457
Computer-calculated lake areas within the specified quad
rangle provide a good example for use of the comparator.
Tables 1 and 2 are retrieval listings that were limited to
only those lakes with surface areas >2.0 km 2 within the
quadrangle. Four lakes had areas >2.0 km 2 . The SLAR lakes
were the largest 3 lakes in the area.
Although usable in its present form, this computer lake
system is not complete and can be refined in the future to
provide more utility with greater simplicity. Three areas
need additional attention. First, a method must be devised
to merge master files from multiple overlapping Landsat
scenes to enable extension of geographic coverage without
duplication of lakes in the overlapping areas. Secondly, an
algorithm should be developed to ease the updating lake data
entry so that specific lakes on the lake file can be ac
cessed and verified with a minimum of centroid position ac
curacy, a maximum of data entry ease, and a minimum po
tential for lake identity error. Finally, spectral thresh
olds used to discriminate between water and land must be
investigated further, and verification of computer-generated
product accuracy, utility, and cost must be completed.
SUMMARY
A computer lake system was developed that uses computer
compatible Landsat data to create a lake information file
uniquely characterizing each lake in the file. The com
puter-calculated latitude and longitude of the centroid for
each lake provide a lake file with a geographic data base
from which lakes can be retrieved for a defined area.
Computer calculations of surface area, perimeter, and
crenulation further characterize each lake. The lake file
provides storage for additional information obtained about a
lake from other sources. Retrieval of lakes on file can be
accomplished through sorting and sieving (filtering) func
tions defined by parameters of interest and values limited
by comparators ( >, =, and <). When a geographic area
containing lakes of interest is specified, the user-defined
area is calculated and printed at the end of the listing.
The sum of all lake surface areas in the listing is also
printed for comparison with the geographic area in which
they exist.
This system has several potential uses. It could be used to
consolidate aquatic data for dynamic retrieval. This can
supplement or replace conventional means of lake data
management and analysis. The system provides worldwide
capability for rapid survey of lakes in relatively flat
terrain and for monitoring their changes in time. Finally,
it provides the capability for lake classification through
sorting and filtering functions applied to the various lake
parameters on file.