Coastal Change Analysis in the Great Lakes The Coastal Change Analysis Program CCAP Nate Herold NOAA

1
Coastal Change Analysis in the Great LakesThe
Coastal Change Analysis Program
(C-CAP)Nate Herold NOAA Coastal Services
CenterGreat Lakes Regional Data Exchange
Conference April, 2006
2
NOAA Coastal Services Center
Charleston, South Carolina
Mission Linking People, Information, and
Technology in the Coastal Zone

• Strategic Focus Areas
• Coastal Hazards
• Coastal Watersheds
• Integrated Ocean Observing Systems

3
Coastal Remote Sensing Program
Linking coastal resource managers with relevant
remote sensing technologies through

• Spatial data development
• Topography, land use/land cover, benthic,
  imagery, etc.
• Capacity Building Coastal Resource Management
• Outreach, training, technical support
• Geospatial tools development software
  applications
• Decision support, data manipulation
• Applied research and emerging technologies
• HABs, water quality, land use mapping, high-res.
  Sensors, etc.
• Outsourcing and contracting
• Private sector partnerships, CSC IDIQ contract

4
The Coastal Change Analysis Program
Monitoring the Nations Coasts

• National coastal land cover and change mapping
  program
• Satellite-based map products
• Standardized data and methods
• Designed to help improve understanding of
  linkages between land change and the environment
  

C-CAP Vision Develop national land cover change
products on a 5-year repeat cycle (or less).
5
The Coastal Change Analysis Program
Monitoring the Great Lakes
6
Monitoring the Great Lakes
New York Example
1996
2001
7
Monitoring the Great Lakes
Michigan Example
Net Change by Land Cover Category
Square Miles
8
Making C-CAP Operational
Retooling

• C-CAP Effectiveness Review Panel (1999)
• 13 recommendations identified, including
• Utilize private sector partnerships
• Need for standard, consistent, timely products
• Leveraging other national efforts (minimize
  duplication of effort)
• Focus on applications customer orientation
  (outreach effort)
• Collaborative discussions with MRLC (2000)
• Consortium focused on nationally consistent
  land cover data
• NOAA, EPA, USFS, USGS are key members
• Decision to modify the C-CAP approach
• NOAA to continue focus on mapping in coastal
  areas

9
Making C-CAP Operational
Retooling

• Embracing MRLC database concept
• Innovative approach
• Common classification scheme
• Automated procedures (decision and regression
  tree processing)
• Standardized and accessible inputs
  (ingredients)
• Metadata tracking of procedures
• Repeatable processes
• Credibility

10
NLCD Database Design
Metadata (5)
Tiled by Mapping Zone
Land cover (4)
Confidence Estimate
Node Map
Derivatives (3)
Image Data (1)
Imperviousness
Decision Rules
IF Node_map 169 spring TC_green lt 73 summer
TC_green gt 90 DEM gt 245 aspect 9 Then
Deciduous
Ancillary DEM Data (2)
Tree Canopy
Cross-Validation Accuracy
11
Value-Added Component of NLCD
Continued Coastal Focus

• Focus on additional coastal categories
  (Wetlands)
• freshwater and estuarine
• Additional processing to meet 85 accuracy target
  (categorical)
• Additional field sampling and validation
• Utilization of change detection to produce
  multiple-date land cover

Time 1
Time 2
Change
Impervious
Tree Canopy
Combined C-CAP/NLCD Product Line
12
Impervious Surfaces
Detroit, Michigan Example
13
C-CAP Classification Scheme

• Developed
• Developed, High Intensity
• Developed, Medium Intensity
• Developed, Low Intensity
• Developed, Open Space
• Agricultural
• Cultivated Crops
• Pasture/Hay
• Rangeland
• Grassland/Herbaceous
• Scrub / Shrub
• Forest Land
• Deciduous Forest
• Evergreen Forest
• Mixed Forest
• Barren Land
• Barren Land
• Unconsolidated Shore

Water Open Water Palustrine Aquatic
Bed Estuarine Aquatic Bed Wetlands
Woody Wetlands Palustrine Forested
Wetland Palustrine Scrub/Shrub Wetland
Estuarine Forested Wetland
Estuarine Scrub/Shrub Wetland Herbaceous
Wetlands Palustrine Emergent Wetland
Estuarine Emergent Wetland Perennial
Ice/Snow Tundra/Alaska Only Classes Dwarf
Scrub Sedge/Herbaceous Lichens
Moss
14
Revised C-CAP Mapping Area
Great Lakes Region
15
1996/2001 Baseline Land Cover
C-CAP Status
Completed
In Progress - Complete 2006
Proposed Mapping 2007 - 2008
16
2005/2006 Updated Land Cover
C-CAP Next Steps
Baseline Mapping Area
In Progress - Complete 2006
Proposed Updates 2007 - 2008
17
C-CAP Vision
Example Trend Analysis
A Small community

18
C-CAP Vision
Example Trend Analysis
A Small community B Local airport C Shipyard
expansion

19
C-CAP Vision
Example Trend Analysis
A Small community B Local airport C Shipyard D
Residential golf course development
E Residential golf course development
20
C-CAP Vision
Example Trend Analysis
A Small community B Local airport C Shipyard D
Residential golf course development
E Residential golf course development F
Residential development G Lake and campground
21
C-CAP Vision
Example Trend Analysis
A Small community B Local airport C Shipyard
expansion D Residential golf course development
E Residential golf course development F
Residential development G Lake and campground H
Land clearing for residential development
22
High Resolution Land Cover
C-CAP Vision
Santa Cruz, CA

• Long term C-CAP vision / need
• and
• Increasingly available, high resolution imagery
  and supporting data
• provide
• New opportunities to
• Introduce new data streams
• Introduce new approaches
• Increase focus on coastal issues
• Provide data at a spatial scale more appropriate
  for use in support of increasingly detailed, site
  specific, management decisions.

23
High Resolution Land Cover
Opportunities or Challenges?

• Budgetary Issues
• Reliable data streams will likely be commercial
  ()
• Wall-to-wall classification over CONUS area
  costly / impractical
• C-CAP will need to prioritize where we work
• C-CAP will need to identify sources of future
  funding
• Technical Issues
• Data storage (data volume)
• Processing requirements (available software and
  limitations)
• Spectral resolution, bandwidth
• Lack of short wave infrared data
• Classification scheme and mapping accuracy

24
High-Resolution Land Cover
High Resolution Land Cover
Landsat ETM Baseline C-CAP
Forested Wetland
Scale 115,000
Impervious Surfaces
Unmanaged Grassland
Scrub/Shrub Wetland
Managed Grassland
Emergent Wetland
Mixed Forest
Unconsolidated Shore
Scrub/Shrub
Bare Land
25
High Resolution Land Cover
83,00000
95000
575
50500
High Resolution Land Cover
Landsat ETM Baseline C-CAP
Santa Cruz Salmonid Recovery Prototype
Project Study Area 165 square miles
26
High Resolution Land Cover

• Image Data Source
• 5 separate sensors
• Satellite Airborne
• Spatial Resolution
• Max 10 meters
• Min 0.5 meter
• Methodology/Techniques
• Various methods
• Various technologies
• - eCognition, Feature Analyst, CART,
  Densification
• Classification Scheme
• Standard C-CAP/NLCD classes
• Added classes

Prototype Summary Information
IRS ResourceSat-1 23.5m multi-spectral 5.8m
pan SPOT5 10m multi-spectral 2.5m
pan-sharpened IKONOS 4m multi-spectral 1m
pan-sharpened Quickbird 0.67m pan-sharpened UltraC
am 0.5m multi-spectral and pan-sharpened
27
High Resolution Land Cover
Image Segmentation Example
28
High Resolution Land Cover
An Iterative Approach
29
High Resolution Land Cover
An Iterative Approach
30
High Resolution Land Cover
An Iterative Approach
Invasive Species (Phragmites australis) mapping
in CT.
31
High Resolution Land Cover
Multi-Resolution Nesting
Area Of High Resolution Land Cover
32
For More Information

• http//csc.noaa.gov/landcover

Contact Information Nate Herold Physical
Scientist / C-CAP Manager Coastal Remote Sensing
Program [phone] Nate.Herold_at_noaa.gov