J.

Eutrophication walier

management

framework for the policy-maker and Sven-Olof (1989). Ryding. use: a new research

(1989). al

Rast, Marjorie

Holland

2. Human investment
Editors: 3. Michael
10

and resource

orientation

the environment/economics Contributing some researcli Malcolm 4. sustained

interface resource

Young and Natarajan and insights

lshwaran. tropics:
(1989). in landscape management (1989). Fournier. approaches l. Dyer. and
10

use in the humid and sub-humid

approaches

Hadley

and Kathrin a proposal

Schreckenberg. research

The role of land/inland and restoration: Editors:

water ecotones

ADVANCED DATA A(QUISITION AND ANALYSIS TECHNOLOGIES FO R SUSTAINABLE DEVELOPMENT

for collaborative
Henri Dcamps of human-impacted Malcolm

Robert 1. Naiman, and restoration

and Frdric

5.

Management Editors:

resources: and Melvin

ecosystem rehabilitation
Kathrin (1990). nutrient (1991). 6. Debt-for-nature potential 7. Carbon, subject proposals Jonathan 8. Economic (1991). Editors: 9. 10. Kathrin exchanges

(1990). Hadley reserves: and biosphere experiences

Schreckenberg,

10/111

E. Es/es Malingreau Raper

Peter Dogs and Bernd von Droste. and water balances management oftropical implications rain forest ecosystems to disturbance: and research

Manfred Ehlers Jean-Paul lonathan

Iati R. Noble
Albert Se liman

M. Anderson and Thomas Spencer. and ecological sustainability oftropical rain forest management

leffrey

L. Star

Jim Weber Schreckenberg and Malcolm Hadley. researcn proposals (1991). scientific issues and collaborative

Biodiversity: 'Les systmes Richard

Otto T. Solbrig. agro-sylvo-pastoraux (1991). Meuret, in tropical [orest areas: plots ( 1992). of permanent diversity Bernard mediterranens : enjeux et rflexions pour une gestion raisonne loffre, 11. Long-term Editor: monitoring Francisco

Huber el Michel of biological and inventory

methods for establishment

Dallmeier.

The designations ernployed and the presentation do not imply the expression concerning of any opinion
the legal status of any country,

of material throughout this publieation

whatsoe ver on the pan city

of UNESCO or

PREFACE

territory,

or area of its authorities.

concerning the delirnitation of its frontiers or boundaries. The opinions expressed in this digest arethose ernployers. of the authors and not neeessarily those of UNESCO or the authors '

John E. Estes Departrneru of Geography University of California, Santa

Barbara,

Jonathan Raper Department of Geography, Birkbeek College, 7-15 Gresse SI. London WIP IPA, UK Albert Sellrnan Director of International Serviees, CIESIN, P.O. Box 134001, Ann Arbor, MI 48113-4001, USA Jeffrey L. Star Departrnent of Geography, University of California, Santa Barbara CA 93106-4060, USA Jim Weber Publie Service Satellite Consortiurn, 1235 Jefferson Davis Highway, Suite 904, Arlington, VA 22202, USA. The MAB Digest publications Series was launched distillations ongoing by UNESCO of the substantive activities in 1989. Severa! findings types of are included: of recent, of MAB activThe tar-

CA 93106-4060, USA

Manfred Ehlers ISPA, Univeristy of Osnabrck Veehta, P.O. Box 1553 W-2848 Vechta, Germany Jean-Paul Malingreau Joint Researeh Centre, Institute for Rernote Sensing Applications of the Commission of European Cornrnunities, Ispra, !taly Ian R. Noble Researeh Sehool of Biologieal Scienees, Australian National University, P.O. Box 4, Canberra, ACT 2600, Australia

About this series ...

ities; overviews
get audience

and planned

within MAB in particactivities. with planners and research in MAB.

ular subject or problem and policy-makers workers

areas; and proposals

for new research

varies from one digest to another. as the main audience

Some are designed personnel

in mind. Others are aimed at collaboraor not they are involved

tors in the MAB Programme.

Still others have technica! of whether

as the target, irrespective

Addresses of the auihors of this reporte Overall direction: Moharned Skouri Series edi/or:Malcorn Hadley Computer assisted layout: Ivette Fabbri Cover designe Jean-Francis Cheriez lllustrntion
011

... and MAB Digest 12

The purpose of this digest is to provide deveJopment modeling technologies, scientific researchers engaged geographical audience in activiinfor-

back cover: NASAIJPL Spaee Shuule Imaging Radar A (SIRA) of fault

zones in western China Suggested citation: Estes. l.E., Ehlers. M .. Malingreau, l.P., Noble, I.R., Raper, l.,
Sellman,

ties reJated to sustainabJe

tion and information mation experts digest, systerns

with an overview

of several

data collecis not the

including

remote sensing,

and advanced

tools. The primary

A., Star. l.L Weber, l. 1992. Advanced Dota Acquisition Sustainoble Development. MAB Digest 12.

and Analysis

Technologirsjor UNESCO, Paris.

Published in 1992 by the United Nations

Educational, Scicmific and Cultural Organiz.uion

and pracritioners of the technologies and approaches described in the but rather members of the broader scientific comrnunity who might wish lO have at hand a synoptic overview of the characteristics and potential of new technologies and their possible applicarion 10 contemporary research challenges. Particular ernphasis is given to rernote sensing and geographic inforrnation systerns. The digest for Sustainable Problems has be en prepared Development by (In ad hoc group on Scientific with support Informntion on body, the sct up in 1989 by the Scierui fic Commiuee eSCOPE). from its parent

7, place de Fontcnoy. 75352 Paris 07 SI' Printed by UNESCO

UNESCO Novcmbcr 1992

Printed in
Francc

01' the Environmenl

International Development UNESCO. Manfred

Council

of Scientfic (UNDP),

Unions

(ICSU),

and from the United specialists: Raper,

Nations and Star for for to

CONTENTS

Programme Jean-Paul An interim

the French

Ministry of seven

of the Environrnent Jeffrey

The ad hoc group Ehlers,

was made-up

John Estes,

Malingreau,

Ian Noble,

Jonathan

and Jim Weber. participants Environmenl

report of the group Conference

was prepared on an Agenda

as background of Science

al the International and Development in November Assembly in its present of applying

into the 21st Century 1991. This report form, the report technologies of SCOPE modern

(ASCEND-21)

organized 1992.

by ICSU in Vienna the VIIIth General Prior mented to publication by examples

was al so made available in January was subsequently to specific

held in Seville (Spain)

comple-

issues and geo-

ecological and

situations.
compiled half by the SCOPE of 1992 group has al so been adapted Environment and in the first Brief as UNESCO

The basic information

pub lished

Development prepared remote

3 entitled New Technologi es: Remate Sensing Geographic Information Systems. This 16-page brief for decision-makers
in cooperation sensing with members of an internal systems, UNESCO and geographic harmonize information led by Robert in remote Missotten sensing

an d
was

task force on of and and volea-

Summary Introduction
Background Purpose of document What follows

11
11
12

the Division

of Earth Sciences. The principal in UNESCO's programmes of geological creation

aim of this task force is to promote advances in such fields as identification origin (landslides, hazard of a database mudflows,

and where appropriate GIS technologies monitoring noes, active faults,

the use ofnew

13

of natural hazards

etc.), risk managent,

management, reserve

mountain

mapping, surface and

on spectral signatures and cJimate

Data
Data vs. information Spatial vs. non-spatal data and information

15
15
16

grour.d water research observation, biosphere

of rocks and soils, ocean observation cal sites, use of cornputer-based data in marine applications,

monitoring modules training

and investigations

planning learning

and management,

study of archaeologi-

Spatial dolo
Remote sensing . Other spatial data

17
17
29

on the use of remote and capacity building.

sensing

and education,

Te(hnologies
Geographic Modeling Expert Systems Information Systems Image processing

35 36
39 42 45

Appli(Olions and examples

Land use planning Estimating Biodiversity/gap Risk assessment in Senegal analyses in Ecuador in the United States crop yields in Italy

47 48 49
52 54

Issues, needs and trends

Scales Infrastructure

57 57 58

Accuracy Standards TimeJiness TechnoJogy trends RoJe of scientists

6J 61 63 63

SUMMARY

64 67

Source material

Throughout human history, technology has been a key factor in facilitating and driving change. Today's technologies can create environtnental change on spatial and temporal scales never before possible. Yet, the wise application oftechnologies can also facilitate investigations and lead to a more complete understanding of human. impact on the environment. Througli appropriate application of remote sensing, geographic information systems and modeling, we can move a significant way towards an~wering questions concerning the spatial and temporal dimensions of variations in environmental resources. These technologies for data collection, manipulation, analysis and information extraction can be used by the scientific community 10: improve our understanding of the global environment; measure, map, monitor and model changes in that environment; and provide decision-makers with the information which they require 10 summarize the options available to decision-rnaking bodies. Remate sensing is used to gather information from a distonce, with sensors recording electro-magnetic energy emitted 01' reflected from the Earth' s surface. Different types of vegetation, soils and other features emit and reftect energy differently. This characteristic makes ir possible 10 measure, map and monitor these [eatures using remote sensing svstenis. Sotellite sensors collect the bulk of remate sensing data C/I the present lime. Most o.f these data are ovailoble in [WO formats: electronic ami analog (i.e. image). Electronic data I11U.I'I generally be enhanced andfiltered by a computer and compored with limited 'ground truth ' data before thev can be used for resource munagctnent. Geographic lnjormation Systems can be defined os comput er systems [or integrating and analysing spatiat injormotion in a decision-making context. Recent advances in GIS technologv include: nel" hardware and software [or digilizing and matiaging dora, extracting information [rom coniplex databuses and reproducing 11I0pS. making these tools more powerful mili easier 10 use;

desk-top producing graphics Models natural models changes

work stations. inexpensive maps, charts

with increased

data-processing

speed

and data storage the quality o] and improved A I/UI1lof of

capabilities;

plotters, lowering and sables; easier

the COSI and improving 10 use GIS software;

INTRODUCTION

and visualiiation 10015.

can take manv [orms and are built Jor a variety have restricted the application of models bUI these problems more clear/y land suitability are being overcome. (e.g. prediction growth, oJ purposes. in the management The tasks required

be;' of factors

resources,

are being stoted in vegeta/ion,

of crop yields, structural models and species sysoJ an

and a suite of successful generic crop /rajectories based systems the problem-solving products of the revival

is being developed

in such fields (sometimes programs

as fores/

ranges.
Expert expert, systems called knowledge able 10 simulate 01' advisory ability tem) are computer intelligence

and are among research.

the 1II0s1 prominent

Al/

in artificial

system can be thought of as a model composed of rules and these rules are often qualitative (e.g. if the amount oJ monoculture cropping in the district is higher than the normfor the region tnen.i.}. This is a majar area of application in remote sensing where rule based models may assist
expert in the initial classification tems is in assisting tools, provided of images. The other majar application of the very powerful, of expert sysbut complex manageby examusers harness the forces GIS. used in combination, crop yield richness as illustrated

As scientists

we strive

constantly

to keep abreast

of new developments increasingly of basic of output increase,

in our informadata anel it

areas of specialization.

This is not an easy task in today's fields

tion-oriented
research increasingly modeling,

society. As the rate and volume

in our own and cognate to follow developments resources. of our global difficult

results

many of us find to irnprove

in a comprehensive being

in those fields seeking Advances

These new information ment, and are increasingly pies on land use planning agricultural production analysis for preservation nologies in physical technological science include

technologies have many uses in resource

in Senegal, determination of natural

our understanding

in climate and atmospheric and our understanding These of the are soin those fields in the area to facilitate at

improvements

in hydrologic Equally

forecasting significant

hazards risk to

carbon cycle are significant. which facilitate of information

are developrnents of dataJinformation. offer the potential and policy

in Ecuador, of species

estima/ion

in Italy, and habitar tech-

the acquisition technologies.

and analysis

in the United Sta/es. of these new informa/ion analysis and policy-making, accuracy standardizaof the role o]

called "hi-tech" areas. In recent the conduct of research, resource

years there has been an "explosin" technologies management

Key issues in (he [urther use and development the question investments of scale in scientific infrastructural for the successful and in interpreta/ion aud clearer in human

These

decision-making

adoption of technologies, and transformation, of access understanding

nieasurement progress.

tion of transfer formats for spatial data, timeliness

and wider and technology affairs.

to data, trends in

scales frorn local to global. The technologies which we believe are critical to improving our understanding of the best paths towarel sustainable econornic development include: geographic information systems (GIS). rernore sensing,
anel moeleling merhodologies . with expert systernancl artificial intelligence-assisted

Background

In 1987 the World Cornmission results Future". of its two ycars A key element

011

Environment

and Dcvcloprncnt

publishcd

the 01'

01' eleliberation 01' this repon

in a rcport

c nt i t lcd "Our rh.u

Common

wns irs cndorscmcnt

01' thc concept to meet thcrn".

"susiainable dcvcloprneru",

defined as "dcvclopmcnt

IllCCIS

rhe nccds of thc

present wiihout comprornising

lile ahilii y 01' fururc gcncr.uions

10 Jl

To carry out this concept the mid-term will require unpreeedented Measurements objeets monitor) Many rently systems even Likewise, resouree and long-term that researehers, quantities

of resource effects planners,

development

with due consideration on the environment have aceess

of T to access data we require in our investigations

10

of such development and decision-makers

more efficiently; and (be the We can, of these the Bronze has been to suit user requirements or decision-maker). rates and scales.

T to store, retrieve, T to provide user a seientist, We are modifying however, changes. debate

manipulate

and analyze

these data more effectively;

and types of resouree locations. Maps

and environmental the spatial depieting

information. dimensions of

a wider range of output environmentalplanner, our environment the specific spatial

products

from specific
al other

sites need to be eombined depieting Models and updated

with data and informato keep traek of (i.e. the range of alterof data cur-

developer

tion on eonditions ehanges

at unprecedented c1imensions,

and phenomena

need lO be produced time intervals. to produce eolleeted in digital

rates and significance revolutions, technology

at varying

As humankind

has progressed change. Today's

from the Stone Age through technologies

natives whieh could result from actions of the systems are eurrently quantities improved required exist and data are being larger of data

ha ve to be tested. these large quantities advaneed now. Other format data eollection ten years. and stages of which can management

and Iron Ages to the agricultural a key factor facilitating change cation of technologies on spatial and temporal

and industrial

can crea te environmental Yet, the wise appliour investigations on our environwe the spatial

scales never before possible. us the ability to facilitate understanding answering of human questions

in the planning tools for eonverting

and development

stages and will produce the next environment

al so offers

within

and leads to a more complete ment. Through and temporal appropriate dimensions can move a significant

impact

these data to relevant

application of variations

of remo te sensing, in environmental and application technical.

GIS and modeling conceming

and poliey-oriented it behoves

information

either exist or are in varying of relevant teehnologies to produce planning,

way towards

development. As scientists enhanee are readily us to keep abreast by individuals our investigations understandable and which can be employed in the resouree results whieh

resources. of these technologies. in many others the for per-

There are still barriers These cases, barriers sociological

to the adoption

are no longer primarily to leam; These others reasons believe

They are institutional Some cite the costs;

in some and legal in others.

and poliey areas.

time required their purposes. computing

that capabi!ities decrease. Systems

are still no! sufficient are easier to leam,

may all be valido Yet, costs of sharing What has not been fully realized ancl cornmercial potential to enhance

data and

today continue

to rapid!y

Purpose of Oocument

fonnance community

is rapid!y increasing.

by the scientific of these technoloof

at large is the scientific

gies. Examples The purpose agers aetivities scientific measure, of this document poliey is to provide seientists, engaged researehers, in sustainable resource mankey science ument, and publie deeision-makers of several extraction. development manipula-

of the use of these technoJogies

the understanding

issues exist; severa! examples

are discussed

in the body of this docthe potentiaJ of the

yet more can and should be done. c1ose, we believe, thought that to exhausting and a willingness together required A great deal needs to be done; is innovative opinion fields seeking to achieve we have only scratched

with an overview eommunity

technologies

for data eolleetion, of the global require

We have not even come these technologies. surface. ment. indivicluals improved What is required in cognate insights, Ir is our eonsidered sensing

tion, analysis

and information

Such technologies

can be used by the environment, and provide the to

to improve

our understanding changes whieh bodies. they

to work with developachieve the can

map, monitor

and model

in that environment,

the goal of sustainable we can resource.

deeision-makers options available A suite achieve

with the information

10

to summarize our ability

working

decision-rnaking

by employing systems, directly

GIS as an integrating we can begin to collect processes

Through c1ata which

of technologies

that has the potential development

to improve available,

use of remote

the goal of sustainable modeling back

is eurrently includes

yet in differsensing, have can

either be translated our understanding oprnent.

into information

or input into models related

which improve devel-

ing staies of maturity. and advanced expanded oprnents improve rapidly dating

This suite of technologies tools. Applications into the last century.

GIS, rernote

of key environmental

to sustainable

of sorne of these technologies Use of these ancl irnages can be derived; technologies

in recent ycars,

while some trace their progression

from devel-

our ability: of key state variables constitucnts from whieh spaon environmental

Whot Follows

T to acquire measurernents
tial information

The following

material

presents

a brief ovcrview

of sorne important

data typcs

12 13

available tainable

10 the science

cornmunity

today which artificial of some research,

could

be used as input lO susof the current and expert stasys-

DATA

developrnenr

studies. processing, needs

This is followed modeling, for further

by a discussion intelligence issues

tus of GIS, image these technologies, development.

tems. The paper concludes

with a review

involved

in the use of future

and tre nds iowards

It is important ronment "face powerful sources value";

to understand

specific

characteristics Too many available

of data types used in enviof us often data." Today, accept data at the however,

and development

programmes.

often it may be the "only that we understand before,

Information analysis for management decision making

too1s we have to collect, as never including

integrate,

manipulate,

and analyze

data make data

it necessary

both the nature and accuracy recognition between spatiaI

of our data and its between data and

of the distinction

and information,

and the distinction

and non-spatiaI

information,

Data

V5.

Information

An essential edge which sions.

goal of the use of inforrnation converting may be used for evaluating is created.

systems

and models

is to convert into knowlmaking decito a process analyzed average The decimean

data into information:

measurements options

and observations as a step toward of as any input a rnonthly

As used in this paper. information. could

data can be thought As such, measurements

where information to provide model's for a given location. output sion process In this process generate validity

could be directly

e.g., daily rainfall could then be employed in turn gcnerare measurerncnts process.

lo produce as information

Measurements

also be used 10 model a process, in a management oiher information careful

which would of making

types (see Fig. 1). the quality the data. upon which ancl the

and observations

which will be usecl 10

inlormation,

one 1l111stbe extrcmely

lo evaluate

01" the conversin

In many instances,

the costs of verifying acquiring

quality of datasets decisions Figure1. Tronsforming dolo lo informolion

can be greater than the original to produce

COSlS 01'

As a par! of our concern

high quality data/inforrnation developmcnt

on the best paths toward susuiinable

can be based, we 1l111S1

14

15

distinguish between accuracy and precision, not only in the raw data, but also in the derived information. Accuracy is concerned with fidelity to a standard, and lack of bias. Precision, in contrast, regards the ability lO make fine distinctions. For example, one could measure annual rainfall lO the nearest 10 cenrimetres in a simple rain gauge, which rnight provide highly accurate (or unbiased) data which are rather imprecise. Such imprecisin, if inappropriately extrapolated over space and lime or across scales, could invalidate the use of these data and lead to incorrect decisions. On the other hand, use of precise but inaccurate (or biased) data can also crea te significant problems for managernent decision-makers.

SPATIAL

DATA

Spotiol vs. Non-Spotiol Dota and Informotion

In applications which involve environmental analysis and sustainable development, we must necessarily consider both spatial and non-spatial data and information. By spatial data, we mean data that can be located on the Earth's surface: a water well, a river, a poltical or adntinistrative district, By non-spatial data, we incJude such information as the date that the water well was dug, the pH and temperarure of the river at a monitoring station, and the name of the political district. In a variety of applications, the volume of the non-spatial data may be much larger than that of the spatial. There is a cJass of non-spatial data associated with spatial datasets as a whole which are called metadata. "Metadata" essentially describethe characteristics of a given data set. Metadata are often described as data about data or data sets. For example, the date at which the dataset was verified for accuracy, and the measurement scale of a dataset (e.g. elevations were recorded in metres, rather than feet) are both forms of rnetadara. Metadata are often key elements of catalogues which can facilitate access to an understanding of spatial data and inforrnation. In the context of this document, we distinguish two broad c1asses of spatial data which are commonly used in the analysis of environmental problems and in the consideration of issues related lO sustainable development. We discuss these data which come from remote sensing systerns separately frorn those ihat do not. Remote sensing iechnology (as described in the fol1owing secuon) provides the means to consistenlly collect data over large areas of the earth Ior a number 01' applications. In addition, rhere are a variety of other data types and thernes described subsequently in this docurnent, so me of which may even represent a product derived frorn ihe analysis of remorely sensecl data that are importanl input lO a wide variety 01' sustainable development relaied studies. Thcse sections provide definitions Ior and examples 01' c1ata types which are being used within geographic information systerns and advanced models LO improve our knowledge of the global enviroument.

Remote Sensing
Since the early days of the various space programmes, there has been an increasing public awareness of the ability of space-derived imagery to capture a sense of our fragi1e planet. While remo te sensing dates from a century before the era of space ftight, there is probably no more familiar remote sensing product than the daily view of clouds and weather pattems presented in weather forecasts on the television news programmes in many countries.

Definition
Remote sensing is the use of systems to gather information on objects and phenomena from a distance. The human eye can be considered a remote sensing systern as can an ordinary camera. Yet, contrary to the popular expression, with rernote sensing what you get is not necessarily what you see. A great deai of what exists in the world around us cannoi be perceived by hurnan sensory organs (see Fig. 2). While one can assurne that distance is the major factor limiting hurnan visin, the spectrum of energy we can see with the unaided eye is also an importan lirnitation. The human eye responds lO light from o~lly a small portion of the electromagnetic spectrurn from bJue lo red, while energy frorn gamma rays lO radio waves swirl around liS unseen. Through the years, however, we have consturuly siriven to expand our ability lo "see" energy fielcls beyond our direct perccprion. 111 this process we have constructed a variety of spccialized insirurnents 10 record infonnation and present it to us in picture-Iike forrns
called images.

16

17

Frequency in cyeles second1

1()1Q

1018

1016

1014

1012

1010

108

102

Remote Sensing Systems

Today, a wide variety of systems patterns inforits atmossurfaces remote acquired and linand are solar multivegsur-

are used to detect

E E eo E
-c

energy provide

(see Fig. 2) which mation phere, The sensing about oceans, bulk

>(

x
M

o ci

o x
M

the earth, and land of satellite scanners

---

-,

~-----' Cameras with '.Solidstote detectors in : infrared ,sensitive film. , sccnners and , , rqdiometers ' Solid slate : deteclors in , scanners and , radiometers , .

, , , , ,
, Side looking , Electromagnelic , : airborne radars. : pulse lechniques'

(see Table

1 and Fig. 3).

: Scintillation , counlers.

,
1

, Gommarays : spectomelers

,,

, Counlers' ' Film in cameras. : with fillered:: , photo " Scanners with , multipliers. , filtered photo, , rnulfipliers : Image orthocon~ " , amI- cameras , , with filtered , , infrared filn ' ,
. '~.,,-;,',,:~;..

,,

data is currently

by multispectral ear array devices near infrared tromagnetic passive radiation spectral

, rf receivers in , scanners and , radiomelers'

, , , ,

, , ,

, '

, , ,

in the visible These

positions spectrum.

of the elec-

Dust & cloud

25%

, ,

Atmospheric absorption = 19%

systems reflected scanners

recording from

from the earth's provide informaand consoils;

surface. Data derived

tion on (among

other things):

Figure 2. The fronsmission 01elecfromognefic energy throuqh the lorth's ofmosphere (below) ond exomples 01sensors thnt imoge in if (obove) (odopfed Irom Esfes, 1974)

etation

type, distribution

dition; geomorphology; These active reflected remote sensor

face waters and river networks. systems into sysof a passive system an active applications. range using laser technologies are being develsystem, as its film records since light energy to its own used in the sun (see Fig. 3). Add a flash attachment it now provides are commonly systems can be broadly systems categorized and passive

Figure 3. Energy Ilow model lor remofe sensing systems

terns. A camera is an example from an existing the camera geological Active depend and it becomes Active and hydrological systerns

source:

source of illumination.

microwave (radar)

in the optical

oped for oceanographic

and Iorestry

on reflected energy

Because active systerns do not frorn the sun for image forrnarion they may be able
as sorne microwave as shorter radiation wavelength ro as great a degree

applicarions.

to acquire data during the day or night. In addition. is not disturbed by the atmosphcre

energy, radar sysicms allow data acquisition in cloucly and rainy weather (see Table 1). This ability lO opcrure day or nighl, and to cre.ue irnages of the surface in spite 01" cloud cover. makes racial' sensors parucularl y auructive tools -for irnproving our understanding 01"boih tropical and polar areas. Thermal intrared data which represent

a record of cmittcd energy Irom surfuces ha ve becn particularly useful in monitoring tires and in irnproving our understanding 01"arcas 01' volcanic and gcorhennat acuvuy. Surluce tcmpcrarure 01" the occan is also

19

Resolution
Resolution is at the core of the successful concept or modeled applied what use of remote modulation sensing transfer normally (GRD). identified data, Yet, resofunctions involves Often as a the this

lution is a very difficult line pairs per millimetre tool. Resolution, question involves describe given important concrete things. Indeed, lutions usefulness "can 1 identify the smallest good here. While stretch

to understand. using

Resolution

can be measured in

as typically

by a user, however resolved contrast. distance"

1 need to identify

on this imagery?"

the use of the term "ground object

GRD is used to on the ground contrast is road on many resothe as used

that can be unambiguously to see white

object to background

Object to background
to see the same size white is thus dependent spectral

it may be possible road.

lines on a black upon

one set of images,

it may not be possible of the same

lines on a

Resolution

the level of matching of the derived

between

the spatial, Spectral

and temporal wil! condition

of the measurements

and the needs of the investigation data/inforrnation. spectrum

range and resolution

here refer to the porton the measurement. thermal olution

-:-v .:-.._~

of the electromagnetic

It must be appropriate

to the questions

which is being used in being addressed, e.g.

Spatial resconcerning of a metre

range for fire detection, conditions

red band for vegetation rule it must correspond identified surface

absorptions.

the level of detail which can be extracted (Le. a resolution features). resolution

objects
or les s is may

~'.~+ ~':,";; :1:t;~. <kP..<"~ v:r:';i.:

in a given
".
.c- _
-J.. -.

scene. As a general

to the size of a typical

f .

0-

'"r...?l -'

. -

~'~~"'-?~":" ~

5ysem"chorocteristics inCluded ore !hos" "'most comrTlonY,empl?yed . ond monogement cpplicotions tody:". ,<"',.' ~ '''''','

in resource pJonning

object which must be separately needed between coverage Finally, to identify be enough into account for studying the volume ocean

..

a single tree, while a resolution and ground of data which

-

of 500 m to a kilometre A necessary of the measurement data - 10-30

2, 6enot~s the otmospheric co~ditio~~ !hot can be penetrqted by ~~ergy in this portion .h'.':'" otthe electrornoqnetic spectrurn where: .;,'\ c::?~ " . "'. = haze; 5 =\moke; 5g ~'smog; F = fog or clouds; R ~rq~n. . Discounting the use oF active opticol systems such as flash units, loser line trocers, or light amplification systems.

compromise must take area m (see also fre-

area to be covered - millions of km2

will be generated

by, say, a large

using

high resolution

Fig. 16, page 60). the frequency of the data acquisition Daily must also match the natural may be needed once every

Toble 1. General rhnruderisfics

01 remate

sensor systems

quency

of change

in the landscape.

observations

for an

assessment related stress therrnal ration. to the dynamics also induces sensors. fields; of coastal waters and currents, detect Over land, plant water can be detected rnagnetic by and 01' the the earth's

of plant evapotranspiration

while observation

year will

be needed for land cover change changes in canopy sensing ternperatures systerns which

assessment,

Other rernote

History 01 Remote Sensing

The first of what would now be calJecl airborne resource remotely management. sensed clata were

gravitational University magnetic inforrnation

these tools are used extensively Santa Barbnra, frorn various location.

in oil and mineral the carnpus ponions

exploacquired 1870's, balloons from baJloons foresters

lO help

Figure 4 is an ex ample of irnagery of California, spectrurn if propcrly over the sarne general designed,

01' the arca around Analyses

of the electroand quality 01'

over Paris in the 1850's. The first pictures

01'

were more art

01' such multispec-

and curiosities

than data for scientific in Germany rnanage timber

Yet, by the late

taken from in increases

tral data can.

incrcase

both the quantity

had begun harvests.

lO use aerial

photographs drarnatic

for given applications.

By the 1920's,

20

21

Figure 4 Ih). Block ond white inlrored photogrophy Figure 4. Four views 01 he oreo oround the campus 01 the Universily 01 Colilornio Sonlo Borboro: (o) Block ond while ponchromo!ic phologrophy
the uses of aerial experience gained uses. Agriculture. ing, soil scienee, 1930's, moved tronie colour beyond sensor phorography during water resources, and mineral the visible systerns began to occur as individuals air photos planning, with military Those who have been involved ogy over the years have watched data types and the analyses sents an atternpt World War 1 began exploration to apply to peace time civil engineerBy the we and 1960's was coined photographs to illustrate with various aspects in the application the increasing in ernploying of rernote sensing technolsystems, 5 repre-

cornplexity

of sensor

undertaken graphically

these data. Figure

the increasing of remo te sensing.

cornplex ity associated We have progressed cornplex prob-

of the developmeru

urban and regional applications

to a point where we are employing lerns associated with global surveys.

muliisensor data in solving

were all pursued.

aerial phorography ponion

was in use. and in the 1950's of the spcctrum images the term remote sensing

to the precursors

of the clccin

Ac(ess lo Remate Sensing Doto

The applications 01" global satellitc covcruge move forward rapidly. However.

of today. Indeed, picture-Iike

the late 1950's because anyrnore were beginning

that could not be called

to come into wider use.

there is still no cornprchcnxive

treaty 011 dal; g;lIhcril1g which

has bccn r.uificd

22

23

Figure 4 (d), Aero-magnelic survey imagery

Figure 5. Hislorical developmenl 01 remole sensing, emphasizing inueasing complexily 01 plntlcrrns, syslems und losks Figure 4 (c).l-band active miuowove imagery

24

.......------------~--~by UN member proposal" been officially nations adopted. regarding rernote sensing States Nations from space. to the Soviet operates The "open Union skies first put forward by the Lnited on Rernote as being has never Key Today the United under the 1986 resofrom Space.

lution entitled:
. Remote resources excludes . Worldwide . Countries

Principles

Sensing of the Earth

for the purpose

points of this resoluton sensing

include:
of improving natural this land use. and the protection satellites from these principies; of data is pemtted systems of whether bear intematonal such actvities and countries aspects as possible related are while ensuring and distribution remo te sensing irrespective agencies of the environment,

s defined

management, surveillance collection operating

that sensed states have access to data of their country; responsibiliare carried forties for their activities, ward by govemment . Further In April applicaton particular United cooperative stronglyencouraged. 1991, the United of the principie account Nations the needs Natons went further. Legal to the into that the exploration of peveloping should and utilization of outer space

or prvate cornpanies; to benefit as many

activities

would be carried

out for the benefit proposed that states

and in the interest countries, concentrate

of all states, taking were put forward. their efforts and equipment

.The

in the fol-

lowing areas among others: . Continued exchange of information, science sensing and technology; . Promotion

Toble 2. Exomples of rurrent solellite ond sensor system parameters

data, materials on space of remotesys-

of easy and low-cost accessibility receiving to promote stations

and availability

data, the ground

and digital image processing the transfer particularly

terns; and . Technical cooperation and expertise ing countrics. lssues urgcncy

and facilitate

of technology

in space science and rechnology,

with the develop-

continu to be addressed with some urgency. This is driven by the realization that a growing number of operational and rcsearch instruments are being planned and that potential users are increasingly Iaced with difficult questions related to the choice of data sources, the availability of appropriate data, the COSof the required coverage, and the availability of the data in a timely fashion. By the end of this cenrury, we will be receiving terof data must certainly of data daily from space. Indeed, if we examine 2002 only the approved systems alone and can proposed

Acronyms for Toble 2 IFOV Instontoneous Field 01 View (resolulionl MSS Multispectrol Sconner 1M 1hematic Mopper AVHRR Advonced Very High Resolulion Radiometer HRV-P High Resolution Visible Ponchromolic HRVXS Multispectrol Mode 01 SPOT Sensor MESSR Multispectrol Electronic Sellsconning Rodiometer SAR Synthetic Aperture Rodar

produce

data in the terabyte

range (see Tables scientists,

2, 3 and 4 and Fig. 6). These managers and policy-rnakers

data as we

abytes!

will be of value to earth resource They will, for example, cover, monitor

United

States

missions

to the year

these

auernpt to improve our understanding 01' the factors that sustain life on the earth. help us measure key aunospheric constituenrs. map land
the expansiOIl of agricultural information of our earth. land and model future wearher patierns.Such clata can provide which may improve our managcment of

l.

/\ terubyic (10") of data is the equivalent 1992 data storagc iechnology.

of o competer

tape approxirnately

5000 km long, givcn

thc rcsources

27

.' li~hlnin~ oclivity 1; eSli;;bies 01 precipilolion

~_ . 7f."" "

~I~)_~~.:;j~'~";Ji5:'f-'' ~~Mulii-Onglerodionce~; olbedo; ,.,."",,:.1 olmospheric oerosols

/.,,::& .:4" _ ,

Vertical proliles 01 ea ond possible CH-4

.F

Toble 4. EOS-BSeries Eorth Sdence Inslrumenls. The inslrumenls lisled ore condidotes lor flight in the EOS-Bseries

OJon sfl.ocewind
... -

7peed'
.

Other Spatial Data

7 GHz dolo uplink copobility

Themes It is well beyond the scope of this paper lo provide a description of all spatial data types which could be used in an analysis related to sustainable economic development. What follows are examples of more cornrnonly employed data types. For each of the following cornrnonly-used geographic data, a sarnple of the derivation of the data is provided.

Toble 3. EOSASeries Instruments. The instruments listed were proposed lor f1ight on the EOSA series plotlorms

28

29

Elevo!on
Data about elevation may be acquired by making measurements at specified locations using Global Positioning Satellite (GPS) receivers plus conventional surveying techniques. Advance in hardware and software for the satellite-based GPSs have revolutionized the field of geodetic surveying and navigalion. lt is now possible lO precisely determine X-, y-, and z-coordinates of any location on earth using a pocket-size GPS receiver. Cornbining digital image processing techniques and GPS technology, remotely sensed information can be quickly and accurately related to absolute .coordinare systems, thus creating improved and efficient ways for creatng topographic inforrnation. Elevation data can also be produced using photogrammetric techniques from overlapping pairs of remotely sensed imagery. From these sets of original source data one may derive contour lines for creating printed maps, and regular arrays of estimated elevations through a mathematical interpolation operator. Interferometry using microwave data is also a promising field of research.

Lond(over, Vegelolion ond Soils

Data about vegetation and soils may be created through field surveys. These survey datasets may then be used as a sampling frame from which data on vegetation and soil are extrapolated across space by incorporating remotely sensed imagery. Biomass and primary productivity can be derived from such databas es using appropriate models.

LondUse
Land use in this context means human interaction wth the land. Land uses include transportaton, structures, agriculture, urban (residental vs. recreation), utilities. Data conceming human use of the land surface may come from many sources, including tax records from a regional govemment agency, maps from planning agencies, engineering drawings and databases from civil engineers and architects, and aerial and satellite data coupled with or collected solely by field surveys.

Meleorology
Data on rainfall, temperature, pressure, humidity and air quality (e.g. percent N02, S02' 03) may come from surface, airborne and space borne instruments, mathematically interpolated lO derive estimated values where measurements are unavailable.

")!'-

::"

1mJIBm!ImJIBImJIm1.lIf1ZJIDtlf.m&m&mmE
l. Joinl with Itoly 2. Joint with Fronce

b~e~Y4~*.H I
Approved, under development, or
operoting

I
Future eorth prob candidotes indu(h .1

3. Russian sotellite

Possib/e extended mission

Proposed mission

Figure 6, right. Exomples 01 turren solellile missions ond those being plonned into the next cenlury

4. Japanese solellile 5. doto purchase 6. Joint with Jopon

ose

missicn

Aristoteles/Grofity , MFE/Mogno/io & TOPO Missions

30

Demographic
Infonnation on the characteristics of people, as a function of their loeation, is , commonly derived from survey data sueh as a periodic census of the population. These are typically recorded as statistieal summaries, aggregated over various spatial scales from pareel to eensus traer to city, county, state, nation and globe.

Cadastral
Property ownership. Information about land parcels and their boundaries and characteristics is typically gathered using conventional surveying practices and compiled in analog or digital map format as well as documents created when properties are sold or exchanged.

Geologic
Lithology of formation and fossil records along with infonnation or structures are often measured at a small number of locations, and spatiaI pattems in these data are then inferred from ancillary data such as remote sensing observations.

Hydrologic
Drainage patterns, flow rates, size of catchments, depth to groundwater. Hydrological data on drainage pattems, flow rates and catchment basins are often derived from ground observations coupled with an analysis of the spatial pattems of lithology and elevation either from existing maps or using remotely sensed data. Data on groundwater are frequently measured at a smalJ number of locations and generalized in space based on correlative infonnation.

Formats and Structures

25 26 27 28 29 30 31
32

For any set of spatial data, there is no single way to record the data values and their variation through space and time. For example, elevation in a region eould be recorded as: T Con tour lines on a map - estimates of the set of loeations which correspond to a predetermined list of elevations above daturn T Entries in a cartographic database - digital data which is designed to store elevation data in a manner which may be used to crea te prnted rnaps

33
34

35
36

37 38 39 40 41
42 43 44

Figure 7, right. Elevolions on o regular grd. The upper portion 01 the ligure shows o ser 01 conlour lines Irom o mop, in which loceions olong o gven line are the some elevolion obove seo level. The orroy 01 numbers in the lower porton 01 Ihe ligure represenl he estimoled elevolion 01 the centre 01 o set 01 squore rosler rells,

45
46 47
48

32

. Values

in a raster array - a regular value is stored on the earth surface network

array of adjacent and the elevation

cel!s where a representaa sheet of graph at the centre of each cel! is captured by creating space;

TECHNOLOGIES

tive elevation paper placed . An irregular an irregular The following In general, digital cautious

for each cel!; one imagines

in the sheet is recorded

(see Fig. 7) - the shape of the earth surface facets located in 3-dimensional

set of 2-dimensional section provides printed albeit additional

where the terrain has linle curvature data suchas

the facets may be large. details of some of these alternatives. may be con verted is required. which to Be

maps and photographs

t,

databases,

with significant generalization: Indeed inforrnation

costs if high precision and the accuracy accuracy using

of cartographic of reality.

a map is not reality but a cartographer's results from the overlay However, when of various it should maps maps and

representation from integrating research many practice products actions resources.

map accuracy

or deriving

or other spatial data types of either known or unknown area with many unresolved users rank currency before questions. accuracy

is a very active be noted that of various cartographic made and

In the last decade developed nologies tant example

an important

set of new technologies Information by rapid The emergence progress

to handle (GIS)

spatial

data has

and matured. of these has been made

Geographic technologies. possible include:

Systems

are an imporin disparate

of this suite of new techin advances

This is often done with the presumption wil! be fol!owed represents in creating to accepted the map product, procedures. the current "state-of-the-practice",

that standard decisions

and that since use of such

fields. Tbese advances . the massive increase

in the power of computer in digital systems; in the techniques spatial

hardware

at a given price; technologies, data modeling for spatially dataand

taken will conforrn

. the rapid deveIopment bases, and information . the significant analysis; and, of highly . the emergence uted phenomena. Today means the technologies to encapsulate as never

data col!ection of spatial

developments

realistic

forms of visualization

distriboffer a can

resulting problems

from this decade with a spatial

of rapid development in models complexity once

dimension

which

represent,

before,

much of the real world's behaviour. Such

and diversity. spatial can being more convaJidated,

These models can then be populated inter-relationships then be explored widely exciting mum path towards applications The material cisely profiling and dynamic

with spatial data so as to incorporate models, tools. These however, systems are already many

by the use of analysis sustainable follows

used to address

many of the challenges dcvelopmern: exist. briefly

associated

with finding we believe

the opti-

for these systems

which

introduces

these key technologies,

their procedures

and potential.

Geographic Information Systems

Certainly

one of the most imporuuu

technologies

for handling

cnvironmental

34

35

data, the Geographic there products compete. ranging

Inforrnation

Systern

(GIS),

is now highly over

developed

and

is a sophisticated

rnarket-place

where

al present

100 commercial
investment in GIS

The last few years have seen an enormous from systerns and are accessible

world-wide, implernented organizations.

which cost less than $200 to those which are via networking across whole and ana-

on large computers

GIS can be defined as "Cornputer in a decision-making means

systerns for integrating

context"

Iyzing spatial information

GIS have been described tions and analyses

as the "biggest step forward

since the introduction for observaby definitime at a given

of the map", The map is the customary of spatial phenomena. record Typically, tion, a static and interpreted cating spatial information.

of communication phenomena paper

The paper map is, however,

of a particular

and place. As such, the traditional drop for plotting When

map is a useful but limited map users employ

tool for communimaps as a backdata have come apparent. their coneasily interactive form:

other kinds of information.

As users of spatial

to make maps by computers, stored in a computer, tent, symbolization and rapidly.

a shift in the role of the map has become maps take on a new and dynarnic can now be changed

and the area covered

relatively

For the first time, users of spatial data have truly flexible, in real time. within therefore, that such maps formed of information

access to their data, any part of which can be updated It can be argued thought tionships. varying of as databases, or collections

Rgure 8. Venn diogrom of overlops in spoliol dolo processing lechnologies

indicate system database costing; that hardware implementation development and software costs, actually costs only account for around

a GIS are better

linked by their spatial relaexpression of some values of a spatially

20% of total
with that the for an

In this conception variable where analysis,

a map can be seen simply as one unique showing the area-by-area refIect chosen a spatial place model

whereas amount

the staff and data costs

associated

of a spatial database,

for example,

to 80% of the total. This implies are thus the most irnportant assets

the categories and is a much

cost of the data and the expertise organization These to protect

required

to run a GIS are a key part of project

kind. In this new spatial data computing tial for spatial information. The functions tial data handling Automated Management of a comprehensive systems which (CARTO), (DBMS).

environment, more efficient

the GIS offers new potento "store" spatial

and that data and expertise

in the long termo begin with a model of the phenomena under study. in various functions); ways, for example, as a set of regions (whose internal spatial by interlocking

AII uses of a GIS should GIS can be traced to a range of early spaevolved Computer in several related (IMPROC). multi-purpose software spatial data this developpackage s sold fields such as Aided Design (CAD), Data Base point samples variation of reality

data models may be specified approximated by simple may be described

Cartography Systerns

or as a ser of identified frorn, the subjects

and Image Processing

objects which litter the plane. The data models used by GIS are usuaIly "phenornenon-based" to specify where using the features in, or derived under study map, and such the model used to represent thern. The best exarnple sueh as the coast. of such a data roads

A comprehensive handling system.

GIS today can be defined

as the union of the functicnality

of each of these systems

to form a true multi-function, of particular

The figure (above right) can be used to visualize on the functionality

model which uses the object a set of features are idemified buildings as points, creation locations srorcd

view is the one c1erivecl from a topographic rivers, by convention. 01' this model. georncrric uscd as thc building are crcated rcfercncing as mctric prirnitives blocks

ment, and lO chart the origin and strengths as GIS. Depending DBMS

on the lanclscape and classified representation are typically Such

offered, most packages

can be plotted on

this diagram: many will plot in the overlap of CAD and CARTO, or IMPROC, and CARTO, but Icw can actually be placed near ihe central GIS category when defined To complete mention: required in this way (see Fig. 8). a description 01"the components Estirnates of a GIS, it is al so necessary human skiIls and management costs of a GIS projccr to

To create a cornputer 01' spatial

lines and arcas objects. using coordinates

for the

reprcsentations This is known nurnbers

by first measuring sysiem: the simples: and is inlormation

based on a spatial

hardware,
lO operar

spaiial data ano [he necessary

the system.

Iorrn is analogous

to an x,y graph.

of the typical

in the Iorrn 01' characters,

and boolcan

valucs

(true or [a lxc).

"

,HBL1O'fECA DE SYEWS DE LA U!I.CH

36

37

since the cornputer the representation, indexed to encode tion): this is known

cannot spatial

handle

spatial

data types such as lines. To complete of the model (referred are known The two alternative must be ordered and to as topological methods as the vector informawhich are and raster

the spatial

components

inter-relationships

as a 'data structure'.

most often used to create data structures, approaches.

Vector surement, straight data structures employ gation. A point can be described line distance between

the principies by its distance

of coordinate

fixing as in navi-

along each of two axes of mea- a vector in the mathematical together in a loop

e.g. north and east, while a line can be described as the shortest
two such points by any number in number of lines joined

sense. Areas can be represented and arranged rnation needed the computer Raster regular squares record is the coordinates and recalled

to reftect the specific

shape of the area. In all cases, the only inforform, which can easily be stored in approach altogether, using

in the right order when needed. however, use a different on a grid. Using such a by sets of controlled keeps a being directly

data structures,

building or pixels,

blocks to build up the shape of a feature the resolution of the representation and therefore allows

grid of very small squares,

a point, line or area can be approximated the computer

by the size of the grid squares. In a raster data structure, of which pixels have which values being represented. of these data structures While raster based and retrieval, when precise vector one or another data storage in location spatial phenomena Using phenomena

form part of one of the the storage typically of spatial allow for for

in the computer.

systems systems

more efficient more specificity the geographical points, Many

offer the potential identifier

coordinates

are included.

To complete to all the data data. or

data base requires

the addition data stored

of a unique in relation

lines and areas represented

so that they al! can be linked LO attribute to the spatial a user to group all geometric

such as a name or other non-spatial

GIS currently

in use permit

and attribute

Figure 9. A geographi( inlormofion syslem m be roncepuolized os o base mop by numerous registe red overloys. lnlormotion in the data base wos derived Irom a voriety 01 sources

data together 'pollurion'. information

for a single subject of study in a single theme such as 'transport' This can be implemented daiabase systern: however, as a set of thematic is an integrated

layers in the database

(see Fig. 9). The cornpleted

spatial and non-spatial

by applying

Imoge Processing
While images frorn remote lites rnay be fascinating, ab1e information estirnating agricultura1 sensing

its potential

can only be unlocked

spatial tools which operate on this structured spatial data. Functions availablc in a typical GTS can easily be frarned by the sequence of operations carried out in the crcation and analysis of a geographical database. Hence typical operarions would include querying the spatial database for given
locations or Ieatures a surface These or network: meeting

sysrems on platforms
10

from aircraft

to satel-

they have yields,

be processecl

and analyzed

to acquire suit-

for applications

such as monitoring on airborne

the paihs of scvere storms. or more reccntly satellite

in sys-

ancl ihe prcdicrion

01" runolT Irorn mountain snow which rcquire

,1

particular

criteria; analyzing of spatial

the characteristics

of

packs. We have moved frorn a reliance

data alone ro accomplish an objcctive incorporating rerns which rernorely can examine
LO

or the intcgr.uion

data frorn different

sources.

thc use 01' methodologics

operarions offer key udvaruages in the use of a GIS as they enable a model of sornc spatial phenornena and gain new insights.

the

sensccl data and many other iypcs 01' information

user to examine

geographic

inforrnntion

quesiions al scalcs frorn local lo global. The use 01' sysrems has grcatly Iaciliuucd rhis trend as sccn in

38

39

Figure 10. In this progression we have also moved from using remotely sensed data for simple identification of forest species to their application in complex problem solving such as their use in models to estimate runoff from given watersheds within specified time intervals. and the use of this information in the projection of available hydropower generator potential (see Fig. 5, page 25). In the past, aerial and space photography was primarily analyzed through human photo-interpretation. However, progress in digital data acquisition and computing technologies has led to computer-assisted and interactive digital image analysis systems. As in many ot~er areas, remote sensing applications have benefitted from the explosion of computing capabilities in the past two decades. Where earlier systems required room s full of computers, technical staff, and custom development of hardware and software, desktop computers and commerciaJ software are now capable of serious project work. Where earlier systems required years of research and development to tailor new tools for users, there are now practical tools for a range of applications. Digital image processing systems can be seen as an example of a rasterbased system where the remo te sensing image is tiled into small (normally square) picture elements or pixels. lmage processing techniques are used in a large number of application areas such as industrial fabrication and control, surveillance systems, television, archiving, and desktop publishing. Specific techniques to process and analyze remotely sensed images include image enhancements (i.e. filtering to smooth images, or contrast stretching), georeferencing and rectification (i.e. registering rernotely sensed data with a geographic coordinate system), automated classification and feature extraction for l arid use mapping/land co~er mapping, and change detection and habitat analysis. In their early stages, digital image processing techniques were primarily concerned with single irnage evaluarion and local applications, and relied to a large degree 011 algorithms based on statistical and mathernatical algorithms or visual analysis. Recently, these techniques have been extended to address topics such as multisensor, multitemporal and multispatial image analyses, integration with geographic information systems, interfacing geoscientific and socio-economic modeling, and the use of expert system technology for knowledge guided image interpretation. For example, aerial photographs have been in operational use for decades to inventory and manage resources in forestry applications. With the advent of cligital photogrammetry. orthoimage maps in digital formal are increasingly being used for this purpose. For countries Iike Callada or Brazil, there is no other way than to rely almost cornpletely on remote sensing and digital image processing techn iques lo rout ine ly monitor ancl manage the i r vast forested areas. Monitoring and quantifying tropical deforestation on a global scale is only now practica! with the availability of modern remate sensing techniques. Recently. colour infrarecl airphotos and scanner imagcs have also been used to analyze ancl assess vegeration stress. cspecially for arcas affecicd by acid rain.

~
( MAtlAGEMENT

Figure 10. Diogrom

01 o GIS os o derision

supporf sysfem

40

41

The critical importance of rernote sensing is of course not restricted to one specific application area. The ability to use digital image processing techniques to extraer rneasurernents (and as such. spatial inforrnation) frorn remotely sensed data has been used in a number of fields concerned with rnapping, monitoring, modeling and managing our environment (see Fig. 10). For exarnple, using sophisticated atmospheric correction models, researchers have been able to relate digital pixel values to energy received from the sun at specific wavelengths at a specific ground location. This has helped users to irnprove their models describing heat flux at the boundary layer between earth and atrnosphere. How else if not with synoptic rernote sensing image data could atmospheric physicists have detected the Antarctic ozone hole or could oceanographers detect and analyze sea surface currents and eddies to assess and refine oceanic global circulation models. It has been shown that ocean colour values as recorded on the Coastal Zone Color Scanner (CZCS) on Nimbus can be correlated with biomass and consequently provide productivity figures for the upper ocean layers. Using SPOT and Landsat satellite data, GPS measurements, and digital photogrammetric and image processing techniques, digital image maps can be produced for desert areas in the Sahara. For the first time up-to-date, detailed, and accurate maps are available for these areas - a basic requirement for planning and decision-rnaking. Expert systems are also being developed 10 improve accuracies of land uselland cover maps. Whereas earlier image analysis techniques were almost solely based on statistical analysis of spectral values at every pixel position, newer developments take into account parameters such as neighbourhood information, texture parameters, or surface shape analysis: researchers are also trying to formalize human interpretation skills into rule-based expert systems in an attempt to improve our ability to extract information from remotely sensed data. Whereas some of the techniques and interfaces are still in the research and development stage, these interdisciplinary approaches already provide promising tools Ior efficient storage, query, information extraction and integration of remotely sensed data (see Fig. I Il.

Figure 11. Combining remole sensing ond GISlechnology inlo so-colled imogeinlegroled GIS (IGISl
A model is a caricature of reality that emphasizes those parts of a system of most relevance to the user while ignoring those components of less importance. The skill in model building is to decide which detail is essential and which can be omitted. Models can take many forrns. Physical models are widely used in areas of engineering such as fluid dynamics. However, the main types of models relevant to sustainable development are analytical models (i.e. a series of mathematical equations that can be sol ved) and sirnulations (i.e. a series of functions expressed as a computer program). Analytical models often oversimplify the systern being modeled to achieve a formal rnathernatical representation, but have the advantage that their behaviour can be efficiently and comprehensively described. Simulation rnodels can capture the cornplexity of many real systerns bUI always al the risk that a large cornplex systern will be replaced with a large complex model - neither of which can be understood. Models are built for a variety 01' purposes. Many are built as experimental tools that explore cornplex hypotheses about the mechanistic basis of systerns. These models are probably better never released from the laboratory, since they oten lead to Irusuation anel disillusionrnent when planners and decision-rnakers attempt ro use them in practice. Here we are more interested in modcls that can be populated with spatial data and used lO explore the cornplex spatinl ami dynamic temporal behaviour 01' systerns. Most models used in resource and environmcntal management are dynamic models. Dynamic rnodels folIow the chunge ihrough lime of particular variables

Modeling

Humans havc always used models. The first were physical models, ranging frorn dolIs la model boats and gliders. The garne i-go (wei-hai originated in China before 600 BC :15 a simularion of the tactics 01' warfare. and in the Thirteenth ceruury AD data frorn ruin and snow gauges were being conveyed to the controllers of river systerns in China to help them preelict flood peaks.

42

43

describing a point or points in space or an aggregated area. Examples of dynarnic models include models of the yield of a crop, the growth and population dynamics of a forest stand, or the production of a herd of animals grazing a specified field. A few models predict the changes of variables in space. These are often used in association with spatial databases to predict the values of variables that are difficult to measure by other means. Gradient modeling is perhaps the best known example of these approaches. In gradient modeling the distribution of species or community properties is correlated directly with more readily measurable descriptors such as altitude, aspect, geological substrate and the like. These can then be used to predict among other things the potential distribution of a crop or species of natural vegetation. There have been very few attempts to combine both model forrns of prediction in time and space. The computational demands of such spato-temporal models were once asevere limitation. However, computing power has increased and more efficient computational algorithms are being developed. Some progress is also being made with the automatic simplification of models. Some forms of models can also be used efficiently in standard analytical and optimization packages, thus allowing the user to examine a wide range of options in seeking better management strategies. An important limitation in developing spatio-temporal models is our inexperience in coupling subsystems of different spatial scales. We have too little experience with this c1ass of models to be certain just which details to include and which to leave out. Mistakes can lead to cumulative errors and spurious chaotic behaviour. A number of factors have limited the more widespread application of models in the management of natural resources. Most models are built for a particular task, often with the mixed goals of prediction and exploration (or explanation) of the system itself. This means that they are often curnbersorne to run and are usable only by the original development team. These problems are being overcome. The tasks required of models are being stated more clearly (e.g. prediction of crop yields, structural changes in vegetation, land suitability). Successful generic models are being developed (e.g. the crop trajectory models of Hall and Badhwar, the JABOW A/FORET suite of forest growth models, BIOCLIM for predicting species ranges) and are currently available in well tested software packages, Also a largor cadre of biologists, agriculturalists, planners and other users with skills in modeling and computing is emerging each day. The users are pushing for more development and better models. This bodes wel! for further advances in modeling.

Expert Systems
Expert systerns (sornetirnes called knowledge based systerns or advisory systems) are among the most prominent products of artificial intelligence research. An expert system is a computer program designed to sirnulate the problem solving ability of an experto The program asks questions and, based on the responses of the user, eventually reaches, or validates, a ooncluson and is able to explain how and why it reached that conclusion. The first expert systerns were developed in the late 1960' S. Many of these early expert systems were designed 10 assist in diagnosis of diseases or in identification or c1assification. A very large literature on the methodology of developing expert systerns has been created and there are numerous software packages to assist in the task of building and using expert systems. Much of the theory behind expert systems draws upon the larger debate about artificial intelligence; however, questions about how humans think or whether a particular program do es show true "intelligence" are not of prime importance in this area. The goal in developing expert systems should be to design and implement useful advisory systerns. What may be of importance are the modes of logic and explanation that people find intuitive so that expert systerns can more readily comrnunicate with their users. The construction of an expert system is a time-consuming process both for the builder (sornetimes called the knowledge engineer) and the experts whose experience is being captured. A cornmon fallacy runs, "We don't know enough about this topic to build a real model so let's build an expert system". This can be the start of a wasteful process. Expert systems should only be built when human expertise is scarce but available. This expertise may be in the forrn of a series of rules or ernpirical relationships, as long as the knowledge encapsulated in these rules or relationships can be cornmunicated in text or pictures. The experts must have the time and willingness to cooperate in the venture. Expert systerns work best in diagnostic or classificatory tasks, especially where the understanding of the task is empirical and the input data are "noisy". They also work most effectively within a wel! prescribed dornain. It is very difficult to capture all the pieces of knowledge that go to make up cornmon sense in a widely defined task. A promising developrnenr in the building of expert systerns is the application of the techniques of rnachine learning or rule induction. In developing a quantitative model, a scientisr may use statistical software to unravel relations and derive mathematical functions frorn data. Machine learning plays a similar role in developing expert systems. New algorithms and software packages are becoming available to assist in extracting information Irom large data sets. The algorithms are especially suitablc for handling qualitative relationships between variables and discontinuities within the data. The information is exprcssed in the forrn of rules 01' decision trees.

44

45

The field of advisory systerns has merged with the more general problern of developing systerns of knowledge handling and delivery. This is becorning increasingly important as the quantity of information and the cornplexity of its interconnectedness increases. Even when information is available electronically. simple searching approaches can still be slow and unsuccessful. The main product in this area is better structured knowledge bases and "intelligent" retrieval systerns available on dernand and on-site. Another related area is rule based models and qualitative modeling. An expert systern can be thought of as a model cornposed of rules and these rules are often qualitative (e.g. if the amount of monoculture cropping in the dislrict is higher than the norrn for the region, then ...). This is a major area of application in rernote sensing where rule based models may assist in the initial c1assification of irnages (e.g. if the ratio of Landsat Thematic Mapper bands three and two is low, then the land cover class is water). Another major application of expert systems is as advisory systems assisting users as they attempt to master the force of the very powerful, but complex, tools provided in a comprehensive GIS (see Fig. 11). Similarly, expert systems can be employed to guide users in the intricacies of reformulating generic models for their particular application, in running the models and in interpreting their models outputs. The fusion of remote sensing and GIS technology into so called image-integrated GIS (IGIS) is especially seen as one of the key research fields in spatial data handling over the next decade. National and intemational research programmes are underway to study the degree to which these two technologies can be combined. It is important that advanced expert systems be developed which can aid users as the anernpt to understand both the potentials and the pitfalls inherent in the integration of these complex technologies.

APPLICATIONS AND EXAMPlES

Each of the technologies discussed in this digest - remote sensing, geographic information systems and environmental modeling - has a variety of current applications both within the research as well as the operational resource management communities. More and more often we find, however, that they are combined to greater or lesser degrees depending upon the type of analysis being conducted. Results are available on the use of remote sensing to provide measurements of environmental parameters such as net primary productivity, atrnospheric concentrations of ozone, or forest stand densities. The use of Geographic Information Systems for facilities management or locating facilities is also readily documented, as is the use of modeling for the prediction of the economic impacts of development. We have chosen to stress applications that bridge these technologies, in providing a perspective on how integration can achieve synergisms that facilitate our understanding and improve our ability to manage our environmental resources. The four exarnples discussed here are drawn frorn four different regions and concern: 'Y Land-Use Planning in Senegal: 'Y Crop Yield Estimation in lraly: 'Y Habita! Analysis for Preservation of Species Richness in the United Stares; and V Deterrninarion of Natural Hazards Risk to Agricultural Production in Ecuador. The first and fourth application exarnples are based on work undertaken within the "Monitoring of Tropical Vegetarion" and "Tropical Ecosystern Environment Monitoring by Satellitc" (TREES) projects of the Joint Research Centre of the Commission of the Europea Cornmunities (lspra. lraly). The second exarnple is a joiut projcci of (he Deparuncnt of Agromereorolog y of the Region of the Venero, ltaly, ami (he Uni versity of California, Santa Barbara. The third

46

47

example is based on cooperation among a range of govemmental State) and non-governrnental agencies in the United States'.

(Federal and

Results The analysis concluded that Senegal could significantly increase its production of rain-fed cereals, without exrensive costs in terms of natural resources, by applying improved technologies and by expanding cultivation. Specific findings included: T Existing rain-fed cereals alone could sustain 3.9 million people (56% of the Senegalese population). T By applying improved agricultural technologies, 5.4 million people (79% of the current population and 37% of the country's projected population for the year 2010) could be sustained from these crops. T By applying improved agriculture technology and expanding cultivation to potentially arable lands, approximately 7.7 million people (more than the current population and some 56% of the projected population for the year 2010) could be sustained, while preserving some natural vegetation. This information was used to define two strategic development goals: (1) sustained efforts in family planning; and (2) an increase in crop productivity through the use of new technology in zones of reliable rainfall. A number of other African countries face the same challenge of increasing food production to feed growing populations. In many cases, the natural resource and socio-economic data needed to carry out GIS analyses similar to the Senegal study are available.

Lond-Use Plcnning in Senegcl

lntroduttion Senegal is experiencing a long-term decline in per capita food production as population growth out-paces agricultural growth. In 1988, Senegal's population was estimated to be 6.9 million. Population is likely to double by 2010. Planners in Senegal face tough land-use decisions. They must weigh the need to increase food production against costs associated with the loss of natural environments to cultivated lands.

Approoch In 1990, the US Agency for International Development, in contact with Senegalese specialists, sought information on options for increasing rain-fed cereal production in Senegal. The EROS Data Center of the US Geological Survey performed the entire analysis on desk-top and mainframe computers. A GIS was used because of the complex spatial component of the data, the need to integrate numerous natural resource and socio-econornic data layers, and the need to analyze a wide range of production scenarios, with the Iikelihood that these analyses would be repeated in future years. The major cost involved was the time analysts spent entering data and maps defined by the GIS model. All of the data came from existing sources, including maps prepared from LANDSAT remotely sensed images, Senegal's national census, climate data and reports. The GIS contained data on natural resources, clirnate, demographics, poli tic al subdivisions, infrastructure, agricultural production and nutrition. Natural resources data included information on soils. vegetation, land-use and water. Agriculture data covered distribution of cropland, major crops, yields and production figures by crop, the caloric value of the major food crops and land-use rotation practices.

Estimcting (rop Yields in Italy

Introduction The Po River Valley of northern Italy is the breadbasket of the country, with crop production in the area exceeding US$ 20 billion per year (see Figs. 12 and 13). The Po River itself forms the southern boundary of the Regione del Veneto (RdV) before it flows into the Adriatic Sea. In 1985, RdV officials became convinced that they would need improved methods for compiling and analyzing agricultural statistics for the Regione in order to make effective agricultural and economic policy decisions in an increasingly complex ecological and economic environment.

1_

The exarnples which tollow in this section draw on material compiled by the authors for UNESCO Environment and Devclopmcnt Brief 3. This brief, entitled New Technologies: Remate Sensng and Geogrophic tnfonnation Systems, \\":1S produced as 3 joint effort by UNESCO and the Scicnrific Committee on Problerns of the Environment (SCOPE). Sorne support of personnel was also providcd by the US Nauonnl Acronautics and Space Administrations Earth Science and Applications Divisin. The l -page colourcd brief was published in mid 1992, and is available in English, French and Spanish frorn rhe Bureau Ior thc Coordination of Environrnent Prograrnmes, UNESCO, 7 place de Fontcnoy. 75352 Pan, 07 SP (France}.

ApproQ(h In 1987, RdV officials institutecl a co-operative research programme with the

48

49

University estimates.

of California, The systern that

Santa

Barbara,

to develop uses digital

better

crop

production image data,

cartographic

data and environmen.al the crop estimaies are required rernote sensing software processed grains

data frorn many other sources for the RdV, a mnimum per year. The project and GIS technology. T TM data crops grown The systern

as input into

was developed

satellite

GIS. To produce mercially available

of four LANDSAT for

TM image acquisitions

has made use of oorn-

The initial investment to provide

Veneto oreo: 1,836,819

ha

project hardware, Extensively information sugar-beet, agronomic, small

and data input has been in the order of US$ 45,000. LAl\DSA and orchards. are used primary in the RdV: com, soybeans, al so uses meteorological, of ground truth

on the five most proritable economic

and field validation

data. The collection

data represents

a signi ficant cost factor in the project.

Figure 12. Veneto reqion ond its provinces

Figure 13. Lond use by province

....
50

g~JfWS DE
51

Results When completed, the system \ViIIprovide estirnates of the total agricultural area within the RdV that is under cultivation for each of the five major crops. These estimates will be available on a biweekly or monthly basis at 75-95% accuracy. The information wil! be used on a regular basis to upgrade region-wide agricultural production forecasts which will become increasingly important as rnembers of the European Community develop closer economic co-operation. In addition, RdV officials will use the system 10 provide farmers with tirnely information on crop production, yields and profitability. Environmental engineers and scientists working with the RdV wil! also benefit from ready access to accurate spatially referenced information on chemical herbicide, pesticide and fertilizer use within the region. Forecasts of ag~icultural production are a key economic indicator. GIS using remo te sensing data for agricultural monitoring are being developed in a number of countries, including Italy and Egypt, and for regional areas such as the European Community. In the future, systems Jike that developed for the RdV are likely to become rnuch more widely used. Their accuracy should continue to increase with experience and as the models they employ are fine-tuned to local conditions.

Gap analysis places resource managers in a proactive rather than reactive mode in the preservation of biological diversity. Gap analysis builds from a prernise that unprotected species and cornmunities represent "gaps in our conservation safety net.

Approoch \York in Hawaii demonstrates the potential of gap analysis in assisting decisionmakers in their efforts to preserve diversity. This work involved cooperation by t he United States Fish and Wildlife Service (USFWS) and the Hawaii Departrnent of Land and Natural Resources. Other cooperating groups included private citizens, the Sierra Club, Hawaii Audubon Society, Conservation Council, Wildlife Federation, and the Nature Conservancy. In the sumrner of 1975, ornithologists in Hawaii were asked to identify the best area in the islands for an endangered forest bird reserve. Hawaiian ornithologists could not provide an adequate answer that would convince decision-makers to cornmit funds for the acquisition of the preserve. While there were several locations where endangered species were known to occur, there were far larger areas where there was no species distribution information. To correct this situation, the Hawaii Forest bird survey was begun in 1976. During the survey, 538 biologists and technicians surveyed 9940 stations and counted nearly 135,000 birds. These studies were conducted along 1401 km of transects systematical!y distributed over the forested areas of five islands. In assessing the need for new preserves, the first step was to plot species richness for endangered forest birds on each island. The next step was lo overlay on this base map a map of existing reserves (see Fig. 14). In doing this simple GIS analysis it was obvious that despite the large surface areas set aside on both Maui and Hawaii, there was little overlap with the ranges of endangered species. Prior lo establishing reserve boundaries, research and rnanagemenr biologists carefully reviewed the floral and faunal distribution and abundance data. Reserve designs were developed and forwarded lO the US Fish and Wildlife Service personnel, based on a nurnber of factors includinz: the distribuiion of current and potential habitats, future land-uses in the gener region. information on rhe prevalence and abundance of suspected limiting factors, and the need for a minimum viable size for the species of concern.

Biodiversity/Gap Analyses in the United States

lntredudien
Loss of species has become an issue attracting world-wide attention. Many times, however, we have focused our efforts on individual, highly visible "flagship" species, possibly at the expense of broader issues. Researchers and resource managers are coming to believe that it is more effective and cost efficient to focus on the preservation of intact functioning ecosystems with their myriad species than to continue to practise "ernergency room" conservation on one endangered species after another or wait until common species becorne endangered before acting to protect them. A challenge to the preservation of biodiversity is to plan future patterns of growth and rnodifications in land use to insure the survival of remaining biological diversity. This goal will not necessarily be reached by rescuing specific endangercd species but by keeping enough of the living world around to supply us with disease-resistant strains of crops. new medicines to fight disease and functioning watersheds that supply water for drinking and for irrigation. Gap analysis is esseruially a GIS based approach to facilitate the achievement 01' these objectives.

Results To date tive reserves have been established on various islands at a total cost 01' more than forty mili ion dollars. An cxample frorn the big islands of Hawaii is

52

53

illustrative densities privately

of the process. reserves, of endangered owned

After analysis it was found

of the map of species that the largest in the Hakalau slopes

richness

overlain

by the existing

nurnbers Forest,

and highest in an area of of

forest birds occurred

parcel on the mid elevation birds at Hakalau plants review including

of Mauna the highest considered

Kea. Densities

Island of
; ~ ..~

the more common area also included gered list. After a thorough proposed of privately project credit refuge of rare and endemic

were arnong several

in the state. The for the endanreview habitar of the protecwith this in

the habitar of the endangered

Hawaii species

hoary bat and a number

Hawaii

Endangered Species Richness 1 Species

of the project and alternatives approved

and a complete the acquisition Individuals in maps

public

KOh:;~-tt.

boundaries

for accomplishing

tion goals, the director owned

of USF\VS depictions

of some associated

113,360 ha

land in the Hakalau to convince

region.

...... ~

the graphic

presented

(such

as that shown

L....1 1972

.4

2 Species Species Preserves Preserves

113 Species

Fig. 14), for helping Such reserves ucts of gap analysis plant and animal now scheduled

their management

of the need for new reserves. Wise use of the prodthe projected losses in States is

D 1987
Hilo

are needed species.

in many areas of the world.

is one way we may be able to reduce By the year 2000, every a statewide gap analysis.

state in the United

to complete

Risk Assessment in Ecuador

Introduction
Ecuador's Gross floods agricultural Product sector, During which provides is vulnerable a high percentage to natural agricultural had become natural employment were mitigation losses of the country's particularly

National

(GNP),

hazards,

and droughts.

1982-83, oods caused

of more than US$ infrastructure. increasingly conwere affecting

110 million in damage to crops and the coumry's By 1990, Ecuador's Ministry of Agriculrure
cerned about the frequency proe!uction officials facilities and extent and relaied to which income, agricultural Although and other hazards teet secroral

hazards

o
Naalehu

10

20

30 km

and investment. roads to to promost vulnerable

were aware

of the problcm, should

they did not know which measures

used for agricultural

distribution implernent

or where the government incorne. cmploymeru

and investrnent.

Approoch
Frorn the outset, sheer nurnber it was clear an autornatcd and qu.uuitv

sysiem was requircd

beca use of the In 1991. with

01' variables

01' data to be analyzed.

Figure 14. Distribution 01 the endongered lorest birds ond preserves on the Islond 01 Howoii in 1987, showing the extent 01 the preserves in 1972 ond the new preserve on the eostern slopes 01 Mouno Keo

54

.55

ISSUES,
support work frorn the Department of American and spreadsheet infrastructures station of Regional States, software Development to analyze facilities, and Environment acquired data a desk-top for Ecuador's systems, and GIS the Organization provinces. networks exchange systern. the Government

N EED S

A N D TRE,NDS

20
road

For each province,

the GIS cornbined - storage erosion, indicators

data on 26 production irrigation f1oods, landslides employment,

systems, and seisforeign into the

14 crops and related mic hazards.

- with data on droughts, earnings, investment

volcanoes,

Data on five economic

- income,

and food security

- were also introduced

Results
The analysis required performed on this database identified 49 critical situations that A growing achieved operational arena. research trends This and number either status needed. of uses of the technologies within resource various within management discussed agencies in this digest community have or and acceptance areas of the scientific

action to mitigate

hazard vulnerability,

including

the following: area that, at the time of (Carchi). in another in Ecuador.

T Erosion hazards were likely to damage a potato-growing

the study, accounted 40% of the ernployment

for 43% of the national income frorn potato production, and 80% of the income in one province (Guayas), and erosion hazards

and the commercial and these technology utilization of these in the technolo-

does not mean The trends,

that there however, to enable section,

are not issues

yet to be addressed

T Flood T Drought
Based eration

hazards impacts

in one province

are very positive we address

(Tungurahua),

posed the greatest threat to agricultural on banana production foreign exchange the Ministry agencies earnings prepared

employment

will, we feel, continue In this closing

even more scientific

in El Oro province

posed the greatest technical coop-

technologies. gies including technologies.

some of the key issues and information

threat to Ecuador's proposal

from agriculture. a US$ 317,000 to finance hazards. assistance threat similar activities activito the to this

future development

and application

of data collection

on this information,

the role of the scientist

in the development

and wise use of these

which several

are now evaluating from natural hazards countries, policies

to reduce

the risks to agricultural forrnulated sector sector.

production natural

In addition,

the Ministry agricultural

new investment Since Latin

and technical analyses

ties in the agricultural

pose a major

Scales

in many

American

one could be used throughout

the region based on GIS technology. Issues Indeed, related to the use of information development analyses cannot and policy systems making technologies from for the studies of sustainable be dissociated the question are intimately between of scale. linkecl ro

scientific

processes

[he level of organization Concerns that analyses urnbrella Regional ulation be made

ro which they apply (see Fig. 15). and with linkages levels (i.e. Montreal its parts require under the etc.). pclluor popmigration, diviat global agreemenrs agreements and that actions Prorocol. of transpon plans, be taken 01' water,

with the Earth systerns

of international quesiions arising policies

[PCC, lGBP, gases, exploitation territorial

in the framework watershed

tarns or of intercountry

with respecl lO resource scales

(rnultinationa]

forest concessions, despire Local needs

etc.) are most often examinecl sions of auihority

al supra-narional of cornperences.

and irrespective

for inforrnation

57 56

about resources are more directly linked to well identified projects and the definition of precise courses of action. At each scale, requirernents for data and information will be very specific. Clearly, if the examination of the factors that enhance sustainable developrnent by their very nature integrates economic, political, institutional, legal, ethical and ecological considerations, then the relative importance of each of those factors will vary according to the adopted perspective. The perspectives of the individual - planners, resource managers, policy-makers - are coloured by the acculturation processes operating in the nations and regions involved. The role of science and technology will, sirnilarly, assume a variable irnportance depending on local, narional, regional and global imperatives. The question of scale also bears heavily upon the resolution or "grain" of the analysis and thus of the necessary data sets required to accomplish adequately a given task. As a rule, local investigations will need finer resolution data than a scientist, resource manager, or policy maker involved in a global analysis. Furthermore, although rapid advances are being made in this area, technology and limitations in our current ability to handle large amounts of data still prevent the systematic application of higher resolution observations and analyses to large regions, continents, hemispheres and the en tire globe. However, there may be specific needs to dissociate scale and resolution when, for exarnple, a highly accurate rneasurernent of a specific chemical compound must be obtained over a large area (ozone, pollutant, etc.). Thus, today, information technologies must still be managed through a set of compromises between scale and detail, and utilize data structures and algorithms which facilitate flexibility in anaJyses. A major advantage of the new information technologies is that the same techniques and instrurnents can, within some limits, work at different scales which can be "nested" into each other (see Fig. 16). This provides new opportunities for examining specific problems "across scales"; a local analysis may thus be easily set in a more regional context and, vice-versa, regional or continental issues can be substantiated using linked local analyses. Together, GIS, remate sensing and advanced modeling offer a means to extend significantly the breadth and depth of considerations brought into the decision-making process.

Geomorphic \ phenomena !

"

Forest management

Figure 15. The spotiol ond temporolscoles 01 vorious environmentol phenomeno

Infrastructure

lnfrasrructural investrnerus represent a key elernent in the successful adoption 01" new information technologies. Such investmenrs, which can be made al different levels 01' sophistication. are neee!ed to Iacilitate access to the tools, networks, systems and concepts discussed in this digest. Again. apart Irom obvious financial considcrations, the scale of concern and the nature 01' the problem to be

tackled wil! govern the choice of technologies acquired by institut ions. Essentially, investrnents in inforrnatics are nceded to acquire, display ane! anaIyze data corning Irorn sources as varied as saiellite sensors, existing maps, field surveys. sratistical tables . The rapid advance 01' cornputing capabilities over the last two decae!es has grcatly facilitnted access lO the new information technologies. Desktop cornputcrs and conunercial software are now available to support most types of projcci work. High data ratc communications nctworks are becoming more cornmon. Yct, training, maintenance ancl updaring of the hardware and software require carelul considcration as they involve rccurring cosis not always rcadily available in institutions. Until a Iew years ago, the balance of invcstments was hcavily slarued towards rhe purchase 01" basic harclware and software. Data acquixition and rraining now rcpresent the bulk of ihe necessary invcstmcrus involvcd in

58

59

ACQUISITION

1-"'--

IMPLEMENTATION

PREPROCESSING

implementing the technologies discussed here. Training is particularly important within the context of the needs of both the scientific and operational applications of these technologies. As both remote sensing and GIS software systems become more sophisticated, the need to understand their potential increases, even as atrernpts are made to make these systems less user hostile. Yet, institutions which must do such training are faced with the complex issues related to maintaining state of the art capabilities in areas of rapidly advancing technologies.

flE'.IlIlUlllll1llj~ SPECIALlZED

Aecuraey

PROCESSING In remote sensing, two components of the accuracy evaluation can be identified. The first one is related to the physical measurement itself, such as the one made by a satellite sensor recording the radiation reflected from the earth surface. The second refers to the accuracy of interpretation or transformation of such physical measurements into information related to either features or processes on the ground. Obviously, both the physical measurements and the derived information are linked. Yet, different types of research are required to improve our knowledge of these processes. Methodological developments in image analysis try to improve constantly the accuracy of interpretation. This accuracy depends, however, upon the type of information sought. For example, land cover maps can generally be produced with an accuracy of more than 90%. National assessments of forest changes can be produced with similar accuracy. Error assessment still represents a major research item in the development and application of the GIS technology. Data used in GIS-based analyses typicalIy must be collected from a variety of sources. Each of these data products will have been compiled to address specific and perhaps conflicting goals with respect to locational and thematic accuracy. Methods must be developed to assess the uncertainty, or error, associated with specific types of data products. Furthermore, a sophisticated understanding must be developed for the accumulation of various types of error as multiple data sources are combined into information products (see Fig. 17). What one must constantly strive to ensure is that the accur acy improves with respect to traditional field methods of data collection.

Figure16. Neltedsroles 01 observofien. Sompledesiqns employingthis conceptore being used to effectivelyextropolotelield doto orrossspore Meosuremenl Preeision 01 rneosurement Density01 observotions lbserver bias. (ompleteness Volidily Relevonce Stotilticolsignilicooce Dotomixingccnstruints Formal Digitolrepresentation Geometricreqistmfion 1nterpolotion/extropolotion Normolizotion Simulolion Simplilicotion Sub-gridsrole proresses Bilurcotion Confidence Senlitivity01 obiedive
[undions

Conlent Mothemolicollogic Spoticlconvolution Clolsificotion system Subiectivity Perceplion Reodobility Approprioteproienion Interprefotion

Stcndnrds

Unduetrust in technology

Standards are neecled for the transfer of spatial data in electronic formo As currcntly structured, spatial data sers typically carry an unaccept able overhead

60

61

LEVEL 1: Global AVHRR

Ileso/u/ion: 1.1 km

LEVEL 11: CONTINENTAL AVHRR Londsof multispecfrol scanner

when rapid error-free exchange between users is needed. More research is required here to facilitate distributed cooperative scientific investigations and sharing of data and information. Standardization of spatial data transfer formats will further encourage the combined use of remote sensing and GIS data sets. Finally, although the problems of operating in heterogeneous computing environments remain, this issue is being seriously addressed by major hardware suppliers. It is our belief that problems currently associated with the exchange of data and processing between different vendor systems will be substantiaJly mitigated within the next few years.

Ileso/u/ion: 1.1 km - 80 m

Timeliness

Generolized Vegelolion Clossificotion

/
LEVEL111: BIOME Londsof multispecfrol themotk mopper scanner

Ileso/u/ion: 80m - 30m

LEVEL IV: REGION Themofic mopper high oltifude oiruoft

~~
Bourdory Woters Conoe Areo

Ileso/u/ion: 30m - 3m+

/
LEVEL V: PLOT High ond low oltifude oiruoft

Ileso/u/ion: 3m+ - 1m+

Timeliness of access to data is a crucial characteristic of an information system. Requirements will obviously vary from one application to another (see Fig. 15). Crop monitoring or a search for locust breeding grounds will lead to more stringent temporal requirements than a general land use survey. Measurements of flooded surfaces will require more frequent data than a national forest inventory. In its early stages, remote sensing may have appeared to respond slowly to operational needs, and requirements of agencies, institutions and comrnercial enterprises. Today, advances in communication or in real-time local satellite data acquisition are constantly improving the potential for the "real enough time" use of these data. Commercialization of portions of the remote sensing sector has also contributed to improve the timeliness of data access even if the pricing structure currently applied to data by the commercial sector may strongly limit access to data for many applications. The problem of data dissernination is thus seen as two-fold: on the one hand, it is up to the users to rnake sure that cornmunication channels with data sources are optirnized for their specific needs. On the other hand, it is also the responsibility of nations and agencies which hold the key to satellite systerns and data archives to reduce as much as possible policy or practical restrictions lo tirnely data access. Another aspect of timeliness in producing the required analyses for sustainable development necessitates the mastering of techniques. This leuds us to ernphasize the need for the training of scientists ami technicians able to cope with evolving technologies. This can also be seen as an infrastructure issue.

LEVEL VI: SAMPlE SITE Surfme meosurements ond observofions

Figure 17 (Left). A((umulofion

01 error in on infegroted

rernote sensing

I GIS processing

flow

63

Te(hnoJogy Trends
Information applied quickly technologies of the kind proposed driven. in this digest may appear to many ingenuity can thus be seen as of may by a that is

as being intensively skills and know-how feel overtaken with advances, state of affairs. wanting, The situation sensing, three resource powerful respect

"high-tech"

Human

first to the design of instrurnents by events. educational

and tools and Iess to the nurturing of incapacity

to use them. The trend is clear and field practitioners

Such general feelings their application, and developmental

to keep up the current

let alone master At this stage, is further

can only be mitigated effort to reverse to work. changes the technology

major and concerted but perhaps

it may not be so much by constant in Table

more our ability to put it properly eompounded

in the technical here (Remote in rhese and a and mearepresent made with

base. Current trends in the three major technology GIS and modeling) amply trends. managers means are surnmarized areas have dernonstrated

areas discussed

5. Techniques
detect changes, technologies

their capability of these

to aid researchers

as they monitor of verifying

the earth surface,

sure and evaluate

The combination of resources

that environmental

commitments

to the conservation

have been rnet. Whether

we like it or nations data which can responsibility and ethical

T T

not, those techniques accountable

offer a possible sustainable

verificationmechanism development social, economic,

for holding

to those commitments.

The quality of intemational is a shared political of information.

help chart the path toward gies have far-reaching

which requires the free circulation scientific, implications.

This topic and these technolo-

Role of Sdentists
Table 5. Curren! !rends in remoJe sensing, GIS ond modeling
It is important widely business ficlds tivcly

that the role of seienee and technology

A corollary

in human

affairs be more is ihat the scito significantly tobe pnaie upgrade its familiarity with tcchniques which have been shown and applying appro10

known and beuer undcrstood. of science is about.

to that statement

eruific community holds a special responsibility

and kceping of investiuatinn. Currcntty.

for communicating

what the
of dis-

most efficient

and effective

in collecring, organizing

up with advances

in a wide variety technologies

data to a parl~cuJar field of investigation. that their resolution at scales \ViII necessitaie which.

One migh: cven venture will increasingly

say

the data and inforrnation Cross dsciplinary uuined

rhat the problerns intractable

of development and environment

become so

cussed in this digest are liulc known and,

smull number 01' scicmisrs. appcaling
10

10 date, are truly understood

by a rela01'

the integrarion and the use of sciliule consider.uion c1evclopto steer

approaches are. as a rule, scientists. This ignorance scientific and rcsourcc or cxcessivc in various rcsponsibility

ence and technology The scicntific

as yct, have reccived

lO

not dirccily the cupability optimism

more classically of tools

community
Finally,

has an additional

rolc lO play in Inaking sure that rhcsc problerns

and limit.uions
lO

by rhe broader

rcchniques
mcnt

are properly

tuned ancl adjustcd

scientists

of xustainahlc
as a communiry,

managcmcnt community

is ocn the source rcality. dcvclopment

of unclue scepticism commuuity involvcd the particular

al all scales.

I11USl endcavour,

elernents

0'- sustainablc

with respect

Thc scicntific

has, thcrelore,

advanced technology human necds.

programmes

of the furure in dircctions

which truly SCI've

64

65

With such a perspective, cance. Indeed, inforrnation in the data analysis not reserved around tainable globally scientists altematives they are increasingly development distributed
10 a particular

technology systems

transfer process.

assumes

a particular

signifiare

SOURCE

MATERIAL

100ls now permit significant Technological scientists among discussed

decentralization advances any more; to susfrorn of

and decision-rnaking

country or a specific

class of scientists

within the reach of individual

objectives would

or institutions others,

the globe. A true contribution in the accessibility

by the world scientific thus derive,

community

improvements

of those techniques

herein to the and field situatechnical to achieve the

science community

as a whole. In retum, the appropriateness

farniliarization

and technicians

with a diversity of scientific concems

arising out of our attempt

tions can only work toward irnproving to the solution of problerns development. goal of sustainable

of proposed

American

Society of Photograrnmetry

and Rernote Sensing.

1991. Special Issue Engineering

on Integration of Rernote Sensing and GIS. Photograrnmetric and Remate Sensing 57(6):641-697. Burrough, P.A. 1986. Principles Assessment. o/ Geographical Information Resources Clarendon Press, Oxford. The Handbook Sensing, Sensing,

Systems for Land of Artificial edition.

Cohen, P.R. & E.A. Feigenbaum (Eds). 1982. lntelligence. Heuris Tech Press, Stnford.

Colwell, R.N. (Ed.).

American Virginia. Society

1983. The Manual of Photogrammetry

of Remate

Second

and Remote information

Falls Church, Journal and

Estes, J.E. 1985. The need for improved of Remate Sensing 11(2): 124-131. Goodchild, M. & S. Gopal. Francis, London. Gennany. Gesellschaft

systems.

Canadian

1989. Accuracy

of Spatial

Databases.

Taylor

fr Technische

Zusammenarbeit. Gesellschaft lmage

1991. Remote

Sensing:

Tools fo r Development. Deutsche Zusammenarbeit (GTZ), Eschbom. Jensen, J.R. 1986. Int roductory Englewood Cliffs, New Jersey. Monmonier, Prospects. M.S. 1982. Prentice Di gital

f r Technische
. Prentice Principies Hall, an d

Processing

Com put er Assisted Sensing Digital

Cartograpliy. Image Analysis: Systems: Commiuee.

Hall, Englewood

Cliffs, New Jersey. An lntroduction, Anlntroduction, 1990. Spatial National

Richards, J.A. 1986. Remate Springer- Verlag. Berlin. Prentice Hall, Englewood

Star, J.L. & J.E. Estes. 1990. Geographic United States Geological Survey

lnformation Science

Cliffs, New Jersey. - Mapping Mapping Prog ram,

Dota Needs: The Future o/ th e National Academy Press, Washington, D.e.

66

Fecha de Vencimiento
United States Geologieal Survey - Mapping Seienee Comrnittee. 1991. Research and Development in the National apping Division, USGS: Trends and Prospects. National Academy Press, Washington, ~.c. Weiss, S.M. & C.A. Kulikowski. 1984. A Practical Guide to Designing Expert Systems. Rowman and Allanhead, Totowa, New Jersey. Winston, P.H. 1984. Artificial Intelligence. Seeond edition. Addison- Wesley, Reading, Massachussetts.

l3 ftt
12;M,
r,
1

'I~~O

I
.\

~ '" 'i uM,U::'

D ~f';n-" '0< H'n "~'()(iJc';;U . ,.... v

eo

"MI1.

..--lUUU

..
I
!

Suggested Reading

!
Aronoff, S. 1989. GIS: A Management Perspective. WDL Pub1ieations, Ottawa. Cowan, DJ. 1988. GIS versus CAD versus DBMS, what are the differenees? Photogrammetric Engineering and Remote Sensing 34:15551-55. Eeologieal Society of Ameriea. 1991. The future of remote sensing in eeologieal studies. Special feature. Ecology72(6): 1917-1945. Fritsch, D. ; Bill, R. 1991. Geoinformationsysteme, Wichman Verlag, Karlsruhe. Guptill, S.C. (Ed.). 1988. A Process for Evaluating Geographical Information Systems. USGS Open File Report 88-105. Federal Interageney Coordinating Committee on Digital Cartography Teehnieal Report l. Mounsey, H.M. (Ed.). 1988. Building Databases for Global Science. Taylor and Franeis, London. Parker, H.D. 1991. International GIS Sourcebook 1991-1992. GIS World Ine., Fort Collins, Colorado. Peuquet, DJ. & D.F. Marble (Eds). 1990. Introductory Readings in G/S. Taylor and Francis, London. Raper, J .F. (Ed.). 1989. Three Dimensional Applications in Geo graphical Information. Systems. Taylor and Francis, London. Rhind, D.W. & N.P.A. Green. 1988. Design of geographieal information systerns for a hererogcncous scieruific eommunity. lmernationol Iornal of Geographical Injorrnatipn Systems 2: 171-190 .. Rouet, P. 1991. Les donnes dans les systmes d'information gographique. Editions Hcrrnes, Pars. Seanvie. J. Y. 1993. Utilisation de la tldtection dans les sciences de la [erre. Manuels el Mthodes, 7. BRGM, Orlans. SIUlh, J.W. & B.G. Lyons (Eds). 1993. Decisin Support Systerns for the Management of Grazing Lands: Emerging lssues. Man and the Biosphere Series 11. UNESCO, Pars, and Parthenon Publishng, Carnforth. UNESCO. 1992. New Technolo g i es: Re m ot e Sen sin g an d Geo g raphi c lnjormation Syst ems. UNESCO Environment and Development Brief 3. UNESCO. Pars.
'c

:
I I

I
1

!
1

i
1

68