Using Satellite Imagery to Analyze Lake Quality

1
Using Satellite Imagery to Analyze Lake Quality

• Matthew J. Kucharski
• Under the direction of Stefan Falke
• And CAPITA
• Washington University in St. Louis
• REU Program
• August 6, 2004

2
Idea behind the Project

• Monitoring of the earth has risen dramatically in
  the last years allowing coverage of large regions
  year round.
• Purpose of Project
• SeaWiFS satellite imagery data processing
• Understand the processing of SeaWiFS data for
  imaging
• Convert raw data to usable formats for air and
  water quality analysis
• Application of satellite imagery to lake water
  quality monitoring
• Compare SeaWiFS and MODIS surface reflectance
  (color) data to lake clarity monitoring data

3
Spectral Characteristics of soil, vegetation, and
water

• Soil
• Reflects Red and wavelengths with higher
  frequency
• Vegetation
• High in green (555 nm) and infrared wavelengths
• Water
• Absorbs most of the light reflected upon it
• Reflects mostly the lower frequency wavelengths
  i.e. blue

Visible Spectrum
Soil
Vegetation
Water
4
Passive remote sensing from Satellite

• Use of the suns radiance
• Absorption and scattering of sunlight in the
  atmosphere, on the land, and on the water

5
CAPITA and Processing SeaWiFS data

• At Center for Air Pollution Impact and Trend
  Analysis (CAPITA), SeaWiFS data is being
  processed for optical thickness of aerosols
  across the United States

6
The SeaWiFS data Process

• Georeferencing
• Splicing and Mosaicing
• Rayleigh correction
• Scattering angle correction
• Process Created by Sean Raffuse at Washington
  University

7
Other Manipulations for SeaWiFS dataCreate Time
Series Plots and Color Time Series
Jan.
June
Nov.
8
Previous attempts at using Satellite for Lake
monitoring

• Environment Remote Sensing Center at University
  of Wisconsin
• Uses Landsat and MODIS imagery to monitor Water
  clarity i.e. Secchi Depth.
• Correlation found between Secchi Depth and Blue
  to Red ratio of the reflectance

9
Using SeaWiFS imagery for Lake Monitoring
Missed Algae Event?
June
Apr.
Oct.
10
Data Used
Fence Lake

• Wisconsin has well extensive monitoring of
  Various size lakes by Wisconsin Department of
  Natural Resources and the Self-Help Lake
  Monitoring Volunteers

Wind Lake
North Twin Lake
Tomahawk
Shawano
Green Lake
Lake Koshkonong

• SeaWiFS surface reflectance data for 2000, 2001,
  and 2002

11
Results of Wisconsin Lakes
Fence Lake
Wind Lake
North Twin Lake
Tomahawk
Shawano
Green Lake
Lake Koshkonong
12
Spectral Characteristics of Lakes

• No Two Lakes are the Same
• Lake size
• Lake chemistry
• Difficult to derive a universal model

13
Problems and Potential with SeaWiFS for Lake
Monitoring

• One km Resolution creates a wide area to monitor
• Not functional with smaller or narrower lakes
• Temporal Alignment
• Atmosphere remaining after processing
• Haze, clouds ect.

• Possible to analyze seasonal changes of lakes
• Color time series gives visual aid to current
  monitoring techniques

14
MODIS vs. SeaWiFS ImageryTable Rock Lake

• SeaWiFS Image 1 km resolution

• MODIS Image 250m resolution

15
Table Rock Lake

• Current research project involving Wash U.
• High phosphorus levels
• Creates Eutrophication
• Monitored using Secchi Disk Depth by the Lakes of
  Missouri Volunteer Program every 20 days
• Of eight time spans for gathering Lake data,
  MODIS had only five Cloud-free days

16
Results of Table Rock Lake

• At 250m resolution
• Examined relationships by day and by site

17
500m Resolution
18
Personal Secchi Depth Sampling

• On July 12-13 2004, I recorded my own Secchi
  depth measurement.
• Secchi disk provided by David Caseletto
• Boat Provided by Table Rock Lake Water Quality,
  Inc.
• Days were perfectly clear for Satellite

• Results
• Definite Human Error
• No longer just a place on a image

19
Future research opportunities of Lake Monitoring

• Lake monitoring via satellite on regional or
  individual scale
• Explore other factors affecting the lakes
  spectral characteristics
• Extent MODIS imagery analysis with similar tools
  as SeaWiFS

20
Acknowledgements

• I would like to acknowledge Dr. Stefan Falke, Dr.
  Rudolf Husar, and Erin Robinson at CAPITA. Dr.
  Lars Angenent and Dr. Dan Giammar for valuable
  insight about lake properties. Also, Gene Bulfin
  with technical support.

21
References

• Giammar, Daniel and Angenent, Lars, 2004.
  Evaluation of Chemical and Biological Tracers
  for Source Appointment of phosphorus in Table
  Rock Lake, on the Missouri- Arkansas Border.
  Proposal
• Introduction to Remote Sensing Environment. 2004.
  www.microimage.com. Lincoln MicroImages, Inc.
  http//www.microimages.com/getstart/pdf/introrse.p
  df
• Kaufman, Y. J., Tanre, D., Gordon, H. R.,
  Nakajima, T., Lenoble, J., Frouin, R., Grassl,
  H., Herman, B. M., King, M. D., and Teillet,
  P.M., (1997), Passive remote sensing of
  tropospheric aerosol and atmospheric correction
  for the aerosol effect, J. Geophys. Res.
  10216,815-16,830.
• Li, F., and Husar, R. B., (1999), Pre-processing
  of SeaWiFS satellite data for aerosol retrieval
  Online. Center for Air Pollution Impact and
  Trend Analysis. Available from
  http//capita.wustl.edu/capita/capitareports/CoRe
  trieval/SeaWiFSPreProcessinghtm. Accessed 27
  July 2004.
• Lillesand, Thomas M. Combining Satellite Remote
  Sensing and Volunteer Secchi Disk Measurement
  for Lake Transparency Monitoring. University of
  Wisconsin
• Radiative Transfer Theory, Atmospheric
  Correction, and Ocean Color. University of
  Miami. June 25, 2004 http//www.physics.miami.ed
  u/chris/envr_optics.html
• Raffuse, Sean M. 2003. Estimation of Daily
  Surface Reflectance over the United States from
  the SeaWifS Sensor. Thesis. Washington
  University In St. Louis
• Schultz, Gert A. Ed., Engman, Edwin T., Ed.
  Remote Sensing in Hydrology and Water
  Management. Springer Heidelberg 2000
• Wisconsin Department of Natural Resources.
  Self-Help Lake Monitoring. July 6, 2004
  http//www.dnr.state.wi.us/org/water/fhp/lakes/se
  lfhelp/index.htm