In the Southeastern Appalachians and Olympic Peninsula, efforts are being directed toward managing the landscape. Such efforts clearly benefit from advances in remote sensing technology and GIS. While improvements in remotely-sensed data have increased our ability to interpret changes in land cover, geographic information systems such as GRASS have simplified the integration of spatial information across disciplines. LUCAS can bring these recent developments into practice by providing a flexible and interactive computing environment for landscape management studies. Future LUCAS-based impact studies for the Little Tennessee River basin and Olympic Peninsula, for example, will address changes in timber output, water quality, and erosion.
Because each sample scenario from Section 5 required
slightly more than three quarters of an hour of elapsed wall-clock
time for 15 replicates (see Table 6) on a 70 MHz
SPARCstation 5 Model 70 with 32 Mb of memory, another version of LUCAS
has been implemented using PVM [3] on a heterogeneous network
of high-performance workstations [6]. The relative speedup
speedup
of the
the distributed version (including start up) over
the serial version tested
was 
. The use of
parallel algorithms [1] for computing map statistics on
machines such as the Thinking Machines CM-5 could also be exploited.
Table 6: Wall-clock execution times of LUCAS on a SPARCstation 5.
Additional information can be obtained by e-mailing to
lucas-info@cs.utk.edu which will automatically reply with a brief
message. For more complete information, the LUCAS World Wide Web page
at URL http://web.eecs.utk.edu/~lucas is available. Images of
the GUI, sample simulations and literature regarding LUCAS can all be
found at this site.