Remote Sensing: Earth Observation Put to Work for You

1
Remote SensingEarth Observation Put to Work for
You

• AMEC Earth EnvironmentalWestford, MA

2
Introduction

• Remote Sensing 101
• Definition
• Image Classification
• Data Sources
• Advantages
• Example Applications

3
Remote Sensing 101

• Definition
• Remote sensing is the science of gathering
  information at a distance, and it provides a
  descriptive, analytical way to identify
  geographic features.
• Remotely sensed data can include aerial
  photographs, satellite imagery, acoustic data,
  and radar imagery (NOAA CSC).

4
Remote Sensing 101

• The use of remotely sensed data is increasing
  thanks to recent advancements in Geographic
  Information Systems (GIS) and image processing
  capabilities.
• Information that previously was available only to
  large organizations is now suitable for most
  desktop computers and is used by all branches of
  government as well as the private and non-profit
  sectors.
• This international, multi-billion dollar industry
  includes the producers, collectors, analysts, and
  sellers of spatial information and the associated
  tools (NOAA CSC).

5
Spectral Image Exploitation

• Materials can be distinguished because of color
  differences.
• Spectral sensors are created to exploit this
  phenomenology.
• Materials can be characterized by their unique
  spectral signature.
• Spectral signatures can then be used to find
  other occurrences of a particular material.

6
Image Classification

• The process of sorting individual image pixels
  into a number of categories according to their
  spectral reflectance characteristics
• Supervised classification
• Unsupervised classification
• Digital image processing versus manual
  digitization

7
Ground Truth

• Need for temporally coincident ground truth data

21 July 2005Shrub / Range
02 November 2005Pasture
8
Change Detection

• Use of multi-temporal data sets to discriminate
  change in land cover or in a phenomenon between
  dates of imaging.
• Many techniques can be used to detect change
  using remotely sensed data including the
  following
• Image Differencing
• Post-Classification Differencing
• Principal Component Analysis
• Tasseled Cap Analysis
• Vegetation Indices
• Image Ratios
• Multi-date Composite Image
• Manual Digitization

9
Sources of Data

• Airborne
• Flexibility in type of sensor, resolution of
  data, and timing of collection
• Satellite
• Pre-determined sensor and data resolutions, fixed
  or scheduled collections

10
Sensor Comparison Chart
11
Sensor Footprint Comparison
12
New and Future Sources of Data

• WorldView I
• Launched September 2007
• Panchromatic (0.45 meter)
• GeoEye-1
• To be launched August 22, 2008
• Panchromatic (0.41 meter)
• 4-band multispectral (1.65 meter)
• WorldView II
• To be launched mid-2009
• Panchromatic (0.46 meter)
• 8-band multispectral (1.8 meter)

13
Advantages

• Remotely sensed satellite data and airborne
  images of the Earth have several important
  advantages compared to ground observations.
• Synoptic view
• Allows for simultaneous regional-scale
  assessments
• Frequent and repetitive coverage
• Allows for easy updating
• Available archived imagery
• Allows for examination of historic conditions
• Worldwide coverage
• Allows for access to remote locations
• Low-cost data

14
Remote Sensing Applications

• Landcover classification for watershed modeling
• Monitoring landcover change associated with
  mining
• Impervious surface mapping for stormwater
  management
• Forest mapping and timber inventorying
• Crop monitoring and vegetation analysis
• Asset management
• Natural disaster damage assessment
• Flood or hurricane inundation mapping
• Oil spill detection and monitoring
• Facility siting, planning, permitting

15
Landcover Classification
Land Cover Categories

• Landcover data can be useful for a wide variety
  of applications
• In this case, landcover data was used in a SWAT
  model to identify critical source areas of
  phosphorus and target BMPs.

WaterForestHigh Biomass PastureLow Biomass
PastureShrub / RangeBare Soil
Clear-cutRock OutcroppingHigh Density UrbanLow
Density UrbanMiningClouds
16
Change Detection
Land Cover Change from 2000 to 2004
Unchanged WaterUnchanged ForestUnchanged High
Biomass PastureUnchanged Low Biomass
PastureUnchanged Shrub/RangeUnchanged Bare
SoilUnchanged Urban
Forest ? Clear-cutBare Soil ? ForestHigh ? Low
Biomass PastureLow ? High Biomass Pasture
CloudsOther
17
Watershed Modeling
Landcover
Topography
Soils
Model Predictions

Weather
Management
Point Sources
18
Monitoring Landscape Change

• Use multi-temporal change detection to monitor
  landcover changes associated with mining
  activities
• For example, identify and locate changes in
  wetlands from de-watering

19
Impervious Surface Mapping

• Frequent and repeated coverage allows for easy
  updating
• High spatial resolution allows for improved
  hydrologic runoff models
• Incorporation into GIS allows for stormwater user
  fees to be set on a per-parcel basis

20
Forestry / Timber Management
21
Delineation of Irrigated Crops

• Irrigated crops are easily distinguished from
  those that are not irrigated within a given year
• Repeated coverage allows for easy updating and
  change analysis

22
Vegetation Analysis
23
Stressed Vegetation as a Potential Indicator of
Natural Gas Seepage

• Locations of anomalously low vegetation density
  located near geologic faults and folds were
  identified using NDVI
• Low vegetation density may be indicative of
  stressed vegetation (i.e., by chlorosis or
  reduced development) because of soil type,
  microclimates, elevation, natural gas seeps, etc.

24
Asset Management

• High resolution satellite imagery, combined with
  limited field verification, can provide a
  cost-effective way to locate and inventory real
  property assets

• A comprehensive, accurate database can be
  developed and updated on a regular basis

25
Disaster Assessment
Before Katrina March 9, 2004
After Katrina August 31, 2005
QuickBird images courtesy of DigitalGlobe.
26
Remote Sensing Benefits

• Synoptic view
• Allows for simultaneous regional-scale
  assessments
• Frequent and repetitive coverage
• Allows for easy updating
• Available archived imagery
• Allows for examination of historic conditions
• Worldwide coverage
• Allows for access to remote locations
• Low-cost data

27
For More Information

• Brenda Berasi
• brenda.berasi_at_amec.com
• Alisa Planson
• alisa.planson_at_amec.com
• AMEC Earth Environmental
• [phone]