A few lessons learned from a pilot project in sustainability science

1
A few lessons learned from a pilot project in
sustainability science

• Nathan Mantua
• Climate Impacts Group
• Center for Science in the Earth System
• JISAO, University of Washington
• Seattle, WA 98195
• May 8, 2006Fisheries 497A

2
El Niño has big Impacts
Accurate El Niño forecasts should be of great
value to people in sensitive regions like the
west coast of the Americas shouldnt they?
3
Origins of the CIG project

• There were high hopes for translating advances in
  climate science into real benefits for society in
  the early 1990s
• The El Niño forecasting problem appeared to be
  solved, but there wasnt a national
  infrastructure for translating climate forecasts
  into useful and useable resource forecast
  information
• Global warming impacts studies were being
  reported by the IPCC at continental scales, but
  what did these mean to real people?

4
The Climate Impacts Group
First of 9 U.S. regional integrated assessment
teams (RISAs).

• Established in 1995
• Based at the University of Washington
  (Seattle) with collaborations in Oregon and Idaho
• Funded largely by the National Oceanic and
  Atmospheric Administrations Climate Program
  Office (NOAA/CPO)

5
NOAAs Climate Program Regional Integrated
Science and Assessments
6
The Climate Impacts Group

• Increase regional resilience to the impacts of
  climate variability and change
• Produce science useful to, and used by, the
  decision-making community

OBJECTIVES
Water Resources Fisheries Forests Coasts Human Health Agriculture
SECTORS

• Climate Variability
• past variations and their impacts
• ability of institutions to respond to extremes
• Climate Change
• regional consequences of global warming
• adaptation/vulnerability to climate change

7
Our early findings

• Virtually no one was using NOAAs climate
  forecasts in the mid-1990s
• They were not accurate enough
• They were not specific enough to particular
  resource issues
• People didnt understand what was meant by the
  probabilistic forecasts

May-June-July 2006 Temperature Forecast
8
LESSON 1

• Resource agencies make forecasts all the time,
  and the research community focuses on improving
  forecasts, but there arent always (often?)
  strong links between these communities

9
How Does CIG Support Adaptation to Climate
Variability and Change?
Decision-support tools Designed to facilitate
use of climate information in operations and
planning
Research Investigating sensitivity and
vulnerability to climate variability and
change Provides the foundation for decision
support and outreach activities
Outreach Designed to develop (and maintain)
ongoing relationships with the stakeholder
community
10
Case study evolution of climate information for
salmon management

• A fishery oceanography study identifies a climate
  impact
• Climate variability explains a large fraction of
  the space-time variations in 20th Century Pacific
  salmon catches (and presumably abundance)
• We (Hare, Mantua, Francis) promote the use of
  climate information for salmon management by
  describing the research results at meetings and
  workshops yet no managers want to use our
  results!
• The response from fishery management staff Your
  work is interesting, but it doesnt suit our
  needs
• We partner with a NOAA fisheries scientist
  involved in salmon management to develop a
  forecast tool they can use
• In the process, we learn how to match the
  space-time scales of climate information with
  those of salmon management, and we learn about
  limits to predicting coho returns

11
A North-South see-saw in salmon production
spring chinook returns to the Columbia River
mouth (1000s)
Alaska pink and sockeye catch (millions)
Warm PDO
Cool PDO
Warm PDO
???
Cool PDO
12
Commercial Sockeye Salmon Catches Since
1883 Bristol Bay, Alaska
Commercial catch (millions)
Composition
Hilborn et al. 2003, PNAS
13
Recruits-per-spawner for Bristol Bay sockeye (by
major river system)
Year
Hilborn et al. 2003, PNAS
14
Lesson 2

• The scales considered in our research were no
  match for the scales most important for salmon
  managers
• Our work was interesting, but unusable

15
OPI (hatchery) coho marine survival
Why? Leading hypothesis changes in ocean
conditions impact the entire marine food-web
16
coastal ocean impacts on coho marine survival
(Logerwell et al. 2003, Fish. Oceanogr.)

• key factors?
• Stratification (SST)
• spring transition date
• alongshore transport (Sea Level)

• key factors?
• Stratification (SST)
• winter winds, downwelling and transport

?
?
1st winter at sea
1st spring at sea
A few to 100 adults in 2nd summer
10s to 100s post-smolts in 1st summer
1000 smolts
17
4 index Ocean Conditions Model hindcasts for
OPI coho marine survival, 1969-1998
Logerwell et al. 2003, Fish. Oc.
R2 .75
18
Correlations and Predictability
SST0 SprTr
Upwelling Winds Spr Tr 0.22
Upw. Winds -0.17 -0.46 SST1
0.15 0.27
-0.16 (1970-1998)

• Implications?
• ocean conditions are the net result of
  essentially random combinations of sometimes
  independent processes

19
LESSON 3

• Environmental predictability for coho is VERY
  LIMITED -- this situation may be more the rule
  than the exception for climate sensitive
  resources

20
Life in uncertain environments

• Bet hedging behaviors one evolutionary response
• diversity of time-space habitat use
• a variety of sensitivities for different streams
  (e.g. Hymer WDFW)
• different ocean sensitivities (e.g. Bottsford et
  al.) for different stocks, incl. Hatchery vs.
  wild fish

21
Coho salmon, at the metapopulation level, hedge
their bets by migrating at different times of the
year
22
fishery management
23
Hatcheries a fish is a fish
Ex smolt migration timing in wild and hatchery
coho
Spring transition date
Wild coho smolt migration
Hatchery coho releases
Mar Apr May June July
24
So what?(what Ive learned)

• Sustaining fish and sustaining a fishery are
  not the same things
• expectations and actions for these two goals are
  often at odds with each other
• right now, fishery managers generally failing to
  deal with climate
• true for year-to-year and decade-to-decade
  variations

25
What are we managing, and why? (McEvoy 1996 The
Fishermans Problem)

• What is a fishery?
• (1) an ecosystem (2) a group of people working,
  and (3) a system of social control

26
Sustainability?

• Saving the fish
• eliminate harvests
• restore diversity
• major hatchery reform, even closures if needed
• restore and protect habitat
• remove barriers to fish passage (remove some
  dams)
• accept variability
• acknowledge a lack of predictability

• Saving the fishery
• keep seasons open as long as possible
• focus on biomass/numbers
• tweak the status quo
• fish passage, hatcheries
• eliminate variability
• use hatcheries, divorce fish production from
  habitat
• emphasize prediction

ECOLOGY
POLITICS-ECONOMICS-ECOLOGY
27
Where predictability matters(Holling 1993
Ecological Applications)

• 1st stream science
• system is predictable, science of parts
• ex the population
• Experimental, seeks explanation and prediction
• implies we need certainty before taking action
• Command and Control Management
• Problem is perceived, a solution for its control
  is developed (e.g. low salmon production, build a
  hatchery)
• Reduce variability to make the system more
  predictable

28
Where Predictability doesnt matter

• 2nd stream science
• Unpredictable, science of integration
• ex the ecosystem, the fishery
• Comparative, seeks understanding, accepts
  inherent unknowability and unpredictability
• The Golden Rule
• Resource management should strive to retain
  critical types and ranges of variations in
  ecosystems (Holling and Meffe 1996)

29
The problem?

• We cant solve 2nd stream problems with 1st
  stream approaches

30
Summary and Conclusions

• climate information has the potential to improve
  resource management
• short term help for salmon fisheries through
  monitoringbiophys models
• Longer range guidance for the trajectory of
  regional climate changes in response to global
  warming
• environmental prediction issues now a source of
  conflict between managing fish and fisheries for
  sustainability
• scientists must own up to the fact that we cannot
  predict the future

31
Saving the fish

• Embrace uncertainty
• wild salmon evolved behaviors that cope with
  environmental uncertainty
• restore natural climate insurance for salmon
• do this by restoring lost diversity of life
  history behaviors this diversity is directly
  linked to availability of healthy, complex
  freshwater habitat
• Save the Fishery
• People must be part of the solution

32
Saving the Fishery

• Save the Fish
• Rethink/revise goals of fishery management
• Industrial fishery model is doomed to failure
  (lots of fish healthy fishery) because it fails
  to deal with the unknowability in the fishery
  system