The USGS Land Cover Institute (LCI)
METADATA FOR THE 1910 TIME PERIOD
Maps with a scale of 1:62,500 were primarily used as the source for the 1910 Urban Data layer. Maps with scales of 1:125,000 and 1:31,680 were also used when available. The following maps were used:
A polygon was drawn to represent any cluster of urban development apparent on the map. Urban clusters usually consist of at least three buildings. The density of urban development represented by the polygons varies and is not qualified by the database. Streets were used as polygon boundaries or guidelines when delineating polygons, but were not a primary consideration in selecting where to draw the polygon lines.
Cemeteries, parks and open space enclosed by urban development were delineated in green ink to identify them as urban other (lulc 17). All other development is classified as mixed urban (lulc 16). An attempt was made to represent waterfront development, areas of industrial activity, shipyards, railyards and other areas containing large and non-residential buildings in separate polygons from standard urban polygons, however a distinction in the database was not made between these polygons: they all reflect the lulc 16 code of mixed urban.
Areas of sparse urban development that carried a place name on the map were included in the data layer by polygon representation even if they did not contain a cluster of three or more buildings. However, place names on the map that had no corresponding buildings (at least one) were not included since the location of the "town" could not be determined accurately. Areas of urban development that contained at least three buildings clustered together were included regardless of the presence of an associated place name.
The mylar sheets containing the urban data were then digitized in ArcEdit as individual coverages. The coverages were named according to the first four letters of the quad name followed by the last two numbers of the year of publication (e.g. Portland 1905 becomes port05).
Once all the mylar sheets were digitized, database development progressed with adding items to qualify the polygons. Two items were added to each coverage, "lulc" to contain the Anderson Classification System Land Use/Land Cover Code, and "city" to contain the place name. Every polygon contains a lulc value of either 16 or 17. However, not every polygon contains a value for "city." Only those polygons clearly represented as part of the place because of proximity to the place name or main city polygon, contain that value in the "city" category. Therefore, some place names contain more than one polygon.
Once digitizing was completed, each coverage was edited to remove dangles and correct any errors or omissions that may have occured during the digitizing process. Topology was then built using the "build" command in Arc.
Once all of the individual quads were digitized and developed into complete, attributed coverages, an AML was generated using a text file containing the name of the coverage and the control point coordinates for the quad. Running the AML produced a projected, appended coverage containing all of the urban data for 1910.
Compiling the Transportation Data Layer for the 1910 Time Period The transportation data layer represents the major transportation routes used during the time period. Every route in use and existence during this time is therefore not included in the database. The idea is to represent the major routes in order to better understand how transportation has both created and been a creation of, urban growth.
The 1910 transportation data layer is not complete, however the methods outlined below will continue to be used in developing the layer.
Major transportation routes are determined by their existence on large- and small-scale maps, especially maps created for the purpose of showing transportation routes, for the time period. Other interpretive factors such as the route's connectivity between developed towns and cities, connectivity to other states or historical significance as a main thoroughfare (e.g. Oregon Trail) play important roles, too. Books and anecdotal information are also used to determine the importance and existence of transportation routes, as well as the dates of construction and operation.
The following small-scale maps were used to develop the 1910 transportation data layer:
USGS. "State of Washington." 1:500,000. 1914. Oregon State Highway Commission. "State of Oregon: Main Traveled Roads." 1921. (Printed in Preston, Ralph N. Early Oregon Atlas. Binford & Mort: 1978.) Rand-McNally. "The Rand-McNally New Commercial Atlas Map of Oregon." 1913. (Printed in Preston, Ralph N. Early Oregon Atlas. Binford & Mort: 1978.) Cram, Geo. F. "Map of Oregon." 1900. Chicago, Illinois.
Transportation routes, once identified, were not digitized from maps, rather a 1:250,000 Digital Line Graph (DLG) from the 1980s was modified to represent the time period. Attributes were added to the DLG to represent each time period (e.g. yr1910 for the 1910 time period). Then a value was calculated for that field if the route was determined to be primary at that time.
There are several places in the DLG where the rail line splits into two adjacent lines for a very short distance (likey for train switching). Only one of the lines was included in the rails coverage rather than both for purposes of simplicity and continuity. Rail yards were left out of the transportation coverage but are included in the urban coverage.
A spatial coverage showing metadata information is created to supplement the data layers (urban, transportation, etc.). A text file is used to create two AMLs which generates a polygon coverage showing the quad boundaries and then adds the following attributes to each polygon:
Publication Date Survey Date(s) Scale Publisher |
U.S. Department of the Interior |
U.S. Geological Survey
URL: http://landcover.usgs.gov
Page Contact Information: LCI Administrator
Page Last Modified: May 2010
