566
permit simple overlaying in grid and polygon format,
which result in composite maps and dissolve
functions which perform the opposite operation by
creating individual maps out of complex multi
attribute maps. Comparison functions also include
more involved statistical calculation of spatial
data relation-ships. Operation include multivariate
analysis, correlation analysis, time series analysis
and canonical correlation analysis.
7.3 Data Output and Report Generation
Before any decision making process can be initiated,
geographical information must be presented to the
decision maker in a clear, concise and comprehensive
format. A GIS must generate graphics and text to
the exact specifications of the user. GIS output
can take the form of images, charts, graphs, maps,
facsimile data, or it can be generated as reports
and tabular data (fig. 8).
8 RECOMMENDED GIS FOR NOAA's CLIMATIC IMPACT
ASSESSMENT
The AISC GIS configuration must consist of an
amalgamation of computer hardware, software, and
peripheral entry and display instruments along with
the information that is to be analyzed. All systems
must function in unison to achieve the specified
goals of AISC in natural disaster monitoring, impact
assessment, and forecasting.
8.1 Delineating Data Variables
The following listing of data variables identifies
inputs that must be integrated into a GIS for
environmental monitoring, environmental and socio
economic impact assessment, and environmental and
socio-economic modeling, forecasting and advanced
disaster warning/preparation.
Weather Station Data
precipitation
temperature
cloud cover (satellite)
precipitation estimates (satellite)
Environmental Observations in Cartographic Form
soils (map)
topography (map)
hydrography (map)
natural vegetation cover (map/satellite)
vegetation response (AVHRR satellite)
Cultural Observations
political units (towns, borders, etc.) (map)
transportation (map)
population density/distribution/character (map)
land cover/use (map/AVHRR satellite)
agriculture
crop suitability index
existing crop regions
production/yield observation
inputs (fertilizer, irrigation, mechanization,
labor)
crop calendars
grazing
livestock density
migration routes
water holes
basic need support
markets
hospitals
good storage depots
refugee camps
8.2 Software Needs
An automated GIS must incorporate a variety of
software capabilities for data encoding, computer
network communication, data entry, data base
management, data analysis and data output. A
complete GIS configuration must include software
that is capable of encoding both spatial and non-
spatial data, along with image data. A GIS must be
capable of processing data through the use of
spatial/statistical analytical procedures for
modeling. Additionally, it must allow for data
storage in an easily accessible and retrievable data
management system. Such a GIS must include the
following integrated software subsystems: 1)
Geographic Subsystem software for data entry of
analog to digitally converted data and data file
creation; 2) Preprocessing software, including image
processing and encoding software to format system
compatible files in vector and/or raster data
structures for analysis; 3) computer networking
communication software for data processing and
access on other network nodes; 4) Data base
management subsystems to facilitate data file access
(editing/update), retrieval (search/extraction/
formatting) and storage/cataloging; 5) Data analysis
software subsystem with spatial/statistical analysis
capability for scenario modeling; 6) Output software
subsystem for peripheral devices, e.g., photocopy,
line plotter, CRT, floppy disk and computer tape.
Particular to AISC needs is that a GIS must be
capabbe of assessing the impact of an event on an
afflicted area in terms of the physical, social and
economic indicators (i.e. vegetation response (NVI),
crop production/yield, food prices, land use
changes). The functions required in this process
would include: combining data sets (overlaying) for
visual interpretation, statistical analysis in
determining variable relationships (i.e. impact),
and various forms of correlation analysis for
determining data redundancy in modeling.
GIS must finally be capable of producing modeling
scenarios for forecasting and determining impact
abating strategies. Examples would include,
specifying in advance areas of impact and intensity
thereof, forecasting crop production/yield,
estimating food shortages, determining disaster
assistant needs, improving policy and food security
management decisions, relocating existing stocks of
food or bumper crops in neighboring regions and
recommending alternate crops capable of growing in
expected weather conditions (Steyaert, 1984).
A complete GIS configuration must
incorporate a variety of software capabilities to
display facsimile data and digital images,
outputting information to a CRT for animation and
image graphics. Hardcopy output should include
graphic representation of maps in grid cell and
polygon format such as contour, choropleth,
isopleth, and dot maps, as well as other graphic
output including pie charts, bar and line graphs.
A GIS configuration must incorporate
software capable of providing computer network
linkage
Further,
with d
Communic
alternat
through
comprehei
must im
and must
Very few
analysis
function!
Those id<
1) Lam
requiremi
complete
and idem
with rem<
base man«
to run oi
2) Sysl
is basic
However,
processii
8600/VMS
$20,000
expanded
this heac
INFO, Ct
However,
set forth
software
been ider
1) ERDi>
2) Chan
8.3 Hardr*
In order
specified
data mus
environme
analysis
maximize
appropria
minimum s
data is s
Therefore
scenario
required
Using
standard
Kilobytes
data se
additions
logical i
operation,
micro com;
were emp
represent!
RAM stor;
These two
large RAM
use of an;
super-mici
II would 1
Further ^
five add:
storage rc
Natural
would necc
sets. Or
variable,
storage wd
manipulatd
variables,
file of