Process_Step:

Process_Description:

These process steps describe, in general, the procedures used to compile the 1992 urban development dataset. The production procedures, hardware, and software used in the collection of urban extent vary depending on the available source materials. The 1992 data was generated by various image processing procedures that combined the spectral information from Landsat data with the dense systematic street pattern derived from the DLG road network, and with the GIRAS land use and land cover data.

Source_Used_Citation_Abbreviation: TM1

Process_Date: 19950601

Process_Step:

Process_Description:

1991 Landsat Thematic Mapper scene to a 30-meter UTM baseline grid. Perform an image to map registration. Locate image and map control point pairs. Compute a 2nd-order transformation matrix. Compute RMS errors and edit control points if necessary. Resample the TM data to the 30-meter UTM grid using a nearest neighbor resampling technique.

Source_Used_Citation_Abbreviation: TM2

Process_Date: 19950601

Process_Step:

Process_Description:

1992 Landsat Thematic Mapper scene to the 30-meter UTM baseline grid. Perform an image to image registration. Locate primary and secondary image control point pairs. Compute a 2nd-order transformation matrix. Compute RMS errors and edit control points if necessary. Resample the TM data to the rectified image base using a nearest neighbor resampling.

Source_Used_Citation_Abbreviation: TM1

Process_Date: 19950601

Process_Step:

Process_Description:

A multispectral classification of the registered 1991 TM data. Compute unsupervised training statistics using a euclidean distance clustering algorithm. Request 30 categories. Use the resulting statistics file to classify the TM data u sing a maximum likelihood classifier. Label as built-up areas all categories whose spectral signature is commonly interpreted as concrete, asphalt, buildings, roads, residential neighborhoods, and commercial buildings.

Source_Used_Citation_Abbreviation: TM2

Process_Date: 19950701

Process_Step:

Process_Description:

a binary composite of the 1991 and 1973 classifications to reduce misclassified pixels. Label new categories in refined classification.

Source_Used_Citation_Abbreviation: TM1, TM2

Process_Date: 19950801

Process_Step:

Process_Description:

Generate a raster image of the road network and create a proximity buffer around all roads. Remove the proximity buffer where the road density was not sufficient to merge adjacent road boundaries. Label remaining areas as developed land.

Source_Used_Citation_Abbreviation: DLG1

Process_Date: 19950801

Process_Step:

Process_Description:

Create a raster image of the USGS land use and land cover data for the study area. Import GIRAS data files into GIS. Append file covering study area together. Convert vector polygon data into raster image.

Source_Used_Citation_Abbreviation: GIRAS1

Process_Date: 19950801

Process_Step:

Process_Description:

Create a map of urban extent by combining the 1991 and 1992 Landsat classification, the GIRAS LULC delineation, and the DLG urban extent data. Combine data using binary addition. Interpret the results and assign values to the appropriate class.

Source_Used_Citation_Abbreviation: TM1, TM2, DLG1, GIRAS1

Process_Date: 19950801