REGULATED RIVERS: RESEARCH & MANAGEMENT, VOL. 12, 617-631 (1996)

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER PRIOR TO COMPLETION OF THE RHINE-RHONE CONNECTION

JEAN-FRANCOIS FRUGET,*’t JEANNE DESSAIX*’T AND SANDRINE PLENETT *ARALEPBP-Biologie Animale, Universitt Lyon I , Brit. 403, 69622 Villeurbanne, France

tESA CNRS 5023 ‘Ecologie des Eaux Douces et des Grands Fleuves’, UniversitP Lyon I , 69622 Villeurbanne, France

ABSTRACT The Rhine-Rh6ne connection project concerns the part of the Doubs River between MontbCliard and Dale (France), i.e. downstream from the confluence of the Allan River. In order to update our knowledge of the benthic macroinvertebrate communities of the Doubs River, three types of aquatic environments of this hydrosystem were sampled (main channel, backwaters and the Freycinet canal). Hydraulic characteristics are the most explanatory environment variables of lotic macroinvertebrate distribution of the Doubs hydrosystem. The aquatic fauna of the Doubs River is varied due to the diversity of biotopes, microhabitats and current velocities. The lotic aspect is well defined. Rheophilic taxa are present in the main stream in riffle areas or located downstream from weirs where the current velocity is rapid ( > 75 cm s-’). The physics of flow (‘stream hydraulics’) create the richness of habitats and thereby the biological diversity. Backwaters have a relatively high richness, similar to that of stations in the main stream located in the same sector. Conservation manage- ment must be oriented towards the present ecological situation of the Doubs River, i.e. that of a lotic ecosystem with active connections with its backwaters. The primary goal from a conservation perspective must be the maintenance and the improvement of areas highly sensitive to disturbances.

KEY WORDS: benthic macroinvertebrates; regulation; floodplain; Doubs River

INTRODUCTION

The Rhine-RhBne connection project concerns the part of the Doubs River between Montbeliard and DBle, i.e. downstream from the confluence of the Allan River. The project is managed by the Compagnie Nationale du Rh6ne (CNR) a river regulation and management agency. This large ship canal project involves the creation of a new waterway which would pass through the valleys of the Doubs, Allan, Bourbeuse, Largue and I11 Rivers. The connection between the Rhine River and the SaBne River will be made via 23 reaches (55 m wide, 4.5 m deep) separated by 24 locks (Figure 1C). A Rhine-RhBne connection (for barges less than 400t) has existed since 1833. The ‘Freycinet canal’ (12m wide, 2m deep), named after its builder, connects the Rhine River to the SaBne River. It partly uses the main channel of the Doubs River between 1’Isle-sur-le-Doubs and D61e. Weirs, locks and sometimes a diversion canal were built on this reach.

In order to appreciate the influence of the new waterway on the ecosystems of the Doubs River, studies were needed to update and to complement the existing biological data. Changes in environmental conditions were reflected in corresponding alterations of aquatic community structure and ecosystem functioning. Functional describers were groups of species descriptive of certain environmental conditions (Bournaud and Amoros, 1984; Amoros et al., 1987a). Plefkin et al. (1989) noted that macroinvertebrate and fish communities were the most used describers. Centofanti et al. (submitted) studies the ichthyofauna of the Doubs River as part of the Rhine-Rh6ne connection project.

Data about the benthic macroinvertebrates of the Doubs River were only available from Verneaux (1973), but these data were limited to insect orders (Plecoptera, Ephemeroptera, Trichoptera and Odonata). Data from water quality surveys of the Water Authorities, used to calculate biological indexes, were not usable for our study due to their low level of identification (family level). Thus, in order to update and complement these data, several aquatic environments of the Doubs River were sampled in September 1992.

The objectives of the present study were: (i) to update the existing taxa list of the benthic macroinvertebrates

CCC 0886-937S/96/060617-15 0 1996 by John Wiley & Sons, Ltd.

Received 30 March 1995 Accepted 8 February 1996

618 J.-F. FRUGET, J. DESSAIX AND S. PLENET

c

0 5 10 15 20km

Aquatic environments :

Doubs River existing Rhine-Rh6ne Freycinet Tributary Large ship canal project cr a Hydrobiological sampling stations :

-! Main channel of the Doubs River

Culrnch

- Freycinet canal and canalized part of A the Allan River

(_> Backwaters

.g 1

1 178 75 - t 229

d Besanpon 1 I I

j F M A M J J A s O N D

f

Chanwa n altitude (m)

1 P B 2

Zd - Length of the waterway: 229 km - Difference in height of the Franche-ComtB side: 157.75 m - Difference in height of the Alsace side: 106.20 m I- - Total difference in height : 263.95 m .22 . - Number of locks: 24

- Number of dams: - on the Doubs River: 12 - on the Allan River: 3

L Z - 7

~- 23

- - - - - - 200 150 100 50

4 0

Figure 1. The Rhine-Rhbne large ship canal project. A; General situation of the Doubs River. B; Location of the sampling stations. C; The Sabne-Rhine connection project. D; Hydrological regime of the Doubs River at Besaneon

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER 619

of the Doubs Rivers, (ii) to improve the understanding of the conservation value of the backwater areas and (iii) to complement the database of the invertebrate communities of the Rhine-Rh6ne Freycinet canal (ARALEPBP, 1993).

MATERIALS AND METHODS

Study area The total length of the future waterway is 229 km, including 160 km of the Doubs River. Sixty kilometres

of the river would be by-passed by an artificial diversion canal. Seventy percent of the riffles (both natural and artificial ones located downstream from the weirs) would be flooded, several meanders would be cut off and several islands (and consequently several kilometres of shoreline, i.e. ecotone areas) would be destroyed.

Low flows are severe in this part of the river, due to the geological structure of the catchment area (Jurassic limestone). Average discharge of the Doubs River at Besanqon is 93.5m3s-' for the period 1952-1990 (Figure 1D). Low flows occur from July to September (discharge < 50m3 s-I) with an extreme discharge of less than 8 m3 s-' (GREBE, 1992). After regulation, nearly 25% of the discharge of the Doubs River would be used to operate the locks during low flows.

Field survey In accord with our three objectives, 17 stations were sampled. They were located in three types of

aquatic environments (functional sets according to Richardot-Coulet et al., 1982 and Amoros et al., 1987a): (i) the main channel of the river, (ii) the backwaters and (iii) the existing Rhine-Rh6ne Freycinet canal. These stations were distributed from the Alsacian side of the Freycinet canal to the Doubs River downstream from D61e (Figure 1B and Table 1). The middle of the channel and the banks were sampled.

Samples in the channel were collected using a modified Surber-type net when the depth was less than 1 m. Beyond this depth, an Irish-type triangular dredge was used (Elliott et al., 1980). The meshes of the dredge and of the Surber net were 500 pm. Samples were collected in the middle of the channel. The definition of the bank was that of Bournaud and Cogerino (1986), i.e. a distance of 1 m from the river side (at low or medium flow) and a maximum depth of 1 m. Macrofauna was collected by microhabitat scraping or picking using

Table I. Name and description of the sampling station

Station no.

Station code

Name and location Sample date

Kilometric point

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17

CA 1 CA2 CA3 DO 1 DO2 DO3 DO4 DO5 DO6 DO7 CA4 AN1 AN2 DO8 AN3 DO9 AN4

~~~ ~

Heidwiller; Freycinet canal Froidefontaine; Freycinet canal Etupes; Canalized part of the Allan River Mathay, Doubs; Unnavigated main channel, Riffles Colombier-Fontaine, Doubs; Unnavigated main channel, Riffles Clerval, Doubs; Unnavigated main channel Baume-les-Dames, Doubs; Unnavigated main channel Vaire-Arcier, Doubs; Main channel, Navigated reach Avanne, Doubs; Unnavigated main channel Rochefort-sur-Nenon, Doubs; Unnavigated main channel, Weir Rochefort-sur-Nenon, Freycinet canal Morte du Pre de 1'Echo; Isolated backwater Corne des Epiciers; Connected backwater Crissey, Doubs; Unnavigated main channel Morte des Tranches; Connected backwater Gevry, Doubs; Unnavigated main channel Corne de Gevry; Connected backwater

25/09 25/09 18/09 18/09 18/09 17/09 17/09 17/09 16/09 16/09 16/09 16/09 16/09 15/09 15/09 15/09 15/09

30 57 67 70 83

107 121 143 163 202 203 206 207 212 214 215 215.5

620 J.-F. FRUGET, J . DESSAIX AND S. PLENET

a dip net. Nine microhabitats (flagstones, boulders, pebbles, gravel, sand, silt, aquatic macrophytes, helophytes and algae) were sampled according to their occurrence.

For each invertebrate sample, water depth, flow velocity and the substrate feature investigated were noted (qualitative classes for water depth and flow velocity, according to Cummins and Lauff s (1969) classification for the substrate feature). These are the most important factors in the aquatic environment that define ‘hydraulic stress’ sensu Statzner (1981) and Statzner et al. (1988). Boulders and pebbles were brushed and discarded, the remaining sediment was put into a box and preserved in 4% formalin. The invertebrates were sorted, preserved in 70% ethanol, identified to species level if possible and counted in the laboratory.

A total of sixty five invertebrate samples were collected between 15th and 25th September 1992 in homogeneous hydrological conditions (low flows of about 10-12 m3 s-’ at the Lougres gauging station). Thirty two samples were taken in the main channel of the Doubs River, 18 in the backwaters and 15 in the Freycinet canal and in the canalized part of the Allan River.

Data analyses The community parameters, i.e. the taxonomic richness, the relative abundance of each taxa and

Shannon’s diversity index, H ’ (Shannon and Weaver, 1949) were calculated for each station and for each type of aquatic environment.

The organization of the macroinvertebrate communities and its significance were analysed using multi- variate analyses. ‘Rare’ taxa were discarded according to the abundance criterion (globally 5 2 individuals) or the occurrence criterion (presence in 5 2 samples). Data were expressed in logz, which allowed us to standardize the estimates of abundance. A combination of multivariate analyses was performed to study the between-stations structure: firstly, a hierarchical cluster analysis (Ward’s agglomerative method, also called the second-order moment method; see Roux, 199 1); and then a between-class correspondence analysis which distinguished between two types of effects (station and microhabitat in our case): only one effect is taken into account whereas the other one is eliminated (Doledec and Chessel, 1989, 1991; Beffy and Doledec, 1991).

RESULTS

Community parameters A total of 149 taxa were recorded in the study, 107 in the main channel, 94 in the backwaters and 72 in the

Freycinet canal and the canalized part of the Allan River, respectively. Of these taxa 26 can be considered as ‘rare’ on the abundance criterion (5 2 specimens) and 44 on the occurrence criterion (5 2 occurrences/65 samples).

The benthic macroinvertebrate communities of the different aquatic environments of the Doubs River showed an important faunistic diversity amongst insect orders, except for Plecoptera, which was only represented by the genus Leuctra. Oligochaeta, Diptera (mainly chironomids) and Mollusca Sphaeriidae were not identified beyond family. Trichoptera and molluscs were the most diverse faunistic groups overall (27 and 22 species, respectively). Coleoptera and Ephemeroptera were also well represented (18 and 17 species). Differences in taxonomic richness and in dominant faunistic groups existed between the environ- ments: Trichoptera, Ephemeroptera, Coleoptera and Diptera were the most diverse groups in the main stream; insect orders were well represented in the backwaters; Diptera and molluscs were the two main faunistic groups in the canals.

Aquatic environments can be classified according to the taxonomic richness and the diversity index of their stations (Figure 2). The Freycinet canal often had a low richness, except for the canalized part of the Allan River (CA3) which is an exception in this category. The high richness of CA3 was due in part to the interference by a weir located a few hundred metres upstream, at the confluence of the Savoureuse River, from where some taxa that had drifted downstream originated. The development of macrophytes near the banks also created a diversification of the substrate in this area where siltation was very intensive. Backwaters had a relatively high richness (close to 50 taxa) and diversity (between 3.5 and 4), except at station AN4. The richness and the diversity of the stations of this functional set were often very similar

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER 62 1

Taxonomic richness

C A I CA2' CA3' D01' D02' D03' D04' DO;

-. = 7- \

D07ICA4' AN1 'AN2'

Shannon's index H'

c

D08' AN3' DO9 ' AN4'

Stations

Figure 2. Total taxonomic richness (histograms) and Shannon's diversity index (lines) of the 17 sampling stations from upstream to downstream

to those of the stations of the main stream located in the same sector. Among this former functional set, DO3 (Clerval) had the highest richness due to an important diversity of biotopes and consequently diversified envir- onmental conditions. The stations of the main stream that exhibited less than 50 taxa (D01, DO2 and D08) were located in riffle sectors with a coarse substrate and a high current velocity, and consequently supported a specialized rheophilic fauna.

The benthic macrofauna was globally dominated by chironomids (38% of the total number of individuals), Trichoptera, Ephemeroptera and Oligochaeta (all close to 13%) and the 'various' taxa (S%, mainly Hydra s.1.). The relative abundance of 22 taxa was above 1% (dominant and resident species according to Bournaud, 1980) (Table 11). Dominant taxa (i.e. > 5%) were various chironomids and oligochaetes. Caenis luctuosa, which is the most eurytopic Caenidae species, was globally abundant. It was found at all the stations except in the isolated backwaters where siltation was high. Several species such as Hydra s.l. , Micronecta sp. or Dugesia tigrina were more especially linked to the ecological conditions of the sampling period (end of the summer, with low flow and low depth which favours the warming of water and the development of algae or macrophytes), which were in accordance with their ecology and their trophic preferences (Bournaud et al., 1987b; Tachet et al., 1987). Rheophilic species (e.g. Hydropsyche exocellata, Cheumatopsyche lepida, Psychomyia pusilla more particularly, or Baetis fuscutus, Hydropsyche contubernalis, Similium sp.) were present at lower abundances and frequencies of occurrence because they were related to definite environmental conditions (coarse substrate with sometimes the development of Fontinulis; average to fast current velocity, i.e. 25 to more than 75 cm s-'; depth less than 1 m). Structure of the macrobenthic fauna

The final data matrix for multivariate analyses were 65 samples x 115 taxa and 17 stations x 115 taxa. The dendrogram resulting from the cluster analysis distinguished three groups of stations (Figure 3A):

(i) the environments with a lentic tendency (backwaters and canals); (ii) the lotic main channel of

622 J.-F. FRUGET, J. DESSAIX AND S. PLENET

Table 11. Relative abundance and ocurrence of major macroinvertebrate taxa (> 1% of the total number of individuals)

Taxa Abundance (YO) Occurrence (x /65 samples)

Orthocladiinae s.1. ‘Other’ (Le. unidentified) Oligochaeta Chironomini Caenis spp. H ydropsychidae Caenis luctuosa Tanytarsini Stylaria Iacusrris Hydra s.1. Coenagrionidae Ceraclea sp. (iuv) Micronecta sp. Dugesia spp. Baetis spp. Chironomidae (n.) Dugesia tigrina Baetis fuscatus Caenis sp. (juv.) Hydropsyche contubernalis Hydropsyche sp. (iuv.) Coenagrionidae cuv.) H ydracarina Simulium sp. (1 + n) Cloeon dipterum Hydropsyche exocellata Cheumatopsyche lepida Psychomyia pusilla

17.5 9.5 8.7 8.5 7.7 6.2 6.0 3.3 3.2 3.1 3.0 2.9 2.9 2.8 2.6 2.1 2.0 1.9 1.9 1.9 1 -7 1.7 1.6 1.5 1.4 1.3 1.2

62 62 59

51 57 30 42

16 30

57 41 23 18 27 24 26 42 17 16 18 17 1 1

the Doubs River; (iii) two ‘isolated’ stations, D01, which is the upstream reference station in the Doubs River, and AN1 , which was an isolated backwater with high siltation and well-developed macrophytes.

The between-stations correspondence analysis confirmed this partition (Figure 3B). The two first axes represented 27.6 and 15% of the variance. The first axis opposes the lotic main channel of the Doubs River (positive coordinates on this axis) with the lentic and less lotic environments (negative coordinates). Among these two groups some stations are clearly separated (e.g. DO1 and ANl) or more moderately separated (e.g. DO2 and CA2) from the stations of the same functional set. The results of the first axis are therefore quite similar to those of the cluster analysis. However, the clustering technique cannot indicate the contribution of invertebrate species to the determination of the different groupings.

The taxa factorial map allows the faunistic spectrum of the alluvial environments of the Doubs River to be drawn (Figure 4), especially for the most significant stations distinguished previously. DO 1, the upper station located upstream from the confluence of the Allan River, had a coarse granulometry and supported the bryophyte Fontinalis. Its macrobenthic fauna showed a rhithronic tendency dominated by rheophilic species. D02, also located in the upper part of the lower Doubs River, supported a similar faunistic composi- tion. Current velocity was high and Fontinalis present on flagstones. Many taxa found in AN1 were characteristic of isolated backwaters with eutrophic characteristics, dense stands of aquatic vegetation and a high load of organic debris (Castella et al., 1984, 1991). This agreed with the environmental conditions of this station. The presence of the planarian Phagocata virtu indicated potential groundwater inputs (Pattee and Gourbault, 1981).

Regulation schemes create homogenization of environmental conditions, particularly of types of microhabitats, as shown for the RhBne River (Fruget, 1991; Dessaix et al., 1995). Thus, in order to study

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER 623

CA1

CA2

CA3

CA4

AN2

AN3

AN4

lenitic environments

Canals a- Backwaters

I P t--- DO1

AN1

1

Main channel of DO8

DO3 the Doubs River

DO4

DO6

DO5 - I 1 1 1 1 0 0.5 1 .o 1.5

Figure 3. Between-stations multivariate analyses. A; Dendrogram resulting from the cluster analysis (Ward’s method) of the macro- invertebrate composition over the 17 stations. B; Stations factorial map F1 x F2 of the between-stations correspondence analysis.

The circles with the stations codes correspond to the centre of classes and the lines correspond to the dispersion of samples

624 J.-F. FRUGET, J . DESSAIX A N D S. PLENET

Figure 4. Taxa factorial map F1 x F2 of the between-stations correspondence analysis. Only the most significant taxa are shown

the habitat influence on the faunistic assemblages of the Doubs River, the data matrix 63 samples x 11 5 taxa was analysed by a between-class correspondence analysis (flagstone and helophyte microhabitats were discarded because they occurred only once). The F1 x F2 factorial map of the samples (44.9 and 19.3% of the variance, respectively) defined three types of microhabitats: (i) some mineral microhabitats, (ii) a type composed of mixed vegetal and mineral microhabitats of lentic environments (macrophytes and silt), and (iii) two microhabitats of lotic environments (made up of pebbles and ‘algae’ (Fontinalis) respectively) (Figure 5B). On the second axis of the F2 x F3 factorial map (19.3 and 15.3% of the variance) a strong clustering between the microhabitats exists, from coarse sediments to vegetation microhabitats (Figure SC). The third axis was mainly characterized by silt which shows an important dispersion of the samples (silt was encountered essentially in the backwaters and the canals). Some taxa were particularly related to certain types of microhabitats which were more or less connected with some types of aquatic environments. Thus, ‘algae’ (Fontinalis) and pebble microhabitats were typical of lotic environments, characterized by taxa of the main channel such as Leuctridae, Rhyacophila sp., Baetis spp., Ephemerellu ignita, Esolus parallelepipedus, Elmis maugetii, Simulium sp., etc. (Figure 5B). On the other hand, macrophytes and silt were connected with taxa found in lentic functional sets (such as Leptocerus tineifornzis, Cloeon dipterum, Caenis robusta, Haliplus spp., Coenagrionidae, etc.), in particular in the backwaters (Figure 5C). These taxa formed an important base for such environments within the Rhbne River floodplain (Castella et al., 1984, 1991).

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER

............ . . . ...._

. M. furcata (HET) . Leuctridae (PLE) - ffhyacophiia sp. (TRI) .8. tuscafus (EPH)

- H. lineatocolis - Ceratopogonrnae (DlP) - Corynoneunnae

~ C. splendens (ODO) - A . aesfivalfs (HETJ . E. maugets (COL)

- Parapoynx spp (LEP) - A . fluviati/is (MOL)

- G hererodifa (WR)

- 8. sowerbyi (OLI) - C. mucedo (BRY) Boulder ‘b ----=a \ I

Signlticant tars of silty envlronments - L. tineifomis (TRI) . P. leachi (HET) . Ceratopogoninae (DIP) - S. lutaria (MEG) - P. carinatus (MOLj - 8. sowe&yi (OLI) . C. mucedo (ERY)

625

Figure 5 . Between-habitats correspondence analysis: eigenvalues graph (A), samples (microhabitats) factorial map FI x F2 (B), samples factorial map F2 x F3 (C) and particularly significant taxa of different environments. The circles correspond to the centre of classes and

the lines correspond to the dispersion of samples

626 J.-F. FRUGET, J. DESSAIX AND S. PLENET

DISCUSSION

Ecological significance of the observed faunistic assemblages Each aquatic environment (i.e. functional set) of the Doubs River supported characteristic species or

assemblages. Macroinvertebrate assemblages are typically good describers for the assessment, at a landscape scale, of aquatic systems within the floodplains (Castella et al., 1991). Figure 6 summarizes the longitudinal distribution of some significant taxa and their distribution within the different types of aquatic environments of the Doubs River. Caddisflies, especially hydropsychids, were present along the entire length of the main stream. Nevertheless, the different species were distributed according to a longitudinal typological gradient defined by current velocity (Bournaud et al., 1982, 1988). The main faunistic groups of the backwaters were mayflies (Cloeon dipterum, Caenis spp.), Odonata Coenagrionidae, some molluscs (Ancylusfluviatilis, Physa acuta) and certain taxa (such as the Heteroptera Micronectu sp). in common with the other lentic environ- ment, the Freycinet canal. Several taxa, such as chironomids, oligochaetes, Asellus and Caenidae, were found in all biotopes but with species changed according to the environment (cf. Caenis).

The relationship between taxa sampled and abiotic and ecological functions was considered with respect to the environmental impact of regulation and water quality. The variables were current velocity, longitudinal distribution (biocenotical classification according to Illies (196 1) and Verneaux, (1973), pollution sensitivity and tolerance to organic matter (saprobity). They allow the environmental structure and the trophic organization of the hydrosystem to be described. Species traits of taxa were determined from autoecological information summarized in the literature by Usseglio-Polatera (1991) and Tachet et al. (1987).

The trophic structure of the different biotopes was connected to their environmental conditions. Col- lector-gatherers were dominant in all aquatic environments because of sedimentation in the backwaters or because of the low current velocity in the canals and in some areas of the main stream. This feeding group was represented by ubiquitous taxa such as some chironomids or oligochaetes. Grazers were also present in all environments because of the development of macrophytes in the backwaters and near the banks in the canals and in the main stream. Shredders and collector-filters were essentially pre- sent in the main channel related to the lotic characteristics of this functional set and the diversity of the microhabitats of its banks. The faunistical assemblages of the aquatic environments of the floodplain of the Doubs River were globally oligotrophic to /3-mesotrophic connected to the average organic load and pro- ductivity of these environments.

Most of the taxa were limno-rheophilic (low to average current velocities, i.e. less than 75cms-I). Rheophilic taxa were abundant in the stations of the main channel with riffles or stations located down- stream from weirs where the current velocity was rapid (in particular stations D01, D02, D08, DO9 and D07).

According to its macrobenthic fauna, the main channel of the Doubs River downstream from the Allan confluence changed from an epipotamon to a metapotamon in the downstream direction (biocenotical classification B7-B9 according to Verneaux, 1973). However some rhthronic taxa (biocenotical classification B4-B6) were present in the upper part of this stretch (Hydropsyche siltalai; Hydropsyche pellucidula, Cheumatopsyche lepida, Psychomyia pusilla and Elmis maugetii).

Flow velocity is the most important structural factor for the macroinvertebrate communities of the Doubs hydrosystem. Bournaud et al. (1987a) stated that discharge is a synthetic key variable in the functioning of the lotic sets of the hydrosystems. Statzner and Higler (1986) and Statzner et al. (1988) suggested that the hydraulic characteristics such as the mean current velocity and the turbulence were the most explanatory environmental variables of the lotic macroinvertebrates distribution. The aquatic fauna of the Doubs River is varied because of the diversity of microhabitats and current velocities. The lotic aspect is well defined. Rheophilic taxa are present, despite being localized.

The faunistic diversity of the main stream reported by Verneaux (1973) was still apparent. The richness of some insect orders are quite similar (Table 111), although the number of taxa in common between the two studies is low. It could be explained by differences between the sampling techniques (in 1973 collection of imagines with traps and capture of adults, beside usual benthos-sampling techniques), differences of identification (H. contubernalis, H . exocellata and H . ornatula were all identified as Hydropsyche ornatula in 1973, for example). However, the ecological status of the assemblages remains the same (see above):

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER 627

- A - H. confubernalis-

. Main channel

Of tho mbl R I ~ rcaraciea sp. -

@ .. I *.a

-H. exocellata - H. pe//ucidu/a -

? 1 C. iepida A. cinereus - H. si/ta/ai -

. .. a

1 Garnrnaridae .E. rnaugetii - Sirnuliurn sp.

. .. I I Iy

- - C. diptervm- r 1 Coenagrionidee -Micmnecta sp. A. //uviafi/is P. acuta -

Msln channel

Douh RIvIr

.. ’

L Fraycln.1 canal

-.@. 1% 5% losb 20% 50%

Fmyclnal canal .

Figure 6. Longitudinal relative abundance of significant taxa. The size of the circles is proportional to the abundance. The stations are separated according to the three types of functional set. A; Taxa only present in the main channel. B; Taxa more abundant in the lentic

environments (backwaters and Freycinet canal). C; Ubiquitous taxa

628 J.-F. FRUGET. J. DESSAIX AND S. PLENET

Table 111. Comparison of species richness of four insect orders from Verneaux’s data (1973) and this study

Faunistic groups Verneaux (1973) ARALEPBP (1993) Taxa in common

Plecoptera Trichoptera Ephemeroptera Odonata

2 22 15 9

1 21 14 6

sensitive and/or rhithronic taxa (such as the mayflies Ecdyonurus and Baetis, the caddisflies Ceraclea annulicornis or Arthripsodes cinereus) are still present.

Predictions concerning the introduction of a new scheme Regulation schemes affect stream hydraulics by homogenizing mesologic conditions and the number of

rheophilic species consequently decreases as is shown in the by-passed sections of the Rh6ne River (Fruget, 1991, 1992; Dessaix et al., 1995). Invertebrate assemblages and their trophic structures typically changed in many regulated rivers worldwide (Krenkel et al., 1979; Stanford and Ward, 1979; Ward, 1982; Lillehammer and Saltveit, 1984; Moreau and Planas, 1984).

For example, a ten year survey period of the macroinvertebrate communities of the impounded French Upper Rhbne (Dessaix et al., 1995) showed that the community changes and the alterations of the benthic macrofauna due to the regulation were mostly induced by the decrease in current velocity. Current velocity was evidently influenced by water discharge, which controlled the other abiotic factors (depth, granulometric composition of the substratum). In the dammed sections, rheophilic and lithophilic species decreased and disappeared. Limnophilic taxa increased with the expansion of macrophytes and the deposition of fine particles. In the by-passed sections, hydrological patterns (flow rate, current velocity, water surface area, depth) governed the presence or absence of species, whereas abundance was greatly influenced by substratum stability, food quality and quantity, development of aquatic macrophytes and periphyton. These environ- mental alterations excluded or limited potamic taxa, such as Hydropsychidae and Heptagenia sulphurea, which are characteristic of lotic rivers. However, some potamic species which are characteristic of lentic environments did occasionally increase (for example Potamanthus luteus, Polycen tropus jlavomaculatus). These results were similar to those obtained in studies on other regulated European rivers (Armitage, 1976, 1978; DCcamps et al., 1979, Henricson and Muller, 1979).

Aron and Smith (1971) emphasized the danger of biological exchanges, and consequently of the introduction of exotic species, due to this type of canal which connects two hydrographic basins (transport of species supplanting autochthonous species, passage of parasitic diseases, etc.). Balon et al. (1982) and Bryson (1992) demonstrated this hazard for the new Rhine-Main-Danube connection. Meurisse-Genin et al. (1987) reported that Atyaephyra desmaresti, a Mediterranean freshwater shrimp invaded the northern part of Europe by the waterways. The problem cannot be excluded on the Doubs River, where the Ponto-Caspian shrimp Corophium curvispinum, a brackish water species acclimatized to freshwater and noted for the first time in the Rhine River in 1987, was collected in the Sa6ne River in 1991 (Genin, 1992). Jazdzewski (1980) pointed out the important role of ships in the transport of this crustacean.

Conservation perspectives and solutions Conservation management must be oriented towards preserving the present aquatic communities of the

Doubs River, i.e. those of a lotic ecosystem with active connections with its backwaters. The primary goal from a conservation perspective must be the maintenance and the improvement of areas highly sensitive to disturbances (such as regulation works). This can be achieved in three ways. (i) Maintenance of at least the average low flow discharge in all the future by-passed sections, i.e. natural sectors. This discharge should be in accordance with the water management policy (French fishery and water laws). (ii) Maintenance of the diversity of the aquatic environments, in particular through the maintenance of river-floodplain

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER 629

connections. A decrease in biological connections (and consequently in biotic diversity) will occur if a decrease of morphological connections happens (Amoros, 199 1; Fruget, 1992). Backwaters have to be kept permanently connected with the main stream to continue to act as refuges, spawning and hatching areas, to avoid silting up and drying, and to permit their maintenance during floods. (iii) Maintenance of bank structure and diversity. These aquatic and terrestrial ecotones are important to the biological equilibrium of rivers. The natural banks show great diversity of microhabitats and act as biological reserves and nurseries for main channel species (Cogerino et al., 1995). The size of this riverine biological stock is connected to the length of natural banks.

However, if aquatic and/or terrestrial areas are lost, similar environments that function similarly must be created or restored to compensate (Cairns, 1986; Petts, 1988). These management recommendations require some changes from an engineering point of view. However, such experiences are now expanding and the techniques are well known (Amoros et al., 1987b; Kusler and Daly, 1989; Gulati et al., 1990; Osborne et al., 1993; Wyant et al., 1995; Henry and Amoros, 1995a, b). Our study provides information at a single point in time and can be regarded only as an instantaneous picture. From an ecological perspective, long- term monitoring must be carried out to study human-made disturbances and restoration measures undertaken in the river floodplains (Sparks et al., 1990). These are necessary to improve our knowledge of the aquatic environments. Biodiversity must now be considered, both at species and at the ecosystem level, as being of economic value (LCdque, 1994).

ACKNOWLEDGEMENTS

This study was supported financially by the CNR (Compagnie Nationale du Rh6ne) in order to update the environmental data of the Rhine-Rh6ne large ship canal project. Many thanks to Guy Collilieux (CNR, Lyon) who provided us with much helpful information and documents, to Pierre-Jean Martinez (GREBE, Lyon) and to Michel Centofanti (ARALEPBP) for their help during field work. Calculations and graphs were carried out using ADE software (Chessel and Doledec, 1993), except for Figure 3, which used MacDendro and Graphmu softwares (J. Thioulouse, URA CNRS 243, Laboratory of Biometry, University of Lyon I).

REFERENCES

Amoros, C. 1991. ‘Changes in side-arm connectivity and implications for river system management’, Rivers, 2 (2), 105-1 12. Amoros, C., Roux, A. L., Reygrobellet, J. L., Bravard, J. P., and Pautou, G. 1987a. ‘A method for applied ecological studies of fluvial

Amoros, C., Rostan, J. C., Pautou, G., and Bravard, J. P. 1987b. ‘The reversible process concept applied to the environmental

ARALEPBP, 1993. ‘Liaison navigable Rhin-RhBne. Actualisation des etudes d’environnement. Etude de la faune benthique’,

Armitage, P. D. 1976. ‘A quantitative study of the invertebrate fauna of the River Tees below Cow Green Reservoir’, Freshwar. Biol., 6

Armitage, P. D. 1978. ‘Downstream changes in the composition, numbers and biomass of bottom fauna in the Tees below Cow Green Reservoir and in an unregulated tributary Maize Beck, in the first five years after impoundment’, Hydrobiologiu, 58 (2), 145-156.

Aron, W. I. and Smith, S. H. 1971. ‘Ship canals and aquatic ecosystems’, Science, 174, 13-20. Balon, E. K., Crawford, S. S. and Lelek, A. 1982. ‘Fish communities of the Upper Danube River (Germany, Austria) prior to the new

Beffy, J. L. and Doledec, S. 1991. ‘Mise en evidence d’une typologie spatiale dans le cas d’un fort effet temporel: un exemple en

Bournaud, M. 1980. ‘Quelques methodes de traitement des donnees en ecologie, 2nd edn. Roneo, Dpt de Biologie Animale et Ecologie,

Bournaud, M. and Amoros, C. 1984. ‘Des indicateurs biologiques aux descripteurs de fonctionnement: quelques exemples dans un

Bournaud, M. and Cogerino, L. 1986. ‘Les microhabitats aquatiques des rives d’un grand cours d’eau: approche faunistique’,

Bournaud, M., Tachet, H. and Perrin, J. F. 1982. ‘Les Hydropsychides (Trichoptera) du haut RhBne entre Geneve et Lyon’,

Bournaud, M., Tachet, H., Roux, A. L., and Auda, Y. 1987a. ‘The effects of seasonal and hydrological influences on the macroinverte-

Bournaud, M., Tachet, H., and Roux, A. L. 1987b. ‘The effects of seasonal and hydrological influences on the macroinvertebrates of the

hydrosystems’, Regul. Riv., 1 ( I ) , 17-36.

management of large river systems’, Environ. Manage., 11, 607-617.

Unpublished Report to CNR. Lyon. 33pp. + appendices.

(3), 229-240.

Rhein-Main-Donau connection’, Environ. Biol. Fish., IS (4), 243-271.

hydrobiologie’, Bull. Ecol., 22 ( I ) , 169-178.

Univ. Lyon I. 104 pp.

systeme fluvial’, Bull. Ecol., IS ( I ) , 57-66.

Ann. Limnol., 22 (3), 285-294.

Ann. Limnol., 18 (I), 61-80.

brates of the RhBne river, France. I. Methodological aspects’, Arch. Hydrobiol., 102 (2), 287-304.

RhBne river, France. 11. Ecological aspects’, Arch. Hydrobiol., Suppl. 76, 112, 25-51.

630 J.-F. FRUGET, J . DESSAIX AND S. PLENET

Bournaud, M., Roux, C., Tachet, H., and Cellot, B. 1988. Les Hydropsyche due RhBne. Rapport A T P CNRS Biologie des Populations.

Bryson, B. 1992. ‘Main-Danube Canal: Linking Europe’s Waterways’, Natl. Geogr., August 1992, 3-31. Cairns, J., Jr, 1986. ‘Restoration, reclamation and regeneration of degraded or destroyed ecosystems’, in Soule, M. E. (Ed.), Conservation

Castella, E., Richardot-Coulet, M., Roux, C., and Richoux, P. 1984. ‘Macroinvertebrates as “describers” of morphological and

Castella, E., Richardot-Coulet, M., Roux, C., and Richoux, P. 1991. ‘Aquatic macroinvertebrate assemblages of two contrasting

Centofanti, M., Fruget, J. F., and Olivier, J. M. (submitted). Ichthyofauna of the Doubs River prior to completion ofthe Rhine-RhBne

Chessel, D. and Doledec, S. 1993. A.D.E. Version 3.6: Hypercard Stacks and QuickBasic Microsofi Programme Library for the Analysis

Cogerino, L., Cellot, B., and Bournaud, M. 1995. ‘Microhabitat diversity and associated macroinvertebrates in aquatic banks of a large

Cummins, K. W. and Lauff, G. H. 1969. ‘The influence of substrate particle size on the microdistribution of stream macrobenthos’,

Decamps, H., Capblancq, J., Casavona, H., and Tourenq, J. N. 1979. ‘Hydrobiology of some regulated rivers in the southwest of

Dessdix, J., Fruget, J. F., Olivier, J. M. and Beffy, J . L. 1995. ‘Changes of the macroinvertebrate communities in the dammed and

Doledec, S. and Chessel, D. 1989. ‘Rythmes saisonniers et composantes stationnelles en milieu aquatique. 11. Prise ne compte et

DolCdec, S. and Chessel, D. 1991. ‘Recent developments in linear ordination methods for environmental sciences’, Advances in Ecology,

Elliott, J . M., Drake, C. M., and Tullett, P. A. 1980. ‘The choice of a suitable sampler for benthic macroinvertebrates in deep rivers’,

Fruget, J. F. 1991. ‘The impact of river regulation on the lotic macroinvertebrate communities of the Lower RhBne, France’, Regul. Riv.,

Fruget, J. F. 1992. ‘Ecology of the Lower RhBne following 200 years of human influence: a review’, Regul. Riv., 7 (3) , 233-246. Genin, B. 1992. ‘Corophium curvispinum Sars, 1895 (Crustacea, Amphipoda) dans la SaBne’, Bull. Sci. Bourg., 45, 27-32. GREBE 1992. Projer de canal a grandgabarit Rhin-RhBne. Etude bibliographique. Physico-chimie et hydrobiologie. Unpublished report to

Gulati, R. D., Lammens, E. H. R. R., Meijer, M. L., and Van Donk, E. (Eds) 1990. Biomanipulation, Tool for Water Management.

Henricson, J. and Muller, K. 1979. ‘Stream regulation in Sweden with some examples from central Europe’, in Ward, J. V. and Stanford,

Henry, C. and Amoros, C. 1995a. ‘Restoration ecology of riverine wetlands: I. A scientific base’, Environ. Manage., 19 (6), 891-902. Henry, C. and Amoros, C. 1995b. ‘Restoration ecology of riverine wetlands: 11. An example in a former channel of the RhBne River’,

Illies, J. 1961. ‘Versuch einer allgemeinen biozonotischen gliederung der fiessgewasser’, Int. Rev. Ges. Hydrobiol., 46, 205-213. Jazdzewski, K. 1980. ‘Range extensions of some gammaridean species in European inland water caused by human activity’,

Krenkel, P. A,, Lee, G. F., and Jones, R. A. 1979. ‘Effects of T.V.A. impoundments on downstream water quality and biota’, in

Kusler, J. A. and Daly, S. 1989. ‘Wetlands and river corridor management’, Proceedings ofan International Symposium, July 5-9,1989,

LevCque, C. 1994. ‘Le concept de biodiversite: de nouveaux regards sur la nature’, Narures-Sciences-Sociit~s, 2 (3) , 243-254. Lillehammer, A. and Saltveit, S. J. 1984. ‘The effect of the regulation on the aquatic macroinvertebrate fauna of the River Suldalslagen,

western Norway’, in Lillehammer, A. and Saltveit, J. (Eds), Regulated Rivers. Universitetsforlaget, Oslo, Norway. pp. 201-210. Meurisse-Genin, M., Reydams-Detollenaere, A., Donatti, O., and Micha, J. C. 1987. ‘Carattristiques biologiques de la crevette d’eau

douce Atyaephyra desmaresti Millet dans la Meuse’, Ann. Limnol., 21 (2), 127-140. Moreau, G. and Planas, D. 1984. ‘Reponse, a court tenne, d’un ecosystime lotique nordique a une forte reduction de debit’, Verh.

Internat. Verein. Limnol., 22 (3), 1776-1785. Osborne, L. L., Bayley, P. B., and Higler, L. W. 1993. ‘Lowland stream restoration: theory and practice’, Freshw. Biol., 29 (2), 187-342. Pattee, E. and Gourbault, N. 1981. ‘Introduction pratique a la systematique des organismes des eaux continentales FranGaises. 1.

Turbellarits Triclades paludicoles (Planaires deau douce)’, Bull. Soc. Linn. Lyon, 50 (9), 279-304. Petts, G. E. 1988. ‘The role of ecotones in aquatic landscape management’, in Naiman, R. J. and Dkamps, H. (Eds), The Ecology and

Management of Aquatic and Terrestrial Ecotones, Vol. 4. Man and Biosphere Series. pp. 227-261. Plefkin, J . L., Barbour, M. T., Porter, K. D., Gross, S. K., and Hughes, R. M. 1989. ‘Rapid bioassessment protocols for use in streams

and rivers’, Rapport US EPA/444/4-89/001. Richardot-Coulet, M., Amoros, C., Reygrobellet, J. L., and Roux, A. L. 1982. ‘Diagnose des ensembles fonctionnels aquatiques

definis sur le Haut-RhBne FranGais. Application a une cartographie ecologique d’un systeme fluvial’, Eau du Quibec, 15 (2),

Roux, M. 1991. ‘Basic procedures in hierarchical cluster analysis’, in Devillers, J. and Karcher, W. (Eds), Applied Multivariate Analysis

Shannon, C. E. and Weaver, W. 1949. The Mathematical Theory ofCommunicarion. Univ. Illinois Press, Urbana. 117 pp.

URA CNRS 367, Univ. Lyon I. 50 pp.

Biology: Science of Scarcity and Diversity. Sinauer Publishers, Sunderland, MA. pp. 46-483.

hydrologicdl types of aquatic ecosystems abandoned by the RhBne River’. Hydrobiologia, 119, 219-225.

floodplains: the RhBne and the Ain rivers, France’, Regul. Riv., 6 (4), 289-300.

connection’, Environ. Manage., in press.

of Environmental Data. Documentation URA CNRS 1974, Univ. Lyon I.

European river’, Hydrobiologia, 304 (2), 103-1 15.

Hvdrobiolugia, 34 (2), 145-181.

France’, in Ward, J. V. and Stanford, J. A. (Eds), The Ecology of Regulated Streams. Plenum Press, New York. pp. 273-288.

by-passed sections of the French Upper RhBne after its regulation’, Regul. Riv., 10 (2-4), 265-279.

elimination d’effets dans un tableau faunistique’, Acta Oecologica, Oecol. Gener., 10 (3) , 207-232.

1, 133-135.

Pollut. Rep. Dep. Environ., 8, 36-44.

6 (4), 241-255.

CNR, Lyon.

Kluwer Academic Publishers, Dordrecht, The Netherlands. 628 pp.

J. A. (Eds), The Ecology of Regulated Streams. Plenum Press, New York. pp. 183-199.

Environ. Manage., 19 (6), 903-913.

Crustaceana, Suppl. 6, 84- 107.

Ward, J . V. and Stanford, J. A. (Eds), The Ecology of Regulated Streams. Plenum Press, New York. pp. 289-306.

Charleston, South Carolina USA.

146- 153.

in SAR and Environmental Studies. ECSC, EEC, EAEC, Brussels and Luxembourg. pp. 115-135.

MACROINVERTEBRATE COMMUNITIES OF THE DOUBS RIVER 63 1

Sparks, R. E., Bayley, P. B., Kohler, S. L., and Osborne, L. L. 1990. ‘Disturbance and recovery of large floodplain rivers’, Environ.

Stanford, J. A. and Ward, J. V. 1979. ‘Stream regulation in North America’, in Ward, J. V. and Stanford, J. A. (Eds), The Ecology of

Statzner, B. 198 1 . ‘The relation between “hydraulic stress” and microdistribution of benthic macroinvertebrates in lowland running

Statzner, B. and Higler, B. 1986. ‘Stream hydraulics as a major determinant of benthic invertebrate zonation patterns’, F r e s h . Biol., 16,

Statzner, B., Gore, J. A. and Resh, V. H. 1988. ‘Hydraulic stream ecology: observed patterns and potential applications’, J. N. Am.

Tachet, H., Bournaud, M., and Richoux, P. 1987. Introduction a l‘itude des macroinverte‘bris des eaux douces, 3rd edn. C.R.D.P. Lyon

Usseglio-Polatera, P. 199 1 . ‘Representation graphique synthetique de la signification icologique d’un peuplement. Application aux

Verneaux, J. 1973. Tours d’eau de Franche-Comtt (massif du Jura) - Recherches tcologiques sur le reseau hydrographique du Doubs.

Ward, J. V. 1976. ‘Effects of flow patterns below large dams on stream benthos: a review’, in Osbom, J. F. and Allman, C. H. (Eds),

Ward, J. V. 1982. ‘Ecological aspects of stream regulation: responses in downstream lotic reaches’, War. Polluf. Manage. Rev.

Wyant, J . G., Meganck, R. A,, and Ham, S. H. 1995. ‘A planning and decision-making framework for ecological restoration’, Environ.

Manage., 14 ( S ) , 699-709.

Regulafed Sfreams. Plenum Press, New York. pp. 215-236.

water system, the Schierenseebrooks (North Germany)’, Arch. Hydrobiol., 91, 192-218.

127-139.

Benfhol. Soc., 7 (4), 307-360.

& A.F.L. Paris. 155 pp.

macroinvertkbres du RhBne a Lyon’, Bull. Ecol., 22 (l), 195-202.

Essai de biotypologie’, Z’hise d’Etat. Universite de Besanyon. 260 pp.

Instream Flow needs Symposium, Vol. 11, Am. Fish. SOC., Bethesda, Maryland. pp. 235-253.

f New Delhi) , 2, 1-26.

Manage., 19 (6), 789-796.