EGIS (1994), copyright EGIS Foundation.
Among the wide range of Geographical Information Systems (GIS) oriented to environmental planning purposes the number of GIS that deals with coastal areas is very limited.
From the user's point of view the confidence is very high in the power of GIS technology to elaborate data with different origins and assemble them in a unique and homogeneous structure as a database. This confidence creates very high expectations from decision-makers but nevertheless «the volumes of data involved, the 'fuzziness' of many geographical data, the choice of methods available which can give different results and the need to do all this economically and to meet (usually tight) deadlines renders the practice of GIS much more complex than the principles» (Rhind, 1991). This seems to be the present limit to environmental GIS: R&D advances are required in this topic.
This contribution is intended to describe and briefly discuss - as case study - the wide and complex range of historical records available regarding the impact of 'coastline changes' and 'mucilage and algae blooms' in an Adriatic coastal area, between the 16th and 19th century.
The analysis is based on a large amount of sources dispersed in many local offices. The study regarding the phenomena makes it possible to improve the understanding of the connections between human actions and the ecological crisis of recent decades. In this dynamic coastal environment we intend to analyse in particular the capacity of various documents and historical records to show the complexity and the frailty of this environment.
Attention is especially devoted to evaluate the existence of different sensitivities in the wide range of documents concerning the phenomena. Can present or coming GIS technology account for that complexity and maintain it for the users?
The theoretic framework of GIS, especially of GIS focused on environmental issues, recognises the role of traditional maps as main sources - in fact «most of the contents of geographical databases continue to be derived from paper maps» (Goodchild, 1991) - but, at the same time, needs to observe and monitor the continuos changing of the phenomena. «An increasing demand for time series analysis of environment phenomena (...) is driving developers to consider adding temporal data to GIS. Current systems are poorly equipped to handle temporal data and models of spatial succession are too general if they exist at all (...). Historical data are difficult to obtain, often collected at varied intervals and varied scale (...)». (Aangeenbrug, 1991).
[End Page 96]
For example, in the state of the art regarding the study of problems of coastal waters and historical shoreline changes, the observation time is a basic concept (Grenier Dubois, 1992; Marchetti, 1992). This situation has been recently enforced as an effect of the adoption of EEC directive 85/377, regarding the introduction of Environmental Impact Assessment (EIA), in many ways dependent on the study of historical sequences, for public and private projects of planning activities.
The use of maps for GIS implies an analysis of the general aspects of the graphic means and of the peculiar features of each map. Special attention must be paid to identify the sources and means used by the cartographer. This assumption seems widely, although not generally, accepted: «the compiler of a database formed by digitising maps may need to consider the purposes for which those maps were created» (Fisher, 1991). Some modern human geographers are deeply involved with such a topic, developing, more or less consciously, some reflections of the Erdkunde (Farinelli, 1992). As Barnes and Duncan wrote: «(...) the purpose of maps, in spite of the rhetoric of many positivistic cartographers, is not mimetic (...), but to communicate ideas within a cultural and political context. Therefore the questions that we should ask (...) are: under what cultural and political conditions did such a representation seem useful, and what rhetorical devices did it use to communicate? I...) It is precisely the problem of representation as it pertains to the geographical world f...I.... geographical writing is not a faithful duplication of an external reality. But how might we read the metaphor?» (Barnes - Duncan, 1992). On the assimilation of the chart with the metaphor, a debate is presently open which seems interesting also for the GIS theory: «a map is a symbolic representation of the world (...). It cannot be an analogy for the world because the relationships it models are those of the real world and no others. Nor we can convert it into a metaphor by adopting a naive cartographic realism» (Graham, 1982). David Harvey suggested a way to 'read' the map: «constructing a map without explicit theory about real-world amounts to stating an a priori model; with explicit theory it amounts to stating an a posteriori model. The a posteriori model, however, can be related only to the domain of the theory which it represents. Using a particular a posteriori model to examine phenomena outside the domain of the theory which it represents amounts either to assuming the domain can be extended to phenomena not initially covered by the theory, or to using the a posteriori model. as an a priori model>> (Harvey, 1969).
Using historical maps for environmental analysis based on GIS, we should add - to the above indicated suggestions - that present GIS «illustrate the effects of temporal change as vignettes (slices of time) for discrete intervals, but we need to show dynamic change over continuos time» (Dobson, 1993).
The coastal zone - as a contact area of different environments and heavily transformed territories having different uses and regulations - seems to be a paradigm representing the gap between the theoretic framework of GIS software and the specificness of the coastal zone (Bartlett, 1991). Such condition requires the availability of «reliable and trusted information (...) to the resolution of environmental conflicts. However, in the present state of the art, coastal zone management still lacks an adequate environmental knowledge-base» (Jollife - Patman, 1985).
In theory there is a widespread confidence in the power of GIS technology to make different data uniform using a database, and to display them with digital cartography.
[End Page 97]
«Modes de levers et de dessin des cartes, sensibilites diverses des auteurs liees a des ecoles differentes et travaillant dans des finalites diverses (par exemple hydrographique ou topographique) creent parfois des dysharmonies meme entres cartes de meme echelle. Le recours a un filtre dimensionnel constant, facile a assurer par l'usage conjugue de la teledetection satellitaire et de l'informatique permet fort heureusement d'eliminer de facon universelle ce risque d'incoherence. II offre en outre l'avantage de faire reflechire au niveau optimal de resolution des donnees que l'on doit croiser avec le images de la Terre, afin d'etudier toutes les interactions ecologiques et economiques dont le littoral est le theatre» (Verger, 1993). This confidence is not restricted to the hypothesis of presenting quantitative data in a uniform grid but is also devoted to the qualitative aspects in view of the integration of the cultural factors in a "unified landscape": «GIS is a technology in extreme need of a cultural perspective during their development. (...) GIS are much more flexible than what can be considered a simple tool - a GIS is not as simple as a hammer - and aims at spatial integration of data. It thus hits problems of cultural differences squarely» (Campari - Frank, 1993).
As far as we know none of the very few GIS devoted to coastal areas seems to examine the problem of the non-geodetic maps, which include a large part of maps of the 16th-18th century (Fleming - Townend, 1989; Townend, 1990; Clark M. J., et Al., 1990; Bartlett, 1990). The same applies to studies which use digital analysis of the movements of the coastal line over long periods of time. The need for computer processing of the information in order to automatically design maps confronting the different coastlines, forces us to use only geometric maps and process them in order to adapt then to a unique geometric frame (Demirpolat et Al., 1989; McBride, 1989). But human geographers (Koeman, 1968; Harley, 1968; 1985) have elaborated an accurate methodology in order to evaluate historical maps with «... a variety of ways to assess map accuracy. Most of these techniques involve the accurate placement of cultural and topographic features on large-scale maps» (Locke - Wyckoff, 1993). This method was used in "traditional" studies on the historical changes of the coastlines, therefore including in its various sources also the pre-geodetic cartography (De Boer - Carr, 1969; Car, 1969; Ren, 1992). The historical approach is also used in the field of biological sea sciences and, in particular, in the multidisciplinary research devoted to the long-term observations of phenomena as "mucilage and algae blooms". (Molin et Al., 1992).
Thus the problem seems to arise when computerised tools are introduced in the attempt to make very different cartographic documents comparable: «map projections may vary and maritime charts may use an altitude datum different from that adopted by the topographic survey (...) disparate definitions of coastline may be adopted by the map and the chart, so that substantial manual intervention is required to bring the two data sources together. (...) The task of producing significant convergence in coastal classification and subsequent data modelling for GIS design (...) involves applied mapping conventions which have defied true standardisation over two decades. (...) Perhaps the ideal of a common basic data structure is unattainable and the compromise of more effective and efficient data transfer may have to take its place, but there is an inevitable loss of advantages in the medium term if this is the case» (Clark et Al., 1990).
This situation has been evaluated in reality on the occasion of the attempt performed in Great Britain by the Ordnance Survey and the Hydrographic Office of meeting
[End Page 98]
<<the need for consistent mapping that bridge land and sea» (Curtis et Al., 1993) A coastal zone map is identified as a basic factor for a global.... This attempt had to be confronted with highly relevant problems, which are mainly induced by the fact that the territory is not a datum but an "elaboration" and that data are "theory laden" (Hanson, 1958; Moretti - Naddeo, 1992). Therefore, in the attempted computerised assemblage of the coastal zone map from the Ordnance Survey G.B. and Admiralty charts, it was acknowledged that: <<Early ideas suggested combining digital data from OS and digitally-extracted chart data from the HO. A digital map was a desirable end product but a paper version was also essential (...). Datum decisions were important, both horizontal and vertical, since the OS maps and the HO charts are constructed from very different bases. (...) Some simplification was necessary when it came to symbology on the prototype to avoid too much detail. Admiralty charts, as sailors know, have a wealth of navigation detail, and the symbols and abbreviations from them are listed and explained in a 75-page publication, Chart 5011 (Curtis et Al., 1993).
Today, the awareness of the limits of modelling, especially in environmental studies, is well rooted: «A major complication is that the environment can rarely be treated as in a laboratory experiment>> (Rhind, 1991). This, also in consideration of the fact that while we need global coverage of data for many of our purposes, however, "even ignoring the effects of humans, earth system processes operate at a variety of spatial and temporal scales" (Rhind, 1991). The case studied attempts a first historical comparison between different natural phenomena of the sea and coast environment, such as 'coastal line changes' and 'mucilage and algae blooms'. The two phenomena are evident in the historical records for the last centuries. The first step of the research has dealt with - on assonance with GIS philosophy - the role of cartographic sources in the understanding of environmental processes.
The research focused on a coastal area of the Northern part of the Adriatic sea. This area has been interested by a strong dynamism of the environmental phenomena, a complex interplay between human activities, coastal environment and hydrography and by an increasing presence of human activities which had and are still having a strong impact on the environment. It seems difficult that the «comprehensive knowledge of the resources and uses of the coastal and marine environment (which) is required before any fair resolution of these conflicts between environmental conservation and developments can be achieved" (Ricketts et Al., 1989) can leave out the extension of the observation time to a period when non geodetic cartographic sources were produced.
A first survey on coastal line changes was carried out on the coast near the Po di Primaro river mouth (since 1767 the Reno river was deviated into the Po di Primaro river bed). The representation of the historical evolution of the coast line of this area using the present knowledges - which are mainly determined by the study of some historical cartography - in a possible coastal GIS, should register a marked advancement of the coast line in the 16th-18th century. A first result of the study in progress, conversely, reveals an alternating trend where extension periods alternate with periods registering a definite erosion of the coastline. Figure la, obtained from a study published in 1601 by G.B. Aleotti, a water expert from Ferrara - shows a remarkable extension towards the sea of the Po di Primaro river mouth. This
[End Page 99]
drawing attracted bitter criticism from the experts who supported competing hydraulic regulation projects, as a voluntary exaggeration of the progress of the river flow towards the sea, while not depicting some important tracts of the coast line. The coast line advancement trend, however, is sustained by the comparison with the historical cartography prior to the above mentioned drawing. From the graphic point of view, the extension represented by G.B. Aleotti shows a large variety of marks, which are not strictly linked to precise geometrical references thus being difficult to manage for a present GIS software; a graphicism aiming at the representation of a not neatly delineated morphological reality.
The profile (Fig. 1b) dated August 20th 1632 shows how, also in the past, different procedures and techniques originating a wealth of information were used in the representation of territories involved in complex anthropical environmental problems. While Aleotti (Fig. la) uses a zenithal perspective, the profile is built following the section criteria. All this is going to make a possible geo-reference operation difficult, but - as it appears from the image - to overlook the information in the drawing would mean to lose an important chronological factor. In fact, if in 1601 the sand banks near the river mouth are represented as not well defined formations emerging from the sea surface (Fig. la), in the drawing dated 1632 (Fig. 1b) these sand banks appear to be submerged, even though moderately if compared to the mean level of the low tide (Fig. 1b line c). It must be stressed that the two drawings are the result of a series of field surveys. The study of the cultural context in which these two drawing originated, however, induces us not to attribute an objective character to the topographic outlines. In addition to the already mentioned reservations on the Aleotti's drawing, as far as the profile is concerned it was the result of technical operations carried out after debate among experts appointed by the Administrations which, at that time, were interested in interventions on the river. Unfortunately, so far, it was not possible to obtain the documents connected to the profile preparation (Fig. 1b). In view of the historical documents examined, however, it can be suggested that the traced lines represent not so much a faithful copy of reality but a compromise amongst different measures and conflicting interests.
These problems are also evidenced by the comparative survey of figures 1c and 1d. The first was obtained by an anonymous map dating back to the last years of the 16th century. The drawing evidences the extension of the Po di Primaro river mouth. The computerised processing of the image by scanner and video graphic processing, in this case, tackled a very complex problem connected to the specific graphic quality of the original drawing. The presence of either sand banks or many sedimentary formations near the river mouth is not marked by borderlines but rather by blue colour changes associated to the use of specific place-names (Fig. 1c: sacha de Primaro, scanarea, la scaia). The dashed lines in the drawing represent a rather inaccurate attempt to delimitate areas using different colours. Information concerning bathymetry and tidal trends near the coastline, at north of the river mouth, are offered in the original drawing by the use of different colours coupled with borderlines and written text. Once more, the computerised processing of such an image, in addition to including judgement ambiguities connected to the vision, evidenced the difficulty of computerised tools to capture and present such a faded information content without simplifying, and thus causing a falsification (Monmonier, 1991). To overlook this source would have caused the exclusion of a wealth of information concerning this section of coast, thus making the content of figure 1d obtained from a large map approximately on scale 1:15000 (dated 1657) - less
[End Page 100]
understandable. Thanks to the work on the written sources it appeared that this drawing was the result of very accurate field surveys which took place over a decade (1647-1656). This map tried to offer a detailed picture of the topographical elements aimed at the negotiations for the composition of a jurisdictional conflict between the Roman Church State and the Dukedom of Modena. The fine drawing represented a starting point for the negotiation: the map was plotted, accepted and officially approved by the two parties. The study of the diplomatic reports and documents which illustrate the negotiations and the topographical work, evidences how, this accurate representation is also the result of choices and mediation which caused the exclusion of some topographical data in favour of others. Moreover, even though not directly perceived from the map, the documented existence of different technical sensitivities should not be overlooked. In fact, a map is not only «a way of graphically expressing the cartographer's mental concepts and images but rather the result of an individual having mastered the rules and conventions and having acquired the considerable skill of map drawing» (Graham, 1982). In this case, luckily enough, the map and related archive documents provide very useful information. The profile of the river mouth is marked by a dotted line which - as explained in the legenda - marks the part of the land eroded by the sea. The moving back of the coastal line is also evidenced by the indication of the previous topographic localisation of the Torre Gregoriana (Figs. la, 1b, 1c, 1d, *) built on the mainland after 1572 and totally submerged by the sea when the map was being drawn up.
The subsequent events concerning this section of coast - which are rather well reconstructed thanks to geodetic cartography and historical research - lead to the situation represented by Fig. 1e.
The case of the sand-banks of the Po di Primaro, with their intermittent presence in the maps, seems to be confirmed by written records: Luigi Ferdinando Marsigli (1715) wrote about the wide presence of sand-banks along the Adriatic coast: his notes were based on direct observation and on reports of fishermen. The attention to the presence of sand-banks continued in the XIX century, when the changes of the water quality in the Adriatic sea and the presence of algae blooms and mucilage were related to forming of sand-banks. In fact, in our opinion, an analysis of the historical evolution of the phenomenon of dirty sea (mucilage) can provide some new elements for use in the evaluation of a rather complex interaction between a natural phenomenology and the alteration of anthropic contributions. On account of the close examination of historical documents we discovered that various aspects of the cultural and social life of these ancient coastal communities contain useful indicators as regards the phenomena that ruined the fishing activities: descriptions of various kinds of special procession and collective rites, supplications, accounting and fiscal books. In this case, the ethnological and cultural anthropological contribution integrated the historiographic research data, the same isn't true of historical cartography contribution.
The two processes we observed seem to have different impact on the maps. Coastline changes are directly related to a topographic, and therefore cartographic, element and consequently they are represented on maps. To that reason we should add others - sometimes more relevant - as political actions related to the control and planning of resources, which, in such a territory, were almost connected with interventions on the hydrographic network. In this context the maps have the role of enforcing specific and often conflicting plans.
[End Page 101]
As a provisional conclusion we can refer to a concept expressed by Wallin: <<In traditional maps, based on the paper as a storage, a distribution and presentation medium, geographic space has appeared as a rather passive arena with no actors and with no play. (...) With the help of modern information technology phenomena could very well be dealt with as in a good story or a great drama...» But, from the study of this case the survey of the historical cartographic sources seems to be not an objective mimesis of reality but as the fruit of both a culture and the specific observation and interpretation mode of the world, thus permitting to trace "actors', "play", "story" and "drama". Likewise, to leave to a technology the attempt to foresee and draw the "next chart" (Gould, 1993; Weizembaum, 1976), besides the serious risk of excluding a wealth of historical information, may induce the situation - imagined by Neil Postman in 1992 - of a technology-born information, so extended, differentiated and dynamic, which does not permit to be organised in a coherent program. Last but not least, from the point-of-view of the researcher confronted by the "fuzziness of the coastal boundaries (which)... introduce another level of difficulty to the coastal data modelling process since movement in space also implies change over time» (Barlett, 1990), the risk exists that «computer-assisted research performed without fantasy and far- sightedness proves to be antithetic to the free exercise of that happy condition called serendity, that is the gift of reaching important results more or less by chance» (Lowell, 1984).
Aangeenbrug R.T. (1991), A critique of GIS, in: D.J. Maguire - M. Goodchild - D. Rhind (eds.), Geographical Information Systems. Principles and Applications, Longman, v. 1 pp. 101-107.
Barnes T.J. - Duncan J.S. (eds.) (1992), Writing Worlds. Discourse, text & metaphor in the representation of landscape, Routledge, London and New York,.
Bartlett D. J. (1990), Spatial Data Structures and Coastal Information Systems, "EGIS `90, First European Conference on GIS", pp. 30-39.
Campari I. - Frank A.U. (1993), Cultural Differences in GIS: a Basic Approach, "EGIS `93, Fourth European Conference and Exhibition on GIS", pp. 10-16.
Clark M.J. - Gurnell A.M. - Edwards P.J. (1990), A GFIS Approach to Management Decision Making for the Coastal Environment, "EGIS `90, First European Conference on GIS", pp. 189-198.
Curtis M. - Harper B. - Fellingham W. (1993), Is there a Mart for these Chaps?, "GISEurope" April, 26-28.
De Boer G. - Carr A.P. (1969), Early maps as historical evidence for coastal change, "Geographical Journal", 135, pp. 17-27.
De Boer G. - Skelton R.A. (1969), The earliest English chart with soundings, "Imago Mundi", XXIII, pp. 9-16.
Demirpolat S. - Tanner W.F. et Al. (1989), High-Precision Study of Florida Shoreline Changes, "Coastal Zone '89, Proceedings of Sixth Symposium on Coastal and Ocean Management", American Soc. of Civil Engineers, New York, pp. 683-697.
[End Page 102]
Dobson G.E. (1993), Commentary: A Conceptual Framework for Integrating Remote Sensing, GIS, and Geography, "Photogrammetric Engineering & Remote Sensing", LIX, n. 10, pp. 1491-1496.
Farinelli F. (1992), I segni del mondo. Immagine cartografica e discorso geografico in eta moderna, La Huova Italia, Firenze.
Fisher P.F. (1991), Spatial data sources and data problems, in: D.J. Maguire - M. Goodchild - D. Rhind (eds.), Geographical Information Systems. Principles and Applications, Longman, v. 1 pp. 175-189.
Fleming.C.A. - Townend I.H. (1989), A Coastal Management Database for East Anglia, "Coastal Zone '89, Proceedings of Sixth Symposium on Coastal and Ocean Management", American Soc. of Civil Engineers, New York, pp. 4092-4107.
Goodchild M.F. (1191), The technological setting of GIS, in: D.J. Maguire - M. Goodchild - D. Rhind (eds.), Geographical Information Systems. Principles and Applications, Longman, v. 1, pp. 45-54.
Gould P. (1993), Why not? The search for spatiotemporal structure, "Environmental and Planning A.", Anniversary Issue, pp. 48-55.
Graham E. (1982), Maps, Metaphors and Muddles, "The Professional Geographer", 34 n. 3, pp. 251-260.
Grenier A. - Dubois J.M.M. (1992), Use of remote sensing for monitoring recent coastal change: an overview of methodology, "Photo-Interpretation", 1990-6, pp. 3-16.
Hanson N. (1958), Patterns of Discovery. An Inquiry into the Conceptual Foundations of Science, Cambridge University Press, Cambridge.
Harley J.B. (1968), The evaluation of early maps: towards a methodology, "Imago Mundi", XXII, pp. 62-74.
Harley J.B. (1985), The Iconology of Early Maps, in: C. CLIVIO MARZOLI (a cura di), Imago et Mensura Mundi, Atti del IX Congresso Internazionale di Storia della Cartografia, Instituto della Enciclopedia Italiana, Roma, pp. 29-38.
Harvey D. (1969), Explanation in Geography, Edward Arnold, London.
Jollife I.P. - Patman C.R. (1985), The Coastal Zone: The Challenge, "Journal of Shoreline Management", v. 1 n. 1, pp. 3-36.
Koeman C. (1968), Levels of historical evidence in early maps (with examples), "Imago Mundi", XXII, pp. 75-80.
Locke W.W. - Wyckoff W.K. (1993) A Method for Assessing the Planimetric Accuracy of Historical Maps: The case of the Colorado-Green River System, "The Professional Geographer", 45, n. 4, pp. 416-424.
Lowell B., cit. in: Postman N. (1992), Technopoly. The Surrender of Culture to Technology, Knopf, Hew York, p. 1 12.
Marchetti R. (1992), The problems of the Emilia Romagna coastal waters: facts and interpretations, in: R.A. Wollenweider - R. Marchetti - R. Viviani (eds.), Marine Coastal Eutrophication, Proceedings of an International Conference, Bologna, Italy (21-24 March 1990), "Science of the Total Environment", Supplement 1992, Elsevier Science Publishers B.V., Amsterdam, pp. 21-33.
Marsili L.F. (1715), Osservazioni di varie cose appartenenti alla storia naturale del mare fatte sulla spiaggia pontiFlcia dell'Adriatico.... Fondo Marsigli, Ms. 71 (a), Biblioteca Universitaria di Bologna.
McBride R.A. (1989), Accurate Computer Mapping of Coastal Change - Bayou Lafourche Shoreline, Louisiana, USA, "Coastal Zone '89, Proceedings of Sixth
[End Page 103]
Symposium on Coastal and Ocean Management", American Soc. of Civil Engineers, New York, pp.707-719.
Molin D. - Guidoboni E. - Lodovisi A. (1992), Mucilage and the phenomena of algae in the history of the Adriatic: Periodization and the anthropic context (17th-20th centuries), in: R.A. Wollenweider - R. Marchetti - R. Viviani (eds.), Marine Coastal Eutrophication, Proceedings of an International Conference, Bologna, Italy (21-24 March 1990),"Science of the Total Environment", Supplement 1992, Elsevier Science Publishers B.V., Amsterdam, pp. 511-524.
Monmonier M. (1991), How to Lie with Maps, The University of Chicago Press, Chicago and London.
Moretti A. - Naddeo D. (1992), Un orizzonte efico per i modelli territoriali, "Archivio di studi urbani e regionali", XXIII, n. 44-45, pp. 187-204.
Postman N. (1992), Technopoly. The Surrender of Culture to Technology, Knopf, New York.
Ren M.-E. (1992), Human Impact on Coastal Landform and Sedimentation - The Yellow River Examples, "GeoJournal", 28, pp. 443-448.
Rhind D. (1991), Geographical Information Systems and Environmental Problems, "International Social Science Journal", 130, pp. 649-668.
Ricketts P.J. - McIver A.R. - Butler M.J.A. (1989), Integrated Information System The Key to Coastal Zone Management, "Coastal Zone '89, Proceedings of Sixth Symposium on Coastal and Ocean Management", American Soc. of Civil Engineers, New York, pp.4138-4150.
Townend I.H. (1990), The application of GIS to coastal zone management, "EGIS `90, First European Conference on GIS", pp. 1098-1105.
Verger F. (1993), La perception geometrigue du littoral, "Mappemonde", 1, pp. 2-3. Wallin E. (1990), The Map as Hypertext, "EGIS '90, First European Conference on GIS", pp. 1125-1133.
Weizembaum J. (1976), Computer Power and Human Reason. From Judgment to Calculation, W. H. Freeman & C., San Francisco.
[End Page 104]
[End Page 105]