Adaptive management

Document Sample
Adaptive management Powered By Docstoc
					Adaptive management


   Dr. e. r. irwin
   FISH 7380
    Managing “Adaptively”
Adaptation defined:
   The adjustment of strategy based on improved
      understanding or observed change

The term “adaptive” predates natural resources by at
least a generation
   First used to describe management of engineering systems

   Based on the fact that you don’t always fully understand the
   system you’re managing
    What ARM is Claimed to Be
   Resource tracking
   Goal-directed management
   Strategic planning
   Sequential decision making
   Assessment of management impacts
   Applied Science
   What I’ve been doing all along
     What ARM is Not
   Not just the doing of science, even if management-
    oriented
   Not just the tracking of resources, or activities, or
    even impacts
   Not strategic planning per se
   Not the identification of goals and objectives
   Not a post-hoc assessment of management
   Most likely not what you’ve been doing all along
What ARM is
   “Managing natural resources in the face of
    uncertainty, with a focus on its reduction”

   Dual management focus
       Achieving the goals of resource management
       Increasing the level of understanding about resource
        dynamics pursuant to these goals

   Emphasis on uncertainty, and the value of reducing
    uncertainty through learning
     Framework for Resource Management
 So what makes good decisions so difficult?
     uncertain
                                    environmental
  resource status
 action                action         action
                                      variation           action



resource              resource          resource         resource
 status                status            status           status
          imprecise
           control

 return                return            return           return

                                                uncertain
ambiguous objectives                       resource processes


                                 time
Conditions for an Adaptive
Approach
   Sequential decision-making
   Agreed-upon management objectives
   Acceptable range of available actions
   Limited understanding of the biological processes
    driving resource dynamics
   Opportunity to improve management through a
    better understanding about these processes
   Opportunity to gain that understanding through
    smart decision-making
    So What’s New?
   Explicit accounting for uncertainty
      Typically through the use of models incorporating
       different hypotheses about system dynamics

   Focus on improving management through improved
    biological understanding

   Use of data accumulated over time
     Involves acquisition of useful data as a goal of
      management
     Involves design (or redesign) of monitoring
      programs specifically to reduce uncertainty
     Adaptive Decision-making
         …            decisiont           decisiont+1         …




                     monitoring          assessment


•   Management objectives guide decision making at each point in time

•   System responses to decisions are predicted with resource models

•   Monitoring used to track actual system responses

•   Actual vs. predicted responses are used to improve understanding

•   Biological status and improvement in understanding are used in the
    next round of decision-making in the next time period
What Makes it Adaptive?
   You account for where you are and what you know at
    each point in time
   You learn by doing, and learn as you go
   You anticipate how well your decisions will
    contribute to both management and understanding
   Management is used to support assessment, just as
    assessment is used to support management
   Basically, the process
       Recognizes competing hypotheses about resource dynamics
       Recognizes uncertainties about which is most appropriate
       Accounts for uncertainties in decision-making, so as to
        reduce them in the future
    Alternatives to ARM
   Ad hoc management
       Seat-of-the pants management
       Based on anecdotal information, absence of stated
        objectives
       Inadequate biological basis for action
   Wait-and-see
       Risk-aversive strategy that seeks to minimize
        management impacts as information accumulates
   Steady-state management
       Attempts to sustain resource system in some targeted
        steady state
   Conventional objective-based management
       Optimal management decisions based on an assumed
        resource model
Example: Adaptive Harvest
Management
   Used for setting annual waterfowl harvest
    regulations over the last decade
      Regulations are used to influence harvest rates,
        which in turn influence population dynamics

   Harvest regulations are set each year based on
     Breeding population status
     Pond conditions on the breeding grounds
     Uncertainty about regulations impacts
What is good for the duck
is good for the darter:
adaptive flow management.



       E. R. Irwin & M. C. Freeman
                 USGS
Adaptive Flow Management
(AFM)
   Iterative approach to management that
    acknowledges uncertainty and the need to learn.
   Process where all stakeholders decide initial flow
    treatment and assessment ensues.
   Return to table to evaluate success of flow
    management.
   Re-prescribe flow treatment if needed; continue
    assessment.
Objectives
   Assess the potential to use adaptive flow
    management to define suitable criteria for
    productive fisheries and community diversity,
    while accommodating economic and societal
    needs.
   Summarize empirical relations among biological
    and hydrological parameters from research in
    regulated Southeastern rivers.
   Stakeholders decide flow              Resource
 regime based on management                          Societal
            goals.
                                             Economic
                          AFM
                           Assessment =
                    management and research
                 to define ecological relations as
                        system is managed

Transfer knowledge
    Approach
   Compiled data from multiple projects to
    determine components of flow regime
    essential for biological processes.
   Quantify changes in flow regime.
    Constructed hypotheses testable in an
    Adaptive Flow Management framework.
What is required for AFM?
   Stakeholders that realize “adaptive” allows
    for adjustment of management regime as
    new information becomes available
   Testable hypotheses with measurable
    objectives to refine management
   Ability to embrace paradigm shifts, radical
    thinking
   Baseline and reference data (?)
Examples of AFM scenarios
   Striped bass in the Roanoke River, VA.
       Long-term flow and juvenile recruitment data were
        evaluated to establish alternative flows from dam.
   Robust Redhorse sucker in Oconee River, GA.
       Spring flows provided to allow for spawning
        windows.
       Flow-advisory team established to monitor success of
        management and potential modifications.
Adaptive management roadmap
    Identify stakeholders with respect to flows
     below Harris Dam
    Meet with potential stakeholders and
     explain the adaptive management process
    Form a workgroup of individuals
     representing all stakeholders
Stakeholders
   Middle Tallapoosa      USFWS
    Property Owners        NPS
   Lake Harris HOBOs
                           USFS
   Alabama Rivers
    Alliance               AL DCNR
   Bass Federation
                           USGS
   Alabama Power
    Company
Next step---Workgroup
   Identify clear, focused management objectives
    that represent all legitimate uses of the river. For
    example:
       Maintain biotic integrity within a certain range in
        specified segments in the river;
       Increase angler catch rates of sport fishes to a certain
        level in specified segments in the river;
       Maintain the economic value of the project at a
        specified percentage of current value;
Setting biological management goals




                         (versus flow
                       management goals)
Establish Management Goals
(versus setting fixed-flow criteria)
   Multiple-use riverine systems; all stakeholders
    goals must be considered.
   Not only a habitat-based approach for
    establishing flow criteria for fishes.
       Fish-habitat relations not linear; species specific.
       We don’t know “how much”, “how variable”or “how
        long.”
   Allows for flexibility in relation to natural flows.
Manipulation/Predicted Response
   Implementation of a            Increase density and
    continuous flow.                diversity of fishes and
                                    invertebrates.
   Provision of stable flows
    and mitigate temperature.      Increased recruitment,
                                    growth, and
                                    abundance of fishes.
   Provide predictable
    boatable flow windows.         Increased recreational
                                    use.
Workgroup
   Identify the array of flow management
    options. For example:
       Provide a baseflow during non-generation
        periods.
       Provide a certain number of contiguous days
        during which flow fluctuations are limited,
        during specified seasons.
       "Ramp" flows up and down at the beginning
        and end of peaking releases.
Workgroup
   Identify limits of acceptable management
    outcomes for APC and for the regulatory
    agencies. What must management achieve
    to be acceptable from all perspectives
    represented in the workgroup?
   Construct a set of meaningful hypotheses
    about relations between management
    objectives and flow parameters
Workgroup
   Incorporate alternative hypotheses into a
    set of models (decision analysis) that that
    predict outcomes with respect to
    management objectives given different
    flow management strategies and observed
    levels of variation in inflow (using
    historical gage data)
Faunal response: e.g. Fish Abundance, IBI



                                                                 a                b




                                                                                          d



                                                                             c




                                            Presen   Threshold
                                            t
                                                     Base flow (during non-generation intervals)
Workgroup
   Estimate the relative likelihood that each
    model (i.e., using alternative hypotheses)
    appropriately describes outcomes as a
    result of a change in flow management
    strategies
     Decision Support Models
• Powerful tools for assessment, learning and
  defining options for management.
• Demonstrate how these models will help us
  decide what to do at R.L. Harris.
• Discuss the methods by which we will build
  the models.
          Bridging the GAP




Conservation               Resource
assessment                management
 Development of Quantitative Planning
    Tools for the Flint River Basin
    Resource Management Decision-Making
        The Traditional “Black Box” Approach


 Resource                            Expected effects
Development                              Habitats

Conservation                           Populations

Restoration
        Quantitative Decision Modeling


Management                          External
                                    Physical
  Actions
                                   Influences

                                                     External
                                                    Biological
                                                    Influences




Stakeholder                  Aquatic
  benefits                  Community

              Explicitly incorporates uncertainty
               Types of Uncertainty
System uncertainty
 due to environmental and demographic variation

Statistical uncertainty
  due to the use of sample data to estimate parameters

Process uncertainty
  due to incomplete understanding of system dynamics


   Factor A            Factor B         Factor A      Factor B

                or                 or
  Population          Population              Population
   response            response                response
       Quantifying Uncertainty


Empirical Models


Expert Judgement



Combination
 Reducing Uncertainty: Bayesian Learning


New Information




Prior Estimate         Posterior Estimate
              Learning How a System Works
                       (Adaptation)
          Infot                                    Infot+1


                                          Actual
Current            Management
                                          future
 state               action
                                           state




                     Model A       Predicted
                   (hypothesis)     State A
                                                   Bayes’
                                                    Rule
                                    Predicted
                      Model B
                                      future
                    (hypothesis)
                                     State B
            4.5
                                                                                                     Etowah River
              4                                >10,050 cfs                                            5.7 fish/PAE
            3.5
              3                                                                                          17 spp.
            2.5
              2
                                                                                                    < 1 ind. = 10 spp.
            1.5                                                                                        24 spp. est.
Max pulse length (d)




              1                                                                                      78 known spp.
            0.5
              0                                                                                        58% recent
              1930             1940     1950       1960          1970          1980      1990     2000



                 25
                                                                                                        Lower
                                                                                                   Tallapoosa River
                 20                            >10,800 cfs
                                                                                                   15.7-20.7 fish/PAE
                 15
                                                                                                        33, 41 spp.
                 10                                                                               < 1 ind. = 25, 35 spp.
                       5                                                                              43, 47 spp. est.
                                                                                                      76 known spp.
                       0
                       1920   1930    1940      1950      1960          1970      1980     1990   2000 86% recent
Redbreast Sunfish Spawning
Success
   156 nests monitored daily (23 May-24 June
    1999).
   Mean daily nest failure was 14% for all life
    stages.
   Nest failure = 32% after 2-unit generation event.
   71% of nests with swim-up fry failed (1-unit).
   Only a total of 3 SUF observed after 2-units.
Daily flow pattern shows loss of stable-flow
      periods in hydropeaking regime
Thermal regimes are altered by Harris Dam
               operations
                                         Spawning Windows
                         100

                          90
Number of YOY/100 PAEs




                          80

                          70

                          60                                                         P. palmaris
                                                                                     Percina sp.
                          50
                                                                                     C. callistia
                          40                                                         C. venusta
                          30

                          20

                          10

                           0
                               0    50      100      150     200      250     300



                           Longest period without hydropeaking July-August (hours)
                                     Years between stable low-flow
                                  periods of >10 days in hydropeaking
                                                reaches
                                              Data for July-September
                              7

                              6
Recurrence Interval (years)




                              5
                                                                        Middle Tallapoosa
                              4                                         Lower Etowah
                                                                        Lower Tallapoosa
                              3                                         Oostanaula
                                                                        Lower Coosawattee
                              2

                              1

                              0
                                    Pre-dam            Post-dam
Flow regime below Harris Dam on the Tallapoosa River
           Hourly flows, April - August 1995
Availability of shallow habitats is high in a
hydropeaking reach of the Tallapoosa River…
PHABSIM data; Freeman, Bowen, Bovee and Irwin, 2001, Ecol. Appl. 11:179-190




80                                          80
60                                          60
40                                          40
20                                       20
                          Habitat availability, April-June,
0                                          0
                             based on hourly flows
     1994    1995     1996 1997                  1994 1995          1996      1997
                 Shallow-fast         Shallow-slow           Deep-fast
But hydropeaking greatly reduces temporal habitat
                    stability
Freeman, Bowen, Bovee and Irwin, 2001, Ecol. Appl. 11:179-190




80                                           80
60                                           60
40                                           40
20                                    20
                  Maximum period of habitat stability, April-
0                      June, based on hourly flows
                                       0
     1994     1995 1996 1997                1994 1995 1996                  1997
                 Shallow-fast          Shallow-slow             Deep-fast
    Reservoir Inflow                           Dam operation
                                         Status quo     0
Wet       22.3
                                         Inc baseNo     0
Normal 31.8                              Inc base W     0
Dry       46.0                           No base W      0




       Continuous non-generati...
                                            Degree days             Slow_Cover amounts
       hours0 100    0
       Hours 100 200 0                 high      100              High         0
       HoursGT 200 100                 moderate    0              Moderate  100
                                       low         0              Low          0




                                     Redbreast sunfish abund...
                                    high             0
                                    moderate      100
                                    low              0
    Reservoir Inflow                           Dam operation
                                         Status quo     0
Wet       34.0
                                         Inc baseNo     0
Normal 42.0                              Inc base W     0
Dry       24.0                           No base W      0




       Continuous non-generati...
                                            Degree days             Slow_Cover amounts
       hours0 100    9.70
       Hours 100 200 13.9              high     41.3              High      33.3
       HoursGT 200 76.4                moderate 29.3              Moderate 33.3
                                       low      29.3              Low       33.3




                                     Redbreast sunfish abund...
                                    high          31.7
                                    moderate      43.7
                                    low           24.5
    Reservoir Inflow                           Dam operation
                                         Status quo     0
Wet       22.3
                                         Inc baseNo     0
Normal 31.8                              Inc base W     0
Dry       46.0                           No base W      0




       Continuous non-generati...
                                            Degree days             Slow_Cover amounts
       hours0 100    0
       Hours 100 200 0                 high      100              High         0
       HoursGT 200 100                 moderate    0              Moderate  100
                                       low         0              Low          0




                                     Redbreast sunfish abund...
                                    high             0
                                    moderate      100
                                    low              0
    Reservoir Inflow                           Dam operation
                                         Status quo     0
Wet       33.4
                                         Inc baseNo     0
Normal 41.6                              Inc base W     0
Dry       25.0                           No base W      0




       Continuous non-generati...
                                            Degree days             Slow_Cover amounts
       hours0 100    8.40
       Hours 100 200 14.2              high     44.7              High      27.7
       HoursGT 200 77.4                moderate 28.2              Moderate 35.5
                                       low      27.1              Low       36.8




                                     Redbreast sunfish abund...
                                    high          22.1
                                    moderate      60.9
                                    low           17.1
What is next?
   Refine models using empirical evidence or expert
    opinion.
   Add to the model.
       All other fundamental objectives.
   To do this we will need to input appropriate data.
   We need to change something at the dam.
   Remember, this is a learn as you go process.
                                                                        Dam operation
                                                                 Status quo      0
                                                                                                              power production
          Reservoir Inflow
                                                                 Inc base ...    0                         high
      Wet       34.0                                             Inc base ...    0                         med
      Normal 42.0                                                Inc base ...    0                         low
      Dry       24.0                                             Inc Base ...    0
                                                                 No base W       0



                                                                                                                              Boatable days
 Shallow   fast amounts                                                                                                  BD GT 50   33.3
High         33.1                                                                                                        BD GT 100 33.3
                                                                                            Slow_Cover amounts
Moderate     33.9            Continuous non-generati...        Degree days                                               BD GT 200 33.3
Low          32.9            hours0 100    18.5
                                                                                          High      33.3
                                                          high     32.5                   Moderate 33.3                          lake_levels
                             Hours 100 200 23.2
                             HoursGT 200 58.3             moderate 35.0                   Low       33.3
                                                          low      32.5                                                    high      33.3
                                                                                                                           moderate 33.3
                                                                                                                           low       33.3


                                                             Redbreast sunfish abund...          Redbreast sunfish stability
                                 Small fish stability
                                                            high          30.6                   high        30.6
  Small fish abundance       high         30.7                                                   moderate    25.4
                             moderate 25.4                  moderate      25.4
high        30.7                                            low           44.0                   low         44.0
moderate 25.4                low          43.9
low         43.9
Workgroup
   Identify a starting point for changing the
    flow regime below Harris Dam; the starting
    point should have a high likelihood
    (according to the models) of achieving
    management objectives. Use models to
    identify an appropriate time-frame for
    assessing whether or not management
    objectives are met
Workgroup and technical advisors
   Design a monitoring program designed to
    assess attainment of management goals
    under a given flow management strategy
   Collect data under new management
    regime for appropriate time-period
Workgroup and technical advisors
   After the agreed-upon period for
    monitoring, use monitoring results to
    assess attainment of management goals.
    Based on the monitoring information,
    revise likelihood estimates for alternative
    models. Reassess the probabilities of
    attaining management objectives under
    alternative management strategies
Workgroup and technical advisors
   If management objectives are not being
    met under the current flow regime, choose
    a new strategy more likely to be successful
    based on the revised models. Return to
    step (k)
Workgroup
   Stakeholders agree to implement the
    change in flow regime, to monitor results
    for the appropriate period, how and when
    attainment of objectives will be assessed,
    and to then further modify the flow regime
    depending on outcomes relative to
    management objectives
Where are we now?
   Utility has provided some data that will be
    incorporated into the model. We need more
    disclosure.
   Utility has been “secretly” testing options at the
    dam.
   The other stakeholders are restless.
   The scientists are frustrated (but still hopeful?)
   A facilitator (or group dynamics psychologist) is
    needed for the next stakeholder meeting.
   Values need to be added.
http://www.freshwaters.org/framework/

The Ecologically Sustainable Water Management (ESWM)
Framework
  Framework:
  1. Define ecosystem flow
  requirements
  develop initial numerical estimates of
  key aspects of river flow necessary to
  sustain native species and natural
  ecosystem functions;
  2. Determine the influence of human
  activities
  accounting for human uses of water,
  both current and future, through
  development of a computerized
  hydrologic simulation model that
  facilitates examination of human-
  induced alterations to river flow
  regimes;
  3. Identify areas of incompatibility
  assessing incompatibilities between
  human and ecosystem needs with
  particular attention to their spatial and
  temporal character;
  4. Search for collaborative solution
  collaboratively searching for solutions
  to resolve incompatibilities;
  5. Conduct water management
  experiments
  design and implement water
  management experiments to resolve
  critical uncertainties that frustrate
  efforts to integrate human and
  ecosystem needs; and
  6. Design and implement an
  adaptive management plan
  using the knowledge gained in steps 1-
  5, create an adaptive management
  program to facilitate ecologically
  sustainable water management for the
  long term.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:2
posted:9/29/2012
language:English
pages:62