The scientific arm of the New Jersey Meadowlands Commission.
Wetland Monitoring and Assessment Using Hyperspectral Remote Sensing


Participants: MERI, Ball State University and Rutgers University-Newark


Problem:
 
The current wetland monitoring techniques used in the Meadowlands are labor intensive, time consuming and very subjective.
 
Objectives:
 
To design and implement a sustainable wetland monitoring and assessment program based on high resolution remote sensing that can be used in the Meadowlands and adopted by other agencies responsible for coastal wetland integrity and reporting.
 

Strategy:

Use both systems (Hyperspectral and LIDAR) to build a model that will link image values from vegetation texture and height to range values of salinity, redox-potential and water table level in the field. After the model is built and tested, only hyperspectral images will be required for the long term monitoring and assessment.

Activities:

  • Obtain approval on the Quality Assurance Project Plan (QAPP) from EPA (completed in March, 2007)
  • Contract Provider for data acquisition. SpecTIR LLC was contracted to fly the sensor covering the submitted areas (same as the LIDAR flight plan) in October, 2006.
  • Hyperspectral imagery was taken in June, 2007.
  • LIDAR imagery was taken in May, 2007.
  • Image Classification - pre-processing was completed by November, 2007. Vegetation maps are still being processed along with the accuracy assessment.
  • Install lysimeters at each sampling site. Four sites were considered representative for the District. At each site three transects were established with three sampling points along each. Lysimeters were installed at each sampling point.
  • Field Campaign – there were three seasonal samplings completed (June, August and October 2007, respectively). Biophysical measurements were carried out in January, 2008.
  • Sample Analysis – Analysis concerning sediment chemistry and water quality were completed after each sampling session.
  • Data Analysis – preliminary, descriptive statistical tests were done after all the field samples were processed.
  • Reporting – ongoing.

Deliverables:

  • Preliminary data was delivered in December, 2007. Geometric and brightness correction were done to reduce the error that occurred during acquisition.
  • The Riverbend Wetland Preserve area was classified using field spectral measurements along with image curves. Based on the classification the 2004 and 2007 vegetation maps were compared.
  • High resolution images (8X8 feet pixels) over time showing changes in species level distribution, water table depths, soil salinity and soil oxygen content due to human and natural disturbances.