Retrieval of Water Properties from Remote Sensing Reflectance

1
Retrieval of Water Properties from Remote Sensing
Reflectance

Presenter Candy Barbaran Martin Van Buren
High school Mentors Professor Fred Moshary, Dr.
Alex Gilerson Department of Electrical
Engineering
2
Introduction

• Ocean color remote sensing is used to monitor the
  earths carbon cycle, ocean water quality and
  contributes to climate research.
• Scattering and absorption change the intensity of
  light as a function of depth, wavelength and
  direction.
• The purpose of detecting all IOPs of the water
  itself and of all its constituents through remote
  sensing is to enable the measurement of
  biochemical parameters more accurately.

3
Long Term Objective

• Improve algorithms for instruments aboard
  satellites in order to make Fluorescence
  measurements of algae more accurate.

4
The Forward Problem

• The forward problem of ocean color remote sensing
  uses assumed values for IOPs which are used to
  predict the shape and or distribution of water
  leaving radiance (Rrs).
• The IOPs themselves are based on particle size,
  index of refraction distribution and shape.

Remote Sensing Reflectance
5
The Forward problem

• Measurements taken by instruments aboard
  satellites need to be compared to measurements
  taken directly off the field to test for
  accuracy.
• It is difficult to compare measurements satellite
  data sets vs In situ data sets.
• Values are assumed for IOPs which are used to
  predict the shape and or distribution of water
  leaving radiance (Rrs).

Hydrolight is used to simulate data sets with
biochemical parameters that vary in concentration
and diverseness which can be found in the field .
Curtis D. Mobely Ludia K. Sundman Sequia
Scientific , Inc
6
MODIS aboard Aqua

• Hardware aboard Aqua can be used to study climate
  change, vegetation, water vapor in the
  atmosphere, clouds and more.
• Aqua has nearly a polar route and makes
  measurements at the same time everyday.

Moderate Resolution Imaging Spectrometer takes 2
days to cover the globe. Has 36 different
spectral bands and acquires 2,330 km swaths.
7
The Inverse Problem

• When working in the inverse want to obtain
  Inherent Optical Properties using the upwelling
  radiance spectrum.
• This is difficult because we only know the water
  leaving radiance for only a few bands.
• SeaDAS outputs the biogeochemical parameters and
  also allows to account for various flags such as
  glint,cloud coverage and other variables that
  obstruct the pure signal to be detected by the
  satellites.

IOCCG Report Number 5 (2006) Remote Sensing of
Inherent Optical Properties Fundamentals, Tests
of Algorithms, and Applications.
8
Procedure

• Satellite imagery was retrieved through the use
  of SeaDAS for the areas of Chesapeake bay and
  Long Island Sound which are some locations from
  which we retrieve in situ data sets.
• Hydrolight was used to simulate Coastal water
  conditions.
• Several runs were made using Hydrolight using
  different absorption coefficients and
  concentration levels for each component.
  (Chlorophyll, CDOM , Mineral and water)
• Data Retrieved from Hydrolight was than graphed
  and analyzed to determine the effects of
  concentration of the various components for
  coastal waters.
• Further work would be to validate the accuracy of
  our algorithms when used in complex waters using
  simulated data sets.

9
SeaDAS Processed Images
Long Island Sound
Chesapeake Bay
10
In Situ spectrum for Remote sensing Reflectance
Vs MODIS Bands
Wavelength (nm)
Wavelength (nm)
11
Absorption of Chlorophyllous particles used for
the simulation of Reflectance's by Hydrolight
,
12
Absorption of CDOM used for the simulation of
Reflectance's by Hydrolight
13
Absorption of water and Total Absorption spectra
used for the simulation of reflectance's by
Hydrolight
14
Backscattering spectra of chlorophyllous
particles and mineral particles used in the
simulation of Reflectances by Hydrolight
Chl backscattering
Mineral backscattering
15
Backscattering spectra of water and total
backscattering used in the simulation of
Reflectances by Hydrolight
,
16
Typical reflectance spectra
Rrs bb/ (a bb) Lu/Ed
17
Conclusions

• Each component has a signature absorption and
  backscattering spectrum
• The steep slope of CDOM absorption has an
  inverse relationship with the magnitude of
  reflectance in the blue area of the
  spectrum(400-500).
• Maximum absorption peaks from 650-700 nm cause
  minimal reflectance in that part of the spectrum.
• There's a red shift for cases of high chlorophyll
  concentration.
• All of these observations need to be taken into
  account when working in the inverse in other
  words when using remote sensing reflectance to
  retrieve Inherent Optical Properties (IOPs).

18
Resources

• IOCCG Report Number 5 (2006) Remote Sensing of
  Inherent Optical Properties Fundamentals, Tests
  of Algorithms, and Applications.
• Curtis D. Mobley 1994 Academic press California
  Light and water (Radiative Transfer in Natural
  Waters)
• Vera De Cauwer, Polytechnic of Namibia MANUAL FOR
  WORKING WITH MODIS IMAGES IN SEADAS 4.9.4
  Draft version 1
• Curtis D. Mobley , Lydia K Sundman, Sequia
  Scientific, Inc
• Hydrolight Users Guide
• http//marine.rutgers.edu/cool/sat_data/?products
  stregionchessnothumbs0
• http//www.wetlabs.com/iopdescript/iopintro.htm

19
Dr. Alex Gilerson Professor Fred Moshary Dr.
Jing Zhou