Scope 47- Long-term Ecological Research, An International Perspective

8.1 INTRODUCTION

Few ecological systems are as variable as fluvial landscapes. The hydrological characteristics which rule these landscapes fluctuate in an as yet often unpredictable manner at different time scales. The settlement of human societies along alluvial plains, which has produced numerous changes in the past, is a constant cause of continuous, and sometimes acute, change. Clearly, the dynamics of a fluvial landscape, with its constant fluctuation and change, can be understood only through long-term research.

This chapter focuses on the ideas that long-term ecological research must take into account to provide an historical perspective of fluvial landscape dynamics, and at the same time to be included in a global and comparative perspective. The first part of this chapter gives the results obtained on the River Garonne, France, as an example of the importance of an historical perspective. The second part discusses the necessity of global and comparative perspectives. The third and fourth parts of the chapter discuss four questions to be addressed by long-term ecological research on fluvial landscapes, and the possibility of monitoring by remote sensing.

8.2 AN HISTORICAL PERSPECTIVE

An historical perspective is particularly necessary in river ecology. Rivers and their floodplains have been utilized by man for a long time, and many human societies have settled and developed along large rivers. Fluvial landscapes were early utilized for transport, water and fish availability, and fertility of the soils in the floodplain. In many parts of the world it is impossible to answer the question 'Why are the rivers like they are now?' without a knowledge of their past utilization by man. The length of time involved in man's utilization of rivers is quite different on different continents. Clearly, Europe has been the most significantly modified for the longest period of time.

8.2.1 THE EXAMPLE OF THE RIVER GARONNE

The River Garonne is 580 km long and drains a basin of 57 000 km². Figure 8.1 shows the diversity of the hydrological network. According to Strahler ( 1957), the Garonne is a seventh-order river after it receives the water from its last Pyrenean tributary (River Ariège) and an eighth-order river after its first tributary from the Central Massif (River Tarn). The hydrology is characterized by large variations of flow between sudden and violent floods in spring and very low waters in summer (Table 8.1).

Table 8.1 Physical characteristics of the River Garonne

Figure 8.1 The River Garonne and its tributaries in south-west France

The choice of the River Garonne as an example is justified by the fact that it is one of the largest West European alluvial rivers in a non-industrialized region. Agriculture is responsible for most of the modifications of the fluvial landscape in the floodplain, and it is, therefore, possible to consider the floodplain as an echo from the past (Heal, 1991). This echo comes from devices for flood and erosion control in the seventeenth century, construction of a navigation channel in the eighteenth and nineteenth centuries, and agricultural and urban developments of the nineteenth and twentieth centuries. The River Garonne, then, gives an example of a fluvial landscape essentially modified by agriculture and, more recently, by urbanization. The main purpose of the research program on the river is to explain the causes and consequences of man's modifications of the fluvial landscape.

The modifications of the fluvial landscape have been of two kinds. The first modification lies in an ever-increasing control of the water dynamics. The control has led to a reduction of the flooded surfaces and so to a transformation of the physio-chemical conditions of soils and waters. Moreover, the isolation of the river from its floodplain has greatly reduced the transport of organic matter to and from the river. The disappearance of numerous secondary channels and backwaters has equally contributed to this reduction. As a consequence, the exchange dynamics within the fluvial landscape in the floodplain has been modified. A second kind of modification, linked to the previous one, lies in the intensification of agriculture and urbanization. This intensification has resulted in a fragmentation of the riparian woods in the floodplain. The consequences of this are many, and among them are a lesser retention capacity of the organic matter in the floodplain forest, an alteration of the succession of plant communities in the riparian area, and a change in the structure of the animal communities, the birds in particular (Décamps et al., 1987).

8.2.2 RIVER CHANGES IN AN URBAN AREA

The morphology of the River Garonne in Toulouse has undergone drastic changes since the end of the seventeenth century. These changes are the result of the interaction between the river flow and the activities of man designed to harness the river through the city of Toulouse (Fortuné, 1988). For example, it has long been impossible to maintain bridges across the River Garonne (Chalande, 1912). The oldest known bridge in Toulouse, constructed in the twelfth century of wood on a Roman foundation, is reported to have been destroyed and reconstructed six times after floods, and finally destroyed in 1523 never to be rebuilt. Such destructions of bridges during floods have been frequent. The first bridge able to resist floods has required long efforts. It was begun in 1543, and inaugurated in 1660, after more than a hundred years in construction (Lotte, 1982).

As in the case of bridges, dams constructed for mill activities were an obstacle to water flow. Mills were built in 1182 and 1190 (Sicard, 1953). Since the Middle Ages, these mills have had the greatest long-term effects on the river bed. In exchange for supplying the region with flour, millers were given the right to manipulate flows and to utilize the banks. Two stone dams were built in Toulouse at the end of the twelfth century to direct the river flow towards the mills. A consequence of the dams was an increase in the sedimentation of alluvial deposits and, therefore, an elevation of the level of the river bed. These modifications rendered the left bank liable to floods, provoking catastrophic destruction in the city during the eighteenth and nineteenth centuries. New embankments were built to prevent erosion and flood damage in the city. This construction and the consequent channel changes resulted again in an increase of flood levels, which required further work for flood and erosion control.

The analysis of ancient maps reveals that the number of islands was reduced from 17 to four over a 2.2-km distance during a 300-year period. At the same time, the shorelength was reduced by 4 km. The degree of these reductions lessened between 1777 and 1874, after the better flood control of the eighteenth century and before the connection of the islets to the banks during the twentieth century (Table 8.2). The historical facts show that man has modified the river bed in the city of Toulouse since at least the twelfth century. Ravaging floods which followed the construction of dams inundated portions of the city and, during the same time, regularly destroyed bridges and dams. Not until 1660, after a hundred-year period of construction, did a secure bridge join the two banks. This success marks the beginning of a period during which man achieved a better control of the river flow. Since this control resulted in increased erosion and sedimentation and an elevation of the level of the water during floods, more channelization was required, leading to a reduction in the number of islands and of the length of the banks in the urban area of Toulouse (Table 8.3).

Table 8.2 Modification of the fluvial landscape

Table 8.3 Changes of morphological characteristics of the river bed as shown by the maps

The ecological consequences of river manipulations appeared gradually during the last three centuries. The number of habitats available to fish and other aquatic fauna diminished because of the reduction in the length of the banks and the replacement of a stony river bed by a uniform rocky bottom. Moreover, the organic matter contribution from the riparian vegetation became practically nil. In the long run, the utilization of the river resulted in economic prosperity, but also in disturbances in the aquatic environment.

8.2.3 CHANGES OF THE FLUVIAL LANDSCAPE

Three successive influences characterize the history of the entire Garonne fluvial landscape. These are navigation, agriculture, and urbanization and industrialization. A flourishing activity developed around navigation in the eighteenth and first part of the nineteenth centuries. This activity reached a maximum in 1840 to 1850, and took several forms as regards to the river and its floodplain. First, the consolidation of the banks and the construction of a towing-path resulted in a transformation of the riparian woods. Elms which had existed since the Middle Ages, for example, were replaced by more flexible cultivated willows, which were regularly cut down. Second, the channel itself was simplified by removal of secondary arms, and by a straightening of some sections. Third, the river bed itself was transformed to allow the passage of boats through rapids. Finally, the consequences of human concentrations in ports appeared along the river, one of which was the acceleration of deforestation along the floodplain.

During the eighteenth and nineteenth centuries, agriculture became a priority in the floodplain. Poplar plantations replaced natural riparian woods in many places, and cultivated areas approached the river. As a consequence, riparian woods became more fragmented along the River Garonne. Another consequence of the development of agriculture was an increasing need of water for irrigation of the alluvial plain, particularly during the twentieth century. Figure 8.2 shows the considerable increase of irrigation during the last few decades.

Finally, during the twentieth century, the river contributed to the development of urbanization in the alluvial plain. Solid material taken from the river bed more than doubled in some parts of the valley from 1966 to 1979 (Table 8.4). A spectacular effect of this removal has been the deepening of the water table (Figure 8.3) and, as a consequence, a further decline of the riparian woods.

Figure 8.2 Increase of equipped areas for irrigation and pumped volume of water in the middle part of the Garonne floodplain

Table 8.4 Extraction of sand and gravel from the River Garonne and its environment within the limits of the administrative department of the Haute-Garonne from 1966 to 1979

Figure 8.3 Diminution of the water level of the River Garonne from 1910 to 1967 at various flows (from Beaudelin, 1987)

8.2.4 CONTRIBUTION OF THE HISTORICAL PERSPECTIVE

The example of the River Garonne illustrates the contribution of a historical perspective to the understanding of the present tendencies in fluvial landscapes. The socio-economic development along the floodplain has led to a channelization of the river. This phenomenon was accompanied by a disappearance of the secondary arms, by a straightening of the river channel and by a stabilization of the banks and of the river bed. A first consequence of this was the increase of the hydrological contrasts between catastrophic spates and more severe low waters. A second consequence was the reduction in length of the boundary between land and water and, therefore, a reduction in the exchanges between the river and the floodplain. It is important to realize that this complex evolution, essentially due to human influences, began, at least significantly, at the end of the seventeenth century. The effects of these disturbances affect the dynamics of the fluvial landscape even to the present.

8.3 GLOBAL AND COMPARATIVE PERSPECTIVES

Global and comparative perspectives are also necessary in floodplain ecology to acquire an understanding of at least four phenomena: socio-economic influences, consequences of changes in the land use of drainage basins, and both fluvial and ecological dynamics.

8.3.1 SOCIO-ECONOMIC DEVELOPMENT

A fluvial valley is a natural as well as a social space, and the result of this is a great complexity of the systems involved. Clearly, a knowledge of the regional socio-economic development is essential to explain the present state of fluvial landscapes, but understanding of this socio-economic development requires a broader context (Richards, 1986). The main difficulty is to place local events in a more general context, at the level of the valley for example. A river like the Garonne is the sum of its geomorphology, hydrology, climate, and man's utilization, from the springs to the mouth. Each sector gives only partial answers to the understanding of the evolution of the whole. Also, each sector maintains with the other sectors complex and diverse relationships. It is necessary to take into account the upstream-downstream sequence of these systems as shown by the river continuum concept (Vannote et al., 1980). It is also necessary to consider these systems in their nesting in larger systems. As an example, it is difficult to understand the evolution of the River Garonne in Toulouse without mentioning the successive constructions of weirs across the river beginning in the twelfth century. In turn, these local events are only understood in terms of economics. The supplying of the entire region depended on the functioning of the mills in the city of Toulouse. The maintenance of these mills and their associated dams, in addition to the flood damage, has deeply marked the fluvial environment at the level of the city.

More generally, from local events repeated over the decades, it is possible to reconstruct economic trends. The history of the river and the economic development have interacted, and together provide an explanation of the observed changes. However, it is necessary to go beyond such studies and make comparisons between diverse fluvial landscapes in order to interpret global and international patterns.

8.3.2 LAND-USE CHANGES

At the level of a drainage basin, changes in land use have proven to be of paramount importance for river ecology. As an example, the River Ain, a tributary of the River Rhone, was studied by Bravard (1987). The lower part of the River Ain is sinking into its sediments, particularly since the beginning of the century .This sinking has been about 1 m in the last 10 years. Several factors explain this phenomenon, among which are the replacement of cultivated areas by forest in the upper drainage basin, the construction of a hydroelectric dam in the middle course of the river 20 years ago, and a change in the river discharge regime. A drop in the base level of the River Rhône due to embankment construction since the nineteenth century has accelerated the tendency of the River Ain to incise its downstream course. Moreover, because of man's activities in the upper part of the drainage basin, the load of solid material has practically disappeared in the River Ain. Therefore, the energy of this river is utilized to incise its own sediments in its lower part. As a consequence, the river does not meander in its alluvial plain and is not able to rejuvenate the riparian communities which characterized the system in early times (Pautou and Girel, 1986). Only an approach that considers the entire basin, taking into account past utilizations of the upper part of the drainage basin, may help to understand the present dynamics of the River Ain system in its lower part.

8.3.3 FLUVIAL DYNAMICS

The fluvial dynamics of most European rivers has been modified by man's activities. Braga and Gervasoni (1983) have reconstituted the natural variations of the River Pô since the sixteenth century. They demonstrated that the meanders of the River Pô moved largely within the floodplain at the level of Plaisance, Italy. Such natural dynamics are now impeded by the riverbanks built to stabilize the river. Similarly, the Rhine was transformed for navigation from a meandering river before 1830 to a straightened channel after 1860, and then to a channelized river after 1950. As a result of these modifications a unique and linear course has replaced a braided, anastomosed and meandering section between Bâle and Mainz (Carbiener et al., 1986). This tendency of channelization and linearization of the water courses under the influence of man is general in developed countries, as illustrated by works on various rivers: the Danube (Bacalbasa-Dobrovici, 1989), the Rhône (Bravard et al., 1986), the Garonne (Fortuné, 1988), the Rio Grande (Boggs, 1940), the Williamette (Sedell and Froggatt, 1984), and various British rivers (Petts, 1988). A comparative approach is needed to understand the various ecological consequences of this general trend.

Clearly, the best theoretical basis for such a comparative approach is given by the river continuum concept proposed by Vannote et al. (1980), and refined by various authors (Ward and Stanford, 1983; Minshall et al., 1983, 1985; Statzner and Higler, 1985; Naiman et al., 1987).

8.3.4 ECOLOGICAL DYNAMICS

Before human influence, riparian systems were characterized by periodic successions dependent upon flood cycles and their severity. Human activities have modified, and in some cases interrupted, the previous succession patterns. Along many large rivers, embankments have isolated the channels from their terrestrial surroundings, and therefore modified the natural dynamics of the riparian communities. To understand the ecological consequences of these modifications, we need to improve our theoretical knowledge of lotic ecotones, in particular as regards to their function in a river system. First, the riparian vegetation has its own dynamics, with successions in time and space as studied along many rivers in the world (Pautou and Décamps, 1985; Bravard et al., 1986; Carbiener et al., 1986; Walker et al., 1986; Salo et al., 1986; Décamps et al., 1988). Second, the riparian vegetation plays a role in the retention of nutrients transported downstream (Ward and Stanford, 1988; Pinay et al., 1989). This retention function of the riparian vegetation is to be considered at the different stages of the succession from recent alluvial deposits to pioneer vegetation and to willow and alder communities. The capacity of retention of these different stages must be characterized as well as the turn-over length of nutrients on the floodplain. For this, as discussed by S. T .A. Pickett in Chapter 5 of this volume, space substitutes for chronosequences are not sufficient, i.e. succession cannot be reconstructed by comparing different sites at different stages at the same time. Comparative long-term studies within a network of sites are necessary to get a general understanding of the role of the different stages of the ecological successions on river banks.

Figure 8.4 Dam construction and reduction of spawning areas for migratory fish in the Garonne network

Figure 8.5 Changes in the specific richness of the fish community in the River Rhine. The species are classified as abundant, rare, and sporadic according to Lelek (1989)

The specific richness of fish communities in rivers has been subject to variations due to human activities. A first example, for the River Garonne, is given by the reduction of spawning sites for migratory fish from the Atlantic Ocean (Figure 8.4). The successive constructions of dams have considerably reduced the ability of salmon and aloses to attain the upper parts of the hydrographic network of this river and its tributaries since the Middle Ages. A second example, for the River Rhine, is given by the changes in the specific richness of the fish community due to pollution during the last decades (Figure 8.5). Lelek (1989) classified the fish species according to their richness as abundant, rare, and sporadic. He was able to show that for more than 50 species recensed, 27 changed to an inferior category of abundance between 1890 and 1950, and 17 between 1951 and 1975, whereas 18 species changed to a superior category of abundance between 1976 and 1985, and only two species diminishing during this last period. This recovery since 1976 is confirmed by other observations made in 1985, even though some species practically disappeared during the 1950s.

8.4 QUESTIONS FOR A LONG-TERM ECOLOGICAL RESEARCH  PERSPECTIVE ON FLUVIAL LANDSCAPES

Four questions are to be considered in order to predict the dynamics of fluvial landscapes for the next 10 to 15 years. These relate to complex, rare, subtle, and slow phenomena which, according to Pickett (1991), are relevant to long-term ecological research.

Interactions between the resources linked to a river produce cascade effects, so that the utilization of a resource affects the use of other potential resources in the future. The interactions are complex and are not completely understood. As shown for the dynamics of the River Garonne in Toulouse, the interactions are due to links between socio-economic development and the ecology of the river. Conflicts of use have been frequent along rivers; for example, between navigation  and agriculture, between industries and fishing, or between the needs of riverine societies in the upper and in the lower parts of the hydrological networks. Because of the complexity of these relationships, the consequences of land-use changes in the ecology of rivers need long-term research. At the same time, various time scales are necessary (Urban et al., 1987), depending upon the subject under study. For a microbial population long-term can be weeks, for invertebrates it can be years, for some fish it can be decades, and for some riparian trees it is centuries.

Due to the interactions, fluvial landscapes appear as systems characterized by strong inertias. Therefore, important environmental tendencies in rivers and their floodplains are practically irreversible over the next decade or two, as stated by Munn (1987) for environmental systems in general. This is the case in the depression of the water table along many alluvial rivers in Europe. For most of these rivers, as yet unrevealed crisis conditions may already be in their initial stages. Because of the surprising character of such crises, a network of research programs is necessary to follow the pattern and cycles of changes in order to predict potential crises, and try to bring them under control.

It must be realized that a number of changes concerning fluvial landscapes appear to be unpredictable because they are rare and sometimes unique. How to adapt to this unpredictability is a very important question in river management. The difficulty is particularly clear when scenarios combining the effects of socio-economic development with environmental effects are to be elaborated at the scale of a fluvial landscape. 

The importance of river degradations during the last decades has resulted in a recognition of the problems of river conservation. The public is becoming increasingly aware of these problems in many countries, even though some reactions are purely emotional and not based on a clear scientific knowledge of the dynamics of the fluvial systems. Two points are to be considered in relation to this. The first important point is to explain to the public the scientific basis of the patterns of change in rivers and fluvial landscapes; the second is to ensure full interaction between the public and the scientists in developing coherent decisions for conservation and management. 

These four major problems - understanding the interactions between resources, inertia to reversal of geomorphic changes, unpredictability, and changes in public opinion - point to the need for a network of long-term ecological research on rivers, and, in addition, for a linked monitoring of the changes in fluvial landscapes.

8.5 MONITORING OF FLUVIAL LANDSCAPES

Remote sensing offers a particularly promising tool for monitoring the ecology of fluvial landscapes. Sudden, as well as longer-term, modifications of these landscapes often signal changes affecting drainage basins on a much broader scale. As an example, the canopies of riparian forests 'can give clues to ecosystem properties that might be precursors for a major change in structure and species composition' (Waring et al., 1986). Remote sensing can be used to map, monitor, and measure such changes or damage along fluvial landscapes. The ecologists need to improve their ability to interpret remote sensing information, in order to define new questions of research compatible with the capabilities of existing satellites, and also to design sensors for detecting environmental problems concerning rivers and their landscapes. As shown by Cummins et al. (1989), a combination of remote sensing and selected on-the-ground measurements may be used to establish patterns of association between riparian vegetation communities and stream shredder population. In this connection, long-term monitoring may make it possible to ask some research questions which cannot otherwise be addressed.

8.6 CONCLUSIONS

The ecology of fluvial landscapes requires long-term research. Several decades are necessary to understand the changes that are taking place, and to predict the results of present tendencies. These predictions will be more precise if they are supported by a good knowledge of the past utilization of the river landscape. Knowledge of the historical effects of human societies is increasingly being taken into account in ecosystems research in North America and in Europe (Sedell and Froggatt, 1984; Bravard, 1987; Fortuné, 1988). Moreover, these predictions have a chance to be successful only if local or regional knowledge is included in a comparative and broader knowledge of the tendencies towards change along various fluvial landscapes in the world (Petts et al., 1989). Only through a network of long-term research can the dynamics of fluvial landscapes be fully understood, and predictive scenarios, particularly of the ecological consequences of socio-economic development, be made.

8.7 ACKNOWLEDGMENTS

This work was supported by the Programme Interdisciplinaire de Recherche sur l'Environnement (PIREN) of the Centre National de la Recherche Scientifique. We thank K. W .Cummins for helpful comments on our manuscript, and A. Lelek for information on the River Rhine.

8.8 REFERENCES

Bacalbasa-Dobrovici, N. (1989). The Danube River and its fisheries. Canadian Special Publication of Fisheries and Aquatic Sciences, 106, 455-468.

Beaudelin, P. (1987). Les méfaits des extractions des galets de la Garonne. Adour Garonne, 34, 10-13.

Boggs, S.W. (1940). International boundaries. Columbia University Press, New York. 

Braga, G. and Gervasoni. S. (1983). Evoluzione storica dell'alveo del fiume Pô nel territorio Lodigiano-Piacentin: rischi idrogeologici connessi. Atti Convegno Nazionale '1l suolo come risorsa' , 60, 59-69.

Bravard, J.P. (1987). Le Rhône du Léman à Lyon. La Manufacture, Lyon.

Bravard, J.P., Amoros, C. and Pautou, G. (1986). Impact of civil engineering works on the successions of communities in a fluvial system: a methodological and predictive approach applied to a section of the Upper Rhône river, France. Oikos, 47, 92-111.

Carbiener, R., Dillman, E., Dister, E. and Schnitzler, A. (1986). Variations de comportement et vicariances écologiques d'espèces ligneuses en zone inondable: l'exemple de la plaine du Rhin. Journées d' Hydrologie de Strasbourg Crues et Inondations, 237-259.

Chalande, J. (1912). Les inondations et les formations alluviales dans le bassin de la Garonne à Toulouse depuis le douzième siècle. Mem. Acad. Sci. Inscript. Belles Lettres Toulouse, 10ème série, 12, 1-16.

Cummins, K.W., Wilzbach, M.A., Gatges, D.M., Perry, J.B. and Taliaferro, W.B. (1989). Shredders and riparian vegetation. BioScience, 39, 24-30.

Décamps, H., Fortuné, M., Gazelle, F. and Pautou, G. (1988). Historical influence of man in the riparian dynamics of a fluvial landscape. Landscape Ecology, 1, 163-173.

Décamps, H., Joachim, J. and Lauga, J. (1987). The importance for birds of the riparian woodlands within the alluvial corridor of the River Garonne, SW France. Regulated Rivers, 1, 301-316.

Décamps, H. and Naiman, R.J. (1989). L'ecologie des fleuves. La Recherche, 208, 310- 319.

Di Castri, F. and Hadley, M. ( 1988). Enhancing the credibility of ecology: interacting along and across hierarchical scales. Geojournal, 17, 5-35.

Fortuné, M. (1988). Historical changes in a large river in an urban area: the Garonne river, Toulouse, France. Regulated Rivers, 2, 179-186.

Heal, O.W. (1991). The role of study sites in long-term ecological research: A UK experience. Chapter 3 in this volume.

Lelek, A. (1989). The Rhine River and some of its tributaries under human impact in the last two centuries. Canadian Special Publication of Fisheries and Aquatic Sciences, 106, 469-487.

Lotte, R. (1982). Construction d'un Pont Sous la Renaissance, le Pont Neuf de Toulouse. Ecole Nationale des Ponts et Chaussées, Paris.

Minshall, G.W., Petersen, R.C., Cummins, K.W., Bott, T.L., Sedell, J.R., Cushing, C.E. and Vannote, R.L. (1983). Interbiome comparison of stream ecosystem dynamics. Ecological Monographs, 53, 1-25.

Minshall, G.W., Cummins, K.W., Petersen, R.C., Cushing, C.E., Bruns, D.A., Sedell, J.R. and Vannote, R.L. (1985). Developments in stream ecosystem theory. Canadian Journal of Fisheries and Aquatic Sciences, 42, 1045-1055.

Munn, R.E. (1987). Environmental prospects for the next century: implications for long-term policy and research strategies. Research Report 15, International Institute for Applied Systems Analysis.

Naiman, R.J., Décamps, H., Pastor, J. and Johnston, C.A. (1988). The potential importance of boundaries to fluvial ecosystems. Journal of the North American Benthological Society , 7, 289-306.

Naiman, R.J., Melillo, J.M., Lock, M.A., Ford, T.E. and Rice, S.R. (1987). Longitudinal patterns of ecosystem processes and community structure in a subarctic river continuum. Ecology, 68, 1138-1156.

Pautou, G. and Décamps, H. (1985). Ecological interactions between the alluvial forests and hydrology of the Upper Rhone. Archiv.f. Hydrobiol, 104, 13-37.

Pautou, G. and Girel. J. ( 1986). La végétation de la basse plaine de l' Ain: organisation spatiale et évolution. Documents de Cartographie Ecologique, 29, 75-96.

Petts, G.E. (1988). Ecological management of regulated rivers: a European perspective. Regulated Rivers, 1, 363-369.

Petts, G.E., MöIler, H. and Roux, A.L. (Eds) (1989). Historical Change of Large Alluvial Rivers. Western Europe. John Wiley, Chichester.

Pickett, S.T.A. (1989). Long-tern studies: past experience and recommendations for the future. Chapter 5 in this volume.

Pinay, G. and Décamps, H. (1988). The role of riparian woods in regulating nitrogen fluxes between the alluvial aquifer and surface water: a conceptual model. Regulated Rivers, 2, 507-516.

Pinay G., Décamps, H., Chauvet, E. and Fustec, E. (1989). Lotic ecotones. In Naiman, R.J. and Décamps, H. (Eds) Land/lnland-Water Ecotones: Strategies for Research and Management. Parthenon Press, Camforth.

Richards, J.F. (1986). World environmental history and economic development. In Clark, W.C. and Munn, R.E. (Eds) Sustainable Development of the Biosphere. Cambridge University Press, Cambridge, 53-74.

Salo, J., Kalliola, R., Hakkinen, I., Makinen, Y., Niemela, P., Puhakka, M. and Coley, A.D. (1986). River dynamics and the diversity of Amazon lowland forest. Nature, 322,  254-258.

Sedell, J .R. and Froggatt, J .L. ( 1984 ). Importance of streamside vegetation to large rivers: the isolation of the Williamette River, Oregon, USA, from its floodplain. Verh. lnternat. Verein. Limnol., 22, 1828-1834.

Sicard, G. ( 1953). Aux Origines des Sociétés Anonymes les Moulins de Toulouse au Moyen Age. Armand Colin, Paris.

Statzner, B. and Higler, B. (1985). Questions and comments on the River Continuum Concept. Canadian Journal of Fisheries and Aquatic Sciences, 42, 1038-1044.

Strahler, A.N. (1957). Quantitative analysis of watershed geomorphology. American Geophys. Union Trans., 38, 913-920.

Urban, D.L., O'Neill, N.V. and Shugart, H.H., Jr. (1987). Landscape ecology. BioScience, 37,119-127.

Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell J.R. and Cushing, C.E. (1980). The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences, 37, 130-137.

Walker, L.R., Zasada J.C. and Chapin III, F.S. (1986). The role of life history processes in primary succession on an Alaskan floodplain. Ecology, 67, 1243-1253.

Ward, J .V. and Stanford, J .A. ( 1983 ). The serial discontinuity concept of lotic ecosystems. In Fontaine, T.D. and Bartell, S.M. (Eds) Dynamics of Lotic Ecosystems. Ann Arbor Science Publishers, Ann Arbor, Michigan, 29-42

Ward, K.V. and Stanford, J.A. (1989). Riverine ecosystems: the influence of man on catchment dynamics and fish ecology. Canadian Special Publication of Fisheries and Aquatic Sciences, 106, 56-64.

Waring, R.H., Aber, J.D., Melillo, J. M. and Moore III, B. (1986). Precursors of change in terrestrial ecosystems. BioScience, 36, 433-438.