Emerging and Contemporary Technologies in Remote Sensing for Ecosystem Assessment and Change Detecti

1
Emerging and Contemporary Technologies in Remote
Sensing for Ecosystem Assessment and Change
Detection on Military Reservations(CS-1098)

• Randall S. Karalus (Project Coordinator)
• USACE Topographic Engineering Center
• Paul T. Tueller, Ph.D. (Principal Investigator)
• University of Nevada Reno (UNR)

BRIEF TO THE SCIENTIFIC ADVISORY BOARD 16 MARCH
1999
1
2
PERFORMERS

• Randall S. Karalus (Proj. Coordinator) USACE
  TEC
• Paul T. Tueller (PI), Ph.D. U. of
  Nevada, Reno
• R. Douglas Ramsey, Ph.D. Utah State
  U., Logan
• Thomas D. Frank, Ph.D. U. of
  Illinois, Urbana
• Scott A. Tweddale
  USACE CERL
• Robert Washington-Allen Oak Ridge
  National Lab
• Carolyn Hunsaker, Ph.D. USDA Forest
  Service
• Laura McCarthy, Ph.D. Utah State U.,
  Logan
• Thomas G. Van Niel Utah State U.,
  Logan
• Others
• UNR Adrienne Breland (Graduate Assistant)
• UIUC Tari Weicherding (Research Ecologist),
  Michelle Suarez (Undergrad. Assistant)
• USU Eli Rodemaker (Graduate Assistant), Nick
  Zimmerman (Doctoral)
• ORNL Brian Beach, Tameka Ivory
• CERL Alana Anderson, Dr. David Price, William
  Jackson
• TEC Rob Fischer, Mike Campbell, Jeffrey Ruby,
  Gery Wakefield

2
3
PROBLEM STATEMENT
DoD alone has military bases with over 25 million
acres of land, more than 11 million of which are
training lands under Integrated Training Area
Management (ITAM).
Todays training/testing needs lead to changes in
the ecosystem that may exceed the current
estimated annual cost of 56M in land repair and
maintenance.
Hence, DoD has a need for efficient tools,
models, and techniques to better characterize
and quantify the carrying capacity of DoD land
resources to support military training and
testing.
3
4
TECHNICAL BACKGROUND

• This research is designed to develop reliable
  remote sensing techniques to relate ecological
  concepts of
• carrying capacity (EDYS ? ATTACC ? LMS),
• vegetation dynamics,
• species diversity
• critical thresholds,
• habitat fragmentation,
• ecosystem response and recovery,
• land degradation, and
• native plant community vigor
• to ecological endpoints.

4
5
TECHNICAL BACKGROUND (continued)

• Ecological endpoints can be more specifically
  defined in terms of the following
• Changes in plant species composition
• Decrease in plant productivity
• Reduction in soil quality
• Accelerated soil erosion and ,
• Changes in landcover and pattern that affect
  ecological function.
• Spectral indicators derived from these endpoints
  measure the response to training activities on
  military installations.

5
6
TECHNICAL OBJECTIVES
Stratify the landscape of individual military
ranges using contemporary and emerging remote
sensing technologies. Identify the fundamental
vegetation and soil attributes of military ranges
as they relate to plant succession in support of
the Ecological Dynamics Simulation (EDYS)
model. (EDYS ? ATTACC ? LMS) Identify the
spatial, spectral and temporal attributes of
remote sensing systems necessary to identify
ecotones. Establish ecosystem response and
recovery in relation to disturbance (land use)
through retrospective studies with
spatially-explicit spectral-based
indices. Develop methods for scaling through
multi-resolution imagery. ( EDYS -- Ecological
Dynamic Simulation model ATTACC -- Army
Training and Testing Area Carrying Capacity
model LMS -- Land Management System)
6
7
TECHNICAL APPROACH
Landscape scale mapping will provide maps which
identify areas of change/disturbance. Ecotone and
disturbance gradient analysis will identify the
sensor attributes necessary to monitor changes in
plant species composition along disturbance
gradients and plant successional
stages. Retrospective analysis will establish
ecological history of each installation in
relation to land use to establish the
ecosystems response and recovery to disturbance.
Multi-scale/Multi-spectral/Multi-temporal remote
sensing systems will complement the traditional
field techniques which alone are inefficient for
characterizing large landscapes.
7
8
TECHNICAL APPROACH (continued)

• This research will utilize contemporary and
  emerging image data, such as
• Landsat MSS and TM
• Indian Remote Sensing (IRS) satellite 5 meter
  data
• Proposed IKONOS Space Imaging satellite 1 meter
  data
• High resolution multispectral and hyperspectral
  airborne imagery and,
• Archived high resolution overflight data.

8
9
TECHNICAL APPROACH (continued)

• Ecotone and Disturbance Gradient Analysis
• Investigators
• UNR, UIUC, CERL, TEC
• Goals Spatial/Spectral
• Assess high resolution systems to identify the
  sensor attributes necessary to monitor changes in
  plant species composition along disturbance
  gradients and plant successional stages.
• Calibration of scales to allow extrapolation over
  larger geographic regions.

• Retrospective Analysis
• Investigators
• ORNL, USU, TEC
• Goals Temporal/Spectral
• Establish ecological history in relation to land
  use to describe how activities affect ecosystem
  and landscape response and recovery, i.e.,
  resilience.
• Identify the range of variation in the
  characteristics of disturbances associated with a
  landscape.
• Determine the existence of thresholds in response
  and recovery to natural and DoD activities.

9
10
TECHNICAL APPROACH (continued)
Ecotone and Disturbance Gradient Analysis
IMAGERY
GIS
IRS 5-m
Veg. from TM
HumVee Tracks
Veg. from Kodak
Kangaroo Rat
CAMIS 1-m
Species List
DATA
Photo
Kodak 0.2-m
Species List
LCTA Data
Photo
Photo
10
11
TECHNICAL APPROACH (continued)

• Retrospective Analysis
• Dry/Wet season anniversary Landsat images, from
  1972 to the present, will provide historical
  context.
• RFMSS and other land management data will
  characterize the land disturbance.
• Spectral Indices of land degradation will relate
  traditional field measurements to their
  respective imagery counterparts.
• Spectral Indices will then be related to a sites
  ecological resilience and critical thresholds,
  thus representing the point where significant
  change in plant species composition occurred.
• Topographic/hydrologic models will relate
  topographic features (e.g., sinks, sources) to
  soil and vegetation indices.

11
12
TECHNICAL APPROACH (continued)
Retrospective Analysis
Landsat 75

State Space Manifold
Landsat 83
Landsat 90
Land sat 97
Vegetation Cover Comparisons
Vegetation Cover
12
13
ACCOMPLISHMENTS

• MILESTONES
• Acquire / pre-process archival and new data.
  
• Acquire high resolution imagery.
• PROGRESS
• Acquired Image Data Sets and Preprocessed
  (e.g., spreadsheets).
• ? Satellite Landsat TM/MSS, IRS, SPOT,
  AVHRR
• ? Aerial Color, Color IR, BW,
  Multispectral, Radar
• ? Others DOQ, IFSAR
• Acquired and/or Built GIS Data Layers.
• ? Soils ? Vegetation ? Climate
• ? Transportation ? Land Use ? ACEC,
  Exclosure Sites
• ? Hydrology ? LCTA, RFMSS ? Grazing
• ? Fire polygons ? Topography ? Climate
• Acquired CAMIS and Kodak high resolution
  imagery.
• ? Coordinated Overflights with Range Control.
• ? Coordinated Ground Efforts.
• ? Flew Fort Bliss and Camp Williams transect
  sites.
• ? Preprocessed frames and documented status
  (e.g., overlap, cloud cover)

13
14
ACCOMPLISHMENTS (continued)
Example Image and Data Resources
IRS Scene mosaic
Image Processing
Fort Bliss
TM/IRS Image Fusion, 5m
CAMIS mosaic, 1m
Kodak mosaic, 0.2m
Field Transect Data
14
15
ACCOMPLISHMENTS (continued)

• MILESTONES
• Establish survey points for ecotone
  identification/analysis.
• Build classifications of ecological sites.
• PROGRESS
• Selected and Prioritized Study Sites
• Commenced field reconnaissance/work and site
  transects.

• Fort Bliss (Army)
• 1. CAMPBELL TANK
• 2. ECHO-3 WAY
• 3. TOY TANK PIOSPHERE
• 4. COPPICE DUNE
• 5. BURN SITE

• Camp Williams (Army National Guard)
• 1. ARTILLERY
• 2. BIVOUAC and TROOP TRAINING
• 3. BURN RECOVERY

• Twentynine Palms (Marine Corps)
• 1. DELTA TRAINING AREA
• 2. LAVA TRAINING AREA
• 3. PDRTSS SITE
• 4. EMERSON LAKE
• 5. BLM WILDERNESS SITE

Note Although not comprehensive, these sites
represent some primary Ecotone and Degradation
Gradients found at the facilities. They
represent sites of interest identified by the
facility environmental managers.
15
16
ACCOMPLISHMENTS (continued)
Ecotone and Degradation Examples

• Camp Williams (Army National Guard)
• BIVOUAC and TROOP TRAINING AREA

• Fort Bliss (Army)
• CAMPBELL TANK PIOSPHERE

16
17
ACCOMPLISHMENTS (continued)

• MILESTONES
• Begin the retrospective analysis.
• PROGRESS
• Acquired Data Sets.
• ? Satellite Landsat TM/MSS (25 years of wet
  and dry season)
• ? Elevation USGS DEM
• ? Meteorological Precipitation and temperature
• Initiated the development of a biophysical
  model/site water balance.
• ? DEM quads merged and cleaned
• ? Researched climate simulation programs
• ? Spatially interpolated daily meteorological
  values
• ? Tested, Corrected, and Improved solar
  radiation programs
• ? Finished development of evapotransporation,
  site water balance and soil wetness models
• Developed Image Processing Algorithms.
• ? Standardized to at-sensor reflectance.
• ? Performed relative atmospheric correction.
• ? Detected response of ecological indicators.
• Developed Prototype Software to Analyze
  Concentration Areas.

17
18
ACCOMPLISHMENTS (continued)
Example Analysis and Results
Soil-Adjusted Vegetation Index 1995
Incr. Heterogeneity
State Space Manifold, Dry Season
18
19
ACCOMPLISHMENTS (continued)

• OTHER PROGRESS (not associated with a milestone)
• Coordinated Research / Maintained Cognizance
• EMAP -- Coordination Plan Dr. Hunsaker
  (USDA), Dr. Bruce Jones (EPA), Dr. Barbara
  Walton (ORD), Dr. Steven Paulsen
• EMAP Symposium on Western Ecological Systems
  Status, Issues and New Approaches April 6 - 8,
  1999. Invited to present.
• The National Environmental Monitoring
  Initiative -- Applicable data, for example, 3rd
  scale of intensive monitoring sites LTER sites
  Jornada and Sevilleta NERP site in Nevada may
  have bearing on 29 Palms.
• International Association Landscape Ecology
  Abstracts
• WARFIGHTER 1

19
20
ACCOMPLISHMENTS (continued)

• EMAP Coordination Progress
• EMAP - 12,600 hexagon across U.S.
• - Data collection is ongoing
• - No EMAP collection in western U.S.
• - Landscape Working Group Strategic Plan in
  May 1999
• - EPAs ORD to monitor western U.S. in
  1999/2000
• Progress
• Established contacts with EMAP Scientist Dr.
  Bruce Jones (EPA), Dr. Barbara Walton (ORD),
  Dr. Steven Paulsen(Technical Workgroup Chair of
  Western EMAP Study).
• Provided project material on a continual basis
  to Dr. Bruce Jones.
• Considered additional relevant research (e.g.,
  LTER).
• Will attend and present at the EMAP Symposium
  on Western U.S. Ecological Systems.

20
21
PROGRAM FUNDING

• Major Project Tasks
  FY99 FY00 FY01
• Data Acquisition
  100K 130K 50K
• Ecosystem Stratification/GIS
  100K 100K 100K
• Retrospective Analysis
  400K 400K 275K
• Ecotone/Degradation Gradient Analysis 400K
  400K 275K
• Transition Plan/ Publication
  50K 100K 190K
• TOTAL 1,050K 1,130K 890K

21
22
PROGRAM PLAN
22
23
DELIVERABLES

• Ecotone and Degradation Gradient Analysis
• Protocols for spatial change detection.
• Methods for scaling through high and low
  resolution data sets for change detection over
  large geographic extents.
• Remote sensing techniques and procedures, and
  sensor requirements, necessary to identify and
  monitor species composition, plant successional
  stages, and soil attributes on military ranges.
• Protocols for linking contemporary and emerging
  sensor data to ecological models.
• Vegetation maps for each study site.
• Data Base of contemporary/emerging high
  resolution imagery and ground data.
• Peer Reviewed Publications.
• Interim Report Contributions.
• NOTE Analysis procedures are to be developed
  using commercial-off-the-shelf software (e.g.,
  ARCView, ERDAS Imagine) and available on CD ROM.

23
24
DELIVERABLES (continued)

• Retrospective Analysis
• Protocols for temporal change detection.
• Ecological Resilience/Behavior Models at critical
  thresholds of the ecosystem.
• Remote Sensing Techniques and Procedures
  necessary to identify and monitor ecosystem
  response and recovery on military ranges.
• Protocols for linking contemporary data to
  ecological models.
• Ecosystem Maps for each facility.
• Data base of contemporary imagery providing
  historical context.
• Enhanced Data Base of RFMSS, unit , and training
  information.
• Peer Reviewed Publications.
• Interim Report Contributions.
• NOTE Analysis procedures are to be developed
  using commercial-off-the-shelf software (e.g.,
  ARCView, ERDAS Imagine) and available on CD ROM.

24
25
TRANSITION PLAN

• Spatial change detection protocols will be
  incorporated into, for example, the Ecological
  Dynamics Simulation (EDYS) model.
• Temporal change detection and uncertainty
  analysis protocols will be demonstrated,
  validated, and implemented within the EDYS
  framework, and would be available for
  implementation at other installations at the
  completion of this effort.
• Technology transfer will occur via scientific and
  technical literature and CD-ROMS prepared as a
  part of this study with specific information for
  each study location.
• Protocols are being incorporated into widely used
  off-the-shelf software (e.g., ESRI ArcView GIS
  ERDAS Imagine image processing software).
• Ecological Dynamic Simulation model (EDYS) will
  be the environmental component of Army Training
  and Testing Area Carrying Capacity Model
  (ATTACC) ATTACC is to become one of the many
  "models" integrated under the Land Management
  System (LMS) umbrella.
• (EDYS ? ATTACC ? LMS)

25
26
ACCOMPLISHMENTS (continued)
PUBLICATIONS and PRESENTATIONS

• Characterization of the ecological integrity of
  commercially grazed rangelands using remote
  sensing-based ecological indicators. Invited
  Paper/Talk Agricultural Research
  Service-National Resources Conservation Service
  (ARS-NRCS) Workshop on Rangeland Health, Las
  Cruces, New Mexico, October 27 - 29, 1998.

• An Automated Method for Atmospheric Correction
  of Satellite Imagery. Poster and Abstract at
  1999 Annual Meeting of the American Society of
  Photogrammetry and Remote Sensing, May 17-21,
  1999 Portland, Oregon. (Also, peer-reviewed
  manuscript to be submitted to Computers and
  Geoscience in 1999.)

• A Remote Sensing-Based Method for Retrospective
  Ecological Risk Assessment of Rangelands Subject
  to Grazing and Military Land Use. Abstract, and
  Oral Presentation to the EMAP Symposium on
  Western Ecological Systems Status, Issues and
  New Approaches, April 6 -8, 1999 in San
  Francisco, California. (Also, peer-reviewed
  manuscript to be submitted to Special Issue of
  Journal of Environmental Monitoring and
  Assessment in 1999.)

• A Spatially-Distributed Method for Calibrating
  SAVI .IALE 99, Landscape Ecology The Science
  the Action. International Association for
  Landscape Ecology 5th World Congress, July 29 -
  August 3, 1999, Snowmass Village, Colorado.

• Approaches to mapping ecotone boundaries using
  emerging remote sensing technology. Proceedings
  The Tenth Wildalnd Shurb Symposium. Ephriam, UT.
  May (Published abstract mansucript in Press)

• Arid Land Degradation with emphasis on Ft.
  Irwin and 29 Palms MCAGGC (Marine Corps Air
  Ground Combat Center). An invited oral
  presentation to the Department of Geography and
  Environmental Engineering at the United States
  Military Academy, West Point, New York April 9,
  1999.

• Paper to be presented at the 17th Biennial
  Workshop on uses of Color Photograpy and
  Videography in the Plant Sciences and Natural
  Resources to be held in Reno this May.

26
27
Army Training and Testing Area Carrying Capacity
(ATTACC)

• Training Policy
• BLTM
• OPRED

Calculate Training Load

• Develop Severity Factors
• Event
• Vehicle

Develop Relationship Between Land Condition and
Training Load

• Installation Data
• Current Land Condition

• Determine Resources () Required to Meet
  Standards
• Develop Land Investment Strategies to Meet
  Standards

Training Land Capacity Standards
27
28
EDYS Requirement

• Problem Military training and testing
  disturbs lands and limits sustained, realistic use

• Objectives Model long-term effects of military
  land disturbance and recovery on ecosystem health
  and plant succession on military installation
• Approach
• Determine the ecological mechanisms controlling
  land recovery
• Model critical relationships for major ecological
  regions
• Translate model results into carrying capacity
  factors for military land use

Pinion Canyon Maneuver Site, CO
28
29
EDYS TECHNICAL APPROACH Plant Species Composition
ATTACC Plant species composition identified as a
critical unit of measure

• CS1102 Incorporate plant species composition
• Utilize the Ecological Dynamics Simulation
  Model to simulate succession. (UTEP, USACERL,
  NRCS, NPS, DoE)
• Utilize remote sensing technologies to map
  plant species to initialize model
• Utilize remote sensing technologies to validate
  model predictions

29