integration. 
Decision Making 
The decision making is frequently 
presented with maps or statistic 
products, derived from remote sensing 
and GIS. In most situations, 
appropriate information concerned to the 
lineage of the thematic data layers and 
accuracy associated to thematic and 
geometry are not given. Besides this, 
the decision making needs the total 
precision estimate and the reliability 
of the products used in the process, 
although decision making is provided 
with little or no acquaintance about the 
potential error sources and no 
information concerning to final accuracy 
and reliability level. 
There is a defined tendency between 
many decision makers of always accept 
products, maps and statistic reports as 
truthful. Bor. the fact. of several 
remote sensing and GIS products being 
thematic products, there is an enormous 
error potential in a decision making, 
for overestimating the accuracy and the 
thematic reliability level of the 
products. It is necessary that the 
remote sensing and GIS communities 
instruct the administrators to better 
understand the error sources associated 
to the products. 
Implementation 
Decisions based on innaccurate data 
and improper reliability level have a 
great . probability of making incorrect 
decisions. The obvious implications of 
an incorrect decision are wrong actions 
of resource administration, which can 
mean serious consequences for the 
resource itself, Causing its loss or 
degradation, adverse impacts on a 
particular ecosystem or its elements, 
impacts on the human health, besides the 
possibility of fines or other punitive 
actions. 
AS products derived from remote 
sensing and GIS are raisingly being used 
as decision basis for resources 
management and as problems regulator, 
there is a high potential for an 
explosion of the number of litigations. 
The biggest challenge for the remote 
sensing and GIS community will be the 
hability of properly portray and defend 
the accuracy and reliability of products 
used by administrators in the process 
implementation. 
CONCLUSION 
It is necessary a considerable amount 
of research and development to be made 
before the errors associated to remote 
sensing and GIS integration can be 
properly quantified and expressed in 
standard formats. 
The objective of standard error 
reports is to provide an evaluation 
method of the adjustment of GIS products 
derived from remote sensing for specific 
applications and to facilitate the 
comparison between several research 
354 
results. 
The present procedures for remote 
sensing errors evaluation were adapted 
from statistic procedures which were not 
Specifically developed for space data. 
These techniques have been adapted and 
have been reasonably good for small 
areas, but their application for 
regional or global grades is not 
economically viable. For the fact : that 
the existent techniques refer to the 
global accuracy, the error Space 
distribution is not evaluated. There 
must be developed techniques to evaluate 
the error space structure. 
The philosophy and the 
recommendations for the acquisition of 
good field data to evaluate the maps 
accuracy has not been well addressed. 
Basic researchs must be developed on the 
accuracy levels associated with 
different ways of field verifying. 
The classified satellite digital 
images (representation  pixel-by-pixel) 
are easily filled in the matrix. format 
but hardly converted into the vectorial 
format. 
Numerous rules have been done in 
order to control the process of matrix 
data conversion into vectorial, but the 
effects on surface, size and accuracy of 
the polygons, when compared to the 
original matrix data, haven't been 
studied with severity vet. ix is 
critical to research the effects of the 
conversion of remote sensing digital 
data. 
Additional information is required on 
the characteristics of the location 
errors in remote sensing and the 
correlation between locations and 
classification errors. It is necessary 
more knowledge on the definition of 
alternatives for remote sensing 
Platforms and on how much the GPS 
technology will contribute for the 
accuracy of remote sensing data 
location. 
The final maps and statistical 
products must be standardized to provide 
information related to the accuracy and 
reliability associated to the specific 
data product. 
REFERENCES 
Aronoff, 3:3 1985. The minimum 
accuracy value as an index of  class- 
ification accuracy. Photogrammetric 
Engineering & Remote Sensing  51(1):99- 
111. 
Blakemore, M., 1984. Generalization 
and error in spatial databases. 
Cartographica 21:131-139. 
Croswell, P.L., 1987. Map accuracy: 
what is, who needs it, and how much is 
enough? Papers from the Annual 
Conference of the Urban and Regional 
Information Systems Association Vol. 
II., Fort Lauderdale, FL, pp. 48-62. 
Dicks, S.E., and Lo, 3T.H.C., 1990. 
Evaluation of thematic map accuracy in a 
land-use and land-cover mapping program. 
Photogrammetric Engineering & Remote 
Sensing 56(9):1247-1252.