IDENTIFYING INFORMATION
Project title: Land Characteristics from Remote Sensing
Geographic area: Southern Florida
Project start date: October 1, 1996
Project end date: September 30, 2000

Project chief: John W. Jones
Region/Division/Team/Section: Eastern/NMD/Mapping Applications Center/Science and Applications Branch
Email: jwjones@usgs.gov
Phone: 703/648-5543
Fax: 703/648-4165
Mail address: 521 National Center Reston, Va 20192

Program(s): Integrated Natural Resources Science Program, South Florida Ecosystem Program

Program element(s)/task(s): Element 2/Task 2.9

BACKGROUND NARRATIVES
Project summary:
Multiple south Florida stakeholders from government, private industry, environmental, and citizen sectors have begun collaborating in an effort to return the Florida Everglades system to its ìnatural stateî. Research into the measurement and modeling of water movement and other hydrologic processes have therefore been identified as a primary science need in support of Everglades restoration. In order to accurately simulate surface water hydrology in South Florida, the variation in vegetation cover and the role vegetation plays in removal of surface water, resistance to surface water flow, and water quality, is necessary. The objective of this research is to develop and apply innovative remote sensing and geographic information system techniques to map the distribution of vegetation and related hydrologic variables such as evaporation through space and over time. This work will provide insights regarding the role south Florida vegetation plays in the redistribution of rainfall and surface flow inputs as well as the cycling of nutrients and other materials in the Everglades waters. It will contribute to our understanding of hydrology at large scales. Finally, it will lay the foundation for monitoring restoration impacts on Everglades flora. These benefits are vital in building the understanding required to properly monitor, simulate, and manage the unique Everglades wetland resource.
 

Project objectives and strategy:
The extrapolation of processes typically measured or modeled at point locations or at micro scales to macro scales is an extremely difficult undertaking. However, this capability is necessary in order to identify the important components of the natural system, quantify the impacts of human activity on the system, forecast system behavior, and monitor the effects of restoration actions. This work will develop the techniques and produce the data sets necessary to conduct hydrological modeling (surface water flow and water budget) at the regional scale. Additionally, success in developing periphyton mapping techniques would produce a critical new tool for biogeochemical and water quality research being conducted in task 4.3/4.5 and other programs within the USGS (for example the WRD Mercury initiative).

In situ and remotely sensed data from numerous platforms and sensors, each possessing different spatial, temporal, and spectral resolution, will be processed, analyzed, and combined to produce fields of information about biophysical variables (e.g., vegetation species composition, vegetation density, vegetation structure, and components of the surface energy balance) and image maps of south Florida. These data will be evaluated for their utility in specific modeling and monitoring contexts. Developed techniques will be transferred to other scientists and management agencies via technical reports, existing program public information outlets and other professional publications and meetings. The data sets developed will be documented using established metadata standards and made available through collaborative research and the results of task 8.2 (Ecosystem Database Development).  Image map products will be developed to convey the results of this work and to portray current vegetation conditions in south Florida to the broadest audience possible.

The overall task has been divided into four primary components with specific objectives and general classes of clients. Component one addresses the extrapolation of point evapotranspiration (ET) measurements and characterization of vegetationís role in ET. Component two is aimed at vegetation density mapping for resistance to flow modeling. Component three is a pilot study of periphyton mapping techniques using hyperspectral remote sensing capabilities. The fourth component is the development of a new, large scale image map of the SICS region. For each component, we will systematically develop, evaluate, and apply the techniques which yield the most appropriate, spatially distributed information possible on the vegetation, climate, and hydrologic variables of interest to south Florida project scientists. The study designs for each component include the combination of field campaigns (to provide information for data calibration and accuracy assessment) and extensive techniques development using state-of-the-art hardware, software, and physical process modeling. Collected data will be processed (e.g., to mitigate atmospheric effects and conduct data fusion to infer biophysical fields) to maximize their information content. Collaborative use of the resulting information products will allow us to evaluate their utility for process modeling and environmental monitoring while facilitating outreach and technology transfer.

Potential impacts and major products:
This work will provide insights regarding the role south Florida vegetation plays in the redistribution of rainfall and surface flow inputs as well as the cycling of nutrients and other materials in the Everglades waters. It will contribute to our understanding of hydrology at large scales. Finally, it will lay the foundation for monitoring restoration impacts on Everglades flora. These benefits are vital in building the understanding required to properly monitor, simulate, and manage the unique Everglades wetland resource. Developed techniques will be transferred to other scientists and management agencies via technical reports, existing program public information outlets and other professional publications and meetings. Select information produced through this effort (e.g., satellite image maps) will be reproduced in sufficient quantity to allow for wide distribution to both specialized users and the general public - providing for increased understanding of the region and the role of the Bureau in addressing science and management needs. Examples of major products are hardcopy maps (see below for details); digital database layers of vegetation and hydrologic variables; CD-ROMs of derived information; and knowledge regarding the scaling properties of various surface features and processes.
 

Collaborators, clients:
Ed German (WRD/USGS) is collaborating through the provision of his point measurements of ET/climate data and the provision of logistical support in ground spectra and image map validation efforts. He is also a client as the maps of ET produced through biophysical remote sensing provide him with insights regarding the adequacy of his sampling network and measurement techniques. Jones is also communicating informally with Dave Stannard (WRD/USGS) regarding ET field measurement techniques and micro-scale ET modeling.

Ray Schraffranek and Eric Swain (WRD/USGS) will make use of the fields of ET and vegetation density generated through this research in their development and operation of a hydrodynamic model for the SICS region.

Data collected through the efforts of Virginia Carter (WRD/USGS) will be used in developing and verifying models of vegetation type and density as well as biomass estimates made using remotely sensed imagery.

Dave Krabbenhoft (WRD/USGS) is a collaborator in the periphyton mapping effort - providing logistical support for field work and expertise in periphyton identification. He and his collaborators (e.g., Jim Hurley) are clients as well in that they will be using maps of periphyton type/distribution and plant biomass generated through this research to better understand periphytonís role in mercury methylation.

Ralph Dubayah (Department of Geography/University of Maryland) is a collaborator in the atmospheric modeling and biophysical remote sensing aspects of the research.
 

WORK PLAN
1999
Evapotranspiration modeling (Jones, German, Dubayah)
Vegetation Density Mapping (Lemeshewsky, Jones, Desmond)
Periphyton Mapping (Jones, Krabbenhoft)
SICS Area Image Mapping (Thomas, Lemeshewsky)

2000
Synthesis and Publication (Jones, Lemewshewsky, Desmond)

FY 1999 activities:
Component 1 - Evapotranspiration Modeling
Jones will communicate and collaborate closely with program scientists (German, Carter, Schraffranek, Swain, and others) to evaluate derived fields of evapotranspiration for suitability in hydrodynamic modeling. Estimates of ET developed using empirical, standard, and new techniques (developed as a result of FY 1998 research activities) will be compared against measured ET values not used in model development. Data sets and techniques developed in FY 1998 will also be used in the hydrodynamic modeling in order to determine their utility. Based on the results of these activities, continued research into ET extrapolation techniques or refinement of previously developed techniques may be necessary to provide improved fields of evapotranspiration. Also, German intends to continue testing the applicability of the Priestly-Taylor method of ET estimation for south Florida. As documented in the literature, this method has been modified to accept remotely-sensed inputs (Kanemasu 1978), although for use in environments quite different than south Florida. It is not clear whether the necessary leaf area index (LAI) variable can be accurately measured in the Everglades environment using remote sensing. Success in measuring LAI in the Everglades may allow for wideóspread use of the Priestly-Taylor method in south Florida and would also contribute to component 2. Development of LAI estimation techniques using remote sensing will therefore be an important research thrust in FY 1999.

Component 2 - Vegetation Mapping
Lemeshewsky will generate vegetation density maps for larger area of Everglades including FL Bay area and demonstrate blind de-mixing technique for finding vegetation mixture density from TM data. Jones and Desmond will investigate vegetation mapping techniques using mixture modeling and conventional statistical classification 7 techniques. Jones will collect ground truth data in support of components 2 and 3 for use in training and validation during field visits for spectra collection.

Component 3 - Periphyton Mapping Pilot
Although no ecosystem funding was provided for this effort in 1998, Jones leveraged 1998 project travel funds with support from NASA, the BRD, and logistical support from Krabbenhoft to acquire archive AVIRIS hyperspectral data and collect ground spectra of periphyton for future use in this effort. Building on the accomplishments of 1998 (see below), in FY 1999 Jones will evaluate methods used to account for heterogeneity of land cover within pixels (e.g, mixture modeling deconvolution through spectral derivative analysis) with the results of Hurleyís HPLC analysis and periphyton spectra collected in the field in FY 1998. Jones will then search for relationships between features in absorbance curves generated from hyperspectral data and information in Digital Multispectral Videography, thematic mapper, and SPOT data collected coincidentally through previous efforts of the land characteristics project. This will determine whether with-in class periphyton mapping and the extrapolation of periphyton characteristics mapped from high-resolution, limited availability imagery to larger areas covered by ìcommonî remote sensing systems are possible. If so, Jones will generate maps of periphyton cycles and distribution using imagery already acquired for the project.

Component 4 - SICS Image Mapping
Thomas and Lemeshewsky will apply state-of-the-art image restoration and mapping techniques to generate numerous image map products. In addition to providing an update of previous image maps, these hard copy map proucts will prove useful in the field (orientation and data compilation), will provide detailed information for interpretation in SICS area modeling, and should be very popular with the ìpublic at largeî - providing an effective outreach mechanism.
 

FY 1999 deliverables/products:
Various reporting mechanisms (e.g., open file reports, input to synthesis reports, and professional papers) and outreach products (e.g., documented digital data sets, image maps, software) will be generated for each component of the project. The following deliverables will be generated specific to each component:

Component 1 - Evapotranspiration Modeling
GIS coverages of ET for use in hydrodynamic modeling.
Limited edition hard copy prints of estimated ET fields.

Component 2 - Vegetation Mapping
Maps of vegetation type and density for a broader area (including Fl. Bay) than generated in 1998.
Reports on blind de-mixing techniques for mapping vegetation density from TM data.

Component 3 - Periphyton Mapping Pilot
The type and amount of products generated by this component will depend on the success of techniques developed. Preliminary maps of periphyton distribution may be produced from the limited, available hyperspectral data. If these products prove useful and methods for fusing other data products for broader (spatial and temporal) mapping of periphyton are developed, additional map, image, visualization, and report products will be generated.

Component 4 - Image Mapping
Five quadrangle image maps at 1:25,000 scale.
Three quadrangle image maps at 1:50,000 scale
One 1:100,000 scale quadrangle image map corresponding to the southern half of the Miami topographic quadrangle and the full Homestead topographic quadrangle.
Four CD-ROMS containing digital image map files for the area corresponding to the three 1:50,000 image maps
(with 5 meter resolution) and the 1:100,000 map (with 15m resolution).
 

FY 1999 outreach:
The research of FY 1999 will be characterized by extensive interaction and collaboration with the primary clients of this research - other project scientists within the program. This work will aid in the interpretation of much of the field data (e.g., ET and vegetation sampling data) which have been collected during the previous two years of the program. The ET modeling and vegetation mapping research are aimed directly at accomplishing spatially-distributed hydrodynamic modeling as precisely as necessary and accurately as possible. While interaction between other project scientists has occurred previously, it is our intention to make this interaction and the associated synthesis and refinement of our techniques and results the primary objective of this years efforts.

To ensure this communication occurs effectively, Jones will initiate periodic teleconferences and meetings with all collaborators. In addition, he will include other project scientists as second and additional authors on professional publications and professional meeting presentations - excellent milestones for success in collaboration.

Much of the work in this fiscal year will be driven by the requirements of the other project scientists, especially the hydrologic modelers, as much of the work in this fiscal year will be directed toward refinement of previously developed algorithms, information, and knowledge, given feedback and better understood requirements on the part of the hydrologic modelers.

Generated hard copy image maps will provide a valuable tool in communicating the complexity and beauty of the Everglades ecosystem to the general public.

New directions or major changes for FY 1999:
Component 3, periphyton mapping pilot and component 4, SICS image map detailed previously represent new directions for FY 1999.

ACCOMPLISHMENTS, OUTCOMES, PRODUCTS, OUTREACH
FY 1998 accomplishments and outcomes, including outreach:
ACSM presentations; IEEE presentation; development of techniques for ET extrapolation. Field work. Greater communications with hydrologists; development of rectified satellite imagery. Coregistration of different data sets. Implementation of software processing capabilities.

Publications (completed):
ìNeural-network based sharpening of Landsat thermal band imagesî by G. Lemeshewsky in SPIE ConfProc: Visual Information Processing VII, 1997.

Presentations (completed):
ìRemote sensing of evapotranspiration in the Evergladesî. John W. Jones. ACSM Annual Meeting - Baltimore. March 31, 1998.

ìNeural-network based classification for Everglades vegetation mappingî. G. Lemeshewsky. ACSM Annual Meeting - Baltimore. March 31, 1998.

Other: (completed)
Developed a tool for reducing speckle noise in SAR data in order to improve machine classification or visual interpretation (Lemeshewsky).

Developed algorithms for extrapolation of in-situ evapotranspiration measurements using statistical summaries of TM data (Jones).

Produced map of evapotranspiration with lOOm resolution for south Florida for the image date of 3/21/96 (Jones).

Tested and eliminated the possibility of transferring TM developed statistical techniques to AVHRR for improved temporal resolution (Jones).

Developed spectra for samples of cattail, sawgrass, periphyton, and open water through in situ measurements (Jones).

Developed co-registered, georeferenced data sets from TM, SPOT, AVHRR, STATSGO SOILS, and climate stations in GIS format (Jones).

Callibration of TM and AVHRR data sets to radiance, reflectance, and apparent surface temperature (Jones).

Development of procedures for statistical sampling and analysis of any georeferenced data set using a combination of GIS, image processing, and advanced statistical software (Jones).

(pending - FYí98)
Examples of vegetation density maps from the integration and classification of multi-spectral and SAR data
(Lemeshewsky).

Algorithms for albedo estimation from both TM and AVHRR (Jones).

Algorithms for evapotranspiration estimation using multiple sensor sources to improve temporal resolution of remotely sensed fields (Jones).

Detailed spectra collection from identified samples of periphyton (Jones).
  Initial processing (georeferencing and callibration) of NASA collected hyperspectral imagery for three study areas in south Florida (Jones).

FY 1998 deliverables, products completed:
Completed:
Map of ET at lOOm resolution for the March ë96 TM image data (Jones).

Pending:
Histogram-matched DOQs written to a special set of CD-ROMS (Jones).
ET coefficients for cells matching those used in hydrodynamic modeling (Jones).
Vegetation density maps (Lemeshewsky).

PROJECT SUPPORT REQUIREMENTS
Key project staff:
1999
John W. Jones, biophysical remote sensing and distributed hydrologic modeling.
George Lemeshewsky, image enhancement and classification.
Greg Desmond, remote sensing and wetlands mapping.
Jean-Claude Thomas, image mapping.

2000
John W. Jones, biophysical remote sensing and distributed hydrologic modeling.
George Lemeshewsky, image enhancement and classification.
Greg Desmond, remote sensing and wetlands mapping.

Other required expertise for which no individual has been identified:
1999
Publication editorial assistance/web publishing.

2000
Publication editorial assistance/web publishing.

Major equipment/facility needs (list by fiscal year for duration of project):
1999
Increased disk storage.
Spectrometer fibre optic upgrade.
 

Back to Project Homepage