METADATA FILE

I.  General description of data set
-----------------------------------

Name
	Population Density for Baltic Sea drainage basin region

Geographic Level
	Regional

Specific Area
	Baltic Sea drainage basin region

Lat/Long Region Boundaries
	NW-X:           0-E     ,       SW-X:           0-E
	NW-Y:           75-N    ,       SW-Y:           45-N
	SE-X:           40-E    ,       NE-X:           40-E
	SE-Y:           45-N    ,       NE-Y:           75-N

Infoterra Keyword(s)
	Human population (16428)
	Patterns of urban growth (16160)
	Urban areas (16476)
	Rural areas (16462)

Other Keyword(s)
	Population density, Census data, Human growth

Primary Data Acquisition Period
	Start Date: 1989
	End Date: 1993

Supplier
	GRID-Arendal. PO Box 1602, Myrene. N-4801 Arendal, Norway.

Contributor
	GRID-Arendal. PO Box 1602, Myrene. N-4801 Arendal, Norway.

Publication Reference
	Langaas, S. 1993.  Some Georeferenced Data Sets for use within 
		Sub-project 1; Land Use and Ecological Carrying Capacity, 
		of the Baltic Drainage Basin Project.  Technical Note. 
		GRID-Arendal. 10 pages + apps.

	Sweitzer, J., Langaas, S., and Folke, C. 1996.  Land Cover and 
		Population Density in the Baltic Sea Drainage Basin: A GIS 
		database. AMBIO 25(3): 191-198.

Documentation (overview)
		The Population Density data set was derived by combining 
	urban and rural population statistics with the Administrative 
	Unit and Land Cover IDRISI data sets, generated for the Baltic Sea 
	drainage basin project.  Population statistics data was collected at
	the smallest administrative unit possible; down to the municipality
	level.  
		Population statistics were used in conjunction with the 
	Administrative Units, and rural and urban data sets. The urban and 
	rural data sets were generated from the Land Cover data set.  The Land
	Cover data set was reclassified into Urban Land and Rural Land.  
	Rural Land was created from Forest, Open Land, and Unknown Land.  The
	Urban Land class already exists in the Land Cover data set.
		Urban populations were distributed throughout urban area 
	with equal probability.  Urban cell values were calculated by 
	dividing urban population, for a particular administrative unit, by
	urban area.  Rural population was derived similarly to urban, but
	rural population was divided by the combined area of open land,
	unknown land, and forest land cover classes.  The combined Urban and
	Rural Population Density data set was aggregated to 5 km x 5 km cells.
		The IDRISI Population Density image was converted into
	ARC/INFO vector coverages, for PC ARC/INFO uses.  Due to ARC/INFO 
	limitations concerning number of arcs per polygon , the IDRISI 5 km
	resolution image was divided into four tiles.  For those ARC/INFO 
	users that are not affected by such limitations (i.e., ARC/INFO 7.0), 
	then we suggest to convert the POPDEN.GIS 5 km resolution ERDAS 
	file into a complete, continuous Population Density vector, polygon 
	coverage.
		County population statistics were provided for both the Czech
	Republic and Slovakia.  The data was not divided into urban and rural
	populations.  Due to the low percentages of rural population for both
	republics (8% and 6% respectively) at the national level, we decided 
	to assign the total county population to the urban grid cells.  When 
	viewing the Population Density data set, for either analysis or 
	graphic display, please make note of the above criteria.

	**** ATTENTION - POPDEN.GIS USERS ****

		The IDRISI Population Density image is in real value format.
	ERDAS files can not support real format.  The image was therefore
	converted into integer, because IDRISI integer values, supported
	by both IDRISI and ERDAS.  However IDRISI integer values can not
	exceed 32,768.  Some cell values were greater than 32,768.  The 
	problem was solved by reclassifying these cells into unique values 
	less than 32,768.  When users convert the POPDEN.GIS ERDAS file into 
	a polygon coverage they should RESELECT the following GRID_CODE 
	values and CALC them to the correct values in TABLES.

		      GRID_CODE          CORRECTED VALUE
			-9999     --->            0 (neg poly)
			30000     --->        35243
			30100     --->        40176
			30200     --->        80138
			30300     --->        93459

		All other values can be left unaltered.  The ARC/INFO export
	files have already been corrected.  Another alternative to 
	creating a continuous Population Density polygon coverage is to
	simply IMPORT the four export files and use MAPJOIN.
	
************************************************************************
Contact information
	Hugo Ahlenius                           phone:  +47-370-35650 
	UNEP/GRID-Arendal                       fax:    +47-370-35050 
	                                        e-mail: mapmaster@grida.no
	
   
II.  Detailed description of data set (members)
-----------------------------------------------

Data set type
	IDRISI 4.1 image file

Description
	POPDEN.IMG - Population Density, whole Baltic region

File Type/Size
	Real/binary: 832 KB

Access
	Free access, unrestricted

Copies (available sites)
	CD-ROM - can be prepared upon request by GRID-Arendal
	World Wide Web (WWW), GRID-Arendal
		URL http://www.grida.no/


III.  Technical information about raster data:
---------------------------------------------

Cell Size
	Height: 5,000 meters
	Width:  5,000 meters

Number of Rows
	520

Number of Columns
	400

Projection Name
	Lambert Azimuthal Equal Area

Projection Parameters
	Units: meters
	Spheroid: International 1909
	Radius: 6,378,388
	Longitude of centre of projection: 20.00.00
	Latitude of centre of projection:  60.00.00
	False Easting: 0
	False Northing: 0

Projection boundaries
	Xmin: -1,000,000        Ymin: -1,300,000 
	Xmax:  1,000,000        Ymax:  1,300,000