CICEET Progress Report for the period 8/1/02 through 1/31/03

Project Title: Remotely Sensed Indices of Land Use Intensity for Watershed-level Monitoring
Principal Investigator(s): Richard G. Lathrop Jr., Robert A. Zampella

Accomplishments
Finish development and assess accuracy of remotely sensed indicators.

Conduct statistical analysis of land use intensity indicators vs. water quality parameters for a selected sample of subwatershed in the Mullica River basin study area.

Progress on Tasks
Our objective is to investigate the utility of satellite remotely sensing techniques to map indicators of urban land use intensity: impervious surface and the managed lawn. Using medium scale imagery (e.g. Landsat Thematic Mapper), individual pixels generally represent a mixture of urban land covers. Thus we are concentrating our efforts at developing un-mixing techniques that estimate subpixel proportions of the land covers of interest. Three different supervised classification methods are being evaluated: traditional maximum-likelihood hard classification, supervised fuzzy c-means (FCM) and Self-Organizing Map (SOM)-Learning Vector Quantization (LVQ) neural network.

Land Use Intensity Estimation: Results
The evaluation of the three different sub-pixel analysis methods is finished and the SOM_LVQ method is the most accurate sub-pixel analysis. Compared to IKONOS and the NJDEP LU/LC map, the SOM_LVQ method estimates the land cover proportion of urban landscape most accurately with little false estimation of pure pixels. In the study area of Hammonton and Egg Harbor City, LMM and FCM estimate false lawn proportion in the pure urban tree pixels, but SOM_LVQ show almost no false estimation in these pure pixels.

The accuracy assessment based on NJDEP LU/LC vector map is shown: the final map of impervious surface estimation, the scatter plot and the table of summarized mean impervious surface percentage for each urban land use. It is only partial accuracy assessment. It does not show the whole picture of the performance of three different methods. The validation with IKONOS or aerial photograph based vector map might provide quite different accuracy assessment. It is already noticed from the validation of aerial photograph based vector map (see Tables).

In the accuracy assessment of the NJDEP LU/LC map, LMM is the worst model and FCM and SOM_LVQ give quite similar accuracies. The zonal mean based accuracy assessment of the NJDEP LU/LC map mask out the subtle difference between FCM and SOM_LVQ in the urban landscape analysis.

In the scatter plot, FCM shows significant clustering pattern. LMM shows somewhat linear relationship of impervious surface and lawn estimation. The sigmoid-like relationship of the impervious surface estimation of SOM_LVQ is the most realistic scatter plot of Landsat and IKONOS. The scatter plots of all methods for urban tree estimation show logarithmic-shape relationship that is the result of the multiple scattering within tree canopy and underlying lawn cover.

In summary, it appears that medium scale resolution image data set provided by Landsat ETM can be used to provide excellent estimates of impervious surface with similar accuracy to conventional methods. Estimates of lawn surface area are more problematic but are possible with acceptable accuracy. Percent tree area is more difficult and requires additional work.

Water Quality Analysis: Progress
Using the results of the SOM model above, we have estimated land use intensity for 26 subwatersheds in the Mullica Rivers basin study area. R. Zampella, project co-investigator, is in the process of conducting the statistical analysis comparing the subwatershed land uses components vs. water quality data. We are also comparing our TM derived estimates with the NJDEP impervious surface estimates.

Difficulties Encountered
Our original schedule ( as laid out in the proposal) had been pushed back by over six months as an official University account was not established until 11/15/2000.

Anticipated Success in Meeting Project Objectives in Scheduled Project Period
We have requested and received a 12 month no-cost extension for the project due to delays in getting the project funding in place. We are now making steady progress and do not anticipate any difficulties in meeting the revised schedule. We are presently wrapping up the project. Sangbum Lee has finished writing his PhD thesis and defends it in late April.

Preliminary Data
Preliminary impervious surface cover maps for the developed areas of the Mullica River basin using Landsat Thematic imagery and based on the three different methodologies. Subwatershed estimates have also been produced.

Tasks and activities for next reporting period

Tasks for the next reporting period
Finish the impervious surface-lawn area vs. water quality analysis.

Prepare manuscripts for peer-reviewed publication. Prepare other publications/web material to communicate project results.

Present results to state level project cooperators for possible implementation.

Work plan to accomplish tasks
S. Lee, the Graduate Assistant assigned to this project, J. Bognar, Project Coordinator, R. Zampella, project Co-I, and R. Lathrop, the PI will continue to work on the above tasks.

Concerns or difficulties
None.

Expenditures
At this point, are in the range anticipated for the work accomplished to date.

List of Presentations on Project
A. Faiz Rahman and Richard Lathrop. 2002. Estimating Urban Land Use Intensity by Combined Use of Aerial Photography, GIS and Spectral Un-mixing of Satellite Imagery. American Association of Geographers Annual Conference, Los Angeles, CA, April, 2002.

Sangbum Lee and Richard Lathrop. 2002. Comparison of three different supervised classification methods in estimating land cover proportions of urban sprawl. American Association of Geographers Annual Conference, Los Angeles, CA, April, 2002.

Sangbum Lee and Richard Lathrop. 2002. Sub-pixel estimation of urban land cover intensity using fuzzy c-means clustering. American Society of Photogrammetry and Remote Sensing Annual Meeting, Washington, DC. April 2002.

Richard Lathrop and Sangbum Lee. 2003. Remotely Sensed Indices of Land Use Intensity for Watershed-level Monitoring. EPA Coastal Tech Transfer Conference, Cocoa Beach, Florida. January 2003.

Publications
Sangbum Lee. 2003. Remotely Sensed Indicators of Human Land Use Intensity. Unpublished PhD thesis. Rutgers University, New Brunswick, NJ. 284 p.