the BiOp may have be nefited from the application of
17
quantita tive VSP met hodologies, it is disputed whether
18
the fail ure to do so represents a breach of accep ted
19
20 scientif ic practice. A court must defer to the a gency in
21 such sci entific disp utes.

22 45. The agen cy is not re quired to generate new
23
studies. For exampl e, in Sou thwest Center for Bi ologi cal
24
Diversit y v. Babbitt, 215 F.3d 58, 60-61 (D .C. Cir.
25
2000), t he district court found the available evi dence
26
regardin g FWS’s deci sion not to list the Queen Ch arlotte
27
28 goshawks “inconclusi ve” and held that the agency was
111
1 obligate d to find be tter data on the specie s’ abundanc e.

2 The D.C. Circuit rev ersed, emphasizing that, alth ough
3
“the dis trict court’ s view has a superficial appe al ...
4
this sup erficial app eal cannot circumvent the sta tute’s
5
clear wo rding: The secretary must make his decis ion as
6
to wheth er to list a species as threatened or end angered
7
8 ‘solely on the basis of the best scientific and
9 commerci al data avai lable to him....’ 16 U.S.C. §
10 1533(b)( 1)(A).” Id. at 61. Requirin g NMFS to adapt t he

11 VSP meth odology to o perate as a quantitative model wou ld
12
be the e quivalent of requiring NMFS to generate d ata.
13
The cour t has no aut hority to do so.
14
46. The same conclusion is required for Plaintiffs’
15
contenti on that NMFS should have engaged in popul ation
16
17 modeling and/or life cycle analysis. Altho ugh su ch
18 modeling is scientif ically preferred, Plaintiffs

19 presente d no evidenc e that they, or anyone else,
20
presente d NMFS with then- existing best avai lable scien ce
21
represen ting appropr iate population or life cycle models
22
for the species of c oncern prior to the issuance of the
23
BiOp. M oreover, the primary purpose of Action IV .2.1 is
24
25 to prote ct outmigrat ing juvenile members of the S SNDG of
26 CV steel head, for wh om no population indices (whe ther
27 absolute or relative ) are available.

28
112
1 b. Correlat ion Between Exports and Effects on

Salmonid Survival.
2
47. NFMS rel ied on a num ber of circumstances to
3
support its general conclusion that salmonid surv ival in
4
the inte rior Delta w as adversely affected by expo rt
5
pumping.
6
7 a. The VAMP data demons trated some obser vable
8 negative impacts, bu t no statistically significan t
9 connecti on, albeit t he lack of statistical signif icance

10 was like ly due to li mitations in the data.
11
b. Figure 1 0 of Appendi x 5 supports the
12
conclusi on that, at least when HORB is in place, there is
13
an obser vable (but n ot statistically significant)
14
negative relationshi p between survival and export s.
15
16 Question s exist whet her it is appropriate to rely on data
17 collecte d when HORB was in place, given that HORB cannot
18 be used under the Sm elt BiOp. However, NMF S pres ented

19
evidence that a work able substitute (the bubble b arrier)
20
for HORB will be uti lized. Plaintiffs have not s uggested
21
the barr ier would be inadequate.
22
c. Highly q uestionable support for the BiOp’s
23
24 conclusi on that expo rts negatively influenc e surv ival
25 derives from a compa rison of exports and adult es capement
26 two and a half years later, from 1951 through 200 3. See
27 BiOp App . 5 at Figur e 11. All parties agreed tha t adult
28
113
1 escapeme nt can be si gnificantly influenced by fac tors

2 such as ocean conditions and harvest. It is undispute d
3
that Fig ure 11 did n ot adjust for these factors.
4
However, NMFS relied on a conceptual model that s uggests
5
because ocean condit ions and harvest were likely to
6
fluctuat e over time, long-ter m downward trends in
7
8 population cou ld be caused by declining freshwater
9 conditio ns.
10 d. NMFS als o relied ext ensively on Newman’s

11 2008 ana lysis of the Delta Action 8 studies, whic h
12
released coded-wire tagged salmon into Georgiana Slough
13
and comp ared their s urvival to coded-wire tagged salmo n
14
released into the ma instem Sacramento River. New man
15
found a statisticall y significant, although weak,
16
17 negative relationshi p between exports and salmoni d
18 survival .
19 e. There is no question that the remaining data
20
connecti ng exports t o reduced salmonid surv ival i s not
21
what NMF S represents it to be. Recognizing that “[w]hen
22
speciali sts express conflicting views, an agency must
23
have dis cretion to r ely on the reasonable o pinion s of its
24
25 own qual ified expert s even if, as an original mat ter, a
26 court might fi nd contrary views more persuasive,” Lands
27 Council, 537 F.3d at 1000 (quoting Ma rsh, 4 90 U.S . 360 )

28
114
1 (emphasi s added), de ference is not required “wher e the

2 agency o ffers an exp lanation for an action that r uns
3
counter to the evide nce before the agency, Tuscon
4
Herpetol ogical Socie ty, 5 66 F .3d at 878. NMFS di d not
5
just rel y on “ambigu ous studies.” Rather, it
6
uncritic ally examine d the body of evidence, somet imes
7
8 disregar ding the exp ress qualifications and reser vations
9 of indep endent studi es, to reach the conclu sion that t he
10 exports negatively i mpact salmonid survival. Thi s

11 conclusi on, although not scientifically unassaila ble, has
12
marginal support in the record.
13
48. NMFS’s o pinion that low Vernalis flow to export
14
ratios t hreaten to a ppreciably increa se the likel ihood
15
that the SSNDG of CV steelhead will become extinc t is
16
17 also bas ed on incomp lete and conflicting evidence .
18 Although no absolute or relative population numbe rs are

19 availabl e for either the SSNDG or the entire ESU, it is
20
undisput ed tha t both are small and imperiled. It is also
21
undisput ed that, pur suant to the VSP approach, ev ery
22
extant p opulation of the CV steelhead must be pro tected.
23
All memb ers of the S SNDG must pass through the in terior
24
25 Delta on their way t o the ocean. As exports incr ease,
26 their ch ances of sur vival decrease. On the whole , the
27 record c orroborates NMFS’s conclusion that planne d

28
115
1 project operations w ill jeopardize the CV steelhe ad. 17

2 49. Other ad verse impact s from toxics, invasive
3
species, predators, in-Delta pumping, and other non-
4
operatio nal hazards were not compared with Projec t
5
operatio ns to determ ine the extent these other st ressors
6
contribu te to the je opardy to the species and the ir
7
8 habitat.
9
c. Did NMFS Adequately Justify the Ratios
10 Imposed?
11 50. The fund amental flaw in NMFS’s justification of
12 Action I V.2.1 is its selection of the specific ra tios

13 imposed under the Ac tion. As discussed in the Fi ndings
14
of Fact, the record reveals no biological explanation why
15
NMFS cho se to impose a 1,500 cfs limit on e xports when
16
flows at Vernalis ar e lower than 6,000 cfs, 18 and a ratio
17
of 4:1, as opposed t o any other ratio, when Verna lis
18
19 flows ar e between 6, 000 cfs and 21,750 cfs. Id. at 71 -
20 72.
21 51. This is a quintessen tial example of arbitrary

22
action. There is no way to know whether these levels are
23
sufficie ntly protect ive, not protective enough, o r far
24
17
It i s no t ne ce ss ar y to no w ex am in e wh et he r NM FS w as j us tif ie d
25 in co nc lu di ng t ha t pl ann ed p ro je ct ope ra ti on s du ri ng t hi s t im e
per io d wi ll j eo pa rd iz e a ny o f th e o the r Li st ed S pe ci es . Ac ti on
26 IV. 2. 1 is d es ig ne d pr ima ri ly t o ai d CV s te el he ad .
18
Th is 1 ,5 00 c fs l im it is t he m in im u m e xp or t le ve l NM FS f oun d
27 nec es sa ry t o ma in ta in he al th a nd s a fet y cr it er ia . B iO p App . 5 at
22. At f lo ws o f 5, 00 0 c fs , fo r ex a mpl e, t he r at io w ou ld be
28 5,0 00 /1 ,5 00 o r ap pr ox ima te ly 3 .3 3: 1 .
116
1 more pro tective than necessary. 19 Particularly in light

2 of the e normous huma n impacts caused by even smal l
3
changes in the flow regime reducing exports, the agency
4
must pro vide a reaso ned and scientifically justif ied
5
basis fo r selecting the specific remedial measure s
6
chosen. They have f ailed to do so.
7
8 52. This con clusion is p articularly justified in
9 light of the concurr ent NEPA violation. Had eith er NMFS
10 or Recla mation perfo rmed a proper NEPA evaluation of the

11 human an d environmen tal impacts of the RPA Action s before
12
implemen ting them, o r if both NMFS and Reclamatio n had
13
worked t ogether to d o so, this would have at leas t forced
14
the agen cies to fully con sider and rational ly bal ance the
15
biologic al need for certain flow levels against t he
16
17 adverse water supply and resulting human impacts those
18 restrict ions effectu ate.

19 53. There is insufficien t record evidence to
20
conclude what altern ative flow/export ratio would be
21
sufficie ntly protect ive of the SSNDG of CV steelh ead, the
22
populati on Action IV .2.1 was designed to protect. NMFS’s
23
scientif ically justi fied conclusion that a low Ve rnalis
24
25 flow to export ratio during the spring threatens to
26 jeopardi ze CV steelh ead m akes it inappropri ate to
27 19
It m ay b e sc ie nt if ic all y ju st if ia b le to b ui ld a m ar gi n of
err or ( i. e. t o ta ke a pr ec au ti on ar y ap pr oa ch ) wh en d es ig nin g an R PA ,
28 but t hi s mu st b e pr op erl y ju st if ie d an d di sc lo se d by t he re co rd .
117
1 complete ly remove an y Vernalis flow to export rat io

2 restrict ion. Plaint iffs offered no scientificall y
3
justifia ble alternat ive except the unjustified ar gument
4
there is no jeopardy caused by project operations and no
5
evidence of peril to the species.
6
7 (5) Challeng es to Action IV.2.3.
8 54. Action I V.2.3 operat es from January 1 through
9
June 15 or until the average daily water temperat ure at
10
Mossdale is greater than 72º F, whichever is earl ier. It
11
limits O MR flows to no more negative than -2,500 to -
12
5,000 cf s, depending on juvenile entrainment leve ls.
13
14 BiOp at 648-52 .
15 55. Plaintif fs and DWR o nly seek an injunction
16 against the -5 ,000 cfs “calendar-based” cei ling.

17
a. Use of P TM for salmo nids.
18
56. Although the PTM mod el, a hydrodynami c
19
20 simulati on used to a ssess the fait of particles a s a
21 function of flow, ti des, project operations, and other

22 factors, has shortco mings, it is an indicator of
23
directio ns of river flows that salmonids follow,
24
recogniz ing their strong swimming ability. NMFS relie d
25
on the P TM studies t o support its conclusions tha t: (a)
26
as expor ts increase, negative OMR flows also incr ease;
27
28 and (b) that at Stat ion 815 (the confluence of th e
118
1 Mokelumn e River and the San Joaquin River), parti cle

2 entrainm ent in creases as negative OMR flows incre ase.
3
Above -5 ,000 cfs, 40 % of particles injected at th at
4
station are entraine d, while 90% are entrained at -7,000
5
cfs.
6
57. Although particles d ecidedly do not mirror the
7
8 behavior of salmonid smolts, which move approximately 3.5
9 times fa ster, they p rovide a very rough approxima tion of
10 salmonid behavior, o ne ground supporting NMFS’s

11 utilizat ion of the P TM as part of its overall rat ionale
12
for Acti on IV.2.1.
13
14 b. Salvage Data.
15 58. NMFS als o relied on salvage data, which
16 demonstr ated that, a s negative OMR flows increase s,

17 salvage increases, a nd that at some point more ne gative
18
than -5, 000 cfs, sal vage increases mu ch more rapidly than
19
at lower levels.
20
59. The data utilized do es not scale salv age to
21
22 populati on size, an undisputed failure to use the best
23 availabl e scientific methods, at least with respe ct to
24 the wint er-run and spring-run , for which population da ta

25 is avail able. Dr. D eriso opined that scaling sal vage to
26
populati on size is s tanda rd a ccepted practice in the
27
field of fisheries s cience. Even from a lay pers pective,
28
119
1 it is ob vious that a bsolute salvage numbers vary

2 dependin g on the siz e of the extant population. NMFS’s
3
reliance on comparis ons of raw salvage numbers to
4
negative OMR f low was clear scientific error and not t he
5
best ava ilable scien ce.
6
60. Action I V.2.3 is als o designed to protect CV
7
8 steelhea d, for which no population data is availa ble. It
9 is less certain whet her NMFS could legitimately a pply
10 comparis ons of raw salvage data to OMR flows to assess

11 the impa ct of negati ve OMR flows on CV steelhead.
12
13 c. Delta Ac tion 8 Studi es.
14 61. As with Action IV.2. 1, NMFS also relied
15 extensiv ely on Newma n’s 2008 analysis of the Delt a Action
16 8 studie s, which rel eased coded-wire tagged salmon int o

17 Georgian a Slough. N ewman found a statistically
18
signific ant, althoug h “weak,” negative relationsh ip
19
between exports and salmonid survival.
20
62. There ar e additional concerns that, as to upper
21
22 Sacramen to River pop ulations, NMFS failed t o cons ider the
23 relative number of f ish that are exposed to condi tions in
24 the inte rior Delta, compared to those that remain in the

25 mainstem of the Sacr amento River. This critique is not
26
relevant to NMFS’s a pplication of the Delta Actio n 8
27
Studies to tho se populations of CV steelhead and spring-
28
120
1 run that originate i n the San Joaquin basin. For those

2 populati ons, the Del ta Action 8 studies support t he
3
conclusi on that the higher the export levels, the lower
4
the chan ce a salmoni d smolt may survive to reach the
5
ocean.
6
7 d. Perry & Skalski and Vogel.
8 63. Perry an d Skalski (2 008) concluded that survival
9
of fish moving into Georgiana Slough and nearby c hannels
10
was redu ced compared to those in the mainstem of the
11
Sacramen to River. 4 /1/10 Tr. 161:20-162:1. However,
12
Perry an d Skalski ob served that “there is limited
13
14 understa nding of how water management actions in the
15 Delta af fect populat ion distribution and route-specific
16 survival of juvenile salmon.” SDLMWA Ex. 227 at 3. Mr.

17 Stuart a dmitted that Perr y an d Skalski 2008 did n ot
18
address water projec t impacts on Delta hydrology, fish
19
behavior , or the ind irect mortality of fish in th e
20
central and southern channels of the Delta. Mr. Stuart
21
22 further admitted tha t he reached his conclusions
23 regardin g water project impacts on Delta hydrology, fish
24 behavior , and indire ct salmonid mortality based u pon his

25 personal extrapolati on from the data contained in Perry
26
and Skal ski 2008, an d not from any conclusions re ached by
27
the stud y. 4/2/10 T r. 19:2 – 21:24. The BiOp and Stu art
28
121
1 used Per ry and Skals ki (2008) to support a propos ition

2 that Per ry and Skals ki themselves disclaimed. Th e BiOp
3
provides no explanat ion to justify this use of Pe rry and
4
Skalski for this pur pose, which is arbitrary and
5
capricio us.
6
64. A simila r problem ex ists with the BiOp’s
7
8 reliance on the Voge l (2004) review of telemetry- tagging
9 data to investigate fish route selection in the c hannels
10 leading to the south Delta. See BiOp at 38 0-81. The

11 BiOp use d Vogel’s wo rk to find that when export l evels
12
were red uced and San Joaquin River flows were inc reased,
13
more fis h stayed in the main channel of the San J oaquin
14
River, h eading downs tream toward the San Francisc o Bay.
15
Id. However, the Vogel study concluded its experiments
16
17 “could n ot explain why so me f ish move off the mai nstem of
18 the San Joaquin Rive r into the south Delta channe ls,”
19 noting t hat “[d]ue t o the wide variation in hydro logic
20
conditio ns” during t he course of the experiments, “it was
21
difficul t to determi ne the principal factors affe cting
22
fish mi gration. Ba sed on the limited data from these
23
studies, it may be t hat a combination of a neap t ide,
24
25 reduced exports, and increased San Joaquin River flows is
26 benefici al for outmi grating smolts, but more rese arch is
27 necessar y.” DWR Ex. 505 at 37 (emphasis ad ded).

28
122
1 65. The BiOp ’s reliance on the Perry and Skalski and

2 Vogel st udies presen ts the same infirmities as in Tucson
3
Herpetol ogical Socie ty, 5 66 F .3d at 879, where th e FWS
4
wrongful ly “affirmat ively relie[d] on ambiguous s tudies.”
5
6 e. Does the Record Supp ort NMFS’s General
Conclusi on that Nega tive OMR Flows
7 Apprecia bly Reduce S almonid Smolts’ Chances
of Survi val?
8
66. There ar e undeniable problems with NMFS’s basis
9
for Acti on IV.2.3. However, the Delta Action 8 s tudies
10
11 support the proposit ion that, for tho se populations
12 spawning entirely wi thin the San Joaquin basin,
13 increasi ng exports n egatively impact salmonid smo lt

14 survival . The highl y disputed PTM studies consti tute the
15
other co lorable supp ort for Action IV.2.3. In su ch a
16
scientif ic dispute, deference is owed unless the Agency
17
is unrea sonably wron g.
18
19 f. Did NMFS Adequately Justify the Calendar-
based -5,000 c fs Cei ling of Action IV.2.3?
20
67. The -5,0 00 cfs OMR ceiling is based, in lar ge
21
22 measure, on speculat ion. It is also based upon BiOp
23 Figures that do not scale salvage to population s ize.

24 This is not the best available science and is arb itrary
25
and capr icious.
26
27 (6) Reclamat ion’s ESA Re sponsibility.
28 68. The ESA regulations require the action agency to

123
1 “determi ne whe ther a nd in what manner to pr oceed with the

2 action i n light of i ts section 7 obligations and the
3
Service’ s biological opinion.” 50 C.F.R. § 402.1 5(a).
4
Prior to accepting a nd implementing the 2009 Salm onid
5
BiOp RPA , Reclamatio n had an independent obligation under
6
ESA sect ion 7(a)(2) to ensure that it “use[d] the best
7
8 scientif ic and comme rcial data available.”
9 69. Reclamat ion, as the federal action agency, “may
10 not rely solely on a FWS biological opinion to es tablish

11 conclusi vely its com pliance w ith its substantive
12
obligati ons under se ction 7(a)(2).” Pyramid Lake Paiu te
13
Tribe of Indians v. U.S. Dept. of the Navy, 898 F .2d
14
1410, 14 15 (9th Cir. 1990). “[T]he action agency must
15
not blin dly adopt th e conclusions of the consulta nt
16
17 agency.” City of Tacoma v. Fed. Energy Reg ulator y
18 Comm’n, 460 F.3d 53, 76 (D.C. Cir. 20 06).

19 70. Reclam ation did no t ens ure that the RPA u tilize d
20
the best available s cience, nor did it independen tly
21
identify and analyze alternative RPA Actions that
22
minimize d jeop ardy to humans and the human environment
23
while pr otecting thr eatened species.
24
25
D. Balancin g of the Har ms.
26
(1) Balancin g of the Har ms in ESA Cases.
27
71. The Supr eme Court he ld in TVA v. Hill, 437 U.S.
28
124
1 153, 194 (1978), tha t Congress struck the balance in

2 favor of affording e ndangered species the highest of
3
prioriti es. In adop ting the ESA, Congress intend ed to
4
“halt an d reverse th e trend toward species’ extin ction,
5
whatever the cost.” Id. at 1 84 (emphasis added). TVA v.
6
Hill con tinues to be viable. See Hom e Builders, 551 U.S.
7
8 at 669-7 1; see also Oakland Cannabis Buyers’ Co-op., 532
9 U.S. 496 -97; A moco Prod. Co. v. Village of Gambell, 480
10 U.S. 531 , 543 n.9 (1 987).

11 72. Winter d oes not modi fy or discuss the TVA v.
12
Hill sta ndard. 20 Alth ough Winter alter ed the Ninth
13
Circuit’ s general pr eliminary injunctive relief s tandard
14
by makin g that stand ard more rigorous, Winter did not
15
address, nor change, the approach to the balancin g of
16
17 economic hardships w here endangered species and t heir
18 critical habitat are jeopardized. See Biodiversity Le gal

19 Found. v . Badgley, 309 F. 3d 1166, 1169 (9th Cir. 2002)
20
(Congres s removed th e courts’ traditional equitab le
21
discreti on to balanc e parties’ competing interest s in ESA
22
injuncti on proceedin gs); Nat’l Wildlife Fed’n v.
23
Burlington N. R.R., Inc., 23 F.3d 1508, 1510-11 (9th Cir.
24
25 1994)(sa me).
26 73. Prior de cisions invo lving the coordinated
27
20
Al th ou gh W in te r in vo lve d ES A- li st e d s pe ci es , th e Wi nt er
28 dec is io n di d no t ad dr ess a ny E SA c l aim s.
125
1 projects ’ operations found that TVA v . Hill and relate d

2 Ninth Ci rcuit author ities foreclose the district court’s
3
traditio nal discretion to balance equities under the E SA.
4
There is no such bar in NEPA injunction proceedin gs.
5
74. Plaintif fs have adva nced a human health and
6
safety e xception and contend that unlike any of t he prior
7
8 cases, t his case jux taposes species’ survival aga inst
9 human we lfare, requi ring a balancing of the BiOp’ s
10 threats of harm to h umans, health, safety and pro tection

11 of affec ted communit ies. No case, including TVA v. Hill,
12
which co ncerned the competing economic interest i n the
13
operatio n of a hydro-electric project, expr essly
14
addresse s whether th e ESA precludes balancing of harms to
15
humans a nd the human environment under the circum stances
16
17 presente d here.
18 75. Even if it is permis sible to balance harm to

19 humans a nd the human environment against Congress’ stated
20
desire t o protect th e Listed Species, doing so in
21
practice is complica ted by the harm caused to oth er human
22
communit ies by the r educed abundance of salmonids , such
23
as to th e salmon fis hing industry and the Winneme m Wintu
24
25 Tribe.
26 76. This c ase is at th e int ersection of harm to
27 threaten ed species a nd humans and their environme nt.

28
126
1 Congress has not nor does TVA v. Hill elevate species

2 protecti on over the health and safety of humans.
3
4 (2) Balancin g the Harms under NEPA.
5 77. Although it is undis puted that all ha rms may be
6 consider ed in evalua ting a claim for injunctive r elief
7 under NE PA, an injun ction should not issue if enj oining

8 such gov ernment acti on would result in more harm to the
9
environm ent than den ying injunctive relief. Save Our
10
Ecosyste ms, 74 7 F.2d at 1 250.
11
78. Here, it app ears t hat i nterim relief is
12
justifie d, if deepen ing of the species’ jeopardy can be
13
14 avoided.
15
E. The Publ ic Interest.
16
79. In adopt ing the ESA, Congress explicitly found
17
that all threatened and e ndangered species “are of
18
esthetic , ecological , educational, historical,
19
20 recreati onal, and sc ientific value to the Nation and its
21 people.” 16 U.S.C. § 1531(a)(3). The ESA advanc es a

22 Congress ional policy to “halt and reverse the tre nd
23
toward s pecies extinction, whatever the cost.” TVA v.
24
Hill, 43 7 U.S. at 184 (emphasis added).
25
80. The publ ic policy un derlying NEPA favors
26
protecti ng the balan ce between humans and the
27
28 environm ent. See 42 U.S. C. § 4321 (declari ng a national
127
1 policy t o “enc ourage productive and enjoyable harmony

2 between man and his environment; to promote effor ts which
3
will pre vent or elim inate damage to the environme nt and
4
biospher e and stimul ate the health and welfare of man;
5
[and] to enrich the understanding of the ecological
6
systems and natural resources important to the
7
8 Nation.. ..”).
9 81. If both these object ives can all be realized by
10 astute m anagement, i t is the government’s obligat ion to

11 do so.
12
82. It is in the publi c int erest that relief be
13
granted to Plaintiff s, who represent a substantia l
14
populati on of water users in California, to enhan ce the
15
water su pply to redu ce the adverse harms of destr uction
16
17 of perma nent crops; fallowed lands; increased gro undwater
18 consumpt ion; land su bsidence; reduction of air qu ality ;

19 destruct ion of famil y and entity farming business es; and
20
social d isruption an d dislocation, such as increa sed
21
property crimes and intra-fam ily crimes of violence,
22
adverse effects on s chools, and increased unemplo yment
23
leading to hunger an d homeles sness. This must be done
24
25 without jeopardizing the species and their critic al
26 habitat.
27
28
128
1 VII. CONCLUSION2.
2 1. Plaintif fs have succ eeded on the merits of their
3 NEPA cla im.
4
a. NEPA req uires that t he responsible agency
5
take a h ard look at the environmental conse quence s of its
6
actions, Robertson v . Methow Valley Citizen’s Cou nsel,
7
490 U.S. 332, 350 (1 989), obligating federal agen cies to
8
9 prepare an environme ntal impact statement (“EIS”) for all
10 “major f ederal actio ns significantly affecting th e
11 quality of the human environment.” 42 U.S.C. §

12 4332(2)( C).
13
b. Federal Defendants a re required to evaluate
14
the impa ct of the co ordinated operations of the C VP and
15
SWP, whi ch constitut es major federal action. The
16
evidence overwhelmin gly establishes s ignifi cant
17
18 detrimen tal effects visited on the quality of the human
19 environm ent by imple mentation of the BiOp’s RPA A ctions,

20 which im pose virtual ly year-round sub stantial
21
restrict ions on the water supply to California to protect
22
the List ed Species.
23
c. Where re quired, an E IS discloses
24
environm ental affect s of a proposed action and co nsiders
25
26 alternat ive courses of action. Id. Here, Federal
27 Defendan ts completel y abdicated their responsibil ity to
28
129
1 consider alternative remedies in formulating RPA Actions

2 that wou ld not only protect the species, but woul d also
3
minimize the adverse impact on humans and the hum an
4
environm ent.
5
d. In consi dering RPA a lternatives, the record
6
shows th e burden of other causes is allocated to the
7
8 water su pply, withou t the required analysis wheth er
9 alternat ives, less h armful to humans and the huma n
10 environm ent, exist.

11 2. Plaintif fs have also shown a likelihood of
12
success on the merit s of their ESA claim. Althou gh the
13
premise underlying t he RPA Actions -- that the species
14
may be j eopardized b y increased negative flows oc casioned
15
by expor t pumping -- has some record suppor t, NMF S has
16
17 failed t o adequately justify by generally recogni zed
18 scientif ic principle s the precise flow prescripti ons

19 imposed by RPA Actio ns IV.2.1 and IV. 2.3. The exact
20
restrict ions imposed , which are inflicting materi al harm
21
to human s and the hu man environment, are not supp orted by
22
the reco rd. Rather, they are product of guesstim ations
23
and atte mpts to try to achieve “equity,” renderin g it
24
25 impossib le to determ ine whether the R PA Actions are
26 adequate ly protectiv e, too protective, or not pro tective
27 enough. Judicial de ference is not owed to such

28
130
1 arbitrar y, capriciou s, and scientifically unreaso nable

2 agency a ction.
3
3. It is hi ghly signifi cant that the co-operat or of
4
the Proj ects, DWR, w ith access to scientific comp etence
5
in the f ields of fis h biology and ecology, and pr oject
6
operatio ns, strongly criticizes some of the scien ce NMFS
7
8 used to justify RPA Action IV.2.3, seeks to enjoi n Action
9 IV.2.3, and do es not oppose enjoining Actio n IV.2 .1
10 4. Under th e balance of hardships analysis,

11 Defendan ts’ contenti on that the ESA, under TVA v. Hill ,
12
preclude s equitable weighing of Plaintiffs’ inter ests is
13
not supp orted by tha t case, as evidence of harm to the
14
human en vironment in the form of social dislocati on,
15
unemploy ment, and ot her threats to human welfare were not
16
17 present in Hil l. They are in this case.
18 5. Defendan ts argue tha t jeopardy to the species

19 cannot b e avoided wi thout continuing substa ntial
20
reductio n of pumping , with resultant reduction of water
21
supply t o Plaintiffs , representing over 20,000,00 0
22
persons, affected co mmunities, and the agricultur al
23
industry in Northern , Central, and Southern Calif ornia.
24
25 Harm to the species has h ad equally detrime ntal e ffect s
26 on the P acific Coast salmon fishing industry and impairs
27 the inte rests of Nat ive Americans. These additio nal

28
131
1 harms ar e deserving of equal protection.

2 6. Congre ss created public exp ectations in the
3
Amended Reclamation Act by instructing Reclamatio n to
4
contract for water s ervice to hundreds of public- entity
5
water se rvice provid ers that supply water to mill ions of
6
people a nd thousands of acres of productive agric ultural
7
8 land. T he agencies have not fully discharged their
9 responsi bility to ef fectively allocate Project wa ter
10 resource s. Federal Defendants have acted arbitra rily and

11 capricio usly in form ulating RPA Actions to protec t
12
threaten ed species u nder the ESA that lack factua l and
13
scientif ic justifica tion, while effec tively ignor ing the
14
irrepara ble harm tho se RPA Actions have inflicted on
15
humans a nd the human environment.
16
17 7. The sp ecies and their c riti cal habitats are
18 entitled to protecti on under the ESA. The specie s have

19 been and will be pro tected. That is the law.
20
Nonethel ess, NMFS an d Reclamation, as the consult ing and
21
action a gencies, mus t take the hard look under NE PA at
22
the drac onian conseq uences visited upon Plaintiff s, the
23
water su pply of Cali fornia, the agricultural indu stry,
24
25 and the reside nts an d com munities devastate d by the wa ter
26 supply l imitations i mposed by the RPA Actions. F ederal
27 Defendan ts have fail ed to comprehensively and com petently

28
132
1 evaluate whether RPA alternatives can be prescrib ed that

2 will be mutually pro tective of all the statutory purpo ses
3
of the P rojects.
4
8. This i s a case of first imp ression. The stakes
5
are high , the harms to the affected human communi ties
6
great, a nd the injur ies unacceptable if they can be
7
8 mitigate d. NMFS and Reclamation have not complie d with
9 NEPA. T his prevente d in- depth analysis of the po tential
10 RPA Acti ons through a properly focused study to i dentify

11 and sele ct alternati ve remedial measures that min imize
12
jeopardy to affected humans and their communities , as
13
well as protecting t he threatened species. No pa rty h as
14
suggeste d that human s and their environment are l ess
15
deservin g of protect ion than the species. Until
16
17 Defendan t Agencies h ave complied with the law, so me
18 injuncti ve relief pe nding NEPA compliance is appr opriate,

19 so long as it will not further jeopardize the species or
20
their ha bitat.
21
9. Injunc tive relief is al so w arranted under the
22
ESA, bec ause, althou gh the general premises under lying
23
Actions IV.2.1 and I V.2.3 find marginal support i n the
24
25 record, the precise flow prescription s imposed on
26 coordina ted project operations as part of Action IV.2.1’s
27 Vernalis flow/export ratio and Action IV.2.3’s -5,000 cfs

28
133
1 “calenda r based” cei ling are not supported by the best

2 availabl e science an d are not explained as the la w
3
requires .
4
10. Injunc tive r elief canno t be imposed witho ut up-
5
to-date eviden ce of the status of the species to assure
6
that alt ered operati ons will not deepen jeopardy to the
7
8 affected species or otherwise violate other laws. The
9 evidence has not suf ficiently focused on remedies to
10 provide a confidence level that completely removi ng the

11 Vernalis flow to exp ort ratio prescriptions of Ac tion
12
IV.2.1 o r permitting negative flows in excess of the
13
-5,000 c fs OMR flow ceiling imposed b y Acti on IV.2.3 t o
14
increase water supply will not jeopardize the con tinued
15
existenc e of the spe cies and/or adversely modify their
16
17 critical habitats.
18 11. Legal and eq uitabl e gro unds for injunctiv e

19 relief h ave otherwis e been established by a prepo nderance
20
of the e vidence.
21
12. A hear ing to addre ss th e proposed injunct ion an d
22
any immi nence of har m to species shall be held Ma y 19,
23
2010 in Courtroom 3 at 10:00 a.m.
24
25 SO ORDER ED
Dated: M ay 18, 2010
26 /s/ O liver W. Wanger
Oliver W. Wang er
27 United States Distri ct Ju dge
28
134

Salmon Findings of Fact Judge Oliver Wanger

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Salmon Findings of Fact Judge Oliver Wanger

Transféré par

Droits d'auteur :

Formats disponibles

Case 1:09-cv-01053-OWW-DLB Document 347 Filed 05/18/2010 Page 1 of 134

1 UNITED STATES DISTRI CT CO URT

16 Injuncti on (“PI”) ag ainst the implementation of

17 Reasonab le and Prude nt Alternative (“RPA”) Action IV.2.1

1 steelhea d, Southern Distinct Population Segment o f Green

9 Californ ia Departmen t of Water Resources (“ DWR”) filed

10 statemen ts of non-opposit ion. Docs. 247, 2 48 & 2 51.

18 joined. Doc. 181. DWR filed a partial joinder i n and

26 represen ted by couns el, as noted on the record in open

1 After co nsideration of the testimony of the

9 injuncti on are enter ed.

10 To the e xtent any fi nding of fact may be interpre ted

15 The 2009 Salmonid Bi Op found that planned coordin ated

16 Project operations w ould jeopardize the continued

23 included numerous el ements for §each of the vario us

24 project divisions an d associated stressors, which NMFS

1 jeopardy and adverse modification.” Id. at 578 ( empha sis

9 • Action I V.2.1, which will limit pumping based on San

10 Joaquin River inflow , measured at Vernalis, from

18 72°F, wh icheve r is earlier.

24 1) the d istrict cour t already found that the Unit ed

1 2) the 2 009 Salmonid BiOp is arbitrary, capriciou s,

9 apply a relevant env ironmental baseline;

10 c) NMFS failed to di stinguish between

18 by the b est availabl e science.

1 365, 376 (2008); Weinberg er v. Romero-Barce lo, 456 U.S.

9 3. The bala nce of equit ies t ips in the moving

10 parties’ favor; and

24 control; second, for irrigation and domestic uses ; and,

25 third, f or power.” 50 St at. 844, 850.

1 the 1937 Act and add ed mitigation, protection, an d

9 4. This cas e presents a critical conflict between

10 these du al legislati ve purposes, providing water service

18 interest that DWR, a co-opera tor and the State

1 7. It is un disputed tha t a March 17, 2010 Order

9 Reclamat ion and DWR, and “carefully avoided presc ribing

10 measures that are no t necessary to meet section 7

18 the exte nt and gravity of the harms to huma ns and the

26 initiate RPA actions when species are present and

27 protecti ons are most needed, and includes specifi c

1 scientif ic studies a nd engineering programs to re fine RPA

9 identify and analyze alternatives that minimize h arm to

10 humans a nd the human environment. Federal Defend ants do

18 co-equal Recla mation Law objective of providing water

21 C. Facts Re levant to ES A Challenges.

22 (1) Current Status of th e Species.

23 a. Sacramen to River Win ter-Run C hinook Salmon.

24 12. Sacramen to River win ter-run C hinook salmon

25 (Oncorhy nchus tshawytscha) (“winter-run”) a re lis ted a s

26 “endange red” under t he ESA. 70 Fed. Reg. 37,160 (June

1 of the S acramento Ri ver and other waters. Histor ical

9 fish) th at persisted into 2008 (2,830 fish). Id. In

10 2009, th ere was a mo dest increase in adult escape ment

18 May or J une. BiOp a t 81, 94; Pac. Co ast Fed’n of

26 14. During t he current e migration year (2009-20 10),

27 juvenile winter-run began entering the Delt a in O ctobe r

1 2009. I d. at ¶46. On Ap ril 1, 2010, Mr. S tuart, an N MFS

9 sets an incidental t ake limit of two percent of t he JPE

10 of winte r-run salmon , or 22,897. BiOp at 7 75; 3/31/10

18 because it “represen ts fish that would probably s how a

26 18. Designat ed critical habitat for winter-run

1 bays to the Golden G ate Bridge. Gutierrez II, 606 F.

6 19. Central Valley sprin g-run Chi nook salmon (O.

7 tshawyts cha) ( “spring-run ”) a re listed as “threat ened”