| Oceanography > Issues > Archive > Volume 17 > Issue 2 |
2004, Oceanography 17(2):76–85, http://dx.doi.org/10.5670/oceanog.2004.50
Bottom Characterization from Hyperspectral Image Data
Authors | First Paragraph | Full Article | Citation
Authors
William Philpot | School of Civil and Environmental Engineering, Cornell University, Ithaca, NY, USA
Curtiss O. Davis | Remote Sensing Division, Naval Research Laboratory, Washington, DC, USA
W. Paul Bissett | Florida Environmental Research Institute, Tampa, FL, USA
Curtis D. Mobley | Sequoia Scientific, Inc., Bellevue, WA, USA
David D.R. Kohler | Florida Environmental Research Institute, Tampa, FL, USA
Zhongping Lee | Naval Research Laboratory, Stennis Space Center, MS, USA
Jeffrey Bowles | Naval Research Laboratory, Washington, DC, USA
Robert G. Steward | Florida Environmental Research Institute, Tampa, FL, USA
Yogesh Agrawal | Sequoia Scientific, Inc., Bellevue, WA, USA
John Trowbridge | Applied Ocean Physics and Engineering, Woods Hole Oceanographic Institution, Woods Hole, MA, USA
Richard W. Gould, Jr. | Ocean Optics Center, Naval Research Laboratory, Stennis Space Center, MS, USA
Robert A. Arnone | Ocean Sciences Branch, Naval Research Laboratory, Stennis Space Center, MS, USA
First Paragraph
In optically shallow waters, i.e., when the bottom is visible through the water, a tantalizing variety and level of detail about bottom characteristics are apparent in aerial imagery (Figure 1a). Some information is relatively easy to extract from true color, 3-band imagery (e.g., the presence and extent of submerged vegetation), but if more precise information is desired (e.g. the species of vegetation), spatial and spectral detail become crucial. That such information is present in hyperspectral imagery is clear from Figure 1b, which illustrates the Remote Sensing Reflectance spectra for several selected points in the image. Spectral discrimination among bottom types will be greatest in shallow, clear water and will decrease as the depth increases and as the optical water quality degrades. Discrimination can also be complicated by the presence of vertical structure in the optical properties of the water, or even if there is a layer of suspended material near the bottom (see Box on opposite page). Despite these difficulties, bottom characterization over the range of depths accessible to remote sensing is important since it corresponds to a significant portion of the photic zone in coastal waters. Mapping bottom types at these depths is useful for applications related to habitat, shipping and recreation. The purpose of this paper is to present the issues affecting bottom characterization and to describe various methods now in use. Given space limitations, we refer the reader to the references for results and examples of bottom type maps.
Full Article
Citation
Philpot, W., C.O. Davis, W.P. Bissett, C.D. Mobley, D.D.R. Kohler, Z. Lee, J. Bowles, R.G. Steward, Y. Agrawal, J. Trowbridge, R.W. Gould, Jr., and R.A. Arnone. 2004. Bottom characterization from hyperspectral image data. Oceanography 17(2):76–85, http://dx.doi.org/10.5670/oceanog.2004.50.