Near-Shore Marine Hyperspectral Case History Using ENVI- Tutorial of ENVI Software - Completely GIS, GPS, and Remote Sensing Lecture Material -
Near-Shore Marine Hyperspectral Case History Using ENVI

Overview of This Tutorial

This tutorial presents a case history for use of hyperspectral techniques for vegetation analysis using 1994 AVIRIS data from Moffett Field, California, USA. It is designed to be a self-directed example using ENVI's complete end-to-end hyperspectral tools to produce image-derived endmember spectra and image maps. For more detail and step-by-step procedures on performing such a hyperspectral analysis, please execute tutorials 7-11 in this booklet prior to attempting this tutorial.


1) To examine application of ENVI end-to-end hyperspectral processing methodology to a near-shore marine case study

2) To give students hands-on experience in actually running the procedures rather than reviewing pre-calculated results (preprocessed results are provided for comparison)

3) To provide students with guidance to perform data exploration in a loosely structured framework

4) To compare analysis results with known ground information.

Files Used in This Tutorial

You must have the ENVI TUTORIALS & DATA CD-ROM mounted on your system to access the files used by this tutorial, or copy the files to your disk.

The files used in this tutorial are contained in the M94AVSUB subdirectory of the ENVIDATA directory on the ENVI TUTORIALS & DATA CD-ROM.

The files listed below are required to run this exercise. Selected data files have been converted to integer format by multiplying the reflectance values by 1000 because of disk space considerations. Values of 1000 in the files represent reflectance values of 1.0.

Required Files (in the Default Spectral Library Directory)

USGS_VEG.SLI	USGS Vegetation Spectral Library
USGS_VEG.HDR	ENVI Header for Above
USGS_MIN.SLI	USGS Mineral Spectral Library
USGS_MIN.HDR	ENVI Header for Above

Required Files (in the M97AVSUB Directory)

M94AV.BIL	AVIRIS ATREM Corrected Data, 500 x 350 x 56 bands
M94AV.HDR	ENVI Header for Above
M94MNF.IMG	VNIR MNF Transformed data
M94MNF.HDR	ENVI Header for Above
M94MNF.ASC	VNIR Eigenvalue plot data
M94PPI.HDR	ENVI Header for Above
M94PPI.ROI	ROI of VNIR PPI threshld
M94_EM.ASC	VNIR ASCII File of Endmember Spectra - all EM
M94_EM.ROI	VNIR ROI File of Endmember Spectra - all EM
M94_EMA.ASC	VNIR ASCII File of Endmember Locations - selected EM
M94_SAM1.IMG	VNIR SAM Classes using M94EM1A.ASC
M94_SAM1.HDR	ENVI Header for Above
M94_RUL1.HDR	ENVI Header for Above
M94_UNM1.IMG	VNIR Unmixing image using M94EM1A.ASC
M94_UNM1.HDR	ENVI Header for above


  1. Examine ATREM-corrected apparent reflectance data and evaluate data characteristics and quality.
  2. Conduct Spatial/Spectral browsing to evaluate data, determine presence and nature of spectral variability, determine linearity of mixing, and to select wavelength range (s) for further analysis
  3. Reduce data dimensionality using MNF transform
  4. Select spectral endmember candidates using PPI
  5. Evaluate linearity and select endmembers using n-D Visualizer
  6. Map endmember distribution and abundance using ENVI mapping methods.

Moffett Field Site Background

  • Launch site for AVIRIS at Moffett Field
  • Remote Sensing Test Site Used By JPL and others since launch of AVIRIS in 1987
  • AVIRIS Standard Datasets for 1992-97
  • Study area for water variability (salt evaporation ponds with algae), urban studies, vegetation.

The salt ponds are highly colored and contain a dense biomass of algae and/or photosynthetic bacteria (Richardson et al., 1994). Accessory bacterial pigments cause distinct spectral signatures that can be detected using AVIRIS data. These include carotenoids, phycocyanin, and cholorphyll a and b. Application of the standardized AVIRIS analysis methods described below should lead to the extraction of endmembers from the data and spatial mapping of their distribution and abundance. There are obvious mixing non-linearities in the data, however, and care must be taken to recognize these. Near-Shore Marine Hyperspectral Case History - ENVI -

Figure 1: Moffett Field AVIRIS True Color Composite Image.

AVIRIS Processing Flow

The following diagram (Figure 2) illustrates an approach for analysis of hyperspectral data that is implemented with ENVI

Near-Shore Marine Hyperspectral Case History - ENVI -

The following outlines in general terms the implementation of this approach. The student is expected to follow the procedures below, referring to previous tutorials and the ENVI User's Guide for guidance in performing specific tasks where required. The purpose of this tutorial isn't to teach you how to run the ENVI tools, but how to apply the methodology and tools to a general hyperspectral remote sensing problem

  • Evaluate the ATREM Correction applied to the JPL-provided AVIRIS spectral radiance to remove the bulk of the solar and atmospheric effects, transforming the data from radiance to apparent surface reflectance. Examine the data using spectral/spatial browsing and color composites to characterize spectral variability and determine residual errors. Extract reflectance signatures for water, vegetation, urban areas, and geologic materials. Compare to spectral libraries.


M94AV.BIL ATREM Apparent Reflectance
USGS_VEG.SLI	Vegetation Spectral Library
USGS_MIN.SLI	Mineral Spectral Library
  • Apply the MNF Transform to the ATREM data to find the data's inherent dimensionality. Review MNF eigenvalue plot(s) to determine break-in-slope and relate to spatial coherency in MNF eigenimages. Determine MNF cut-off between "signal" and "noise" for further analysis.

Near-Shore Marine Hyperspectral Case History - ENVI -

Figure 3: MNF Eigenvalue Plot

Files: Make your own MNF-Transformed dataset or review the results in the files below

M94AV.BIL ATREM Apparent Reflectance
M94MNF.ASC	VNIR Eigenvalue ASCII Data
M94MNF.IMG	VNIR MNF Eigenimages
Near-Shore Marine Hyperspectral Case History - ENVI -
Near-Shore Marine Hyperspectral Case History - ENVI -

Figure 4: Top MNF Band 1, Bottom MNF Band 20.

  • Apply PPI Analysis to the MNF output to rank the pixels based on relative purity and spectral extremity. Use the FAST PPI option to perform calculations quickly in system memory, creating the PPI image. Display the PPI image, examine the histogram, and threshold to create a list of the purest pixels, spatially compressing the data. Near-Shore Marine Hyperspectral Case History - ENVI -

Figure 5: PPI Image.

Files: Generate your own PPI results and ROIs or review the results in the files below

M94MNF.IMG	VNIR MNF Eigenimages
M94PPI.ROI	VNIR PPI Threshold Results
  • Perform n-Dimensional Visualization of the high PPI value pixels, using the high signal MNF data bands to cluster the purest pixels into image-derived endmembers. Rotate the MNF data interactively in 3-D, or spin in 3-or-more dimensions and "paint" pixels that occur on the "points" (extremities) of the scatterplot. Use Z-Profiles connected to the ATREM apparent reflectance data and the n-D Visualizer to evaluate spectral classes. Use class collapsing to iteratively find all of the endmembers. Pay particular attention to the linearity of water mixtures, variability, and endmembers. Save your n-D results to a save state file (.ndv). Export classes to ROIs and extract mean spectra. Compare mean spectra to spectral libraries. Use spectral/spatial browsing to compare image spectra to ROI means.

Files: Extract endmembers and make your own ROIs or review the results below

M94MNF.IMG	VNIR MNF Eigenimages
M94AV.BIL ATREM Apparent Reflectance 
M94_EM.ASC	VNIR Saved ASCII Endmember Spectra (all)
M94_EMA.ASC	Selected VNIR Saved ASCII Endmembers
USGS_VEG.SLI	Vegetation Spectral Library
USGS_MIN.SLI	Mineral Spectral Library
  • Use ENVI's wide variety of mapping methods to map the spatial occurrence and abundance of materials in the Moffett Field scene. As a minimum, try the Spectral Angle Mapper (SAM) and Unconstrained Linear Unmixing. Use SAM to determine spectral similarity to image endmember spectra Perform your own SAM Classification or review the results below. If time and space permit, try a SAM classification using one of the Spectral Libraries. Be sure to evaluate the Rule Images. Use the Unconstrained Linear Unmixing to determine material abundances or review the results below. Be sure to examine the RMS error image and evaluate linearity and whether the physical constrains of non-negative and sum to unity (1) or less have been satisfied. Iterate if time and space permit. Compare abundance image results to the endmember spectra and spectral libraries using spatial/spectral browsing. If time and space permit, try running Mixture-Tuned Matched filtering and/or Spectral Feature Fitting. Near-Shore Marine Hyperspectral Case History - ENVI -

Figure 6. Spectral Unmixing Results: Red Pigment (UL), Green Pigment (LL),
Vegetation 1(UR), Vegetation 2(LR)


M94_EM.ASC	VNIR Saved ASCII Endmember Spectra
M94_EMA.ASC	Selected Saved ASCII Endmembers
M94AV.BIL ATREM Apparent Reflectance 
M94_UNM1.IMG	VNIR Linear Spectral Unmixing Results
USGS_VEG.SLI	Vegetation Spectral Library
USGS_MIN.SLI	Mineral Spectral Library 

Selected References

Richardson, L.L., 1996, Remote Sensing of Algal Bloom Dynamics: BioScience, V. 46, No. 7, p. 492 - 501.

Richardson, L.L, Buison, D., Lui, C.J., and Ambrosia, V., 1994, The detection of algal photosynthetic accessory pibgments using Airborne Visible-Infrared imaging Spectrometer (AVIRIS) Spectral Data: Marine Technology Society Journal, V. 28, p. 10-21.