Transcript March 25, 2009 - Cornell Computer Science

February 10, 2010
Green Infrastructure:
Linking Landscapes and Communities
By Ole M. Amundsen III
Strategic Conservation Program Manager
The Conservation Fund
Green Infrastructure – Linking Landscapes and Communities
The Conservation Fund
Mission Statement
The Conservation Fund forges partnerships to
conserve America’s legacy of land and water
resources.
Through land acquisition, community and economic
development, and training and education, the Fund
and its partners demonstrate balanced
conservation solutions that emphasize the
integration of economic and environmental goals
Green Infrastructure – Linking Landscapes and Communities
Strategic Conservation Services
•
•
•
Green
Infrastructure
Approach
Identify best lands
to conserve and to
accommodate
development
Implement
strategies for
smarter,
sustainable land
uses
IslandPress.org
Greeninfrastructure.Net
Green Infrastructure – Linking Landscapes and Communities
Why “Infrastructure”?
Infrastructure – “the substructure or underlying
foundation on which the continuance and growth of a
community depends” Webster’s New World Dictionary
 Network connections are required for functioning
infrastructure
Infrastructure features and processes need to be
managed to maintain essential services
 Funding for infrastructure needs to be provided upfront
 Infrastructure is a necessity not an amenity
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure – What is it?
• A strategically
planned and
managed network of
natural lands,
working landscapes,
and other open
spaces that
conserves ecosystem
values and functions
and provides
associated benefits
to human
populations
Green Infrastructure – Linking Landscapes and Communities
Nashville/Davidson County Open Space Plan
Public – Private Partnership
Land Trust for Tennessee
Manager of project
December 2010 target for
completion
Nashville Crayfish key
species, river front
development and
conservation
Green Infrastructure – Linking Landscapes and Communities
Land Use (2001)
Green Infrastructure – Linking Landscapes and Communities
Network Design Goals and Objectives –
NiSource Project
Identify protection priorities
for impacted species
identified in the Multiple
Species Habitat
Conservation Plan
Protect core species
habitat
Provide a framework for
avoidance, minimization,
and mitigation
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure Network Design
Core Areas:
• Contain fully functional
natural ecosystems
• Provide high-quality
habitat for native plants
and animals
Hubs:
• Slightly fragmented
aggregations of core
areas, plus contiguous
natural cover
Corridors:
• Link core areas together
• Allow animal movement
and seed and pollen
transfer between core
areas
9
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure Steps
Green Infrastructure Network Design
ID LANDSCAPE
TYPES/GATHER AND
PROCESS DATA
DEVELOP DESIGN
GOALS & OBJECTIVES/
SELECT DESIRED
GI ATTRIBUTES
SEEK
“OUTSIDE”
REVIEW &
INPUT
ASSESS NETWORK
COMPONENTS /
SET PRIORITIES FOR
ACTION
IDENTIFY & CONNECT
NETWORK
COMPONENTS
(HUBS, LINKS & SITES)
Green Infrastructure – Linking Landscapes and Communities
Step 1.
Green Infrastructure Network Design
DEVELOP DESIGN
GOALS & OBJECTIVES/
SELECT DESIRED
GI ATTRIBUTES
Green Infrastructure – Linking Landscapes and Communities
Central Indiana Land Trust GI Project
 Conserve significant contiguous
natural habitat
 Identify and protect a network of
stream and land corridors for
wildlife movement and human
enjoyment
 Help local planning become
more environmentally sensitive
 Increase public awareness of
the multiple benefits of green
infrastructure
 Increase public support for
green infrastructure
 Increase the coordination of
green and gray infrastructure
projects to maximize the
benefits for nature and people
Green Infrastructure – Linking Landscapes and Communities
Central Indiana and Public Perception
 Indianapolis ranks 99th out of 100 metropolitan areas
in per capita carbon emissions from transportation
and residential energy use
 Marion county ranks national at 568th out of 592 in air
quality according to the American Lung Association
 Forbes.com ranks Indiana 49th out of 50 states in its
America’s Greenest States
 Indiana has the highest amount of toxic discharges
into water bodies according to the US EPA
 Indiana ranks 16th for adult obesity (27%)
 Indiana ranks 7th in the US for loss of farmland
Green Infrastructure – Linking Landscapes and Communities
Resource Inventory, Analysis, and Synthesis
Gathering
information on
current natural
resource
conservation and
green space
programs and
policies at the
federal, state,
regional and local
level
Green Infrastructure – Linking Landscapes and Communities
Step 2.
Green Infrastructure Network Design
DEVELOP DESIGN
GOALS & OBJECTIVES/
SELECT DESIRED
GI ATTRIBUTES
ID LANDSCAPE
TYPES/GATHER AND
PROCESS DATA
Green Infrastructure – Linking Landscapes and Communities
Landscape Attributes
Identifying the landscape attributes or features present in the
study area provides direction as to the diversity of data you
will need to collect for network design.
Ecological Network
Landscape Type
Landscape Attribute
Forest
Large blocks of undisturbed forest
Wetland
High-quality wetland complexes
Freshwater aquatics
Minimally impaired stream reaches
plus associated riparian
vegetation, within minimally
impacted watersheds
Green Infrastructure – Linking Landscapes and Communities
Identify Landscapes – Central Indiana
In the Central Indiana there were several landscapes
of interest including:
• Forests
• Freshwater aquatic
• Wetlands
Green Infrastructure – Linking Landscapes and Communities
Identify Landscape Attributes – Central Indiana
Features Include
Trails
Historic Resources
Landcover Types
Stream-River Corridors
Lakes
Floodplains
Wetlands
Rare, Threatened and
Endangered Species
Green Infrastructure – Linking Landscapes and Communities
Create a Cultural/Recreation Asset Map
 Many regions do not have cultural and recreational assets
mapped so you may need to make your own map.
 The information does exist but most be gathered together.
 Official sources include:
National, State, Local Historic sites
National and State scenic byways
National and State Wild and Scenic rivers
 Local canoe launch sites
 Local trail heads and trails, bike paths
Other types of assets may benefit from asset mapping exercise:
Local farm stands
Local artisan studios
Swimming holes, bird watching and favor fishing spots
Water Towers
Green Infrastructure – Linking Landscapes and Communities
Assets, Needs, and Opportunities Exercise
• 109
opportunities
• 38 are within core
areas
Green Infrastructure – Linking Landscapes and Communities
Examine the Resource
Hawaiian
Honeycreepers
Source: Charles Smith, Cornell University, Dept. of Natural Resources
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure – Linking Landscapes and Communities
GAPS – High
species diversity,
but no preserve.
Green Infrastructure – Linking Landscapes and Communities
Step 3.
Green Infrastructure Network Design
SELECT Desired
GI ATTRIBUTES /
DEVELOP DESIGN
GOALS & OBJECTIVES
ID LANDSCAPE
TYPES/GATHER AND
PROCESS DATA
IDENTIFY & CONNECT
NETWORK
COMPONENTS
(HUBS, LINKS & SITES)
Green Infrastructure – Linking Landscapes and Communities
GI Network Design Guidelines
• Species well distributed across their
native range are less susceptible to
extinction than species confined to
small portions of their range.
• Large blocks of habitat, containing large
populations of a target species, are
superior to small blocks of habitat
containing small populations.
Better
Worse
• Blocks of habitat close together are
better than blocks far apart.
Better
Worse
Cont’d…
Green Infrastructure – Linking Landscapes and Communities
GI Network Design Guidelines
• Habitat in contiguous blocks is better than
fragmented habitat.
Better
Worse
• Interconnected blocks of habitat are better
than isolated blocks; corridors or linkages
function better when habitat within them
resembles that preferred by target
species.
Better
Worse
• Blocks of habitat that are roadless or
otherwise inaccessible to humans are
better than roaded and accessible habitat
blocks.
Green Infrastructure – Linking Landscapes and Communities
Core Areas
• Fully functional natural
ecosystems
• High-quality habitat for
native species
• Nucleus of the network
Green Infrastructure – Linking Landscapes and Communities
How Large Are Core Areas? It Depends…..
Example: Wetland Buffers and Functions
From Planners Guide to Wetland Buffers For Local Government” published by the Environmental Law Institute, 2008
Green Infrastructure – Linking Landscapes and Communities
Focal Species
• What are potential focal species for Davidson
County?
• Could you name species for forest, wetland and
aquatic resources?
USFWS Photo
Steven Wayne Rotsch/Painet Inc
US NPS
USFWS Photo
Robert Barber/Painet Inc.
Green Infrastructure – Linking Landscapes and Communities
Central Indiana: Core Forest Lands
American Woodcock (Scolopax minor)
Ovenbird (Seiurus aurocapillus)
• Deciduous successional
• Mid-late
shrubland
successional
• 20 acres - 3 acres
deciduous forest
• Forest interior bird
• Area sensitive
Yellow-breasted chat (Icteria virens)
• Forest patches >
•Till plains deciduous
750 acres
successional shrubland
• Shrubby openings of any
shape > 12.5 acres
Wood Thrush (Hylocichla mustelina)
• Till plains deciduous
forest
• Added Marion County Forest Blocks/woodlots
• 247 acres
> 12.5 acres
• Roads
• 100 meter edge
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure – Linking Landscapes and Communities
Central Indiana: Core Wetlands
King Rail (Rallus elegans)
• Emergent
wetlands/marshes
• > 20 acres
Indiana Bat (Myotis sodalist)
MaxEnt modeling
• Climate data
• Land use
• Other layers
Green Infrastructure – Linking Landscapes and Communities
Maximum Entropy (MaxEnt) Models
MaxEnt is a machine learning technique that can be
used to predict plant or animal distribution.
Compare species occurrences over a study area to a set
of relevant environmental factors, such as vegetation or soil.
 Estimates spatial distribution of the species by assuming
nothing about which is unknown (maximizing entropy) but
by matching the occurrence data with underlying
environmental variables.
 Useful for rare and endangered species since absence
data is not required and the model needs relatively small
number (50) occurrence points.
Green Infrastructure – Linking Landscapes and Communities
Indiana Bat MaxEnt Model
Indiana Bat Summer habitat preferences:
• Forested habitat, near water, wetlands
• Roosting sites under sloughing bark of large live or dead
trees with direct sun exposure
Winter hibernacula (caves) could not be satisfactorily
modeled, but most cave locations are known. Protecting
known hibernacula and combating white-nose syndrome
are the highest priority for this species.
For Indiana both the GAP data and National Land Cover
Data were used and compared
Green Infrastructure – Linking Landscapes and Communities
Indiana Bat Model for the State of IN
Model Inputs
• Point locations of focal species: A total of 208
element occurrences (both individuals and colonies)
were used for Indiana.
• Grid (raster layers) representing relevant
environmental factors: A total of 30 Environmental
Variables were used for Indiana.
Green Infrastructure – Linking Landscapes and Communities
Scaling Environmental Variables
Based on habitat needs, GIS layers were
assembled.
• Some layers needed to include a scale to
provide useful information
• Some variables used a 1 km scale based on
the average foraging range of 11 individuals
tracked by Sparks et al. (2005), and
• Other variables used a 3 km scale based on
their averaged maximum linear distance from
roost.
Green Infrastructure – Linking Landscapes and Communities
Table 1 Indiana Bat Environmental Factors
Green Infrastructure – Linking Landscapes and Communities
Model Output
Spatial layer
shaded between 0
and 1 (higher
values = more
suitable habitat)
Run the model 10
times for cross
validation
Heuristic Estimate of Relative Contributions
of Environmental Variables
The cross validation provides
“clues” as to which
environmental variables
predict species occurrences
to the greatest degree
1. Percentage forested
Wetlands with 1km
2. Length of unchannelized
streams or rivers in
deciduous forests within
1km
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure – Linking Landscapes and Communities
Central Indiana: Aquatic Core Areas
River Otter (Lontra canadensis)
High quality streams
• Known occurrences plus
50km of high quality stream
• Hilsenhoff Biotic Index or HBI
<= 5.50
• Qualitative Habitat Evaluation
Index or QHEI >= 50 (mean
score = 65.7 and sd = 14.24)
Fresh water mussels
Photo credit: Illinois State Museum
Photo credit: US FWS
Riparian buffers
• Streams with riparian
buffer on both sides (50
meters)
• Use expert knowledge
to delineate best streams
for mussel habitat
Green Infrastructure – Linking Landscapes and Communities
Green Infrastructure – Linking Landscapes and Communities
Hubs
• Slightly fragmented
aggregations of core
areas, plus contiguous
natural cover
• Large enough to support
populations of native
species and serve as
sources for emigration
into surrounding
landscape
Green Infrastructure – Linking Landscapes and Communities
How Big Are Hubs?
It depends…
Disturbance
Example Forest - In 300
Years:
-15 Fires
-4 Hurricanes
-7 Windstorms
-Infestations
-Blight
-Disturbances the Rule,
Not the Exception
Source Mark Anderson, TNC,
LTA Rally 2004
How Large Are Forest Hubs?
Source Mark Anderson, TNC,
LTA Rally 2004
Central Indiana Hubs
HUBs
 Hubs are aggregations of core areas
Significant sites as delineated by IN Natural Heritage
Program.
Pasture lands and hay fields surrounding core areas
with 300 meters (1,186 feet) of core edge.
Combine all and extract those hubs that are larger
than 100 hectares (247 acres).
Green Infrastructure – Linking Landscapes and Communities
Corridors
• Linear features linking core areas
• Wind through human-dominated land like agriculture or development
• Allow animal, seed, and pollen movement between core areas
Green Infrastructure – Linking Landscapes and Communities
How wide to make corridors? It depends…
Studies on corridor width vary;
best functioning corridors are
wide enough for interior
conditions and species
(as shown at right)
Narrow corridor with no interior
conditions
Narrow corridor with eroded
stream banks and invasive plants
Green Infrastructure – Linking Landscapes and Communities
Forested Buffers for Streams
No rules of thumb, however, recent studies recommend
a minimum buffer width of 115 feet of forested land cover
to provide basic physical and chemical buffering.
 For many eastern states, wildlife movement facilitated
by buffers at 650 feet in fragmented areas to 1,100 feet
(500 ft interior forests with 300 ft buffer on either side) in
less fragmented areas.
 Can consider reforestation in fragmented areas to
achieve adequate widths for corridors.
Green Infrastructure – Linking Landscapes and Communities
Least Cost Path GIS Model
Calculates the least accumulative cost distance
for each cell to the nearest source (e.g., between
one core area and others) over a cost surface.
A series of “cost” or “impedance” layers are
created.
 The higher the cost, the more difficult it is for
a species to move across the landscape
Model outputs are one cell width
(30 meter) paths
Paths are expanded to a minimum
size as required by focal species
Green Infrastructure – Linking Landscapes and Communities
Least Cost Path for Forest Corridors
Cost layer for forest is a composite of the following layers:
 Land cover (degree of forested land)
 Impervious surfaces
 Roads (remoteness, fragmentation)
Focal species:
Bobcat, grey fox, turkey, and
eastern box turtle:
 Corridor width set at 100 meters
(328 feet)
Green Infrastructure – Linking Landscapes and Communities
Greening the Crossroads
Green Infrastructure – Linking Landscapes and Communities
Step 4.
Green Infrastructure Network Design
SELECT Desired
GI ATTRIBUTES /
DEVELOP DESIGN
GOALS & OBJECTIVES
ID LANDSCAPE
TYPES/GATHER AND
PROCESS DATA
ASSESS NETWORK
COMPONENTS /
SET PRIORITIES FOR
ACTION
IDENTIFY & CONNECT
NETWORK
COMPONENTS
(HUBS, LINKS & SITES)
Green Infrastructure – Linking Landscapes and Communities
Characterizing a green infrastructure network
• Resources may be too limited to protect an entire network in the
short term, so usually need to prioritize efforts.
• Different ways to characterize and prioritize network elements:
– Overall ecological value
– Development risk
– Functional (hydrologic control, nutrient cycling, wildlife habitat,
etc.)
– Ecosystem services (water quality, groundwater recharge, flood
protection, pollination, carbon sequestration, genetic bank,
ecotourism, etc.)
– Combination (e.g., value and risk)
Green Infrastructure – Linking Landscapes and Communities
Restoration targeting
•
Can prioritize areas within green infrastructure for restoration efforts:
– Agricultural (if marginal), cleared, or mined areas;
– Wetlands impacted by dredging, draining, filling, etc.;
– Stream reaches without riparian forest cover;
– Ditches affecting wetland and stream hydrology;
– Stream blockages preventing fish migration;
– Other stream and floodplain degradation and alteration (channelization,
erosion, flashiness, etc.);
– Invasion by exotic species;
– Structures such as underpasses or bridges can facilitate wildlife
movement where roadways and railways cross corridors and hubs.
•
Potential projects can be prioritized according to benefits, costs, and
feasibility.
Requires on-site ground assessment and design.
•
Green Infrastructure – Linking Landscapes and Communities
Assess Network Components –
Corridors
Restoration Needs - Inadequate Network Width
Green Infrastructure – Linking Landscapes and Communities
Step 5.
Green Infrastructure Network Design
ID LANDSCAPE TYPES/
GATHER AND
PROCESS DATA
SELECT Desired
GI ATTRIBUTES /
DEVELOP DESIGN
GOALS & OBJECTIVES
SEEK
“OUTSIDE”
REVIEW &
INPUT
ASSESS NETWORK
COMPONENTS /
SET PRIORITIES FOR
ACTION
IDENTIFY & CONNECT
NETWORK
COMPONENTS
(HUBS, LINKS & SITES)
Green Infrastructure – Linking Landscapes and Communities
Public Involvement
• Over 80 stakeholders attended
Leadership forums
• 129 stakeholders attended one of
four focus group meetings on the
following topics:
– Working Lands
– People, Greenways and Wildlife
Corridors
– Forest Interiors
– Water Quality and Quantity
• Technical review team of 26 experts
Green Infrastructure – Linking Landscapes and Communities
Potential Student Projects
Institute for Computational Sustainability
Current Opportunities
• Optimization of land conservation opportunities
within the Central Indiana GI network or with the
land trust focus areas
• Optimization of restoration opportunities for the GI
network for Central Indiana
Future Opportunities
• Green Infrastructure Plan for Wildlife Management
Areas and Refuge System for State of Louisiana
• Nashville/Davidson County Open Space Plan
Green Infrastructure – Linking Landscapes and Communities
Additional Information
Ole M. Amundsen III
Strategic Conservation Program Manager
[email protected] phone 607-277-0999
 The Conservation Fund, Strategic Conservation Program Case Studies Website:
http://www.conservationfund.org/node/686
 Green Infrastructure Network website for Case Studies:
http://www.greeninfrastructure.net/projects
 The Conservation Fund Courses
http://www.conservationfund.org/training_education/upcoming_training_courses
 Weber, Ted, “Development and application of a statewide conservation network in
Delaware.” Journal of Conservation, Vol 3 (2007) pp. 17-46
http://www.journalconsplanning.org/
 Mark Anderson, How much is enough? Conservation Planning at Multiple Scales, TNC,
http://sweetwatertrust.org/forestreserves
 Planner’s Guide to Wetland Buffers for Local Government, Environmental Law Institute,
2008. See http://www.elistore.org/reports_detail.asp?ID=11272
 Conservation Thresholds for Land Use Planners, Environmental Law Institute, 2003.
See http://www.elistore.org/reports_detail.asp?ID=10839
Green Infrastructure – Linking Landscapes and Communities