dynamics (year 1992) and the possible incidence of zooplankton grazing ... aspects of the phytoplankton of the River Meuse were studied during 1992, at a point.
Hydrohiologia 289: 179-191, 1994. J.-P. Descy, C. S. Reynolds & J. Padisdk (eds), Phytoplankton i Turbid Environments: Rivers and Shallow Lakes. ( 1994. Kluwer Academic Publishers. Printed in Belgium
179
The phytoplankton community of the River Meuse, Belgium: seasonal dynamics (year 1992) and the possible incidence of zooplankton grazing V6ronique Gosselain', Jean-Pierre Descy' & Etienne Everbecq 2 i Unit of FreshwaterEcology, Department of Biology, F.U.N.D.P., rue de Bruxelles 61, B-5000 Namur, Belgium; 2 Centre Environment, University of Liege, Sart Tilman, B-4000 Liege, Belgium
Key words: potamoplankton, community structure, European rivers, grazing
Abstract Qualitative and quantitative aspects of the phytoplankton of the River Meuse were studied during 1992, at a point 537 km from the source. The phytoplankton was dominated by diatoms and green algae. The Stephanodiscus hantzschii-groupwas especially prominent. Other important taxa were Cyclotella meneghiniana,small Cyclotella and Thalassiosira, Aulacoseira ambigua and Nitzschia acicularis. Cell abundances varied from less than 1000 units ml- l to about 25 000 - 30 000 units ml- during the blooms. The Stephanodiscushantzchii-groupconstituted almost entirely the first spring bloom. During the summer period, small Thalassiosiraceae developed markedly and large Thalassiosiraweissflogii appeared. During this period, green algae dominated diatoms as expressed in cell abundances. The main Chlorococcales were Scenedesmus quadricauda,Scenedesmus div. sp., Dictyosphaerium ehrenhergianumand Pediastrumduplex. Dinophyceae contributed a significant biomass during the summer period. Total biomass varied between 100 and 3 650 ttg Cl ' . As previously observed (Descy, 1987), the factors regulating the phytoplankton growth were clearly physical variables: discharge, temperature and irradiance. However, in the summer period, low abundances might indicate a regulation by biotic factors. The impact of grazing by zooplankton is discussed, on the basis of observations of zooplankton development in the River Meuse and on the basis of simulation by a mathematical model. A comparison is carried out with recent data of phytoplankton in large European rivers. Introduction In a previous publication, Descy (1987) produced a synthesis on the potamoplankton of the River Meuse in its Belgian part, with special reference to (i) factors controlling the phytoplankton dynamics, (ii) comparison with data from other reaches of the same river and (iii) comparison with other large European rivers. The literature review showed that algal communities seemed to develop in a characteristic way from the 'relatively fast-flowing headwaters' to the downstream reaches. In the first river type, the suspended algal assemblage was characterized by lower abundances and dominance by pennate diatoms and green algae, accompanied by some Chrysophyceae and Cryptomonads, with a significant contribution of tychoplankton, i.e. from algae originally growing on
the bottom or associated substrates. In the second type, referred to as 'large lowland rivers', the community is typically dominated by centric diatoms throughout the year, although Chlorococcales may be an important component in summer conditions. The comparison with other data from the same river tended to confirm the downstream increase in diatom dominance, which was tentatively attributed to tolerance of decreasing available light in the water column. Most large European rivers, for which sufficient data were available, present a similar phytoplankton community structure and dynamics, controlled by physical factors (discharge, light, temperature). However, in well-documented cases (i.e. River Danube, Schmidt & Virts, 1980), it seemed certain that the nutrient enrichment from human activities over the
180 past decades have brought about significant changes of biomass and composition of the phytoplankton. Since this modest contribution to potamoplankton ecology and composition, which succeeded, among others, pioneer studies in England (e.g. Swale, 1969; Lack, 1971) or in Hungary (Szemes, 1967), interest in riverine phytoplankton has continued to grow. For instance, the synthesis on river algae by Reynolds (1992) gave a full account (i) of the importance of time and spatial variation of discharge on the transition between diatoms and green algae, (ii) of the response to light limitation in a circulating turbid water column, as well as (iii) of the role of slow-flowing reaches for the survival of some taxa or (iv) of the selection of 'R-strategists' (Reynolds, 1988) in these disturbed and nutrient-rich environments. We present here an updated account of the potamplankton community of the River Meuse, including some taxonomic details, and discuss the possible role of biotic interactions in influencing the phytoplankton community. In addition, some recent data on other rivers are compared.
Description of the site studied The River Meuse (Fig. 1) rises in the east of France and flows through Belgium and the Netherlands, where it meets the lower Rhine, forming the Dutch Delta, which opens in the North Sea. The total length of the river is 885 km and its catchment area is about 36 000 km 2, 40% of it being in Belgian territory. In all of its Belgian course, the River Meuse has been regulated for navigation, with weirs and locks distributed along its length. The site studied is situated at a point, 537 km from the source. It belongs to the less polluted part of the River Meuse in Belgium. At this site, the mean depth is 3.95 m and the mean width is 100 m. The year 1992 presented a typical discharge pattern, with a large variation between the winter (max 480 m3 s- j) and the summer period (min 28 m3 s- 1), with a low-flow period of several months (Fig. 2). The water temperature varied between 1 and 25°C (Fig. 2). The River Meuse has alkaline nutrient-rich waters. Some variations in the nutrient content occur over an annual cycle, due to inputs from the drainage area (N, Si), from sewage (mostly P) and to uptake by primary producers. However, nutrients are not depleted to levels where they could be considered to limit phytoplankton growth. More details can be found in Descy et al. (1987, 1988).
Materials and methods From the end of January 1992 to mid-October 1992, 25 water samples were taken with a 101 polyethylene container in the subsurface zone. Sampling frequency was more or less once a week but only one sample was taken in August and none in September. One or two litres of each sample were fixed with Lugol's iodine. The suspended matter was concentrated by successive settling in glass cylinders of decreasing volumes, to a final volume of 10 ml. The suspensions were then mounted on a Burker cell for counting under a standard microscope (DIAPLAN) and the number of units per ml was determined for every taxon present. Precise identification of the unicellular centric diatoms required the oxidation of the organic matter to examine the siliceous walls. To this end, suspensions were 'cleaned' either by burning or by H202 treatment and the material was mounted with Naphrax. The relative proportions of these centrics were then determined using and 100x oilimmersion objective. Specimens were identified and named from Krammer & Lange-Bertalot (1991) and by reference to the taxonomic literature cited in Descy & Willems (1991). At the present state of our observations, some uncertainties concerning the names of the small Cyclotella species still remain. Although we do not have precise cell counts for every taxon, we give, in Table 1, the list of unicellular centric diatoms we have identified in the samples of the year 1992. In the counts, Cyclotella pseudostelligeraand C. stelligera Cleve & Grunow (in Van Heurck) were not distinguished. Nevertheless, C. stelligera has been previously identified in the upper Meuse. The species formerly identified as Stephanodiscus rotula (Kutzing) Hendey has to be classified as S. neoastraea. Among the most important diagnostic criteria are the presence (or absence) and different distribution of valve fultoportules on the valve faces (Casper & Klee, 1992). Concerning the species S. minutulus and S. parvus, we have separated them according to Krammer & Lange-Bertalot (1991), despite the fact that there is no general agreement on their classification. The counting unit was the cell, except for green and blue-green algae, for which the counting unit was a colony or a trichome. In every case, the cell or unit biovolume was determined by measurements of the relevant dimensions and by calculations using the formula of the closest simple solid (Lewis, personal communication). Biomass (carbon content) was calculated from the mean biovolume of each taxon using the
181
THE NETHERLANDS
BELGIUM "La Plante" Studied site
FRANCE
B
Fig. 1. Map of the basin of the River Meuse, with the location of the studied site at 'La Plante' (Namur); the shaded area is the Belgian part of the basin.
Strathmann equation (1967, in Smayda, 1978). Total biomass was also estimated from chlorophyll a values using a conversion factor C vs chlorophyll a of 37 (Descy & Gosselain, 1994).
Water analyses were made on each day of sampling. The measured variables were temperature, pH, dissoved oxygen, conductivity, total alkalinity, nitrate, ammonia, nitrite, soluble reactive phosphorus (SRP),
182 25
800 700
20 600 '
U 4-
,
15
E
500
E
400
cm
300
,
r-
10 as
200 5 100
J
F
M
A
M
J
J
Time
A
S
0
N
D
(months)
3 Fig. 2. Water discharge (m s- 1) and temperature (C) in the River Meuse during the year 1992. Data from the CIBE at Tailfer.
Table I. List of unicellular centric diatoms found in the River Meuse in 1992 Bacillariophyceae - unicellular centric diatoms Actinocyclus normanii (Gregory ex Greville) Hustedt Cyclostephanos dubius (Fricke) Round Cyclotella atomus Hustedt Cyclotella caspia Grunow Cyclotella meneghiniana Kiltzing Cyclotella ocellata Pantocsek Cyclotella pseudostelligera Hustedt Cyclotella radiosa (Grunow) Lemmermann, syn. Cyclortella coma (Ehrenberg) Kiltzing Stephanodiscus hantzschii Grunow Stephanodiscus hantzschii f. renuis (Hustedt) HAkansson & Stoermer Stephanodiscus imisitatus Hohn & Hellerman Stephanodiscus minutulus (Kutzing) Cleve & Moller Stephanodiscus neostraea HAkansson & Hickel Stephanodiscus palrus Stoermer & HAkansson Thalassiosira weissflogii (Grunow) Fryxell & Hasle
reactive silica. All these measurements were made by means of standard methods. In addition, particulate carbon, nitrogen and phosphorus were determined as described in Descy & Gosselain (1994). Data on solar energy were provided by the Royal Meteorological
Institute and discharge by the CIBE (Compagnie Intercommunale Bruxelloise des Eaux) at Tailfer (km 530). Chlorophyll a was measured by the acetone/methanol 5:1 method (PNchar, 1987), and calcu- lated using
183 Lorenzen (1967) equations, according to Marker et al. (1980). The mean light available in the water column (I) (Fig. 3) was calculated using the following formula: I = Io/lkz',
where I'o is the total daily irradiance (PAR) in /E m- 2 d- 1 , k the vertical attenuation coefficient of light (m - l) and z the mixing depth (m), equal to mean depth of the channel. The mathematical model of the River Meuse (Billen et al., 1985; Descy et al., 1987) was run by the Environment Centre of the University of Liege, in order to simulate the phytoplankton growth and the losses attributable to sedimentation and zooplankton grazing. This simulation used the actual values of the physical variables (discharge, solar energy, temperature) observed in 1992.
Results Composition of the phytoplankton As in former observations along the same stretch of the River Meuse (Descy, op. cit.), the 1992 data confirmed the dominance of diatoms and green algae, representing respectively 81% and 15% of the total numbers of algae encountered. Other algal groups were: Cynaobacteria, Cryptophyceae, Euglenophyceae, Dinophyceae and Chrysophyceae. Among diatoms, the Stephanodiscus hantzschiigroup (which includes S. hantzschii, S. hantzschii f. tenuis, S. parvus and S. invisitatus) was largely dominant. Then we find Cyclotella meneghiniana, small Cyclotella and Thalassiosirataxa (called 'small Thalassiosiraceae'), Aulacoseira ambigua (Grunow) Simonsen and Nitzschia acicularis W. Smith. Centric diatoms largely dominated on pennates throughout the year. Algae abundance and seasonalvariations The total phytoplankton abundance varied from less than 1000 units ml - i to more than 30000 units ml - ' during the blooms. The maximum was reached in March with 34000 units ml- . The most striking fact in the seasonal variation is the alternation of diatoms and green algae, the latter becoming dominant (expressed as cell abundance) in the summer period.
Development over the year is summarized in Fig. 4. Plankton was scarce until II March. Characteristic taxa of this period were Aulacoseira ambigua, Cyclotella meneghiniana, C. radiosa,Stephanodiscus hantzschii-group(dominant), S. neoastraea,Nitzschia acicularisand some Naviculae lineolatae. The limited growth at this period was clearly related to the high flow occuring in spring. The first bloom, on 11 March, developed quickly during a short period of low flow (Fig. 2) and consisted almost entirely of S. hantzschii. Then the discharge increased again and the phytoplankton numbers remained low with the same characteristic taxa as previously found. Also appearing were Navicula lanceolata (C. Agardh) Ehrenberg, Synedra spp. and some Chrysophyceae and, among the green algae, Coelastrum microporum Nag.. From 14 April to 15 May, a second diatom spring bloom occured, which comprised mainly Stephanodiscus spp. but also Cyclotella spp., Nitzschia acicularis, Aulacoseira ambigua. Some Chlorophytes began to develop: Coelastrum microporum, Chlamydomonas sp. and then various Scenedesmus taxa. During the spring period we also found sporadically some blue-green algae (Oscillatoria sp.) and some Cryptophyceae (mainly a minor species referred to as Chroomonasacuta Uterm6hl). The summer period was characterized by a greater proportion of green algae (Scenedesmus quadricauda [Turp.] Br6b. sensu Chod. and Scenedesmus spp., Dictyosphaerium ehrenbergianum Nag., Pediastrum duplex Meyen and a decrease in the total abundance of the phytoplankton. Some diatoms developed and showed relatively marked optima: small Thalassiosiraceae and large Thalassiosiraweissflogii, Aulacoseira ambigua, Skeletonema potamos (Weber) Hasle. Dinophyceae (Peridiniumspp.) developed markedly in late summer. From the end of May and throughout the summer period, the phytoplankton biomass remained low. During spring and summer periods some bluegreens developed, mainly Chroococcus minutus (Kitz.) Nag., C. dispersus (Keissl.) Lemm. and Oscillatoria limnetica Lemm.. Among the Cryptophyceae, Chroomonasacuta and Cryptomonasovata Ehrenberg occasionally reached relatively high cell numbers. The fall period was characterized by a qualitative reversal to the spring situation, with diatom dominance, but with a still reduced algal abundance. In Fig. 5, we present the relative proportions of the main taxa of the unicellar centric diatoms (dominant group).
184
10000 9000 C
8000
O
7000
E
'3:
o
6000 5000
a
.-
4000
'-
3000
.M
2000 1000
0 (months)
Time
-2 Fig. 3. Available light in the water column (E m d- ),calculated from solar energy, vertical attenuation of light and depth, at the studied site of the River Meuse during the year 1992. Solar energy data from the CIBE at Tailfer.
35000
* Other groups 30000
E Diatoms
_
i
i
*
20000
e
15000
0
E
25000
Chlorophyceae
~ I
i:
10000
i
i
Iq:W 7
k
5000
F716A L -0
O
'1 I IVJ
I4 R C:z ION 0 1 0 a,
NI I QR 01 I 01 0
, N
I
_
ii: , i
IIC
'C Sc rC c0
~R ~
CO ON Ot C
O
RzNt
N
rIN 0 4
~rveQ -
-0 o 0l4
ON I
R
-
1 0 4 0 '0 '0
C
rQ
-
x_
0 1 0~
EI r r
'
r
o o
I I
svl
*01III
,
!
j
·
j i
eq
R
o
_
Time (Sampling days) Fig.4. Variations of the main algal groups at the studied site of the River Meuse during the year 1992, expressed as abundances (units ml- ).
I 100% 90% =
80%
. in
70%
Thalassiostraweissfiogii C Stephanodiscusneoastraea
60
'IV l
o
50%
E small Thalassiosiraceae
8 :
Cyclostephanos dubius
-
d
40%
Li Cvclotellameneghinana
Z
Stephanodiscus gr. hantzschii
20% 10% 0% o 0
C:
;.
h N> c
_ __
2caCC
ccc c
N
R
Z
E
_cc
=
rat _
0
(N
_ t
_
_
C 00 NC _
-
r-t -,l Ps N
N
cL
C
Time (Sampling day) cc Time (Sampling day)
c
zt.
i.0 -
-
-
Fig. 5. Relative abundances of the main unicellular centric diatoms at the studied site of the River Meuse during the year 1992.
Biomass development If the biomass (calculated using the Strathmann equation, 1967, in Smayda, 1978) is considered (Fig. 6) instead of cell abundances, it is noticeable that diatoms remained largely dominant throughout the year. Dinophyceae, which have large cells, may at times represent a significant biomass, even more important than that of the green algae. The estimation of the total biomass from chlorophyll a agrees satisfactorily with the calculation of carbon from biovolumes (Fig. 7). However, conversion of chlorophyll a to algal carbon did not take into account a possible variation in cell size (Vbros & Padisik, 1991 ), nor a possible variation of chlorophyll a per cell attributable to light acclimation. In the present case, we applied the average C:Chla ratio for the River Meuse (Descy & Gosselain, this volume). This simplified conversion may account for some local discrepancy between algal C calculated from biovolumes and from chlorophyll a. Physical and chemicalfactors Some physical and chemical characteristics of the site studied in 1992 are presented in Figs 2, 3, 8 and 9. As has already been suggested, discharge seems to be the first factor controlling phytoplankton biomass.
It is worth noticing that some species (mainly 'small Stephanodiscus') can develop at low temperatures and low irradiances, as on 1I March, but they need to seize the opportunity of a low-flow episode. Indeed, when the flow increased after the 11 March, phytoplankton could not maintain high biomass levels. It is also striking that some green algae (small colonies of Coelastrummicroporum) participated in the first spring bloom. Later on, the second spring bloom occurred when discharge decreased again. At that time, irradiance and temperature were increasing, and both factors permitted the development of green algae and their progressive dominance (expressed as cell abundances) over diatoms. By contrast, nutrients do not seem to have any influence, even if some concentrations varied according to the algal biomass. For instance, during the second spring bloom, the depletion of SRP showed an inverse relationship with chlorophyll concentration, which is attributable to the algal uptake (Fig. 8). Similarly, the low silica level recorded in spring (AprilMay) is related to the demand for diatom growth, as well shown in Fig. 9. Such an impoverishment in Si in the river water is insufficient to support the conclusion that Si was 'limiting', although it might nevertheless have been a factor favouring green algal development, which is also promoted by increasing irradiance and temperature.
186 -
AOo
1 Dinophyceae
3500
C Cryptophyceae 3000
* Cyanobacteria 2500
:
[
Diatoms
M Chlorophytes
2000
.= *M 1500
1000
500
0 N
20
_
-
2
N c c N
c c
6
_C g
N-
