Mixing Smelt, Salmon, and People

• Author: Craig M. Pease
• Position: Ph.D., a research scientist, teaches at the Vermont Law School Environmental Law Center
• Pages: 18-18

Page 18 ❧ THE ENVIRONMENTAL FORUM Copyright © 2010, Environmental Law Institute®, Washington, D.C. www.eli.org.

Reprinted by permission from The Environmental Forum®, May/June 2010

By Craig M. Pease

Mixing Smelt,

Salmon, and People

In California’s Sacramento–San

Joaquin Delta, natural ecosystems

and human institutions have become

hopelessly intertwined. e delta is

habitat for the endangered winter-run

chinook salmon and threatened delta

smelt, and it also provides drinking

water for two-thirds of California’s 38

million people. Natural processes (e.g.

snowpack in the Sierras) and the man-

agement actions of government agen-

cies and courts (administering a com-

plex system of dams, canals, and levees)

simultaneously control the water f‌lows

critical to both f‌ish and people.

e entanglement of the natural

and human greatly increases the com-

plexity of managing the delta. is was

brought home forcefully in two recent

decisions from the Eastern District of

California, one earlier this spring in

the Consolidated Salmonid Cases, and

one last fall in the Delta Smelt Consoli-

dated Cases. Both disputes arose when

the Bureau of Reclamation wanted to

make management changes in the Cen-

tral Valley Project, its sprawling hydro-

logical system. But f‌irst, as required by

the Endangered Species Act, it had to

formally consult with the U.S. Fish and

Wildlife Service (for the smelt) and Na-

tional Marine Fisheries Service (for the

salmon). e services’ reviews found

that the bureau’s planned changes

would jeopardize these f‌ish and, as re-

quired by the ESA, they recommended

reasonable and prudent alternatives to

mitigate the harm. Not surprisingly,

these alternatives would restrict water

withdrawals for human use.

Several California districts that pur-

chase water from the bureau and supply

it to municipalities and agricultural us-

ers sued, on the grounds that no envi-

ronmental impact statement had been

completed. ey complained particu-

larly of the lack of any analysis of the

impact of the recommended reasonable

and prudent alternatives on the hu-

man environment. In both cases, Judge

Wanger ruled in favor of the water dis-

tricts. e agencies started out focused

on f‌ish, but found that the law required

them to look simultaneously at f‌ish and

human needs.

is web of natural and human pro-

cesses has also ensnarled science. In the

delta, it is no longer sensible or even

possible to study entirely natural phe-

nomena (e.g. delta smelt distribution

under natural water f‌low and salinity

regimes). Rather, most relevant science

is done against a back-

drop of water f‌low

regimes controlled si-

multaneously by natu-

ral processes and hu-

man institutions (e.g.

ef‌fect on delta smelt of

withdrawing a certain

amount of water, from a certain loca-

tion, at a particular time of the year).

Science has merged with adaptive man-

agement, characterized succinctly by

Kai Lee: “Policies are experiments; learn

from them.”

Strictly from the standpoint of sci-

ence, adaptive management is emi-

nently sensible. Both natural and hu-

man systems are hugely complex. It is

impossible to know a priori what man-

agement options are best. But just how

much management f‌lexibility does the

law allow, and in exactly what circum-

stances?

In the delta, the ability of the agen-

cies to pursue adaptive management is

limited both by standards on agency ac-

tion imposed by environmental statutes

and by binding enforceable provisions

of water delivery contracts between the

bureau and water districts. e case law

on both continues to evolve.

Especially instructive are two earlier

cases, also in front of Judge Wanger, and

also involving ESA consultations be-

tween the bureau and services over the

impact on smelt and salmon of planned

changes in the Central Valley Project.

In 2007, in NRDC v. Kempthorne, the

court ruled illegal the opened-ended rea-

sonable and prudent alternative recom-

mended by the Fish and Wildlife Ser-

vice, whereby a team of scientists would

periodically review delta smelt data and

recommend management changes, but

which provided no guarantees of any

real and enforceable agency action if

smelt numbers decreased. is was too

much management f‌lexibility. en just

a year later, in Pacif‌ic Coast Federation of

Fishermen’s Associations v. Gutierrez, a

parallel case involving chinook salmon,

the same court found that a reasonable

and prudent alternative establishing

fairly broad upper and lower bounds on

permissible agency actions (e.g., on wa-

ter f‌low rates) passed

muster, even though

the bureau reserved

some management

f‌lexibility within these

bounds.

Alas, there is one

critical, if somewhat

ancillary point. In theory, increased

management f‌lexibility should lead to

improved science, more knowledge,

and better results. In practice, it often

also opens the door to political inf‌lu-

ence on natural resource management.

is all too frequently runs counter to

good science.

As the natural and human worlds

have become increasingly entangled, so

to have science and law. Now scientists

work, not surrounded by nature, but

deep within human institutions. e

evolving case law that sets limits on the

decisionmaking f‌lexibility of human

institutions is having a wider impact on

science than is commonly understood.

Craig M. Pease, Ph.D., a research scien-

tist, teaches at the Verm ont Law School En-

vironmental L aw Center. He can be reached

at cpease@ve rmontlaw.edu.

S   L

In California, a web

of natural and human

processes has also

ensnarled scienc e