Environmental vegulation in African shallow lakes and wetlands

Environmental vegulation in African shallow lakes and wetlands

,Jick F. TALLING (1)

ABSTRACT

An illustrated account is given of environmental conditions and their controlling mechanisms in African shallow lakes and associated wetlands. The subject is introduced by a relatively well-studied case-example, from the productive Lake George in Uganda; here inieractions belween physical, chemicaI and biotic factor-systems are emphasized. There is particular interest in wind-generated patierns of horizontal disfribution in a basin of near-uniform morphometry, in hydrological buffering against any development of a flooding-drying cycle, and in the pervasive influence of diel cycles in a seasonally equable environment.

Individual components of ihe physical and chemical factor-syslems are then treated on a comparative, pan- African basis. Seasonal cycles of wafer temperature are related 10 the systemaiic variation of solar radiation input with latitude, with some imporianf modifications associated with factors fhat control ofher components of the energy budget. The influence of altifude upon waier temperature is quaniified. Persistent temperaiureldensiiy siraiificafion is generally ill-developed in fhese shallow waters, but cari be governed by a seasonal wind régime (e.g. harmattan) and displaced by a predominant solute[density stratification in saline waters. The range of toial solute concentration is very wide, according 10 available inputs and opportunities for evaporative concentration. With if are correlated, positively or negatively, the concentrations of tnany chemical components. Of the major nutrienf elements, Si is typically present - by world standards - in considerable concentration but is liable 10 depletion by diatom growth; forms of N (NHJ-N, Nos-N) are typically in low concentration, with some influence from (femperature-sensifive) denitrification; and P is oflen in considerable concentration, especially in many - but no1 a11 - soda lakes. A guide is given, with examples, 10 ihe quantities incorporated in the biomass of algal or macrophyte communities. Further examples illustrate factors controlling the varying concentrations of dissolved gases (especially 02, CO2 : biogenic processes), of heavy metals (especially Fe, Mn, CU) and also of man-made eutrophication of some shallow waters. Exatnples of chetnical budgets are also given, especially the more complete ones for lakes Chad and Naivasha wilh variable chemical incorporation into bhe seditnents. Qualitative and quantita&e aspecis of sediment generation are briefly reviewed.

Three types of cyclic environmenfal change in lime are described, fhaf involve concurrent changes in many variables. The diel (24 h) cyrle is impressed by the daily short-wave radiation but often modified bq wind-stress; it involves heat accumulation, temperature rise, density stratificafion with ils constraints on vertical distributions, and aspects of photobiology that include Os/COs exchange. The annual (12 mo.) cycle cari have radically differenl forms depending on wheiher it is or is net accompanied by extensive volume change of a flooding-drying cycle. The latter is well-developed in many African river floodplains. In many shallow lakes subject 10 irregular rainfall (e.g. Chad, Chilwa) there tends 10 be annual carry-over and cycles of longer period. The coniraction phase is one of terrestrial recovery followed by aquatic stress that may include biological effects of higher salinity. Reflooding during the

(1) Freshwafer Biological Association, Ambleside, Cumbria LA22 OLP, U.K.

Rerr. hydrobiol. trop. 25 (2) : 87-M (1992).

Sd .J. F. TALLING

expansion phaw cari induce a flush of nutrient release as rvell as a longer-ferm snlinity change and aquatic recoloniza- [ion.

KEY WOHI)S : Tropical - Swamps - Radiation - Temperature - St.ratification - Sa1init.y - Nutrients - Srdimrnts - Ri.1dget.s - Cyc~les.

LE RÙLE DES CONDITIONS DE MILIEU DANS LES ZONES HUMIDES ET LES LACS PEU PROFONDS D’AFRIQUE

Les rondifiotts de milieu ut leur inpuence dans les lacs peu profonds d’Afrique et les marais assoc& sont dPcrites. L’e.rample du lac Ceorgr, en Ouganda, est d’abord donnd en itztroducfion, en soulignanl les inferactions entre lrs eomposantrs physiques. chimiquw ei biologiques. Dans ce lac, sont particulièrement soulignées les disfributions horizonfales yèni>rc>~s par le vtwt dans un bassin quasi-circulaire, la stabilifé du niveau de l’eau et l’influence des cycles nycth6rnc+aux dans un climat dépourvu de variations .saisonniércs sensibles.

Les diffhwtes combinaisons de facteurs physiques et chimiques sonf ensuite abordées sur une base comparutiue pan-africaine. Lrs cycles saisonniers de la température de l’eau sont mis en relation avec les variations de l’kergie solaire incidentt~, et les influences sensibles des autres composants du bilan énergétique. L’effet de l’altitude est quantifil. Ilne stratification de températureldensité est en général peu développ+e dans les milieux peu profonds, mais peut être r~gul~e par la suisonalité des vents (par exemple l’harmatfan) ou déplacée par une prédominance des effets de lu salinité dans certains tnilirux. Les gammes des concenfrations de solutés est très large, el foncfion des apport el des condiiions d’évaporation. Beaucoup de concentrations des divers éléments dissous sont corrélées, posiiivement ou négafivrmtwf, ù In charge dissoute totale. Parmi les t;léments nufriiifs, le Si asi en général présent en très fortes cortwtn’rations, wlntivernent au,r tnoyennes mondiales, mais esi aussi susceptihlP de fortes baisses liées à la croissance des diatomées. Lrs formes de l’azote (NHd-IV et NO.y-N) sont, de façon caracfèrislique, en faible concentration, avec une influence dP la dènifrification liée uux lempératures èlevées. Le phosphore est souvent en forte concentration, particulièremenf dans beaucoup de lacs sodés. Un diagramme est proposé. avec des exemples, des quantifés incorpo- rèes dans la bornasse micro- ou macrophytique. D’autres ezemples illusfrent les facteurs qui contrôlenl les concenfra- fions dtas yaz dissotus (particulièrement Os ei CO2 liés uux processus biologiques), des métaux lourds (Fe, Mn, CU) et aussi dr i’eutrol-‘hisation d’origine anthropique. Des bilans bio-géochimiques sont également présenfés, notamment pour les lacs les tnieux &diés comme le Tchad ou le L. Naivasha où les sédiments joue& un rôle importanf. La &»gL;r~6sr tif3 sf%Iimetits, sous ses aspects qualitatif ef quuntifatif, est rapidement passée en reuue.

Trois types d’èvolulion cyclique de l’environnemenf sont décrits. Le cycle nyclhèm,+al (24 h) dépendant des radiafions dp rourte longueur d’onde esf souvent modult; par les effets du venf ; il lnclui l’accumulation de chaleur, l’cr~~~~rnc~t~f«tion de fempérature, la stratification thertniqw avec ses effets sur les répartitions verticales, et des effets de photobiologie incluant les échanges O~jCOs. Le cycle annuel (12 mois) peui être très diffèreni selon qu’il est associé, ou non, à utw forte variation de volutne liée au cycle hydrologique de crue-décrue. Ce dernier aspect est surtout sensible dans lrs plaitws d’inondation. Dans beaucoup de lacs peu profonds soumis à un rt?gime pluviomètrique instable (L. Trhad, L. (~,‘hilwa), des effets de rémanente se font sentir avec des c!jcles de période plus longue. La phase de rtkession esf urw phase de recolonisation terresire suivie d’une perfurbatlon aquafique, et peul faire intervenir des taffets biolo@ques lit;s à une augmentation de salin&. La remise en eau peuf induiw une retnise en solution d’éléments rtrrfrififs arnsi qu’une modification durable de la salinif&.

hIo~s CLÉS : Environnement. tropical - Marais - Lacs - Afrique - Conditions de milieu - Phgsico-chimie.

1. INTR~~DIJCTION

In the &allnw inland water-bodies of -Ifrica, as in frrkhwaters elsewhere, environmental conditions are drterminrd by an interacting array of external and inknal influences. In thesr -physical and chemical, abiot-ic and biotir. processcs part.icipatP (Fig. 1). Thrir oppration encounters recurrent constraink

from the conditions of shallow basin morphometry, involving limited storage capacity. Further restric- tions arë set, by the prevailing tropical and subkopi- cal climat.es.

Although individual factors are numerous, they are ultirnately traceable to input-output relat,ionships of energy (e.g. radiant or sensible forms) and of mat.erials (e.g. water, solutes, particulates, gases),

AFRICAN SHALLOW LAKES 89

CLIMATIC DETERMINANTS

ENERGY EXCHANGE

* / \

\ WATER EXCHANGE

precipitation

BIOGENIC GEOLOGlCAL

DETERMINANTS DETERMINANTS I :

FIG. 1. - Diagrammat.ic represrntation of components involved in the control of envirorunenk by climatic-at.mospheric, geological and biogenic determinants.

Schéma des composants du sysfème aquafique. inpuencés par les facfeurs climnfique-(lfmosphéïigrte, géologique et biolo,yiquc.

and to their interna1 transformations and storage, Al1 contribute to t-he variability in time and space reviewed in this account, yet such variability is often dominated br a relatively small number of components. Their influence may derive from their magnitude, their variability, or their involvement of a qua- litatively crit.ic.al constituent. Examples include the component of short-wave solar radiation in the energy balance and hence temperat,ure regime; of the geological availability of leachable substrata for solut,e inputs and hence ionic composition; of the incidenc.e and magnit,ude of rainfall and hence inflow terms in the hydrological cycle ; of extreme values of the surface outflow t,erm, as when large in relation to storage volume (low retention time) or small in relation to inflow (salinization) ; of t.he frequency of wind stress in excess of values required for vertical mixing in open water; and of biological reactions upon the wat,er-mass and sediments in situations of high areal density of biomass.

This review is primarily concerned with natural water-bodies of mean depth < 5 m. A single case- study is first used to illust.rate important factors and t.heir interact.ion. A comparative survey is then made of t.he variation of individual environmental factors

Heu. hydrohiol. trop. 25 (2) : 87-M (1992).

in shallow waters over Africa as a whole. Chemical budgets are illustrat.ed by the two more complete examples from African shallow lakes, and modes of net t‘ransfer of material tel sediments are outlined. !Mention is finally given to some major systems of change in time, which involve many concurrent variations and are of especial importance for shallow water-bodies. Biological implicat,ions are notbd briefly ; they are examined more fully by DLIWONT (in press).

Although there is no previous integrated and specific treatment of the present topic, much relevant information (*an be found in the pan-African surveys of TALLING and TALLING (1965), GRIFFITHS (1972), BEADLE (1981), SY*IOENG et al. (1981), LIVINGSTONE and MELACK (1984), and DENNY (1985a). Some important. water-bodies (‘case st.udies ‘) are described in recent monographs or symposia, notably Lake George (GREENWOOD and LUND, 1973), t.he Nile system (RZOSKA, 19X), Lake Chilwa (KALK et al., 1979), Lake Sibaya (ALLANSON, 1979), and Lake Chad (CARILIOUZE, DURAND and LÉ~&QuE, 1953). Fig. 2 shows the location of t.hese and other waters to which frequent mention is made in the text.

J. F. TALLING

L. ICHKEUL

L. MARIUT

/ R. NIL,\\,,

t

L.NUBtA-NASSER

L. KAINJ, L. CHAD GEEEL AULIA nksn - _-.. - , 7 \ P

--SA SEIRES RESR

i \ SUDD SWAMPS

'\ \ .

BISHOFTU LAKES)

i L. VOLTA

L. VICTORIA

L. TANGANYIKA L. EDWARD --

L. MALAWI (NYASA)

L. SANGWEULU --“y \1

CASORA SASSA RESR

PRETORIA SALT PAN

30”-

L. MclLWAINE

FIG. 2. - Distribution of important wat.er-hodies in Africa. Names of shallow lakes and wet.lands ment.icmed frequently in the text are underlined. From TALIJNG (1990).

1~s $att~ ti’euu impwtantr d’.4friyue. Les noms des milieux peu profonds cif& souvent dnns le texte sont ,sozzlipés. D’après TALLING l!UO.

Lakr George (Fig. 3) is a shallow water-body of considerable &e (area ‘250 km”, mean depth 2.4 m) but simpltb morphometry, situated on the equat.or in \Vest. Uganda on the floor of t.,he Western Rift Val- lfky. The following information is largely derived from thr intensive studies of an IBP Team during 1967-19X!, for which the symposium volume organi- zrd by GREEN~• OD and LUND (1973) and the revitws of GREENWOOD (1976) and BURGIS (1978 - wit.h detailed bibliography) - provide convenient and int.egrated summaries. The lake contras& with ot.her case-studies, as of L. Chad, L. Chilwa, and the

Pongolo floodplain, in the reduced variability of temporal change, excepting that of the diel cycle. Such stabi1it.y has several modes of origin, from relatively invariant. inputs and from local buffering mec.hanisms.

The seasonal climate at. L. George is such that. sys- t,ematic variations of tIhe short-wave solar radiaCon input are small. Acc.ording to VINER and SMITH (1973), IO-day means have an annual Variat(ion bet,- ween 1720 and 2210 J (411 and 529 cal) cm-2 day-1. The estimated mean rate of evaporation, 5 mm day-1, is equivalent. t.o a loss of 1446 .J (346 cal) cm-2 day-1 and involves little seasonal variation (Fig. 4aj. The corresponding varititions in other major compo-

I%I*. hydrobiol. trop. 2.5 (2) : 87-1~2 (1992)

2.6 2 2.5 5

m 2.2 0

1965--- 1966~- 1967--- 1966o ,969-c,-

Months

R~U. hydrobiol. trop. 25 (2) : 87-144 (1992).

650 Ë "0 L

FIG. 3. - Map of Lake George, Uganda, with adjacent. water- bodies and relief. From VINER and ShuTH (1973).

Le lac George, Ouganda, awc les lacs et maricages associés, ef le relief. D’après tFlivER rt SWTH, 1973.

FIG. -4. - Lake George: monthly changes of characteristics associated wit.h the water-halance, including (a) evaporation (b) lake level (c) direct, rainfall ont.0 the lake (0) and lake volume (II), showing out.-of-phase relation. Adaptrd from

VINER ami SMITH (1973). Le lac George : variations mensuelles de uariables associées au bilan hydrique : a) évaporation, b) niveau du lac ef c) déphasage entre la pluie directe sur le lac (0) ef le volume lacustre (El).

D’après VINER et SMITH, 1973.

.J. F. TALLING

Ci) (ii)

Temperature (“Cl Chlorophyll p (mg rnT3)

(b) 30

El 9 120

cl ‘O;ui ai 210

Cc) 30

06.00 14.00 20.00

(a) 30

(d) 30

(b) 30

oa. 14.00 20.00

FIG. 5. - Lake George : examples of daily (diel) cycles in t.he dept.h-timr dist.ribution and iayering of (i) t.emperature (iso- therms in “C) and (ii) chlorophyll a content, of phytoplankton (isoplet.hs in mg 111.3 = pg I-1). Dates : (a) 18 May 1967, (b) 5 July 1967, (c) 26 Blarrh 1968, (d) 5 February 1968. Adapt.ed

from GANF (1974~). Le lac (ieorgr : exemple de cycles journaliers de répartifion de la femp>rature (isofhermes en “C) ef en concenfration de la chlorophylle (isopliples en mg rn;‘). Dates: a) 18 mai 1967, b) 5 juillet 1967, c) 26 mars 1968, d) 5 fbrrrier 1968. D’après GANF, 1974b.

nrnts of energy balance (net hack-radiation, sensible hrat transfer) are not known, but are also likely to he amall, as the lake bottom temperature varies only hetwrm 23 and 25 C.

In t-he Upper half of the wat,er-column. however, the diurnal change in solar radiation flux leads to net rlnergy Aorage of about 2.6 k,J (620 cal) cm-2 within an absorpt.ive shallow layer and hence elevated tem- pfbraturrs there. The latter net uncommonly reach

afternoon maxima of > 30 “C. Thus a pattern of temperature and density stratification, with barriers t,o vertical exchange, is set up with a diel frequency (Fig. 5). The resulting subdivision of the water-mass makes a corresponding impress on other entities in dynamic flux, including the concentrations of oxygen (Fig. 6b) and carbon dioxide with pH (Fig. 612) influenced by photosynthesis with respiration and surface exchange, and the biomass concentrations of positively or negatively buoyant. phytoplankton (Fig. 5). The daily stratification is ended by increased vertical mixing due mainly to a combination of reduced radiation input, increased wind strength late

0.8 - l-j ‘<fi

Ï .5 E .is

. pi ;....

i 5 0.4 - i

I'

z I l

.-. p .,... **-.

k....-\ '..

-.. 0" " " " " " L.'

06.00 12.00 16.00 20.00

(b1 ’

100

Ê 200 * * . . * I , 0 I

. . I

I 0 a . s--ï-- 70

0 06.00 12.00 16.00

(c) 100

12.00

Time (h)

FIG. 6. - ILake George : diel variation during 26 March 1968 in (a) incident. solar radiat.ion and, as depth-t.ime diagrams, (b) percentage oxygen sat.urat.ion and (c) pH. Adapt.ed from GANF

and HORNE (1975). For temperature see Fig. 5c. Le lac George : variations journalières le 26 mars 1968 de a) rayonnement solaire incidenf et 6) réparfition espace-temps du pourcentuge de saturation en orrygène et c) du pH. D’après GANF

et HORNE. 19Y.5. Pour la tempéraiure, voir la figure .5ç.


a) Current velocity (cm s-‘1

0 5 10 15 ,.--

i : velocity

r--7-

temp.

Eh40

Temperature PC)

standard station

FIG. 7. -Lake George : examples of (a) depth-distribution of current velocity and t.emperature, and (b) the horizontal distribution of current direction, on days when light winds were blowing, with possible consequences in concentric di&ibution patterns of (c) phytoplankton density as mg chlorophyll a m-3 (= p 9 1-l) and (d) density of zooplankt,on Crustacra as pg dry mass 1-l. Adapted from

VINER and SMITH (1973) and BURGIS et al. (1973). Le lac George : exemples a) de répartition oerticale de la oifesse du courani, b) de répartition horizonfale des direcfions du courant par oent léger aoec c) leur conséquences possibles sur la distribufion concentrique de la chlorophylle (mg.m”) et de la densité de crusiacés du

zooplancton (mg.m-3 de poids sec). D’après VINER et SMITH, 1973 et BL~RGKJ et. al., 197.3.

in the day, and transfer of sensible heat to a cooler nocturnal atmosphere with mean night minimum 16.5 “C.

hlthough t.he wind regime has a diel component of variability (VINEH and SMITH, 1973 Fig. 7) there is net. a pronounced seasonality as the surrounding region largely ascapes the July impact of the S.E. Trades influential elsewhere in East and Central Africa. Inpu& of wind energy tend, as noted above, to counter by vertical turbulence the stratifying influence of near-surface accumulations of heat. They also evoke horizontal transfers of water, whi.ch were s~udied by SMITH (in VINER and SMITH, 1973) who int.roduced current measurement,s by alumi- nium drogues suspended at various depths. These demonstrated a rapid decline in the uppermost 10 cm of c.urre& velocity set, up by light winds (Fig. 7a). Implications for disturbance of bottom sedirnents were also explored theoretically ; a critical water velocity for disturbance of about 25-30 cm s-1 was estimated to c.orrespond to an observed long- term sediment disturbance depth of 25 cm (frequent

R~U. hydrnbiol. irop. 25 (2) : 87-144 (1992).

disturbance to only c. 5 c.m. on chemical grounds - GANF and VINER, 1973, and Fig. 11). Relationships were deduced between wind-speed of 1 h, length of its over-lake travel or ’ fetch ‘, and theoretical water velocities (Uny cm S-1 at, the sediment-water interface near 2.5 m depth (Table T). From the incidence of wind speeds above 12.5 m s-1, about 0.3 7; of the time, Smith estimat,ed that. significant areas of the lake bed are likelp to be rlist,urbed about once every three weeks. Another important finding was a wind-

TABLE 1

Beaufort wind scale Wind 5 6 7 8

velocity (= 9.5 m sl)(*iZS m s-1) (= 15 m 5-l) (= 19ms-1) fetch (km) UD UD UD UD

2.5 9.8 16.8 26.1 36.9 5.0 12.9 22.3 50.4 7.5 14.1 24.5 $:8 57.2

10.0 14.3 25.6 37.5 61.9

9-t .J. F. TALLING

genrrat ion of rotat.ory wat.er movements over the lake, asti-clockwise under northerly winds (Fig. 7h) and probahl~ cloc*kwise under southerly winds. Such watrr rotat-ion has (‘onsequences for the often roncentric distribution of phytoplankt,on and zooplankton densities (Fig. 7c.,d) (GANF, 1974a). GANF (1971b) also II& the physical data t.o examine stahi- lity relationships in a st.ratified water column, employing as crit.erion est.imat.es of the Ric.hardson number, i;l relation t.o vertical movements of phyt.o- planktcm.

The hydrological balance is such that pronounced seasonai chnngrs in water level do not occur, the typical annual‘variation of levrl being net. more than f 0.2 m (Fig. 4~). However t.here is a small bimodal variat.ion (peaks about. June and December) whic.h (‘an be rrlated. with lag, to a corresponding variation in rainfall (peaks about April and October). Suc.h bimodality of rainfall, rharactrristic. of much of East. Africa, causes water input t.» be extended seasonally. Extension also results locally by t,he supply of water from generally higher and prolonged rainfall on t.he acljac&t Ruwenzori Mountains. Finally, a peculiar olltflow channel (Kazinga Channel : see Fig. 3) of very slight gradient but. considerable volurne, and thr downstrgarn Lekr> Edward, act. hydrostatically as furthrr local buffers to level fluctuation. The following ç4irnatps of the main quantities in the annual hydmlogical budget, wpre made by VINEn and SMITH (1973) :

irltlon (R,) 1948 (X 100 m”) dirtvt rainfall «n lake cv 205 vvafloratiorr from lakr CE) 456 wrfacr outt1ow (hy differenctb) (Ro) IfY-7 underground drainage (aswmed) (GJ 0

As the mean lake volume is 600 x 10” n$, t.he mean rt-btent.ion time (as \Tolurne/outflow) is approxi- mately 4.3 mont.hs but would vary seasonally by an order of magnitude. During thr drier seasons evaporation is calimulated (as c. 1.25 x 106 m3 day-l) to be larger than cruttlow (as c. 0.8 x 106 ni3 day-1) as a source of water loss.

Thr last. t-ypr of diffrrcnce, when more pronounced in clthrxr Afiiçan lakes, cari induce appreciable seaso- na1 or long-terni increase in t.he total ionic concentration or sa1init.y (also indicated by electrical conductivity). Such tluctuation is small in L. George (VINEH, 1969 ; see Fig. S), where t8he ionic concentration is close to the averaged characteristics of the intlows. Drainage to t.he latter does not extend from the vr’ry distinctive geological sources of Na+, K+, and Mg?+ in the Virunga volcanic field, which increase bot#h the t.otal salinity and t.he proportions of t.hrsr ions in many rivers and ot,her lakes of the

Hw. hydrotkd. trop. %5 (2) : 87-144 (1!#2).

os 0.2 0 t 1

’ I~I~~~I~~~~~~~~~IDIJIF~M~A~M~

Months

FIG. 8. - Lake Gporgc : seasonal c.llanges in lake wat.er of ronduc.ttivity and t.he concent.rations of major ions, in relation to mean monthly rainfall, 1X57-8. Adapted from VINER (1969). Le lac George: chunyements saisonniers de lu conductivité et des roncenfrationn en ions majeurs en fonction de la pluviomt’frie

mr?nsue/le en 1967-68. Lj’nprès MINER, 1969.

Western Rift Valley (TALLING and TALLING, 1965). Tnstead, the ions Ca2-t and HCOa- predominate in bot>h inflows and lake (Fig. S), and the latter is of relatively low salinity and conduct.ivity (at 20 “C, circa 220 & 20 PS cm-l).

Some properties or constit,uents of the lake water are likely t-o control, or be controlled by, the dense phytoplankt,onic biomass present. The latter, plus a considerable bac.kground c.oncentration of filter-pas- sing organic material (10-15 mg C and 0.5-0.8 mg N 1-l : GANF and VINER, 1973), produce a high and spectrally selec.t.ive attenuation of light (Fig. 9). Phytoplankton accounts for most of the total content of nitrogen and especially of phosphorus (Fig. 10), with forms of inorganic nitrogen (NHd-N, NOx-N) and soluble reactive phosphate (PO1-P) existing in very low residual quantities that are presumably in active turnover (VINER, 1973, 1977~ ; GANF


(a)

400 500 800

Wavelength (nm)

(b)

Wavelength (nml

0 300 600 DO0

Chlorophyll x1rn.g m?

and VINER, 1973). TO these fluxes, hact.eria and also zooplankton contribut.c st-rongly (GANF and BLAZKA, 1974) ; est,imates of N HI-N excretion by fish are an order of magnitude lnwer (GANF and VINER, 1973). Input fluxes of N and P to the lake inc.lude considerable atmospheric component.s, based on solutes in rainwater (VINEH and SMITH, 1973; GaNF and VINER, 1973; see Table III) anrl Na-fixation by blue- green algae (HORNE and VINER, 1971 ; GANF and HORNE, 1975; Section 4.4). The concentration of anot.her plant nutrient, silicon, is SO high (c. 9 mg l-1) as to be little affected by the Si-consuming diat.om populations present.. In the upprr illuminated layer, phot.osynt.hetic. act,ivity induces a diurnal inc.rease (t,o super-sa’turation) 0; oxygen and decline of tot,al carbon dioxide, the latter being linked t.o an increase

FIG. 9. - Lake George : somc characteristics of underwater light. attenuation, includinp (a) spertral variat.ion of the vertical attenuation coefficient. E. (b) relative energy distribut.ion at. various depths, (c) relationship brtween chlorophyll a concentration and the minimum value, over t,hr spectrum. of the vert.ical attenuation coefficient. C”lin. Adapted from GINF

(1974). Le lac George : caractérisiiqutx optiques de Veau a) specfre du coefficient E d’affenuafion owficale dr, la lumière, b) réparfifion sprcfrale de l’énergie à différentes profondeurs et, c) relation enfre la concentration en chlorophylle a A +,#in le coefficient d’affenua-

fion verticale minimal sur le apecfrr. D’apris GANF 1974~.

Particulate + soluble

- 0.6

- 0.4 1 P

0.2 F

-0

FIG. 10. - Lake George : changes in concPnt.rat.ions (mg 1-I) of chlorophyll a, particulate N, part.iculate P and total (part.i- culate+soluble) P during 1967-8. From VINER (1977~).

Le lac George : oariafions de la concenfrafion en chlorophylle (mg.~-I), de l’azote parficulaire, du phosphore parficulaire et total (parficulaire+ dissous) en 1.967-68. D’après VINER, 1977c.

Reo. hydrobiol. trop. 25 (2) : X7-144 (1992).

.J. F. TALLING

NH,-N (mg 1-l)

0 10 20 I I I I 1

PO,-P (mg 1-l)

0 2 4 6

FIG. 11. - Lakr George : dept.h-distribution in trffshore sedi- rnrnt colurnns of soluble, H&-washable, PO.r-P and NHJ-N, &ou+ne reduct-ion in the upper zone. (ca. O-5 cm) affected bÿ frrquent. turhulerrcr. hdapted frorn GANF and VINER (1973).

Le lac George : répartition oerficak dans les .sédimenfs du P-P04 et :\‘-.1’Hd soluble. extraif par l’eau. Les faibles teneurs dans les 5 premiers centimèfres correspondenf à l’épaisseur soumise à la

furbulence. D’apr& GANF d I*INER, 1973.

of pH wlrich cari exc*ped 10. There is evidence for increased nucleation of solid carbonate (presumably largely C:aC&) in surh alkaline water (TALLING. unpul:rlished). whç~ the solubilit,y pr0duc.t of CaC:& is often escreded (GOLTERMAN and KOUWE, 1980, Fig. 4.aj. A further combinat.ion with phosphate to fopm hydroxy-apatit.r is also possible (VINER, 197%).

Thesc and reverse changes at. depth, where respiration prrdominates (GANF, 1974d), are controlled in Sp”Cf’ and time by the int.ermittent daily temperaturr-densit.y stratifir.ation and wind-induc.ed turbu- lenrr. IXsturbancr of bottom sediments by the latt,er cari be rel>ogniset.l in their rhemic.al profiles, e.g. of NHs-N and FWJ-P content. (Fig. 11); conversely, an uppc’r layer mal- he recruited from racrntly sedimen- trd phytoplankton. These are examples of a wider range of sediment~-water int.eractions. Some sediments, brought in b;v inflow st.reams, contribute tn a limited drvelopment of swamps off t,he northern sho- relinc. Elsrwhrrc thc shorelines are marked by an abrupt fa11 in the soft subst.ratum, and a shelving littoral zone is not devrloped.

<4 44-h or diel cycle of change (see Fig. 6) is predominant for many features mentioned above (temperature, 1%. pH. phytoplankton layering) and also for various types of biological activity shown in Fig. 39.

Although changes over the longer time-scales (< 24 h) are genorally small and indicat,e a considerable - and, for sballow lakes, exceptional - stabi- 1it.y under an eyuable climat.e, there is evidence that this condition is delic.at,ely poised and SO easily dis- t,urbed during anal aftrr episodes of unusually prolonged and enhancad st,ratification which occasio- nally develop during calm weat,her. Tt is likely that t,he consequences for some gases, especially deoxyge- nation, lead to biologic.al dist.urbances that include sporadic fish-kills (GANF and VINER, 1973).

3. REGULATION OF THE PHYSICAL ENVIRONMENT

The physical environment. is dominated by inputs of energy (e.g. from solar radiaCon, wind) and their transformat.ions within the water-mass.

3.1. Solar radiation input

The dist.ribution of solar radiation over the land- mass of -4fric.a is determined partly by direct.ional or ’ geomet.rical ’ fact.ors of beam-incidence related t.o latit-ude and season, and part.ly t.o more locally variable modifications due to atmospheric t,urbidity. These factor-groups cari be summarised, for a given 24 h period, by frartional transmission values of cg. and E, respectivel\-. Their mutual product with the solar const.ant (1.36 k,T m-2 s-1 or 1.95 cal c.m-? min-l) determinrs the daily radiation flux on a horizontal

50

I I Il 1 11 1 I ” JFMAMJJASOND’

FIG. 12. - Sea,onal variation, at. four lat.it.udes Nort.h, of the geometrical or dircctionel factor .z~ governing the daily flux of short-wavr solar radiation on a horizontal u;urface. From hION-

TEITH (1:)72). Variafions saisonnières, pour 4 lafifudes de l’hémisphére nord, du coefficient ge’om&ique cg déierminanf le flux solaire (ondes

courtes) sur un plan horkonfal. D’après MVNTE~TH, 1972.


(b) 1.0

1

(a) 1.0

1 Air

0.6

Ea

0.6

Air Air

FIG. 13. - Annual variation in at.mospheric transmission fact.ors E, for solar radiation, including the overall actual transmission factor ea (-), that applicable to a cloudless du&-free atmosphere (- . -), and that obtained from mean monthly radiation maxima and applicable to a near-cloudless but dust-containing atmosphere (---); (a) Sumaru, Nigeria (h) Kinshasa, Zaïre (c) Muguga, Kenya.

Adapted from MONTEITH (1972). L4lriafion annuelle du coefficient de transmission atmosphérique ~a. En trait continu, coefficient global réel; en trait mixfr pour une afmospk8re sans nuages ef sans poussière; en tireté, pour la radiation maximale moyenne mensuelle, sans nuage mais uuec poussikre. a) :

Sumaru (Nigeria), b) : Kinskasa (Zaïre) et c) : Muguga (Kenya). D’après MONTEITH, 1972.

surface (Q,). Thus if Es = 0.3 and E, = 0.5, the daily flux

Qr = 1.36 x 10-S x 0.3 x 0.5 x 3600 x 24 = 17.6 MJ m-2 day-1 = 1.76 kJ c.m-2 day-1 (422 cal cm-2 day-1). The two fractional transmission values are illustrated by MONTEITH (1972) in diagrams shown here as Figs 12 and 13. The first shows the dependence of the ’ geometrical fac.t.or ’ upon latitude and season, wit.h similar values (circa 0.3) over low latitudes and many seasons and a marked winter reduction at higher latitudes. Comparison was made of calculated daily fluxes at the top of t,he atmosphere with those measured at three African sites, deriving an at.mospheric transmission factor to which calculat,ed cont.ributions by air, fine dust particles (aerosol) and cloud were distinguished (Fig. 13). The last two cont.ribut.ions are often of similar magnitude, expres- sive of t.he prevalence of smoke and soil-derived dust during dry seasons.

The following discussion deals c.hiefly wit.h day- integrat,ed measures of short-wave radiation on a horizontal surface (Qr), expressed in MJ m-” (or kJ cm--) day-1 or cal c.m-2 day-1 (1 cal c.m-2 = 41.8 kJ n+) .

Diurnal variability is fundament,ally controlled by day-length and maximum (solar-noon) solar eleva- Con. The variat.ion of these is minimal at the equa-

Heu. hydrohiol. trop. 25 (2) : 87-M (1992).

ter, and values of bot.h cari l-je found from meteorological t.ables, such as t.hc Smithsonian bIet,eorological Tables (1951). The samr Tables provide valups of Qp as a funct.ion of latitude and month, both uncorrec- ted for atmospheric absorption and corrected for various values of atmospheric t,ransmission per unit air mass. A value of 0.X for the laat fact,or was used by WOOD et al. (1976) t.o estimate potential clear-sky radiation incomr for t-he Bishoftu lakes (Ethiopia) and compare with observecl values which wrre markedly lower during the rainy season.

Published measuren1rnt.s of Qr at sites in Africa are not numerous and still frwrxr relate to freshwater sites. Some examples of its seasonal variability are shown in Figs. 14 and 17, with minima attributable to the latit.ude-dependent, geometric.al fac.tor or to pronounced seasonal c.loud caver. Thesp patterns are further diseussed by TALLIN~; (1990) in relation to temperature cycles. hlorr geographically comprehen- sive but. idealized cont.our-maps of Qi have been c.onstructed from calculated and measured values. The sets (Fig. 15) illustrated hece, from BLACK (1954) and THOMPSON (1965). show latitudinal swings of maxima and minima with season t,ogether with local modificat.ions. For the last. cloud caver is important ; an example-distribution, in May, is seen from space in Fig. 16.

98 .J. F. TALLING

ICHKEUL 137’N) NAKURU ion1

Ï 2

CHAD f13”Nl CHILWA (15’Sl 2.6 2.6

Y l ..-., 0 2.0 l ./ - 2.0

2 3,0 ~ KILOTES i9’NI

1 I ‘J’F’M’A’M’J’J’A’S’O’N D

1 1 I I 1 ‘J’F’M’A’M’J’J’AS 0 N D

SWARTVLEI (34%

FIG. 11. - Xunual changes of incident solar radiation (horizontal surface) for eight shallow African lakes; (a) L. Ichkeul (from LEMOALLE, pers. ctrmm.), (b) L. Chad (from CARMOUZE ef

CI/., 1983). (c) L. Kilot,es (from WOOD et a[., 1976), (d) the Ebr% lagoon (from P.4c;Ès rt al., 1981), (e) L. George (from VINER and SMITH, 1973), (f) Nakuru (from VARESCHI, 1982), (6) L. Chilwa (from KAL.K ef ul., 1979), (h) S wartvlei (from HOWARD-WIL-

LIA~ and ALLANBON, 1981). Iwariniion.s annuellrs de l’énergie solaire incidente (sur un plan horizontal) pour S lacs peu profonds; a) L. Ichkeul (J. LEMOALLE, comm. pers.), b) L. Tchad (CARMOUZE et al., 1983). c) L. Kilotes (!V@O0 et al.. 1976). d) Lagune Ebrii (PAGES ut al., 19X1), e) L. George ( VINEA et SMITH, 1973), f) L. Nakuru ( TARERCHI, 1982), y) L. Chilma (KALK et. al., 29ï9)

ef h) Swarfolri (Hou'AR~)-~~~II,LIA~~~s et ALLANSON, 1981).

3.2. Temperature

As in other cnnt.inents, t.hc seasonal progression of water surface temperature at higher lat,it,udes broa- dly follows that of solar radiation input, usually with some Iag indicative of the energy-accumulative cha- ractrr of wat.er temperature (Fig. 17). This pat.t.ern is expressed in a series of seasonal curves from north to sout,h (Fig. 18). Although most. of t.he lakes involved are ‘ drrp ’ (exceptions : L. Chad, L. Bangweulu) the groas fchatures of seasonal pat.t.ern would also apply to shallow lakes. More examples for the latter are given in Fig. 19. Among these, pronounced tempera- turc minima arise during the mont.hs June-August for diffrrrnt reaaons in t.he Ebrié lagoon at. 50 N

(çloudy-rainy season) and Lac Ihotry at 21005’s (winter season with tradr winds). At. lakes George (00) and Ihema (1050 S) t.he seasonal differences do net. exceed 2 *C, whereas at Swart-vlei (34O 03 ’ S) (Fig. 17), and L. Tchkeul (370 10’ N) there is a pronounced winter minimum related t.o the latitudinal- geomet-rical fact-or of solar radiation.

The fa11 in surface t,Pmperature with altitude is most easily depic.t-ed for eyuatorial watrr-bodies in which seasonal variation is small. Jt. is illustrat,ed in

(a)

Jun

(b) Im ’

FIG. 15. - The distrihut,ion «ver Afrira of incident, solar radia- t.ion on a horizont.al surface, exprrssed (a) as calculat.ed mean daily values (cal cm-’ day-1) in January and June, adapted from BLACK (1956), (h) as mean values (kcal cm-2 yr-‘), adapted

from THOM~SON (1965). Répartition de l’énergie solaire incidente en Afrique a) moyennes mensuelles calculéea en janaier et juin (d’après BLACK, 19.56) et b) moyennes annuelles (kcal. cm-s.an-1) d’après THOM~SON. 1965.

Rtw. hqdrohiul. trop. 2.5 (2) : Hi-144 (1992).


FIG. 16. - Distribution of cloud over the African continent at 11.55 GMT on 31 May 1978. as recorded from the European Space Agency’s geostationary weather satellite Meteosat 1, Photo by courtesy of the Meteorological Office, U.K.

La eouverfure nuageuse sur l’Afrique le 31 mai 1978 à 11 h 55 GMT, vue par le satellite géostationnaire de l’Agence Spatiale Européenne, Meteosat 1. Image aimablemenf communiquée par /‘Office de Météorologie du Royaume Uni.

1976 1977 1978

FIG. 17. - (a) Annual variation of daily solar radiation (as monthly mean values) at Swart.vlei, S. Africa, and its relation to (b) the temperature of surface water measured in the open lake and the littoral macrophyte-rich zone. From HOWARD-

WILLIAMS and ALLANSON (1981a). Moyennes mensuelles de l’énergie solaire incidente journalière à Swarfvlei, Afrique du Sud, et fempérature de surface dans les eaux libres et dans la zone littorale à macrophytes. D’après

HOWARD-WILLIAMS et ALLANSON, 1981a.

R~U. hydrobiol. trop, 25 (2) : 87-144 (1992).

100 J. F. TALLING

1 ‘C latitude lake

FIG. 18. - Annual variat.ion of surface temperat.ure in a series of 14 African lakes (shallow underlined), arranged by latitude. Successive curves are displaced downwards, and t.be common temperaturr scale (left) refers to differences only. Absolute values cari he located using t.he single temperat.ure marked on

P?ar each curve. From TALLING (1969). Variations annuelles des températures de surface dans 18 lacs

\ u 33-N

24’

AGUELMAME SIDI AU 1951-Z africains (le nom des lacs peu profonds esf souligné). Les dif- firentes courbes ,çonf déplacies successivement vers le bas, l’échelle

AÇWAN RESERVOIR 1942 de guuche n’indiquant que l’amplitude des variations. Les valeurs absolues peuvent être déduites de la température de rt!fgrence indi-

CHAD 1964-5 qut;e sur chaque courbe. D’après TALLING. 1969.

PAWLO 1965

ALBERT 1953

EDWARD 1953

VICTORIA 1953

KIVU 1953

TANGANYIKA 1956

N’ZILO 1958

BANGWEULU 1958

MALAWI (NYASA) 1958

KARIBA 1961

FLORIDA L. 192.6

TEMPERATURE L’C)

27 -

25 -

23 -

21 -

ICHKEUL ‘9 -

FIG. 19. - Annual variation of surface temperature in seven 31 - shallow hfrican waters. Ebata from LEMOALLE [pers. comm.), GORGY (1!1!59), TALZJNG (unpubl.), WOOII et al. (1976), DUHAND

29 - .

‘/ay,*‘~

.-• and CHANTRAINE (1982), DE KIMPE (1964). and TV~OREAU IHOTRY 27-

(1982). (22%) 2s \ . ./

I.ariations unnurlles des températures de surface dans sept lacs 23 - IHOTRY

peu profonds d’,4 frique et Madagascar. Données de LEMOALLE (Comm. prrs.), (~)RGY (195.9). TALLING (non publié), ~VOOD et.

21 - \/ .-.

al. (1976). 13rimN1-1 et CH.~~~TRAINE (1982). DE KI*~PE (1964) et 11, 1 , , II, IL MOREAU (1982). ‘J F M A M J J A S Cl ND’

Heu. hydrobiol. trop. 25 (2) : U-144 (1992).

5000

4000

3000

E

CI 2 5 a 2000

1000

0

AFRICAN SHALLOW LAKES

0 10 20 30

.

. Bujuku L.

0 Ellis L.

0 L. Mahoma

+ L. Lungwe

L. Kilotes L. Naivasha

L. Nakuru

‘L. Tana

. L. Abaya

- Winam Gulf (Victoria)

-.- L. George

-e L. Tumba

t 1 I t I I I

101

0 10 20 30

FIG. 20. - Decrease with altitude of near-bottom temperatures measured in various tropical African water-bodies. Bars indicate seasonal ranges. Values from L~FFLER (1964, 1968), DUBOIS (1955), Woou et al. (1976). GAUDET (1979), VARESCHI (19SZ), MORANDINI

(1910), TALLING (unpubl.). VTNER and SMITH (1973). and DUBOIS (1959). From TALLING (1990). Décroissunee de la température près du fond en fonction de l’alfifude des plans d’eau africains. Les traits horizontaux indiquent l’amplitude saisonnière de variation. D’après les données de L~FFLER (1964, 1968), DUB~E (1955), Woon et al. (1976), GAUDET (1979), V.RESCHI

(1.982). MOIMNDINI /1940), TALLING (non publié), VINER ef SMITH (1973), ef DUBOIS (1959). D’après TALLINCI (1990).

Fig. 20, and by L~FFLER (1968) and GASSE et al. (1983). There are few high-altitude waters on which seasonal measurements of temperature are adequate, but the observations of MORANDINI (1940) on L. Tana (1820 m), of DUBOIS (1955) on L. Lungwe (2710 m), and the much more detailed st.udies on the Bishoftu crater lakes (1870-2000 m) of WOOD et al. (1976), cari be mentioned.

Shallow lakes are thermaily distinctive bu the usual lack of any persistent temperature straiifica- tion, because mixing by wind-induced turbulence and convection freyuently affects the entire water- column. However, diel cycles of stratification are typically well developed (see Sect.ions 2, 6.1). As energy t.ransfer is predominantly across t,he water

surface, a shallow wat.er-column of limited heat- c.apacity per unit area tends to accentuate both diel and seasonal maxima and minima of temperature. The low capacity factor is ac,centuated when much daytime energy consumption is confined to a near- surface layer by high t.urbidity. This situation favours the diurnal development of very strong but superfic.ial temperature gradients, as recorded by WOOD et al. (1976) in phytoplankton-rich lakes of Ethiopia. The diel differences of temperature in surface water of L. Chad, relative to the seasonal diffe- rentes, are shown in Fig. 21. In shallow African lakes, as elsewhere, horizontal differences of temperature and its stsatification cari develop in several ways. Thus a cooler or warmer inflow may locally

Rea. hgcirobiol. trop. 2.5 (2) : 87-144 (1992).

J. F. TALLING

30

25

20

15

l I I I I I J'F'M'A'M'J'.' A S 0 N D

Months

FIG. 21. - Annual variation in L. Chad at, Bol of surface t,emperat.ure measured near the doily minima (7 II) and maxima (16 h), and ei;pressetI as mean values «ver 5 consecutive days. Adapt.ed from CARM«UZE, CHANTRAINE & LEMOALLE (1983).

Temphttww de surfhrr drr L. Tchnd à Bof, près du minimum (Y h) et maximum (16) journnliera. Moyennes sur 5 joors consécufifs. D’aprè.~ C.AR.WOUZE, CHANTRAINK ef LEMOAI.I.E, 1983.

imprrss it.s rffect ; shallow marginal areas cari oft,en grow warmer by day (Fig. 17b) and cooler by night, wit.1~ profile-bound densit.y currents possible from the latter (T~LLING, 1963, 1969) ; the thermocline (even a dirl one) may bc tilted down-wind, as measured in L. Iiilot.rs (Woor) et nl., 1976) ; and vegetation caver, as flctatin~ mats or weed-beds, may lead t,o local t.emperature incrrase by energy int.erc.ept.ion and limitation of water transfer. The last situaCon is analysed in detail by RWM and KEHR (1983) for t-he float,ing wced Sulninia molestu.

1Vhere a lawrr more saline layer exists in a very shallow fresh water, t;hr t.emperat,ure of that layer may be considerably elevated by solar heat.ing. Examplrs art’ given by BEAI>LE (1943) from Algeria, MELACK and KILHAM (1972) from Uganda (L. Mahega), and by ASHTON and SCHOEMAN (1983, 1988) for the Pretoria Salt Pan where an additional warm rniddle layer develops daily (Fig. 22). Ins- tances of salinity st,rat.ification wit.hout suc.h t,emperat-urr inc.rease are not uncommon in coastal lakes wit.h some input of sea water (e.g. Swartvlei : ROB.UTS and ALLANSON, 1977; ALLANSON and H~~ARD-WILLIAM~, 1984).

3.3. Turbidity and light attenuation Light att.enuation under wat,er is governed mainly

by particulate material and by more dispersed ’ colour ’ (’ gelhstoff ’ or L gilvin ‘). The latter is st.rongly drveloped in most. swamp waters and in not a frw shallow lakrs, innluding Chilwa (Mass and MOSS,

1969), Kilotes (TALLING et af., 1973), Chad (LEMOALLE, 1979). Tana (TALLING, unpubl.), Abaya (WOOD eI ul., 1978), and Swart.vlei (ALLANSON and HOWARD-\VILLIAMS, 1984). However, high or variable turbidity, assooiat,ed with part,iculate cont.ent., is the most dominating influence in most shallow Afric.an waters. It may arise from a heavy development of plankt,on, usually predominantly phytoplankton, or from t,he suspension of sediment and plant. det.ritus. The commonest. measure of turbidity, or its inverse of transparerzcv, is the depth of visibility of a white disc (Secchi disc), but gravime- trie rneasurements of suspended solids are sometimes available. For phytoplankton-dominated waters, estimations of chlorophyll a concentrations are also appropriate. In L. Chad it. bas been possible to obtain large-scale synoptic information from remote (satellit,e) sensing, cross-referenced and interpreted in t.erms of Secchi disc transparency (LEMOALLE, 1979).

Generation of t,urbidity by the wind-disturbance of bot.tom sediments is widespread in waters less than circa 5 m deep, and is accentuated when the wind-fet,ch is long (cf. L. George, Table 1). Episodes of high turbidity then follow strong winds. Episodes cari also result from land run-off in rainy seasons, when silt-rich flood-water of bigh turbidity moves down a drainage system and may be injec,ted into reservoirs (e.g. ENTZ, 197$), lagoons or pans (e.g. ROGERS and BREEN, 1980; AICHURST and BREEN, 1988). r)et.ailed studios exist. for the White and Blue Niles (TALLING and RZ~SKA, 1967), where (as often

H~O. hydrohiol. trop. 25 (2) : 87-144 (1992)

AFRICAN SHALLOW LAI<ES 103

(a) Tema. PC)

- 1100 h

25 30 35 40

F~G. 2’2. - Stratification in the Pretoria Salt. Pan on 29 December 1979 : (a) daily change in temperature at a shallow station, and (b, c) typical depth profiles of temperature, pH, and dissolved oxygen at a shallow and a deep st,ation. Adapted from ASHTON and

SCHOEMAN (1983). Stratificution dans le Pretoria Sali Pan, le 29 décembre 1979 : ( a ) variaiion journalière dans une station peu profonde. ei (b et c), profils iypiques de iempérature, pH et oxygène dissous dans une siafion profonde et une station peu profonde. D’aprh ASHTON et SCHOEMW

1983.

elsewhere) the turbidity of shallow impoundments and lagoons is determined by two opposing influences - the sedimenlation of larger silt. par- tic.les and the produc.tion of denser phytoplankton. In Lake Chad, the reduction of turbulence associat,ed with t,he enhanced growth of mac,rophytes at low water levels led in some areas t.o greater sedimentation and higher transparency (LEMOALLE, 1979 ; &nafouzE et a[., 1983). However the low-level phase could also be accompanied by low transparency (circa 0.1 m) from algal growth and resuspension of

clay sediments (CARMOUZE ef al., 1983). There is also evidence elsewhere (MCLACHLAN and MCLACHLAN, 1976; AKHURST and BREEN, 1988) that, clay precipitation c.an be acceleratpd at higher levels of salinity and conduct.ivity.

Where phytoplankton is an important, cause of t.urbidit.y, its persist.ence in t:ime Will be governed by biological cycles or fJt,her fluctuat,ions. Even in waters with a persistent,ly dense phytoplankton, such as L. George, the variabi1it.y may be sufficient to allow a relationship between biomass concentra-

R~U. hydrobial. frop. 25 (2) : 87-M (1992).

101 J. F. TALLING

tion and light at-trnuation (or transparency) to be rstablishrd (Fig. 9c : GANF, 1971~). Many soda lakes are turbitl with phytoplankton but at. least. some bave a consid~~rablc background attenuat,ion (e.g. L. Kilotrs : TALLING rf al., 1973; L. Nakuru : VARESCHI, 19W). often from soluble or dispersed organiç matrrial. The same is truc for some relati- ~riy di1ut.c shallow lakes (e.g. L. Abaya = Marghe- rita, I,. C%arne, L. Baringo) ric.h in both blue-green alga~ and tolal iron (TALLIN~. and TALLING, 1965; \VO~» arrd TALLING, 1988).

Quantit.ativr relationships between light, at.tenua- tien, S~chi disc transparency, and concentraCons of absorbing components (e.g. c%hlorophyll a, silt) have brt% espkrrd in se\-eral shallow wat.ers - not.ably L. George (Fig. 9c : GANF, 1974c), L. Chad (LEMOALLE, l%). 1983b). L. McIlwaine (ROBARTS, 19ï9j and the Ebrié lagoon (DUFOUR, 198%~). Spec- tral components of differing at.tenuat,ion bave been distinguishrd in thesr artd other (chiefly East African - TALLIN~;, 1957a, 1965) lakc and river wattars. whic,h demonstrate t.he ‘ rtsd shift-’ in maximum transmission - i.r. its displacement t,owards longe1 wavrlengt.hs - ctharac.tt+&ie of most turbid waters. For one wat-er, L. George, an experimental approach using high resolution epectroradiometry has been made (T~LLING, 1970) and Pstimated spectral distributions of underwater radiation (Fig. 9b) have been published (GANF, 19741~ ; GANF and HORNE, IB75). In one t.urbid South African impoundment such spac- tral distributions, shifting with dept,h, have been measnrrd directly by submersible spect.roradiometer (\vAthWLEY Pt fd., 1!%?k)).

A ~Y~~IIII~~~ but approximate guide t.o the depth- extension of nppreciable photosynthetic activity is the euphotic zone limit (z,,,) set by the 1 9; level of tt1e sl.irfatrc--pt~ri~t.rating ad photosynt.hetically active radiation (PAR). Exarnples for shallow waters are @\-en by TALLING (1%X), TALLING et af. (1973), GANF (1075), LEMOAL,LE (1973, 1983b), MELACK (197h. h, 1!181), ROHARTS (1979) and PAGÈS et al. (1979). \Vhen the euphotic depth (z,“) is only a small fraction ( > (‘. Cl.?) of t he mixed depth (z,iJ the pho- t.ot.rophic: growth of phytoplankton is likely t.o be inipeded e\‘en under tropical conditions of high insolation. (;~NF and VINER (1973) consider this hmit,a- t ion with referrnc.e to depth-relations found in L. George (zmix = 2.1 rn, zp, z 0.75 m) and t.he eco- Iogictal stability of its phytoplankton. From this and othrr t~amples it. would appear t.hat shallow lakes ran bt, optically quite deep, wit.h implication for light-limitation of the net, production and growth rates of phytoplankton and submersed aquatic macrophytt~s. Thus a near-absence of the latt.rr from Iakes George anrl Chilwa has been attributed to high light att.rnuation associated wit,h dense phytoplank-

ton and fine suspended sediment? respectively. An observed derlinp of macrophyt,es in Swartvlei may also be due to incrrased light attenuation, associated with run-off from a det-eriorating catchment- (TAY- LOR, 1982 : ALLANSON and HOWARD-WILLIAMS, 1984).

3.4. Water movement

Water mnvenients originate in relation to gravity- flow (streams, rivers) or t.o wind-stress on a wat.er surface. They are obviously import.ant for horizont,al transport, for vert.ical turbulence, and for ir&erac.- Cons at t.he water-air and water-sediment, surfaces. Most observaCons are indirect and qualitative, as from t,hp di&ribution (and deposition) of suspended object.s, of temperature or chemic.al composition, or of correlat,ed wind-stress. Specific examples include : the noc.t.urnal destruction of thermal stratification in L. George (GANF, 1974b ; secL Figs. 5,6) ; the complex horizontal distribution of Na+ concentration and conduct-ivity in L. Chad indicative of a predominantly nort.hwards wat.er transfer from the southern inflow (CARM~~~E, 197’3, 19%; CARMOUZE and LEMOAT,LE, 1983 : sec Fig. 13) ; and the local reduction of horizontal flow during t,he phase of macrophyt.e abundance in L. Chad (CARM~UZE et al., 1983). Swamp vegetat.ion is a widespread impeding factor for water movement, operat.ing partly through shel- t.er of t.he water surface from wind. Its influence on wat,er movement. at- L. Chilwa is described by HOWARD-WILLIAMS and LENTON (1975).

Direct rneasurrments in shallow standing waters of the velocity of movement, its Spat>ial pattern, and its relation to wind stress, are few. One notable study on L. George has been described in the case- study (Section 2). Simple drogues we~re also used by DENNY (1978) on t.he Nyurnba ya bIungu reservoir, Tanzania. By measurements from a hydrologist’s mult.i-vane current meter, RZ&KA (1974, Fig. 2) described the horizontal gradients in flow velocity bet.- wcen rnid-river and adjacent swamps of t.he Upper Whit,e Nile, and also (RZ~SKA, 1976) the reduction of velocity downstream along the axis of a Nile reservoir in relat.ion t.o plankton development,. Low velocities of wat.er movement have also been measured in East African lakes by a hot-wire current. meter (~~CINT~RE, 1981), especially for the meromictic though shallow L. Sonachi (MCINTYRE and MELACK, 1982), and by Ekman and Aanderaâ current meters in t:he Ebrié lagoon (GALLARDO, 1978).

Circulation gyres, of t.he t,ype illust.rated for lakes George (Fig. 7) and Chad (Fig. 23), are especially likely to develop in shallow lakes under wind st,ress. This is bec.ause a wind-driven transport. of surface


early JUN

JAN-FE6

_*-- **--- **--

VL FEE 1970

FIG. 23. -Lake Chad : (above) seasonal changes of circulation patterns in the water-mass; (below) distribution of Na+ concentrations (in meq 1-l) in 1970. Adapted from CARMOUZE

and LEMOALLE (1983) and CARMOUZE (1972). Le lac Tchad. En haut, évoluiion saisonnière de la circulation générale; en bas. répartition de l’ion Na-t (en méq. 1-l) durant l’année 1970. D’après CARMOLTZE et LEMOALLE (1983) et CAR-

MOVZE (1972).

TABLE II

Lake Date cal cms2

Qr QS Qb Qe Qc Tana a) 24h, 2-5 March 1937 - 293 -

b) night 12h only 0 -ii&3

16 147 95

Qarun 24h, July 1950

Kilotes 24h, 1 Jan. 1966

612 + 233 517 -188

548 294 (talc.) 33 221 + 90 (abs.)

Chad a) 24h, 15 Dec. 1970 545 0 216 114 -11 b) day 12h only 525 585 108 57 -10 c) night 12h only 0 -585 108 57 -1

+ tmumed nealigible.

water is c.ompensat.ed by a ret.urn flow not at. dept,h, as in deeper lakrs, but in horizontally adjacent areas.

3.5. Energy budgets Modes of energy transfer to and from a water-

body include short.-wave (solar) radiation input (Qi) corrected for a mean surface loss by reflection plus scattering of c. 10 Y/0 (Qi) ; long-wave (((thermal ))) radiation both incoming and outgoing, often combined as net back-radiation (Qb) ; sensible heat transfer of variable direc,tion down thermal gradients between wat.er and atmosphere (Qc) ; and the latent, heat of evaporation loss (QR). For waters in low latitudes, attempts to partition t.hem with energy storage (QS) within an energy budget.

Vi-' = Qs+Qh+Qe+Qr are few and, in Africa, rudimentary ; see however studies of GORGY (1959) on L. Qarun, WOOD ef af. (1976) on the Ethiopian Bishoft,u lakes, EGGERS and TETZLAFF (1978) on L. Chad and TALLING (1990) on the Jebel Auliya (Nile) reservoir; of NEUMANN on L. Kinneret in Israel (SERR~~A, 1978), and of CAR- MOUZE and AQUISE (1983) and TAYLOR and AQUISE (1984) on L. Titicaca in the Andes. Some rough est.i- mates for four shallow African lakes (Tana MORAN- DINI, 1940; Qarun GORGY, 1959; Kilot,es WOOD et al. 1976; Chad Eggers and Tefzlaff, 1978) are given in Table II. In a bot dry African çlimate as is found in the Sahel and Sahara, an evaporation rate of 9 mm day-l cari be reached ; this implies an enexgy transfer of about 622 cal (2.6 kJ) cm-2 day-1, a similar order of magnitude to that. transferred in solar radiation (as also at. L. George, Section 2; L. Qarun, Table II). Evaporative cooling is therefore of quantitative importance, especially in shallow waters. Its energy- relations are considered in detail by GORGY (1959) for L. Qarun, Egypt. Here its severe reduction in winter (GORGY, 1959 ; MESHAL and A!ARCOS, 1981) will limit the cooling then observed in the lake water. In t.he Bishoftu lakes of Et.hiopia, including the shallow

RE~. hydrobiol. trop. 25 (2) : 87-l& (1992).

1 Otj ,J. F. TALLING

L. Mot.es. evaporat-ivr loss was probably mainly rrsponsiblr for the non-coincidence bet.ween periods of coolirig and minimal insolation (WOOD et al., 1976). Hrre concomitant. seasonal measurements on four adjacrnt lakes of var?ing dept.h showed that nrar-surface temperature m t.he shallowest, lake (KilotPa, z,~,,, = fi.5 m) was generally lower by seve- rai degrces than that in the dceper lakes ; enhanced hrat loss per unit volume by back-radiation, and by evaporation at a more exposed sit-e, may be suspec- trd. Seaaonal energy storage as hcat per unit. area (t.he so-called ‘ annual heat budget ‘) of a shallow water-mass t.pnds to be rest.ricted for purely morpho- nletric reasons. as to the value of 1.3X 10” cal cm-‘j for L. Kilotes. Hrat storage in sediments bas apparently not bren evaluated, and alt,hough relatively larger in shallow t,han deep lakes is likely to be a stnatt component of thr daily and annual energy budget.s. On t he diurnal time scale water-column storage cari be of similar magnitude to t.he daily input of &ort-wave radiation. often c.irca -1-W cal cm-? (= 2 ‘JC; rise over O-2 m, or 1 “(1 rise over O-4 m). This is approaching the magnitude of the seasonal heat, storage citer1 abovr, instead of being about 1 order of magnitude lower as in many deep t,ropical African lakes (T.\L.LING, 1990) and nearly 2 orders of magnitude Inwer as is comrnonplace in deep temperate lakrs.

4. REGULATION ENVIRONMENT

OF THE CHEMICAL

1.1. Chemical sources

Inland water-bodies receive material input,s (other than water) in three main ways - from t,heir inflows and seepage, from at.mospheric precipitation both wet and dry, and from gaseous exc.hange wit,h the atmosphere. The last involves no distinctively Afri- cari features, although specific forms of swamp vegetation (e.g. floating swamp) create particular barriers to vert.ical exchange.

The significance of t.he second pathway, atmospheric precipitation, is little known in the African cont.ext. Near the sea toast a considerable contribution of ’ cyclic’ sea Salt., via sea-spray, cari be anti- cipated but. quantit.ative assessments appear to be lacking. Even well inland, as in East Africa and Ethiopia, the contribution is appretiable and for various ions cari be gauged from the Cl- content (BAUDET and MELA&, 1981 ; WOOD and TALLING, 1988). Very varied resu1t.s have been report,ed from the chemical analysis of rainwater (VISSER, 1961 ;

TALLING and T~LLING, 1965; LEMOALLE, 1973b; HEMENS ef al., 19’77 ; RODHE et al., 1981 ; LEMASSON and PA~+Ès. 1982), both as regards major ions and such plant nutrients as forms of N and P. This varia-

Constituent L. Naivasha N’Djamena L. George East Africa L. Ebrié (1) Chad (2) (3) (4) (5)

NOS-N+ NlQ-N km-2 yrl) - 0.47 - -* 1.11 1.05

NOS-N tigl-1) (sm-2 yrl)

NH4-N (pg l-1 1 (gm-2yrl)

Total N cl& l-1) (gm-2yrl)

PO4-P (pg l-1 ) (gm-2yr1)

PH CS2+ (mg l-1) W+ (mg l-1 1 Na+ (mg 1-l 1 K+ (mg 1-l 1 Cl- (mg 1-I 1 SO42- (mg 1-l 1

Conductivity, k25 (@ cm-l)

-

:

-

-

c.5.7 0.19 0.23 0.54 0.31 0.41 0.72

-

181 0.12

535 0.35

37 0.024

<20-890

130-4100

4-1700 0.48

6.4 0.39 0.08 0.63 0.30 0.60 0.75 10.0

274 0.59

216 0.46

1440 3.08

111 (c.lo-800) 0.23

(11 GALIDKI snd MF.I.A(.K 1981, (2) LEYOALL.F 1973, V~ER and Snrrm 197.7, GANF and VINER 1’37.1, (4) RODHE r.l af. 19111, (5) LEMASSON and PacÈs 1YS2.

Hetr. h.ydrohml. lrop. 2.5 (2) : Y7-I&i (1992).

AFRICAN SHALLow LAKES 107

bility may include secondary changes during collec- tion and storage, t,echnical errer, and a changing contribution from atmospheric du&. Some analyses are given (as means or medians) in Table III.

The contribution from inflows depends, quantitatively and yualitat,ively, upon the general hydrology and the geochemic.al character of t.he cat,chment, plus its biological development as by cult.ivation, fringing swamp, floodplain, and human set,tlement. In a few instances attempts have been made (GAU- DET and MELACK, 1981; WOOD and TQLLING, 1988) to distinguish the surface-denudative component from atmospheric input, the latt.er obtained from data on precipit.at.ion and estimates of its evaporative concent.rat.ion. Thus GAUDET and MELACK (1981) estimated the Malewa R. input t,o L. Nai- vasha t.o carry a total solute load of 91 kg ha-1 day-1, of which in 19734 only 21 7; was derived by surface chemical denudat,ion.

Effects of human settlement, influencing organic balance, dissolved oxygen, plant nutrients, and phytoplankton production, are well illustrated in regions of the elongate Ebrié lagoon subject to urban pollution from Abidjan (DUFOUR, 1982b; DUFOUR and DURAND, 1952). Another coastal water in which the chemical and biological consequences of heavy pollution have been studied is Lake Mariut and the adjoi- ning Hydrodrome, that receive sewage from Alexan- dria (ELSTER and VOLLENWEIDER, 1961; BANOUB and WAHBY, 1961; ALEEM and SAMAAN, 1969; SAAD, 1973, 1980; WAHBY ei al., 1973; WAHBY and ABDEL-M• NEIM, 1979; SAMAAN and ABDALLAH, 1981). More generally in the Delta Lakes of Egypt, chemiçal conditions c.hange with the relative influence of marine ingress and of drainage water from agricultural areas, t,he latter nutrient-rich and inçreased in the post-High Dam era. Examples inc.lude L. Edfu (SAAD, 1976, 1975; BANOUB, 1979, 1983) and L. Manzalah (EL-HEHYAWI, 1977; SHA- IIEEN & YOSEF, 1978). For the former, estimat,es of net annual consumption of N, P, and Si in the lake have been made (BANOUB, 1983).

111 regions of high rainfall and runoff, such as the Zaïre basin, the effects of dilution and past leaching t,end to produce dilute inflow waters. The sarne consequence may result by drainage from ancient and long-weathered land surfaces, widespre-.d throughout Africa ; the Upper Zambezi catchment is an example from a region of lowey and strongly seasonal rainfall. In general, freshwaters of low ionic content (conduct.ivity < 70 FS cm-l) are muçh more common in West t.han East Africa (VISSER, 1974; VISSER and VILLENEUVE, 1975; JOHN, 1986). The geological influence is obviously enhanced in areas dominated by hard rocks su& as granitse, as in the Upper drainage of the Niger and the Senegal River

R~U. hydrobiol. trop. 25 (2) : 87-144 (1992).

(GROVE, 1972). Weil-leached sandy or lateritic soils cari also bear very t:lilut,e wat.er-bodies; SYhIoENs (1968) and HARE and CARTER (1984) give examples of conductivity < 15 PS cm-l. Conversely, soft roc.ks and formations rich in soluble salt deposits cari lead to more saline inflows of distinctive ionic composition. Examples include areas of limestone (eg. t,he Blue Nile gorge), of fossil evaporites of marine origin (e.g. the Afar (nanakil) depression), and the K+ - and Mg2+ - rich rocks of the Virunga volcanic region. Nevertheless, most. strongly saline waters are of more secondary origin and owe their development to their situation in basins of closed drainage and a predominance of evaporat,ion in the hydrological budget.

Some peculiarly biological pat;hways of chemical input. and output. dexerve mention.

Mobile animais cari transfer material horizontally and vertically, as by the spatial scparation of gra- zing and excretion by hippopotami (VINER, 1975a; 1<ILHAiX, 1982), e1ephant.s and ungulates (S.M. MCLACHLAN, 1970, 1971), and by t.he mass emer- gences of chironomids and chaoborids or ‘ lake-flies ’ (examples in HZ~SKA, 1976).

Nufrienf upfake by rooted swamp vegefation, subject to growth and decay. çan trensfer solutes from the substratum to surface Rater. Although an active t.ransfer through t.he living plants has been postula- t.ed, ther’e is no subst-ant.ive evidence for this (Denny, 19S5b), but. a ’ nutrirnt pllmp ’ movement sediment. -+ macrophyte -+ water is completed by senescence and decay of vegetation. For one element, carbon, the sequence is atmosphere + emergpnt macrophyte -+ water, with net. C& transfer. The stock per unit area (Table IV), subject to gr0wt.h and decay, of such elements as K, Si, N and C cari be high in reed- swamps (e.g. of Phruymites australis in L. Chad : CARhlouzE et al., 1978; C;ARN~~~E, 1983). Compara- tive examples for various dense stands of reedswamp, floating and submerged macrophytes, and a dense phytoplankton cari be found in Fig. 46. For some macrophytes the t.ime-course of release by decay has been followed (e.g. Cynodon dacfylon : FURNESS and BREEN, 1982; Potamogefon crisprrs : ROGERS and BREEN. 1983; Pofamogeton pecfinnfus : HOWARD-WILLIAILZS and DAVIES, 1979; Typha domingensis : HOWAR»-WILLIAMS and HOWARD- WILLIAM~, 1978; H~~.~I~~-WILLIA~I~, 1979: and ~~~"PUS ,uapyrus : &~»ET, 1977).

Interactions between swamp and opttn water also include the horizonfuI franslocation of chemical components. Horizontal gradient,s for many consti- t.uent.s have often been demonstrated in transects, as by BEADLE (1932), CARTER (1955), TALLING (1957, and in BZ&KA 1974, H~~ARD-WILLIANS (1972),

108 ,J. F. TALLING

Plant

Cyperus ppyrudl) (y = young shoots o = old shoots)

Place Biomass Relative content Stand-stock density (W dry weight) Cg m-2)

(p;drywtm-2) N P K Na Si02 N P K Na Si02

L. George 5020 y 1.76 0.105 3.88 0.39 - 61.6 5.43 103.2 20.9 - 01.06 0.106 1.54 0.43 -

Phragmites uustrdis(2) L. Chad (a) subaerial (b) underground (cl total

3100 - - 2.05 0.025 3.0 - - 63.5 0.77 93 4340 - - 0.34 0.04 19.1 - - 14.8 1.74 829 7440 - - 1.05 0.03 12.4 - - 78.3 2.51 922

Typha domingensis(3) L. Chilwa 2537 0.55 0.083 0.55 1.19 - 22.7 2.1 13.9 30.3 -

Potnmogeton pectinntus (4) (dominant)

Swartvlei 1027 1.2 0.12 - - - 13 1.3 - - -

Potamogeton crispus(5) Tete Pan 41 2.9 0.6 1.8 2.0 - 1.2 0.3 0.8 0.8 - (seasonal max.) (Dongo R.)

H~WARD-WILLIAM and LENTON (1975). and GAU- t)E’r (11)X). Swamp-water concentrations are typically higher (e.g. for total ionic c.ontent and conductivity, fiCX&- -alkalinity, 02, NHJ-N, POJ-P, total Fe and Mn)? with the obvious and not.able exception

- of ()w. (.:orrrsponding chemic.al influence on through- flowing rivrr systems bas been studied for t.he Sudd swaml&. of the \Vhite Nile (TALLING, 1957a ; BISHAI, l\W) atld for t.he Makwa River inflow to L. Nai- vasha (C;AVL)ET, 1978, 1979). For other shallow lakes thr> work of Howard-Williams and co-workrrs on the L. Chilwa-swamp system, summarized by HOWARD- \VILLIAMS (1X9), is particularly notable. HOWARW WILLIAMS and TANTON (1975) gave examples of hori- UJIltal chrmical exchange induced by wind stress and by strram inflow; they also drew attention t.o thr probable effet+. of swamp anoxia on sediment + watpr transfer of solutes and nrganic inputs to the open watrr. Littoral solute contribut.ions are generally likely to be favoured from sheltered depositio- nal as opposed to exposcd erosiona shores, as the former faveur 1:)oth swamp deuelopment. and accumulation of fine sediment.

As swamp vegetatinn cari incorporate relatively large st.orxks per unit. area of the elements C, N, P, K and Si, a major secondary supply by re-cycling - as opposecl t.o a primary or ‘ new ’ input - is set up. Grnrralized aspects of input-out.put, St-orage, and loading rrlationships for N and P in swamp systems are rtxviewed by HOWARD-WJLLIAILIS (1985).

Br$i+s set t ing up chemical storage and secondary inputs, biological accumulation often determines çhemic~al concentrations by inducing complementary nutrient depletions in the water-mass. Examples are

given lat.er in relation to the e1ement.s C, N, P, and Si. Biological storage post mortem occurs extensively in the sediments of shallow lakes and swamps (see 5.3). The rote of sedin1ent.s as secondary chemical sources would be expected t.o be stronger’ in shallow wat.er-bodies, but. high biological produc.t.ivit,y in the wat,er-tnass may accentuate a ‘ short.-circuit. ’ recycling there (e.g. L. George : GANF and VINER, 1973; VINER, 1977a,b,c).

4.2. Total solutes

Shallow wat.ers (Table V) encompass virtually all of the very wide range of ionic content or salinity established by a general chemical survey of African lakes (TALLING antl TALLING, 1965; Fig. 25). They range from very di1ut.e wat-er-bodies on hard moun- t,ain rocks (e.g. t.arns of Mt. Kenya : LOFFLER, 1964) to evaporation pans where t.he saturation limit depends mainly upon the predominant anions. Thus, using the convenient index of elect.rical conduct.ivity measured at. 20 *C (kyo), or 25 OC (kt:,, z k20x 1.12), carbonate-bicarbonate dominated waters (e.g. L. Magadi) attain conductivity values (k20) of about 1 x 10” $4 cm-l, and chloride waters (e.g. L. Afrera) even higher. Almost a11 saline lakes are shallow (MELACE;, 1983) ; in Africa L. Shala (Et,hiopia) is the most outstanding exception.

There is an obvinus connection between high sali- nities in inland waters and the hydrological water balance when nut.flow both surface (R,) and underground (G,) is small and evaporation (E) relatively high. Evaporative concent.rat.ion of salts supplied

Lake Grmtry

Nabugabo Uganda

date kzo z Z Na+ K+ Gaz+ Mgz+ alk Cl- X0-4 ta.~ p04-P si tot.Fe pH Reference (p m-1) cations anions (m-l l-9 @gl-1) (mi+‘) (pgl-1)

Jun67 25 0.198 0.199 0.090 0.028 0.064 0.020 0.140 0.040 0.019 - - - - 7.0-8.2 Beadle. 1981 Tumba Z&e BallgWeulLI Zambia OP* Nigeria

55 24-32 - 60 - 0285

Jan-Feb 80 - 0.315 May80 153 -

Jul61 76 1.03 Mar 64 137 1.68 Jan 56 178 - Jun.51 201 237 Feb 76 (1%; 2.63 Jun 61 260 2.94 Jlm61 330 3.92

Mar 64 32’2 3.72 Dec79 530 6.3

Jul76 (565) 6.66 Aug76 (45) 0.55 May - 7.47

Jul61 785 8.60 Feb 64 623 9.1 Mar 76 (187) 3.18 Oct 76 (720) 11.04 Nov69 978 8.58

Jul66 - 10.8 Jan70 X00 12.85

Dec70 2500 35.85 Dec79 4770 58.6

66- 59.1 61 5120 51.7

Apr63 - 75.7 Dec 79 105oo 139.0

Jd69 11700 172 Mar 64 158OQ 228.5 Aug69 23500 301 Jun78 - 616 Jan 70 57400 1245

7880 (52000) 1264 May 61 725CQ 784

Jun61 94OCQ 937 Feb 61 16OLHlO 1666

May 71 (111300) 2879

0.293 0.113 0.033 O.CI 0.075 0.100

0.02 0.066 0.0%

0 0.260 0.03

- _ 0.02 - - - - - -33.1 - 0.052 -

- - 100

4.9 - 6.8 - 11 - 8.5 2%

15.9 48 15.2 500 21.1 - 14.0 5410 11.8 - 4.5 - 4.1 - 16.1 - 18.7 -

18 -

32 530

54 - 15.0 - 66 620 83 - 60 - 8.4 9.5

4.5-5.0

0.113 0.049

Dubois. 1959 Talling & TaBing. 1965 Hare & Carta. 1984

Mwau TaXIa Ras Amer George Kabara Mulehe Naivasha Zd Baringo Cbad. N Chad. SE Mohasi Kltangiri Abaya Tete pan

.Zanibia&!aire Etbiopa Sudan Uganda Mali Uganda Kenya Ethiopa Kenya Chad-Nigeria

1.05 020 0.032 1.62 024 0.040

0.418 Od5

0.141 0.044

15

239 2.55 3.09 3.97 3.80 6.11

0.25 023 412 0.48 037 - 034 0.65 272 0.41 025 122 024 022 - 0.82 036 70

30 2im <18

220

6.5

8.4 9.1 9.6

8.0

8.0

8.7 7.7

Talhg C Talling. 1965 Wood & TaBine. 1988 Tding. unpubi: -@lling & Talling. 1965 Dumont et Id.. 1981 Talling L Talling. 1965 Talling dr Talling. 1965 Wood & Talling. 1988 Talling & Ri=. unpubl. Carmouze et or.. 1983

Uganda Tanzania Ethiopa s. Atiica

7.19 9.15 9.1

Hippo Pool Uganda Chamo Ethiopa chllwa M&Wi

8.25 11.7

0.59 0.11

0.40 037 0.470 0246 1.96 0.58 2.11 030 4.85 033 1.87 0.76 0.12 0.07 3.791 0.235 6.74 0.123 7.70 Odl 1.70 0.03 6.70 0.05 0.65 2.61 9.1 036 113 035 33.9 0.59 53.4 4.41 45.w 1.45 49.6 2.17 70.5 4.5 136.0 29.6 165 7.3 222 6.5 xl0 024 493 6.1 1235 9.9 12M) 3.3 774 10.4 935 2.4 1652 13.7 2565 302

0.375 0.945 1.20 1.01 1.30 1.085 0.76 0.70 0.70 2.22 020 1.390 1.205 0.76 030 1.26 260 0.70 0.60 0.66 0.33 280 CU.05 0.7 0.05 8.1 4.1 0.15 23.7 <o.o CU.05 CU.15 CU.5 cQ.5 0.76

0.1 0.83 1.52 0.81 1.91 1.70 2.18 331 334 4.93 627 0.46 3.10 6.65 7dl

0.66 0.56 1.131 0.63 0.615 035 1.81 0.16 2.05 0.55 022 1.15 3.03 2.72 0.64 0.60 0.70 0.44 9.83 do8 8.6 0.01 4.1 CO.1 0.16 933 0.18 dl.1 cQ.6 Q.5 Q.5 11.0

4.06

1.80

1.10

_ - 0.022 - 0.10-0.711020 0.60 128

carmouze et al.. 1983 Damas. 1954

527 282 9.4 1.66 6.7 7.89 19.0 14.51 526 4.41 523 45.15 533 10.79 63.4 13.6 107.0 253 107.0 55.5 166.5 51.5 116.4 186.5 3.6 181 965 180 Km 845 580 154.6 8% 244 1180 637 150 1450

4 34 1120 14 5100

5500

_ - 0.15 - 0.62 - - - - - 2.91 4% 8.88 - 3.44 4500 0.4 - 6.7 6% 2.8 - 225 - 17.3 - 347 191 4.5 - 5.0 9000 97.5 11000 47.5 ti5oœ 50 llooo 1270 -

Talling & Talling. 1965 WC& & Talling. 1988 Rogers & Braen. 1980

Kilham. 1982 a Wood L Talling. 1988 McLacblan. 1979 E

Sonachi Kenya Mariut. Sta.1 Egypt Rukwa N Tanzania Kilotes Etbiopa Nakuru Kenya Elmenteita Kenya Abiata Ethiopa Eyasi Tanzania QWlXl WP BQpd. oknnington) Kenya PIemil Lb Pan s. Ahica Metahara Ethiopa Manyara Kenya Magadi Kellp Mahega Uganda

6.4 8.9 8.5 8.8

59.9 593 67.7 77.4 139.0 182 240.5 324 532 1205 1249 831 1097 1867 2870

9.6

9.4 103 9.5

7000 122 - 120 -

50-J 8.9 - 117 - 13 -

10.6 10.4 9.9

m TaBing & Rigg. unpubl. m

El-Wakeel et al.. 1970a.b Talling SS Talling. 1965 Wood & Tallina. 1988 Talling & Rigg.-upubl. Hackv & Kilbam. 1973 Wooi dz Talling. 1988 Heckv C Kilbam. 1973 Tallir;g & Rigg unpubl. Heckv & Kiham. 1973 Asht& & Scboeman. 1983 Wood & Talling. 1988

9600 10.1 10.9

TaBing & Talling. 1965 TaUing & Talling. 1965 Melack & Kilbam. 1972 . . . ~~~

~aar &VS% Natml Egypt AUg76 - - 5620 5959 34.8 - - 220 49w 500 - 4120 - - 1mhott et Id.. 1979

( ) conduçtivity rnrrected from 18 or 25 “C.

110 J. F. TALLING

80

A-1 -A- , L--Y

Si02

I I

._ I

8 P”

7

6

I M’J’J A’S’O’N’D J’F’M’A

Months

FIG. 2-1. - Monthly c:hangps in (abovr) thr discharge of the main inflow to L. Chad and their rc?lati«n t.n (helow) chemical concentra- tians in thr npyr lake basin. Fr«m GAC (1981)).

l‘trritzfionr menxttcllrs des apporfs principnns cm ftzc Tchtzd par le Chari (en haut) ef leur relation atrec les cnracf6risfiqnes chimiques df l’tatz dans le bassin oerstznt.

rminly in surface run-off (Ri) then occ~urs. This situation is widesprt~ati in endorheic lakes of closed hasins (R,, = O), in 1vhic.h ealinity varies in t.ime with varying rainfall and hence Ri. Such variation also occurs drrring t he brirf existence of t,emporary rain tmols and their often rcmarkable fauna (eg. RZ~SKA, I!bh7). E~tirnates of evaporative concent:rat~ion factors cari somrtirnrs be obtained with respect to time or hyrlrological interface (e.g. inflow/lake) from rat.ios of Cl- cc:,rlrentratic,n, where local sources are slight. (e.g, Ethiopian waters : WOOD and TALLING, 198P).

The hydrological relationship has bren directly illust.rat,ed, from many-year records, for lakes Chilwa (KALK et (II., 1979), Nakuru (VARESCHI, 1982), and Chad (GAc, 198C) ; CAHMOIIZE rt al., 1983). In the last

lake, salinity (and SO conduc.t.ivity) increases north- wards away from t.he main inflow, whose seasonal discharge contributes to the spatial and temporal patterns (Figs. 23, 24). Here, and in L. Naivasha, an underground afflux (CI,) limits the expect.ed increase in salinity, augment-ed by chamical incorporation int.o the sediments (se? 4.7). The sait balance may be


.

. cd .F

8 ï .L

* a ANIONS cn

.

F- CATIONS

v-e I I I 1 I 0.1 1 0.1 1 0.1 1 10 0.1 1 10 100 1000

Ca” Mg”

(meq 1-l)

FIG. 25. - Concenlxations of (a) major anions and (b) major cations in relation to conductivity in a series of East. and Central African lake waters. Distributions of two groups of phyt.oplankters, Melosira (Aulacoseira) spp. and Spirulina fusiformis (‘plafensis ‘), are

also shown. From FRYER and TALLING (1986). Concentrations en ions majeurs, anions et cafions, en fonction de la conductivité dans une série de lacs d’,-lfrique Centrale ef de l’Est. La distribution de deus groupes de phgfoplancfon, Melosira (Aulacoseira) spp. ef Spirulina fusiformis (platensis), est aussi figurée. D’aprts

FRYER et TALLINC, 1986.

complicat-ed by underground reserves of saline water supplying the surface rvaporating basin as discussed by GUEST and STEVENS (1951), BAKEH (1958), and GOOD and TALLING (1958) for some East, Rift soda lakes, by HEEG et al. (1978) and HEEG and BREEN

(1982) for the Pongolo pans, and represented (Maglione, 1976) in t.he temporary waters of Kanem (Chad). It is notewort.hy t.hat t-he special case of a largely atmospheric ront-rol of the water budget, in large lakes, wit.h direct prec.ipitat;ion (P) = evapora-

R~U. hydrobiol. frop. 2.5 (2) : 87-M (1992).

Il? J. F. TALLING

tien (E) > inflow (Ri), is compatible with net. inccrnsictt,rable solnte c,oncentration as evidenced by lakps Vict.oria, Tana and Malawi.

Annther special case is constitut.ed by shallow coastal w;rtc>rs (v.g. lagoons) wit-h a limit.ed and sometimes intermittent connectZion to t.he sea, from which saline water enters. Thus the horizontal distribution nf salinitp is often unctven. Studied examples include SUIIIP Drlta lakes of Egypt (Section 4.1); Lake Ich- kfxul, Tunisia (LEMOALI.E, 1983a) ; the elongate Ebrie lagoon (Ivory Coast : DIJF~LI~, 1982a; ~~RANCI and &IANTR.~INE, la*?); Lake St Lucia (Sout,h Africa); and Swartvlei (South Africa) in which the denser salinr inputs cari st.abilizr a deep-wat,er layer (HOBARTS and ALL~NSON, 1977). Wit.h anot.her type of lagoon rut off from.a- relatively dilute lake. salinity ma y develop int.ernal1-y from rvaporative ctonçr,ntration. An example off L. Tanganyika is des- cnribed by (:AL.JoN (1 987).

Solid Salt. drposit,s (sec Section 5.2) are a conspi- mc~uj feature of the margins of mari?? shallow saline lakrs. Some pclriodic removal by wmd may occur. The relat ively insoluble carbonates are particularly (~ommon, notably trona (Na&O3.NaHCO:~.2H@) and Irss frequently a tufa of calcium and magnesium Ixarbonates (~.g. Gae, 1980). Sodium chloride occurs orrasionally (e.g. L. Katwe in W. Uganda).

1.3. Major ions and pH

Almost a11 thr ionic cont.ent is cont,ribut,ed by four major cat.ions (Ca”+, Mg’+. Na+, K+) and t,hree major anions or anion-pairs (HCOs + CO$-, Cl-, ~C)AP-). Variation of their concentration is shown in Table \’ and furthrr illustrated for many East and Crnt.ral African lakes, shallow and deep, in Fig. 25; these are largely bicarbonate-carbonate dominated waters. A wide range of wat,er-types is represented in the chf~mical surveys of LIVINGSTONE (1963), SyMoENs (1468), KILHAM (1971), IIECKY and KIL- HAM (1973), VISSER (1974, VISSER and VILLENEUVE (l975), \'INER (1975a), GASSE et al. (1983), and Woou and TALLING (1988). as well as in det.ailed descrip- tions of individua lakr basins as by VINER (1969) for L. George, CHANTHAINE (1978), GAC (1980) and C;AR- MOUZE (1983) for L. (Chad, and GAUDET and MELACK (1981) for L. Naivasha.

r:hl«ridc-durriin~t,e(~ waters are most strongly rfspresented in c,oastal or near-coastal sit.uations with presrnt or past ingress of sea-water. They include many lagoons of nort,h and west Africa (e.g. Ebrié lagon rn : DUFOIJR, 1982a); L. Asal (C;ASSE et ul., 1983. Wooo and TALLING. 1988) and L. Afrera (= L. Giulietti) (MARTINI, 1969; GONFIANTINI et al., 19’73) npar Djibouti : Swart.vlei (ROBARTS and

ALLANSON, 1977), L. Sibaya (ALLANSON, 1979) and some other coast.aI lakes (ALLANSON and VAN WYIC, 1969) in Sout.h Afric.a ; and L. Qarun in Egypt (MES- HAL and &LOR~~S, 1980) - t.he la& with many marine organisms (e.g. plankton : El MAGHRABY and DOWIDAR, 1969). There are also some more inland sites with local sources of c.hloride as with L. Katwe in Uganda (TALLING and TALLING, 1965; ARAD and MORTON, 1969) and L. Mohasi in Ruanda (DAMAS, 1954)) L. Ihotri; in Madagascar (MOREAU, 1982), and various Saharan sait, lakes (e.g. Dawada, Kufra).

In the crJmmonest.. bicarbonat.e-carbonate, type of African inland water, increase in salinity (and conductivity) is associated (Fig. 25) with a st.eady rise in the concent.rat.ions of Na+, rather less regular rise (affect.ed by loc.al sources) in those of K+, and the final loss by prec’ipitation of Ca2-t and Mg”+. Thus the rat,io of monovalent to divalent. cat.ions increases, and in saline waters Na+ is predominant. This t.rend has hpcn furt-her analysed by TALLING and TALLING (la@>) and Woon and TALLING (1988) from comparisons of many African lake waters, and by GAC (1980) from bot,h observations and experiments involving the concentration with time of water from L. Chad and its main inflow (Fig. 26). In the survey Table ‘I’ all lakes of high salinity (c.ationic or anionic content > 100 rney 1-l) have Na+ as t.he dominant- cation, but. if wit.h low alkalinitv (e.g. L. Ihotry) t.he concentrations of Ca’+ and Mg;+ cari be considerable - as in seawater. A c.ontrasting situat.ion of very low &I$f concentration (< 0.02 meq 1-I) was encountered bg Loffler (1964) in high alt.it.ude lakes on the East, African mountains Ruwenzori and Mt.. Kenya. Although many biological consequences of t.he overall salinity series are well documented, the influenc,e of individual ions and ion ratios is more controversial (BEAnLE, 1981). Thus WOOD and TALLING (1988) t.entat.ively related a strong representat.ion of some phvtoflagellates with lower monovalent to divalent caGon ratios.

Given a predon1inanc.e of bicarbonate+carbonate (= alkalinity), t.he more saline wat.ers are also inten- sely alkaline in reaction. a feature atcentuated by the phot,osynthet.ic removal of CO2 by their freyuen- t.ly dense phytoplankton. 111 this respect. it is int.er- esting t,o compare the productive East Afric.an lakes Nakuru and George. If air-equilihrium prevailed with respect. to CO2, their expected pH would be about, 9.9 and 8.2 respectively (TALLING and TAL- LING, 1965, Fig. 3); in reality that of L. Nakuru lies bet,ween 10 aid 11 (TUITE, 1981 ; VARESCHI, 1982), and that of t.he less buffered, low alkalinity waters of L. George is typically between 9 and 10. Thus carbon deficits, relative to air-equilibrium conditions, are set, up in productive waters. Much lower values of pH are usual in saline, chloride-dominated waters

AFRICAN SHALLOW LAKE~ 113

1 I

10’ 103

Concentration factor

FIG. 26. - Evaporation experiment. with water from the inflow Chari River to L. Chad, showine differential changes with t.ime of chemical concerkations in relation to t.he overall volumet.ric concentration factor. From GAC (1980).

pvaporufion expérimentale de l’eau du Chari qui parvienf au lac Tchad. Certains ions suivent une évolution diffkenfe lorsque le facteur de concenirafion kuaporitique augmenie. D’après G.kc, 1980.

of relatively low alkalinity, such as t,hose influenced by sea water. As alkalinity (and HCOs- concentra- t,ion) fa11 towards zero, t.ypically in waters of very low ionic content,, pH approaches c. 5. Still lower values of pH are associat,ed with free acidity, aIthough the Iatter is only rarely analysed directly - as for the water of L. Tumba on the Zaïre river systcm with pH 4.5-4.9, free acidity 0.1-0.7 meq I-1, conductivity (18 OC) 24-32 PS cm-1 (DUBOIS, 1959).

At very high (> 10) or very low (<4) values of pH t,he ions OH- and H+ respec.tively exceed c. 0.1 mmol 1-l and then cari influence conductivity appre- ciably, altbough their proportion of the total ionic. content is usually small. BERG (1961) gives examples from acidic waters of the Congo (Zaïre) basin ; there, at pH 3.5, a conduetivity value of c. 130 FS cm-1 would be expected. Below pH 4.2 waters were inhibi-

tory to t.he invasive water hyacinth. Eichhornia crassipes.

In some East. African waters, usually of high alkalinity, the fluoride ion exceeds 10 mg 1-l (sometimes > 1 g 1-l). It. may then have adverse effects on certain freshwat,er organisms (KILHAM and HECKY, 1973) and on human utilizat.ion for drinking water.

4.4. Plant nutrients

Aspects of the dist.ribution, and especially the exchanges, of nutrients in African freshwaters have been reviewpd hy THORNTON (1986) and (for phosphorus) MELACK and MAC INTYRE (in press). Most major ions are also plant. nutrients, but are usually presumed to be present in excess of growth-limiting

R~U. hydrobio[. trop. 25 (2) : 87-1~3 (1992).

ELMENTEITA 1980

MWERU

BANGWEULU

Si (mg 1-l)

I

I I I I I

1 10 100 1000 10 000 100 000

total P (pg 1-l)

FIG. ‘27. - C.orrczrrt.rat.iorls of soluble reactive silicon and total phosphorus in relat,ion to ronduct,ivity in a series of shallow East. and Central African lake waters. Adapted from TALLING and TALLING (1965), KALFF (1983), and WOOD and TALLING (1988).

Concrnfration du silicium soluble réactif et du phosphore iota1 en fonction de la conductivité dans une série de laes peu profonds d’Afrique Centrale et de 1’Esf. D’après T.ALLING et TALLIN~+ (196.5). KAI.FF (1983) et WOOD ef TALLING (1088).

concentretions - a view challenged by BEAUCHAMP (1953) for sulphate and possibly questionable for the lower concentrations of pot.assium encountered in some Afriran waters. Here we are concerned with forms of t he elements N, P, and Si. When variability in their conccrkrations is viewed against total ionic c*oncentration Or conductivity (Table V and Fig. 27) there is little regularity - although saline carbonate lakes typically have high concentrations of Si and total P. Besides their biological significance in the bulk-water phase, their presence and uptake from sediments is important for many rooted macro-

phytes (see Dtlnny, 1985a). Carbon is another plant nut,rient, derived from free CO2 (section 4.5) or HCO:j-+COS~- (section 1.3); it cari also provide a comparat,ive measure of biomass and organic detritus, as in studies of Lake George (BURGIS et al., 1973) and the shallow Wuras Dam (GROBBELAAH, 1985; GROBBELAAR and TOERIEN, 1985).

Silicon, present; in solution as silicic ac.id (Si(OH)J or silic.ate, is of significance for the growt.h of diatoms. In most ‘4frican shallow wat.ers its concentration is greatly in exc.ess of those levels (< rirca 0.3 mg or 10 pmol Si 1-l) at which a possibility of growth

Heur. hydrobiol. lrop. 25 (2) : X7-1J4 (1992).


30

20

-ï 10 E

22 0

ô 0 z

x .Z z

B

8 ,- Fi 5 10 a B

5

0

30 1

E

N

10 g

FIG. 28. - Large depletions of soluble reactive Si (expressed as SiO?) during hvo episodes of diatom growth at t.he st.ation Bol

on L. Clhad in (a) 1973 (b) 1974. From LEMOAI.LE (1978). Fortes variations de la silice riactive (exprimée en Si Os) au cours de drus kpisodes de développemenf important des diatomées dans le lac Tchad à Bol en 19%3(a) et 1974(b). D’après LEMOALLE,

1978.

rate-limitation might be suspected. For wat,er of the Ebrié lagoon (Ivory Coast), application of a biological assay proc.edure gave no indication of a limiting role in phyt.oplankton production (DUFOUR and SLE- POUKHA, 1951). The suggestion of I<ILHAM (1971) of a qualitative influence of higher concentrations on dia- t.om floras was not clearly supported by the survey of C;ASSE el nl. (1983). In many shallow lakes (and rivers) concentrations of 5-20 mg 1-l are found, which are high by world standards ancl probably influenced by tropical weat,hering. Examples include L. George (TALLING and TALLING, 1965; VINER, 1969), L. Naivasha (TALLING and TALLING, 1965; GAUDET and MELhcK, l%l), L. Chad (LEMOALLE, 1978; GAC, 1980; CARMouzE, 1983), and L. Tana, L. Pibaya, L. Chamo, and L. Baringo (TALLING and TALLING, 1965; ~VOOD and TALLING, 1988). There

are few series of seasonal analyses in relation to the abundance of diat-om populations, but. in L. Chad (LEMOALLE, 1978 - sw Fig. 28) and in shallow inshore wat.ers of L. Victoria (TALLING, 1966) some periodic depletion cari occur. However in these lakes, as in t,he Ebrit lagoon. availability of Si is generally unlikely to limit- the growtdl of diatom populat.ions. Soda lakes of high alkalinity (e.g. Nakuru, Elmen- teita. Magadi, Kilot-es, L4biata) are invariably of very high Si content (TALLIN~; and TALLING, 1965; WOOD and TALLING, 1988) ; its abiogenic transformation from solute to solid (sedimentary) phases has been studied for L. Magadi (EUGSTER, 1967; EUGSTER and JONES, 1968), as ~~11 UP for thr relatively low alkalinity waters of L. Chad (CARMOIJZE, 1983).

Of the several measures of phosphorus concentration, that of tot.al phosphorus (obt.ained by digestion) is probably t.hc most informat.ive for c.omparative purposes. In silt-rirh wat.er, however, it cari merely indicat,e the abundance of a variable particulate

G - George Na - Nakuru a - Boaorla

/ -” ii - Sciachi N - Naivashs 0 - Oloidie” id - Winam Gulf

01 10

I I I I

50 100 500 1000 5000

total P (mg mM3)

Frc. 29. - Phyt.oplanktou abnndance. as indicated by mean values of chlorophyll LI concent2rat.ion, in relation to the associated concent.rat.ions of tot.al phosphorus in various African lake waters. The inserted line is a reeression from DILLON and RIGLER (1974, relat.infr the winter-spring roncent,rat.ions of total P wit.h the mean summcr concent.rati«ns of chlorophyll a in various t.emperate lakzs. Adapted from KALFF (1983), with

correct,ion for L. Victoria. ilhondance du phytoplancton (exprimée par la concentrafion molyenne en chlorophylle a) en fonction du phosphore total dans divers lacs d’Afrique. La droite correspond à la relation de DIL-

LON et RIGLER (1974) reliant P total en hiver-printemps et la chlorophylle moyenne d’éfé dans diffbents lacs de zone tempérée. Modifié de KALFF, 1983 avec une correction pour le lac Victoria.

Reo. Irycirobiol. frop. 25 (2) : 87-M (2992).

116 .J. F. TALLING

phase. Unfortunat~elp it is avaiIabIe in few çhemical survr’pa of African waters (e.g. TALLING and TAL- LIN~;, 1966 ; set\ Table V and Fig. 27), and most- ana- lyses are of soluble reactive phosphate. St.ill fewer analyses refer to particulate phosphorus : here detailed studies on shallow lakes inc.lude VINER (1977) for L. Geqe (Fig. lO), BAUDET (1976. 1979) and KALFF (1983) for L. Naivasha, HOWARD-WILLIAMS (1977) and H~~ARD-WILLIAMS and ALLANSON (1981) for Swartvlei (Fi-s. 30, 31), and DUFOUS (1984) for the Ehrié lagoon.

Analyses of tot.al phosphorus show that. - presumably for geological reasons - the content in many shallow African waters is considerable (> 30 pg 1-l) by world st.andards. Comparat.ive examples are given in Fifz . 29 and Table V. and in a recent. review by N~EL~CK C% ~~AC~NTYRE (in press). There is a ten- dency. as elsewhere. for higher concent.rations per Imit. volume to be present in shallow than in deep lakes. anb also in more saline than in dilute waters. Howef.rr analyses for t.he last class - which include tbr import-ant L. Bangweulu and most West. Afric.an waters - seem especially deficient.. Phosphate may ahare a phase of progressive concentration, during hydrfJhgic.al contraction, wit.h total ionic conrent.ration (e.g. in L. Chilwa : MCLACHLAN et al., 1973). The many shallow lakes which brar dense and long-las- ting phytoplankton are inevitably high in total phosphorus as the converse of dense phytoplankton but low total phosphorus is excluded. In such waters most phosphorus cari be incorporat.ed as particulat,e phosphorus in t-he algal populations (Fig. IO), with cmly a amall fraction of soluble reactive phosphorus (e.g. < 5, pg 1-l in L. George : GANF & VINER, 1973; I'JNER, 1977~). Turnover of the latter fraction cari thrn be expected t,o he rapid, as demonstrated expe- rimr~nt,ally wit.h POd-P uptakr for L. George (VINER, 1973, 1977~) and with n2P by PETERS and MCINTYRE (1976) for L. Elmenteita and by KALFF (1483) for laktbs ( )loidic~n and Sonachi. III t.he last. lake there is rvidence of P- limitation from t,he response of phyto- planktnn t.o addrd phosphate, but, not ammonium. and from low ratios of sestonic P/C and P/N (MELACK et (II., 1982). In other shallow lakes (e.g. Nakuru) wit-b a grrat,er exress of soluble reactive phosphate, t.he experiment.al turnover of radio-phosphorus was slower. The situation in L. Naivasha. sturlipd by KALFF (1983), appeared intermediate and includrd marked seasonal variation. The partitioning of radio-phosphorus has also been followed in 3.5 TII - deep isolation columns within the Midmar reserboir. Natal (TWINCH and BREEN, 1984). Water from this lakr. and if,s cc~lu~nns, has been used for syst-rniatic enrichment and assay experiments (TWINCH and EREEN, 1981. 1982). ‘4lgal responses \-arird with season or prier enrichment., but, positive

H~O. hytirnhiol. frop. 2.5 (2) : 87-144 (lB?.2).

~ (alstandingstock ofi\

0.5

0

14 (b) P uptake rates 7 in situ

2 0 10

I E 6 CI

2-

I I I I I I I I I I l I I I JJASONDJJFMA~JJASON

1965

Months 1966

FIG. 30. - Seasonal chanbt f ‘P of P quant,ities in Swartvlei, sho- ring t.ime-relationships of (a) t.he maximum in standing stock of plant hiomass with (h) &imated rat.es of uptake and (c) release as soluble P. nedrawn from ~IOWARD-\VILLIAMS and

hLI..ANWN (19Xlh). l?uoluiion saisonnière de dioerses formes du phosphore, qui monfre les 6rvolrcfion.s successioes a) du maximum du stock dans la biomasse primaire, b) des r&imafions des rlitesses d’assimilation ef c) de lu remise en solution du phosphore. Modifié d’après

response to P enrichment was frequent even when the total phosphorus already exceeded 30 pg 1-I. In lakes with dense weed-beds of submerged macrophytes, the latter may dominate P-uptake and recycling. This occurred during t.he seasonal growth and decline of Pofamogeton pectinatus (+associat.ed Cla- dophora) in Swartvlei (HOWARD-WILLIAMS, 1981 ; HOWARD-WILLIAMS and ALLANSON, 1981 b ; se8 Fig. 30). In this lake the P- fractions are also stron-


E m Particulate P m ‘“’ ] Soluble 0 SRP

1976

Months

FIG. 31. - Rlonthiy variation in Swart.vlei, over 3 years, of the concentrations of total P and it.s cornponent fractions of soluble react.ive P @RP), soluble unreactive P (SUP), and particulate P, showing the relationship of maxima t.o episodes of river floods and

wind dist,urbance. Redrawn from HOWARD-WILLIAMS and ALLANSON (1981). Concentrations mensuelles de diverses formes du phosphore dans le Srvartvlei au cours de 3 années successives, avec les inf[uences des crues des rivières et des turbulences dues au vent. SRP = phosphore réactif dissous, SUR = phosphore non réactif dissous. Parficulata P = pkos-

phore particulaire. La somme constitue le P total. Redessiné d’après HOIYARD-IT~ILLIA~~S ei AILANSON. 1981.

gly influenced by inflowing flooclwater and by wind- dist,urbance of sediments (Fig. 31).

In shallow lakes the biological importance of sediment-water exchange of phosphate (and ammonium) is enhanced by the large ratio of sediment. area to water volume, the short.-lived (usual diel) character of thermal barriers, and possibly by wind-disturbance of sediments. Detailed studies of suc.h exchange have been made for L. George by VINER (1975d,e, 1977a,b) and for the Ebrié lagoon by LEMASSON ef al. (1982); VINER believed that in the very productive L. George most recycling of N and P occurred in the water-column rather than the sediments. For L. Kioga a modelling approach to sediment-water exchange has been used (KAMP-NIELSEN e1 al., 1980). Suspended sediment with adsorbed phospl1at.e is abundant, in many shallow waters, and no doubt, plays a role as a nutrient reserve. In swamps with rooted macrophytes the nutrient-P pathway from sediments ta water via plant uptake and decay may be quantitatively important (DENNY 1985a).

Although the phosphorus content. of plant, biomass is variable, some indicat.ion of the quantities (pg l-1)

Rev. hydrobiol. frop. 25 (2) : 87-14Y (1992).

likely to be incorporat~ed in phytoplankton crops of various densities cari be obtained from Fig. 32. Detailed seasonal and experimental observations have been made 011 L. George (VINER, 1973, 1977c) and inthe Ebrii: lagoon (DUFOUR, CREMOUX & SLE- POUKHA, 1981; DUFOUR, LEMASSON & CREMOUX, 1981) where the mean interna1 cellular subsistence quota for P was estimated as a ratio 0.0055 : 1 by atoms to sestonit C (i.e. C/P quotient of 182). Nitrogen. The following forms or fractions of combined nitrogen cari be distinguished : nitrate-N, ammonium-N, nitrite-N, hydroxylamine-N, dissolved organic N, particulat,e N, and total N. Most ana- lytical information for hfrican shallow waters refers to the first t.wo of these. Nit.rite-N is generally negligible quantitatively, although it is an important intermediate and has been used experimentally by VINER (1973, 1977~) to t,race N-uptake by phyt.o- plankt.on in L. George. 1t.s seasonal and spatial distribution in L. Edku, Eept, has been followed by SAAD (1978). Hydroxylamme-N has been specifically studied only in some Ethiopian lakes, where it was probably a significant, N çomponent (BA~TER ef al., 1973; PITWELL, 1975; Woon et al., 1984). For sur-

118 -1. F. TALLING

FIG. :+Y?. - .\pprosimate int,errelations, read horizontally a~ross vwtical logarithmic wales over 1 orders of magnitude, cif c‘orlI~rrrtrat.ioll~ of two indices of phyt.oplankton hiomass (cell volnn~e, chl<lrophyll u) ohs~rved in Y African waters (shallow undrrlined), and &imates of associated quant.ities of wllular Ci, N, arr11 P. Thr int.crrrlat.ionr aw hased on a chl-n cont,ent prr unit 1~~11 vr)lume of 4 pg mm.“, a Cjrhl-a mass ratin of 40, and C;/?I ;III~~ C/P ratios that are (a) the mean vaiues for L. GHJI-~P t-~«rtid by YINER (1977), (h) t.he generalized Hed-

fit~ld VHIIICS. Adapted from TALLIN~; (1981). lnterreic~iicm.s, en khrllc logarithmiques sur -1 ordres de yrandcur, de di0w.w cnrncférisfiques phyfoplancioniqllex : Aume nlgal, concrnfrtrfion en chlorophylle CI, ef esfimatinn des quantifés carres pondczntes dr I,‘, N ef P. Les correspondances sont calculées sur lu hase de (:hltrjwlumr cellulaire de 4 py, rnm3, Clchla en masse de 40, Clx ef C/l’ t1’aprP.c 11) \.INEA (1977) pour le lac George ef 2>) le

rupprf de Redfirld. D’uprk T.4 LLING, 1.981.

face watrrs gerrrrally, nitrat.e-N or dissolvrd organiç N are usually llrcdominant in t.he total N. In hfric.a dissol\-rd organic N is t.he more likely, although thrrc-b are only few support-ing analyses of it or of total N (e.g. Du~o~s, 1959; VINER, 1977c; GAUDET, 1979: i\s~&~, 1981: DIJFO~H, CREMOUX and SLE-

POUKHA, 1%1 ; E<ALFF. 1983).

The concentrations and dynamics of particu1at.e N are perhaps best knc-nvn &om studies of swamp regictins doruinated by macrophytrs as at L. Chilwa (H~u.AR~,-~S'ILLI~M~ and H~~ARD-WILLIAMS, lf178: h1CLIC:HLAN. 1979). the littoral of L. Kariba (s. bI. RIC:LACHLAN, 1970). L. Naivasha (GAIJDET,

1979: (.;AUDET and MUTHURI, I%ls, b), Swartvlei (~-I~~~.~Hu-~~~ILLI~~Is, 1977), and L. Victoria (C;A~-

DET. 1976). In such vckgetation stands the amounts of plant--N presrnt per unit. area (Fig. 46) are much hig-

her than those in most- phyt.oplankton communities, for which t,he probable magnitude of concentraCons per unit, X-olumr cm be located from Fig. 32. Parti- culate N concent,rat.ions and dynamics have been studied direc,tly by VINEH (1973, 1977c) for the dense phyt.oplankton of L. George, where in 1967-8 concent,rat.ions varied between 1 and 5 mg N l-1 (Fig. lO), and hy DUFOUR, LEMASSON and CREMOUX (1981) for t.he less produc,ttive Ebrié lagoon where they ranged gencrally bet;ween 150 and 400 Fg N 1-l but. rose to 700 Fg 1-I in rutrophic areas.

Such high concentrations are rarrly reached for inorganic nitrogen in the surface region of any productive shallow wat.er. In L. George, for example, concent.rat.ions of NHd-N are typically < 10 pg 1-l with NOa-N and NC&N uxually undetect<able (GANF and VINEH, 1973 : VINER, 1977c). For N therefore, as for P, turnover by the met.abolic.ally active phyt,o- plankton must be rapid and rates of regeneration crucial. Profiles of NH,I-N accumulation in t.he bot- t.om sediments (GANF and VINER, 1973 ; see Fig. Il) indicated periodic disturbance and release to the overlying wat.er, but, VINER (1975c, 1977b) believed t,hat the sediment contribution to overall N-cycling was small.

The lack of prolonged thermal (clensity) strat,iflca- tion in tha open watrr of most. shallow lakes generally eliminates accumulation of NHd-N in deeper water, although VINEH found that- some small increase occurs cm a diel tirne-scale even in t-lie shallow L. George. In sheltered. often anoxic swamp water, ric.h in (and oft-cn overlain by) veget.ation in various st.ages of decay, the situation is quite otherwise. Transects of various African swamps have shown considerable (often > 1 mg N 1-l) accumulations of NHd-N away from t,he open margin (CARTER, 1955; HowAR»-WILLIAM~, 1972; GAUDET, 1976, 1979; review in HOWARD-WILLIAMS and GAUDET, 1985), and release to t.he open water may significan- tly influence the latter (e.g. L. Chilwa : HOWARD- WILLIAhis. 1972; HOWARD-WILLIAMS and LENTON, 1975; HOWAR»-WILLIAMS and H~~ARD-WILLIAMS, 1978). AI~ accumulat.ion of NHd-N (> 1 mg 1-l) is also known in anoxic, wet,er below a salinity-density barrier of the shallow Ehrie lagoon (DUFOUR, 1984).

In shallow temperate waters it is common to find a seasonal (usually winter-spring) phase of higher nitrate concent,rat.ion. This is generally lacking in shallow African waters, excepting some at. higher latitudes (e.g. L. iCIariut., Egypt.; ALEEM and SAblAAN, 1969; t.hr Midmar reservoir, Natal : TWINCH and HHEEN, 1981), unless subject, to seaso- na1 flood-water rich in nit.rate. The Blue Nile flood (TALLIN~; and HZOSKA, 1967) provides one example ; another is t,he elevat,ion of nitrat.e concent,rat,ions in an estuarine region of t.he Ebrié lagoon early in the


rainy season (DUFOUR and DURAND, 1982). A t,emporary flush of nitrate on reflooding of dry lake sediment (see Section 6.3) is probably widespread, as is a flush of nitrate in savanna soils and their run-off after the beginning of a rainy season (see, e.g., VINER, 1975a).

Fixation of gaseous nitrogen (Nz) by certain bacteria and cyanophytes is undoubtedly widespread in African waters, but has been st.udied quantitatively in very few, including L. George (HORNE and VINER, 1971, and Fig. 39 ; GANF and HORNE, 1975) and Rietvlei (ASHTON, 1979, 1981), where fixation by cyanophyt.es was est.imated t.o be an appreciable contribution ta the N- economy of the lakes. There are rep0rt.s of N-fixation by microbes associated wit,h various tloating African macrophytes (eg. Saluinia molesta, Eichhornia crassipes, Cyperus papyrus), but its quantitative significance is generally uncert.ain. The subject is reviewed in DENNY (1985a).

The converse process of denitrification, mediated by bacteria and liberating N2, has rarely received specitlc study in Afric.an fresh wat.ers. Promoted by higher temperature, it. is possibly crucial in maintai- ning the generally low levels of nit.rat.e, and hence t.he inorganic N/P rat,io of interest in relation to the control of plant growth (eg. TALLING, 1966; KALFF, 1983). Direct measurements of t.he denitrification of added nit,rate have been made by VINER (1982) for sediments from L. Naivasha and nearby wat,ers.

4.5. Dissolved gases

The gaseous content of shallow waters is, to a great extent,, intluenced by the exchange pat,hway açross the water/air interface. If such exchange were the dominant. influence on gaseous concentrations, the latter would be set. by the partial pressures of atmospheric gases, their intrinsic solubilities in water, and the modifying influences of temperature and salinity. In reality, large modifications c.an be introduced by barriers to vertical exchange and by biological act.rvit.ies. The latt.er include the met,abolic eschanges of 02 and CO2 in photosynt.hesis and respi- rat,ion, and - in 02 depleted zones - the bacterial production of H$ and C;H4 as reduced end-products.

Dense and illuminated populations of submerged aquatic plants, planktonic or macrophytic, generate concentrations of oxygen in excess of air-equilibrium values. The accompanying consumption of CO2 is rarely measured directly (as for L. George - GANF and MILBURN, 1971) but is reflected in increase of pH (e.g. Figs 66, 22). As daylight is discont.inuous, diel oscillat,ions are set up in the concentrations of 02 and total CO2 (cornprising free COz, HCOs-, and CO$-) whic.1~ are inversely relat,ed, and in pH which

Reu. hydrobiol. trop. 25 (2) : U-144 (1992).

varies inversely with the total CO2 content. In highly buffered waters, such as L. Nakuru (VARESCHI, 1982), the diel pH changes are insigniticant. The diel cycle usually includes nocturnal 02 levels below air- saturation, whereas the pH and CO2 levels in such productive waters are t.ypically indicative of main- t.ained Cor- deficiency relative to t.he air-equilibrium state. This difference in behaviour of the t.wo gases is influenced by the relatively low partial pressure of CO2 in t.he at.mosphere and the equilibria uniquely relating this gas to ionic reserves (HCOg, COaz-). The shallow African waters in which such diel fluctua- t.ions haue been studied include a lagoon and reservoir on the Whit,e Nile (TALLING, 1957b), bays of L. Victoria (WORTHINGTON, 1930; TALLING, 1957b), L. George in [Jganda (GANF, 1974b, GANF and HORNE, 1973 : Fig. 6), L. Nakuru in Kenya (MELACK and KILHAM, 1974), L. Kilotes in Ethiopia (TALLING et al., 1973), the Ebrié lagoon (VARLET, 1978), and the Pretoria Salt. Pan (ASHTON and SCHOE~~AN, 1983 : Fig. 22). In a11 these waters the diel gaseous changes were prirnarily due to dense phytoplankton.

Barriers to vertic.al exchange cari develop from temperature-densit,y and salinity-density st.ratification, or from layers of macrophytes, especially when these form float.ing mat.s (e.g. MUSIL et al., 1976). Although temperature-density stratification is short- lived (,often diurnal) in most shallow wat,ers, when combined wit.h int,ense biological activity it cari result, in steep vert,ical gradients of oxygen, carbon dioxide, and pH. Examples are provided by the diel studies cited above, here illustrated in Figs. 6,’ 22 38. It. is uncommon, alt.hough known (TALLING ei al., 1973; ASHTON and SCHOEMAN, 1983; Fig.38), for complete anoxia to develop or persist in wat,er below the diurnal density barrier. Production of CH4 and H$S has apparently not bren recorded in this context. They are known to accumulat,e below a salinity-density barrier in coastal lakes or lagoons t,o which seawat.er has entered, as in parts of t,he Ebrié lagoon and in Swartvlei.

The situation is quitr otherwise for swamp condi- t,ions, characterised by a large input of decomposable organic material and t.he common development of floating plant, mats, such as the rhizome-layer of Cyperus papyrus. Anoxia is widespread in the water beneath such mats, except. very close to outer exposed margins or when st.reams renew the water rapi- dly (e.g. at L. Naivasha : GAIIDET, 1978, 1979). An onshore wind Iras been shown to drive oxygenat.ed lake water deep inside fringing swamps of Typha domingensis at. L. Chilwa (HOWARD-WILLIA~IS and LENTON, 1975). Tllustrative sections t.hrough papyrus swamps bordering rivers or lakes have been made by various workers (e.g. BEADLE, 1932a ; CARTER,

120 J. F. TALLING

(a)

(b) 0

FI~;. 33. - Transe4 at. L. Kariba from deeper lake to shallow rstnary of the inflowing Mwenda Hiver. showing (a) t.he vertical distribution of dissolvsd oxygen in relation t.o (b) temperature strafifiration and t.he rxtent of a surface mat, of Suloinia rnolestu. Tbe intrusion of more oxygenat.ed lake wat.er underneath thr mat is probably due t.o compensaCon currents resul- tinp from a wind-induced movement. of surface wat.er away

from t.he mat. From BOWMAKER (1976). Section dans le lac liariba des zones profondes vers I’estuaire peu profond de lu rivih Mmenda. Répartition de l’oxygène (a) en rrlation avec cellr de la température (b) et l’exfention d’un tapis tic Salvinia molesta. L’infrusion d’eau lacustre mieux oxygénée SOUS le tapis de Salvinia részclte probablement de courants de retour compensant un courant de surface s’éloignant des rives sous

l’action du vent. D’après BOWATAKER, 1976.

1955: T.~LLIN(T in HZOSKA, 1974; CA~DET, 1979). CcmsiclerabIr accumulations of dissolved C3& cari O~:~UI‘ mder thr mats, and in some cases CH4 also. There is, however, little rvidence for prolonged accum~~lnt.ions c.jf H& although t#hrb rernoval of sul- phatr undoubt-rdly occurs (e.g: in the Sudd swamps : TALLIN~;, 195ïa). For further information or1 t,hese aild other c~hemi~al aspects of swamp environments in Afric,a. the reader is refc>rred to GAUUET (1976, I!X)), BEADLE (I%l), THOMPSON and HAMILTON (lCN3), HOWAHD-WILLIAMS and GAIIDET (1985), and "PENNY (l%&).

More diffuse free-floating mats of plant aggre- gatths, widespread in both shallow and deeper waters of ,Africa, include thp spec.ies Pistia stratiotes, Eich- hornicl crussipes. and Salviniu molesta. Eichhornia

and Salvittia mats in particular cari lead t,o reduced concentrations of oxygen in t.he wat.er below. Excellent seasonal examples, for Snlvinia, are provided by t.he sturlies of BOWMAKEH (1976) on t,he estuary of a river-inflow t,o L. Kariba (Fig. 33). Experiment.4 studies of ASHTON (1977) showed how tloating mats of dzolla filiculoides could alter the characteristics (e.g. pH) of water underneath. Conversely, in dense weed-beds of submersed aquatic.s, local conditions of high 02 content. and pH often develop during daytime (e.g. MUSIL et al., 1976).

4.6. Metals and organic complexes

Local geochemical and man-made sources deter- mine appreciable inputs of heavy metals t.o some shallow African lakes, such as of CU to L. George (BUGENYI, 1979, 1982) and possibly several Kenyan waters (KOEMAN ef nl.. 1972 ; WANDIGA, 1981). Their significancr lies in possible biologiral toxicit-y, also conditioned and lcssened by c.omplexing behaviour. Thus, for L. Nakuru, K~LLQUIST and MEADOWS (1978) showed clear t,oxic effects of experiment,ally added CU on mtifers and the blue-green Spirulina fusifortnis (formerly identifled as &‘. platensis) in quantities > 100 pg l-1, much in excess of the pre- existing amounts.

Contents of thr t.wo most. abundant metals in shallow water-bodies, Fe and Mn, usually depend strongly on ot,her interna1 factors - their susceptibility t.o mobilization 011 chemical reduction and t,o stabiliza- tion as c.oloured organic complexes in dispersed and çolloidal phases. In L. Chad there is a significant flux of Fe from water column t.o sediments (LEMOALLE, 1973a, 1979 : CARMOUZE, 1983). In productive waters it is comrnon for sediments t.o be highly reducing and for the content. of ’ dissolved ’ (or colloidal) organic matter t.o be high. Uuring a wide survey of African lake waters (TALLING and TALLING, 1965), t.he higher concentrat.ions of total Fe (> 500 pg 1-l) and t,otal Mn (> 100 pg 1-I) were generally found in shallow productive lakes. Concentrat,ions of total Fe reached 5000 pg 1-l or more even in well-oxygenated surface waters of the Eastern Rift, lakes Baringo, Zwei, Abaya, and Langano, where a reddish-brown coloration could be seen in situ.

h relatively high cont.ent of ‘ dissolved’ organic matter in a fresh water is usually associated with yellow coloration, humic. and ful;ic acid fractions, and high spectrophotometric absorbante in the blue and ultra-violet regions. Such water cari be contribu- ted from swamps to adjacent- river or lake syst,ems, as scen in exarnples on the IJpper White Nile (TAL- LING, 1957a) and at L. Chilwa (Mass and Mass, 1969). It is especially prevalent, in water-bodies of

Ilcv. hydrobiul. trop. 2.5 (!2) : X7-I&I (1992).

AFRICAN SHALLOW LAKE.3 121

TARLE VI

Water volume Na+ K+ Ca2+ Mg2+ HCO-3 Si(OH)q (109 lx-?)

A. Hydrology (i) lake volume

(ii) component flux, yr1 river discharge

rainfall

72 (range 42.5-91)

41.5 (range 20.1-57.2)

6.3 (range 2.7-8.7)

lake evaporation seepage - out

B. Mean quantity of solutes in lake South Basin : concentration (mM)

stock (109 Mol)

44 3.8

0.355 0.110 2.205 0.15 1.15 0.595 8.95 2.82 5.22 3.86 29.2 15.0

North Basin : c&centration (mM) 2.05 0.605 0.85 0.70 5.65 0.88 stock (109 Mol) 96.2 28.3 39.8 32.8 263.0 41.1

Total lake stock (109 Mol) 105.2 31.1 45.0 36.6 292.2 56.1 C. Volute fluxes (109Mol) South Basin inputs : river discharge 5.60 2.05 4.35 3.30 22.4 16.1

outputs : seepage out 0.46 0.14 0.27 0.20’ 1.50 0.77 sedirnentations 0* 0.15 0.99 0.60 3.23 3.30

North Basin input : river discharge 5.14 1.76 3.09 2.50 17.6 9.03 (via South Basin)

outputs : seepage - out sedimentations

* ûssumed

5.15 1.51 2.13 1.75 14.05 2.19 0* 0.25 0.96 0.75 3.60 6.83

low ionic content, as in the Zaïre basin (BERG, 1961). Besides an ecological role in metal-binding (e.g. with Fe in L. Tumba, Zaïre : DUBOIS, 1959), and possibly in Ca-organic c.omplexes, dissolved organic material is an important determinant of light attenuance and henc.e photosynthetic activity in many waters. Lake Kilotes, Ethiopia, is one example (TALLING et al., 1973); Swartvlei, South Africa, is another (ALLAN- SON and HOWARD-WILLIAMS, 1984).

4.7. Chemical budgets

A variety of chemical interrelationships in African water-bodies have been discussed under the heading of ‘ chemical budgets’ or ’ nutrient budgets’ (e.g. VINER et al., 1981). Examples include an assessed component of a total flux (e.g. T& N fixation in N income : HORNE and VINER, 1971), the relationship between input. load and final concentration (THORN- TON and WALMSLEY, 1982), t,he partitioning of ‘ export ’ of nutrient elements (C, N, P) between out- slow and sediments (VINER, 1977c), the comparison of phytoplankton incorporation and nutrient-solute

depletion (PRO~SE and TALLING, 1958; DUFOUR, CREMOUX and SLEPOUKHA, 1981), and the use of concentrat.ion data to calculate changes per unit area. As THORNTON (1986) has point.ed out, there is very limited information available for African waters on the dynamics of nutrient transfer between compartments. This is especially true of interna1 nutrient loading and recycling. A true budget involves the quantit.ative comparison and equating of inputs and out,puts plus storage for a system. As applied to a range of inorganic chemical components, such is apparently available only for lakes Chad and Naivasha among African shallow lakes, although the relatively deep L. Turkana has received parallel treatment (YURETICH and CERLING, 1983). Partial information on the nitrogen and phosphorus budgets of Lake George is summarized by LIVINGSTONE and MELACK (1984). A detailed but l-component budget, exists for chloride in a shallow lake of Senegal, lac de Guiers. subject to marine incursions (COGELS and GAC, 1982; 1983). Likewise, a nitrogen budget- for the Rietvlei Dam, S. Africa (ASHTON, 1981) deserves mention, with increments in outflow above inflow influenced by N-fixation; also

R~U. hydrobiol. trop. 25 (2) : 87-M (1992)

J. F. TALLING

TARLE VII

Water volume, 106 IX? 1973 1974 1975 Na+ K+ Ca2+ M$+ HC03 SO24 Cl- F- Si02

A. Hydrology (i) lake volume (Dec’74) (ii) component flux, yr-1 surface runoff river discharge rainfall seepage - in total input

680

0.6 0.7 0.4 90.8 204.0 260.5 106.1 114.2 77.1 37.0 42.3 50.8 234.5 361.2 388.8

evapotranspiration 14.3 13.2 13.3 Lake evaporation 309.5 276.0 278.2 seepage - out 17.6 36.6 78.3 irrigation off-take 7.0 14.0 15.0 total output 348.4 339.8 384.8

change in storage B. Mean concentrations of solutes : min water Malewa River Gilgil River seenaee -in

-113.9 +21.4 +4.0 mg l-1

0.54 0.31 0.19 0.23 1.2 0.72 0.41 - - 9.0 4.3 8.0 3.0 70 6.2 4.3 0.4 17.2 16.1 7.4 4.4 2.2 75 9.6 3.9 0.8 18.2 44 31 23 10.2 227 11 28 1.7 47

1 v

seepage - out main lake C. Relative solute flux, as %

40 22 22 6.7 203 3 13 1.3 16 40 20 21 6.4 192 6.2 14 1.5 34

% total input or total output:

Inputs : rainfall 0.5 0.9 0.3 0.5 0 1.2 1.4 - - river discharge 11 7 12 21 25 11 10 5 14 seepage - in 17 30 13 26 24 7 24 8 12 sediment exchange 72 61 79 52 51 81 73 86 74

outputs : seepage - out 15 11 8 11 16 2 10 9 3 sediment exchange 85 89 92 89 84 98 90 91 97

another for t.he Ilartheespoort Dam (ASHTON, 1985). A mean annual solute budget for L. Chad (1954-

197”) is described by CARMOUZE (1983) ; it is summarized, with related hydrological quantities. in Table VI. Chloride and sulphate are not included. ~Mthough the estimated inputs and outputs balance, this is generally not an independent check on the budget validity as some components are estimated by difference. Sedimentation fluxes are c.alculated from ionic ratios and the assumption that the sedimentation of Na is negligible. Considerable between- year variation exists behind many of the mean values çit-rd (e.g. of solute concentrations).

The budget indicat,es how a lake of relatively low salinity cari exist without. surface out.let in a tropical rrgion wit.h high open-water evaporation. Although solute concentration in the river input (predominantly thr Chari H.) is low, t.his would net, prevent a progressive rise of salinity in the absence of losses nther than evaporation. The other important net losses are by seepage and sedimentation. The first is unseleçt.ive wit.h respect. to solute-component.s, but

occurs chiefly from t,he deeper northern basin whose input is already with ionic content increased and qualitat.ively modified from that of the main southern inflow river. Sedimentation is partly by biological agents, with CaCO3 deposition by molluscs, Si and K incorporat,ion by macrophyt,es (CARMOUZE et

2 1978), and Si incorporation by diatoms

EMOALLE, 1978; CARMOUZE, 1983) a11 quantitatively important. For exampIe, LÉVÈQUE (1972) has est.imated t.hat annual production by a rich mollus- cari benthos of 1970 would remove 7 x 105 t. of Ca, equal t,o four times the annual river input or half the dissolved stock in the lake. Dissolution and recycling of Ca are t-herefore important. Non-biological transfer to sediments occurs on a large scale by transformations of sediment minerals (= neoformation of clay smect.ites, i reverse weathering ‘) favoured by an alkaline medium rich in soluble silicate. The latter is consumed, t.ogether with quantities of Gaz+ and IV~$+. and some HC03- tranformed to COe. Finally, some precipitation of Calcit>e, CaCOz, occurs especially in the northern basin.

A solute budget for anot,her lake without surface outflow, L. Naivasha, has been estimated by GAU- DET and MELACK (1981). It. is summarized, with related hydrologic.al quantit,ies, in Table VII. Here also some components (water seepage-out, sediment exchange of solut.es) were estimated by difference; a11 major ions, plus F-, were included. The relatively low salinity of t.he lake is ascribed to the major dilute inputs of river wat.er and direct rainfall, an appreciable unselective loss by seepage (I-irrigation off-take), and a selective net, accumulation of solutes in sediments. Estimated by difference, in which constituent-errors may be compounded, the absolut,e magnitude of the last. and especially its resolution into (? overestimated) input and output components must be uncertain. There is good evidence for uptake and sedimentation of the silicon stock by diatoms, but appreciable effects from sedimentary neoformation (‘ reverse weat,hering ‘) were considered unlikely. The study was further notable for a direct use of seepage meters, and a Cl=based assessment of t.he proportion of cyclic sea-salt in the river solute input,, and hence by difference the proportion (- 30%) attributable t,o chemical denudation.

5. GENERATION OF SEDTMENTS

Interactions between water-mass and sediment.s, exemplified in the discussion of chemical budgets, are likely to be of particular import.ance in shallow water-bodies. Sediments c.an act as both source and sink of solutes and particulates, in bulk accretion as morphometric det.erminants, in permeability as hydrologie agents, and in int.erface properties and texture as biological determinants. Their generation - illustrated schematically in Fig. 34 - cari be t.ra- ced to t.hree types of processes.

OUTFLOW floating

t


5.1. Allochthonous transfer of particles

Such transfer from outside the water-body is predominantly as suspended sediment, in communic.a- ting river-systems. WALLEN (1984) has reviewed estimates of annual yield per unit area of catchment and its variation over Africa. A wide range of about. l-4000 t km-2 yr-i is likely, with only limited areas of > 100 t km-2 y’-*. There is also a small component of aerial dust. that includes pollen grains as potential indicators of biological history. Sediment types cari also be specific to dist-ant ongms, as in those contri- buted from the Et.hiopian highlands to sediments along the Blue Nile and Main Nile below (MCDOUGAL ei al., 1975; RztbK.4, 1976). This system also illus- trates how heavy loads are carried during a short seasonal period of flood-watrr and, in deposition, have provided the framework for the shallow De1t.a lakes and the recent massive sediment. accumulations in the reservoirs of southern Lake Nubia (ENTZ, 1976, 1978) and Sennar. The quantiCes of allochthonous sediment. dcposited per unit area in a lake typically decrease horizontally from t.he point of entry, conspicuously in examples of some fringing swamps (as at L. George : Case example), deltas (as of the Omo River in L. Turkana and of t.he Semliki River in L. Albert,), and between successive basins as in L. Chad.

5.2. Chemical precipitation

Abiogenic chemiçal precipit,at.ion may or may not be related to an evaporat-ive conrentrat,ion of saline waters. This process cari go to complet.ion in evaporites around the receding margins of shallow lakes, leaving conspicuous white deposits as of trona (Na&Oa. NaHCO:<. 3II&) or - less commonly -

INFLOW

allochthonous Particles

histbrical stratigraphy sediment neoformation. smectites

FIG. 34. - Diagrammatic representation of con1ponent.s of wdiment. eenwation. Schéma des divers processus contribuant b la sédimentafion.

Reu. hydrobiol. trop. 25 (2) : 87-144 (1992).

1% J. F. TALLING

halite (NaCI). Gaylussite has also been detec,t.ed in trona drposits of Kanem (MAGLIONE, 1968). In the case of Lake Magadi, Kenya, almost all the lake basin is occupird by thick deposit,s of trona which are exploited commerçially ; the chemical origin of thr preceding brines and the derivation of solid deposits are traccld by EUGSTER (1970), JONES et al. (1977) and MONNIN and SCHOTT (1984). In Lake Chad, the marginal sait deposits formed during rec.essiorr constituted a salt loss which had an appre- ciablr buffrring effect on the salinity of the residual lake water (CARMOUZE. 1983).

At a much rarlier st,age in such evaporat.ive concentration of a water with HCOz-/CO+?- predominant among anions, t.he divalent cations Ca’+ and Mg2+ are likelv to be depleted by precipit.at.ion as carbonat.rs. This cari be inferred ‘from correlat.ions between ionin composition and salinity (Fig. 25), demonstrated in mode1 experiments (GAc, 1980 ; set3 Fig. W), or t-racetl from t.he identification of the carbonates in rrcrntly deposit.ed lake sediments. Thus, in L. Chad; t.here is normallp no appreciable sedimentation of (&C:C);~ in the less saline southern basin, but in t-hr saline northern basin t.he sediments contain (1. 5-1O oo of calcite (CARMOUZE, 1983). In this lakr C~.4RMOLIzE (1976) believed t.hat precipita- tien began when t.he product. of t,he activities of free C:a-+ and fier COS’-, {Ca’+}. (CO$). exceeded 18 fimes t he soluhi1it.y product of c.alcite. Elsewhere, iIldepeIIdeJltly of salinization, the photosynthet.ic elevation of pH cari cause a periodic precipitation of C:aCX&, as on the leaves or stems of aquatic macrophyt es.

One form of ahiogenic sedimentary accretion is rrpresented by the neoformation of smect.ites with chemical transfer from solutes. Such ’ reverse weathering ’ is probably widespread in alkaline, silicate- rich waters. It- is a significant component of the chemical budget of L. Chad (see Section 4.7).

5.3. Biological deposition

The biologically derived components of lake sedi- mr~nts oft-cri predominate in t.he more dilute waters distant from intlows (e.g. offshore L. Victoria). Bath organic matter and inorganic skeletal c0nstituent.s (*an bc involved. ïrnport.ant among the lat.t.er are the calcareous shells of molluscs and the silic.eous frustules of diaf.oms. In some waters the faecal pellets of rnic~ro-~rustacea are a significant fract,ion (e.g. L. Tanganyika : HABERYAN, 1985).

The chemical budget of L. Chad (Section 4.7) has alrrady illustrated a large annual incorporation of CaW:~ by an abundant. molluscan fauna on the sediments. It also indicated dissolution and re-cycling of this coniponent. from sediment to water column.

Rw. hgdrobiol. trop. 2.5 (2) : 87-l& (199.2).

Such quantitative est,imates do net seem to exist for other African lakes with a ric.h molluscan zooben- thos, which include the deep L. Albert and the shallow L. Tana. However ident.ifiable remains of molluscan shells are widespread in Plfric.an lake sediments and some (e.g. in the Kaiso beds of L. Albert-L. Edward) are important indic,ators.

The bio-sediment,ation of silica as diatom frustules has also few quant.itative chemical est.imat.es. Never- theless massive or prolonged fluxes are indicated by extensive deposits of diat.omite relat,ed to various shallow lakes, past or present., in Africa, and most lake sedimant,s bave detectable - if less massive - diatom remains. Diatom records from past sediments have been used widely in Africa to reconstruct lake history, as for the shallow waters of the Faiyum Lake (ALEEM, 1959), L. Naivasha (RICHARDSON and RICHARDSON, IC)?I), L. George (HAWORTH, 1977), Pilkington Bay of L. Victoria (KENDALL, 1969), L. Chad (SERVANT-VILDARY, 19%; SERVANT and SERVANT, 19X3), L. Manyara (HOLDSHIP, 1976; RICHARDSON et ul., 1978), and lakes Abhe, Asal, Afrera, and Gamari in the Afar-Danakil region (C;ASSE, 197-I-a-b, 1375, 1977; GASSE and DELIBRIAS, 1976).

Organic deposition invariably accompanies bio- sedimentation. Thus VINER (1977c) estimated a current deposition rate of 15-20 g organic matter m-2 yr-1 in the productive Lake George. The relative content of organic matter in the final sediment varies widely (MCLACHLAN, 1974), being low in allochthonous silt deposit,s and high in peat.s. Rela- ted to the lat,ter, but. distinctive in location, is the deposit of ’ sludge ’ whic.h develops below floating mats of papyrus (GAUDET, 1976). Recent algal sedimentation, as of blue-greens in L. George (GANF and VINEH, 1973; GANF, 1974b), cari generate very high organic contents in superficial sediments which are partly composed of living c.ells. Rapid decomposition cari occur (VINER, 1975b) but, is not inevitable even at high tropical temperatures, as illustrated by observations (REALJCHAMP, 1958) and experiments (HESSE, 1958) on the highly organic near-surface sediment,s of shallow inshore bays of L. Victoria. The possibility that. organic residues may be sources of petroleum from African lakes is current.ly under investigation.

6. TIME-RELATE» SYSTEhI CHANGES

Shallow waters are subject to c.yclic patterns of change wit,h time, on various time-scales, in which many factors vary concurrently. Three examples are described below, that vary in period from 24 h to


1 year or longer, and are set up by different evoking factors.

6.1. Diel (24 h) cycle

This type of cycle has been introduced by the case-example of L. George (Section 2, Figs 5, 6). The dominant, evoking factor is the diel flux of short- wave solar-radiation. As described already (Section 3.1), its daily amplit,ude is potentially higher in the trop& than at higher latit.udes, although subjec.t to modification by atmospheric t.urbidity. In combination with other less variable camponents of energy balance (including long-wave bac.k-radiat.ion, favoured by clear night skies), it induces with some lag a corresponding cycle of energy slorage, viz. tempera- t.ure. Examples for two shallow water-bodies (Lake Chad, Jebel Auliya reservoir) are described by TAL- LING (1990), with resolution of component flux densi- tics. There may be indirect effects upon wind- regime, although the coupling is very variable. Thus there is a diel component, of on- and off-shore breezes in t.he shallow marginal areas of L. Victoria (Fish, 1957; Evans, 1961), and in other examples wind stress was greatest at night (Jebel Auliya reservoir : TALLING, 1957b), morning (L. Chad : CARMOUZE et al., 1983), or 1at.e in the afternoon (L. George : VINER and GITH, 1973).

The diel temperature cycle is susceptible to day- to-day and seasonal changes of weat.her, as illustrated in Fig. 35 for a Kanem soda lake and reservoir at Nairobi (latitude 1020’ S), and by HARE and CARTER (1984, Fig. 6) for a small lake m Nigeria (lat,itude 6045‘ N). Interactions between the t,emperature cycle, wind regime, and wat,er depth further deter- mine whet.her a diel pattern of density stratification will develop. Areas of plant caver which obstruct water movement also tend to promote stratification. It may be absent in very shallow waters (e.g. a Nile lagoon : TALLING, 1957b) or with persistent wind st.ress or low insolat;ion. Variations with weather have been described by TALLING (1957b) on a day-to- day basis for a bay of L. Victoria and on a seasonal basis for the Jebel Auliya reservoir. In the latter, more pronounced diel stratification during a phase of high water temperat.ure (- 30 “C) is probably influenced by the then greater rate of change of density with temperature. The same non-linear relationship between density and temperature Will tend to generally accent,uate diel cycles of stratification in warm tropical waters. In waters deeper than a few metres, it. is net. uncommon for a superficial stratum with a pronounced diel stratification to overlie deeper wat,er with more persistent st.ratification. L. Sonachi, a crater lake in Kenya (MELACK, 1981) (Fig. 36), L. Aranguadi, one in Ethiopia (TALLING et

Heu. hgdrobiol. trop. 2.5 (2) : 87-144 (1992).

(a)

40

Time (hl

c (b)

FIG. 35. - Seasonal shifts in diel temperature cycles measured (a) by continuous recordinn at. 0.6 m dept.h in a reservoir (!vell- çome Dam) near Nairobi.‘in or near the last. t,wo days of the months indicated during 1971-2 (* = short. rains, ** = long rains) (b) by separate surfa<:e det.erminations during three mont.hs on a Kanem lake in Char1 (mare de Lat.ir). i\dapt,ed

from YOUNG (1975) and TLTI~ (1969). Évolution des cercles diurnes de la fempérnture de l’eau au cours de l’année a) pour les différents mois de norrembre 1971 à nooembre 1.972 dans un lac de barrage, Wellcome Dam prks de Nairobi, à 0,6 m sous la surface; enregisfrement continu b) pour 3 mois dif- fërents dans la mare de Latir, dans le Kanem au Tchad. U’aprés

I*O”rVG (197.5) et ILTIS (1.969).

al., 1973), and the Pretoria Salt Pan (ASHTON and SCHOEMAN, 1983, 1988) (Figs 22, 38), are well-studied examples. In a11 these, t,he diel strat.ification patt,erns are probably enhanced by wind-shelter from the crater rims - a feature yuantified by MELACK (1978).

126 .J. F. TALLING

o-

1 -

2 -

E

.c 3- E

c1

4 -

5 -

Temperature ( OC)

.sz24. . .b

. . . . . . . . . . . . . .

. . . . . . . . . . . . . . -2’ - . . . . . . . . . . . . . .

"l-k,:: ,*- ,!.: .., .I-:,,:.! ,,,, L#.

06.00 12.00 18.00 24.00 06.00 12.00 18.00 24.00 06.00 12.00 18.00

Time (h)

FIG. 36. - Lake Sonachi, Kenya : t.he dept.h-time dist.ribution of t,emperat.ure during 5-7 hlarch 1973, . rhowing t.he regular genera- t.ion of diel st.rat.itîcation cycles. From bfELACK (1981).

1-p lac $onachi, Krnycz : diagramme profondeur - temps des tempérutzzres du 5 au 7 murs 1973, montrunt l’établissemenf régulier des sfraiifications journalières. D’après &fElACK, 1.973.

WEIR (I%I~) has describrd how a daily stratification, constituenta t.hat. are in rapid change. Examples with xurfacc temperatures > 35 C. c.an develop in include dissolved oxygen (Figs. 6, 38), carhon vrry shallow pans unless dist.urbed by visiting herds dioxide, and pH in productive wat,ers (Sect#ions 2, of ungulates (Fig. 37). 4.3); and the population der1sit.y of positively or

C)nce estahlished, a diel densit.y st.ratification is negatjvely buoyant-, or migratory, plankters (Sec- likrly to constrain the distribut-ion pat.terns of other tien 2, Fig. 5~). Diel cycles of environmental origin

45

40

ô 0

- 35

L 5)

m 30 L

E ? 25

12.00 14.00 16.00

Time (h)

FIG. 37. - Temperature values measured at. various times during January in pools of t.he Wankie Nat,ional Park, Zimbabwe, showing diurnnl divergence between surface wat,er (0) and bottom mud (0) in one pool dnring werm cloudless days, of carresponding values on overcast rainy days (A 1 A) and of corresponding values on warm cloudless days- in a pool st.irred by herds of visiting

ungulates (0, W). From WEIR (1969). Temptktures mesrzrées à différents moments de la journée, dans différentes mares du Parc Nationa; \~&zkie, Zimbabwe; les mesures indiqrzent la dizrergezzce entre la surface (0), le sédiment superficiel (0) dans une mare pendant des jozzrs ensoleillés, et pendant des jours plzzoiezzr ozz nzzngeur (A, A), et par jour clair ef chaud mais azrec perturbation par des troupeaux d’antilopes (0, n ). D?après WEIR,

1969.


0

10

40

2400 1200 2400

Time (h)

1200 2400

FIG. 38. - Pretoria Sait Pan : the depth-t.ime distribution of dissolved oxygen indicated hy isopleths in mg 1-I during 28-29 Decem- her 1979, based on hourly sampling at the depths indicakd. From ASHTON and PCHOEMAN (1983).

Pretoria Salt Pan : diagramme profondeur - temps de I’oxygÈne dissous (en mg. 1-l) les B-29 dkembre 1979. Les mesures ont été faites chaque heure, aux profondeurs indiquées. D’après ASHTON et PCHOE.WW, 1983.

may also interact with biologic,al ones that have a marked endogenous control. This is exemplified (Fig. 39) by the diel variability of ingestion, digestion and excretion by a c.opepod (Thermocyclops hyalinus) and two fishes (Sarotherodon (Tilapia) niloti- tus, Haplochromis nigripinnis) in L. George (MORIARTY et al., 1973; GANF and BLA~KA, 1974).

The destruction of diel stratified structure by vertical mixing may be induced by strong winds at any time, but. most often occ,urs at. night when the wat.er column has a negative energy balance (cf. Table II, L. Tana). This nocturnal mixing pattern may be often reinforced by the diel wind regime in some waters (e.g. Jebel Auliya reservoir : TALLING, 1957b) but not in others (e.g. L. Chad : ROBINSON, 1968). Especially near dawn, surface heat loss may lead to an unstable inverse stratification with near-surface water cooled below that at depth (e.g. Fig. 5(i)d). The resulting convec.tion may cause significant redistri- butions, as of dissolved oxygen, in waters of both lakes (L. Naivasha : BEADLE, 1932b; L. Nyumba ya Mungu : DENNY et al., 1975) and swamps (BEADLE, 1932a).

In a small number of examples, diel cycles have been analysed in terms of the changing content of oxygen below unit area of surface in relation t,o pri-

Reo. hydrobiol. trop. 25 (2) : 87-M (1992).

mary produc,tion (TALLIN~-, 1957b; TALLING et al., 1973; MELACK and KILHAM, 1974; GANF, 1975).

6.2. Seasonal cycle with small volume change

Cycles of this kind are familiar and much-studied among lakes generally. For Afric.an shallow lakes and wetlands their freyuency is reduced by t,he common intermittency of ramfall combined wit.h basin t.opography and limit.ed storage capacity. Known sit.e-examples are associated with situations of seasonally ext.ended wat.er input and/or large size relative to the input,-output. fluxes. Thus Lake George (sec Case stndy, Section 2) has a regime of ext.ended and seasonally bimodal rainfall, permanent inflow streams from adjacent mount,ains, and a peculiar outflow--buffer channel. The last feature re- appears in anot.her form in many brackish coast.al lagoons with some connrction to the sea (e.g. Ebrié lagoon), through which outflow may reverse to inflow. A completely freshwater analogue would be a shallow and semi-enclosed gulf of a large lake, such as t.he Nyanza (Winam, Kavironrloj Gulf of L. Victo- ria. Small lakes in a morf: seasonal climate, such as the Bishoftu crater lakrs of Et.hiopia, may have

138 J. F. TALLING

1.2 Solar radiation

m

5

2 4 6 8 lo 12 14 18 20 22 24

TIME (h)

FI~;. 39. - Comparative examples of diel rycles at L. George. including solar radiation, fixation of C and N wit.h evolut.ion of 0.2 hy phyt.oplankt.on, N and P exc.rrtion by zooplankton, and ingestion of food by a zooplankter (Thermocyclops hyafinus, as C: per individuai) and t.wo species of fish (Sarufherodon niloficzzs, Haplochromis nigripinnis, as st.omach weight). From BURGIS

(1978). Errmples dc cycles nycfhémPraux au lac George : rayonnement incident, cwsimilafiorz de C et Ns et production de Os par le phyfo- plancfon, escréfion de N rf P par le zooplancfon, ingestion de nozzrrifure par fi nzicrocrzzsfacé Thermocyclops hyalinus, expri- mée en C par indizridzz) et par deus espèces de poisson (Sarothe- rodon niloticus, Haplochromis nigripinnis, exprimé en poids

d’esfomac). »‘après BURG~S, 1g78.

maintainA seepqe-input. from a relatively stable water-table. Large shallow lakes of long ret.ention and fairly stable level include L. Tana and L. Abaya (= Margherita) in Ethiopia, L. Bangweulu and L. Mweru (DE E(rMPE, 1964) in Zambia-Zaire, and L. Baringo in Kenya.

In t.he deeper Afric.an lakes, numerous conditions in the water-mass are normally associat.ed in an annual periodicity t.hat. is linked to a seasonal stratification cycle, itself conditioned by varying energy exchanges including the wind regime. Such cycles are widespread even near the equator (TALLING, 1969). The common denominator of seasonal persistent st.ratification is characteristically lacking from shallow lakes. Its place does not seem to be taken by any other general ‘ master-system ’ of seasonal variability ; perhaps as a consequence, in a hydrologically stable equatorial lake like L. George, there is a lack of pronounced seasonal changes in the physical and chemical environment. However, in extra-t.ropical Africa the increased amplitude of sea- eonal radiation and t,emperature changes possibly dominat‘e, or - cspecially in water-bodies of small ret,ent.ion time - t-he varying quant.ity and quality (solutes, silt c.ont,ent) of inflows. In c.ontrast to L. George, there are few extended seasonal studies of

(b) o

1

E 5 2 P 0

3

1979 1980

FIG. 40. - Opi Lake, Nigeria : depth-time distribution over 13 months of (a) t,ernperature and (b) dissolved oxygen, showing changes of temperature stratification and associat.ed oxygen depletion in deeper water. From HARE and CARTER (1984). Lac Opi, Nigeria : diagrammes profondeur-femps durant 13 mois de la iempérafure (a) ef de l’oxygène dissozzs (b) montrant la strafificafion thermique et l’éoolufion associée de l’oxygène dans

l’hypolimnion. D’après HARE et CARTER, 1984.

Fkcr. hyfrobiol. trop. 2~5 (2) : X7-14.$ (1X92).

AFRICAN SHALLoW LAKES

(a)

129

- 160 4 (b) - 120 5

- 60 -2 5

ô 33

0 31

2 a

29

L x

27

E 25 fi 23 k 4 21

19

4 4

(dl

0” I I

1978 1979 1980

FIG. 41. - Opi Lake, Nigeria : seasonal variation, over 2 years, of (a) rainfall (b) daily wind run and water depth (c) air temperature (d) water temperature measured at the surface (S) and botkom (B) at. OS.00 and 17.00 h (e) ox;ygen content in surface (S) and bottom

(B) water. Arrows indicat.e onset of t.he harmattan wind regime. Adapt.ed from HARE and CARTER (1984). Lac Opi, Nigeria : évolution, au cours de deux années, de la pluvioméfrie (a), du vent total journalier ef de la profondeur (b), des fempératures de l’air ef de l’eau en surface (S) et au fond (B) à 8 h 00 et 17 h 00 (d) et. en bas, de l’oxygène en surface et au fond (e). Les

Pèches indiquant l’établissemenf de la période d’harmatfan. Modifié de HARE et C.ARTER, 1984.

such water-bodies in Africa. That of HOWARD-WIL- LIAMS and ALLANSON (1981) on the coastal lake Swartvlei (S. Afric.a) is not.able for the impact of a seasonal cycle of macrophyte (Potamogeion pectima- tus) growth and decay on phosphate depletion and regeneration in the lake (Fig. 30). The important class of shallow coastal lakes is also exemplified by the Delta lakes of Egypt (see RZ~SKA, 1976) and of the Ebrié lagoon in West Africa (e.g. DURAND and CHANTRAINE, 1982; DUFOUR, 1982a), a11 of variable salinity. Observations of BOWMAKER (1962) on a lagoon at L. Bangweulu also deserve mention; here

too variable hydrological inputs dominate seasonal change.

Aspects of met.eorological control, and some unu- sua1 fac,t.or-c.ombinations, are illustrat-ed in the seaso- na1 cycle of Opi Lake A (HARE and CARTER, 1984), a very small (area 1.4-2.0 ha) and shallow (max. depth 2.4-3.95 m) lake in the Nigerian savanna at latit,ude 6045’ N. Here t-he seasonal range of temperat,ure is little greater than t.he typiral diel range (3-4 OC) in surfac,e water. In spite ot this diel variability and t.he shallow depth, vertical mixing is apparently incomplete in the rainy season and anoxia then

Reo. hydrobiol. trop. 25 (2) : U-144 (1992).

130 J. F. TALLING

drvelops in neer-hot.tom wat-er (Fig. 40). Overall, eeasonal change is mainly evoked not by change in solar radiat,ion but, by the alternation of humid mari- time ant-l dry continental (Saharan) air masses, the latter marked by t.he nort.herly Harmattan wind regimr. Onset of t-he Harmattan in Nov.-Dec. (Fig. 41) induf*f>s a steep fa11 of water temperature, and enhenced mixing which ends anoxia, in part by t.he increase in evaporative heat 10s~. Thus the lake water C:C& at, a period of high solar insolation and da.ily-maximum air temperature, but lowered daily- IZIIIIII~UI~ air temperature. The alt.ernation of wet, and dry seasons is associated wit.h a 2 ‘/z-fold range of water volume. a discontinuous surface outflow and marked changes in depth and area.

ti 3. Flooding-drying cycle

In a shallow hasin of gently shelving morphome- t.ry, large changes of water volume due to hydrological fact,ors lead to corresponding large horizont.al excursions of water level as well as changes in mean dept.11. Such excursions may take numerous forms.

A water-body of simple outline may undergo predominantly radial extensions and contractions, whe- t.her it is a rain-pool fed by intermit.tent precipitation or a deeper river-fed lake such as L. Nakuru. L. Chilwa is a more c.omplex case, with markedly asymmetric (N-S) distribution of river inflow and shallows plus swamp whic.h led to a ‘ freshwater ring ’ of water during refilling after drought (HOWARD- WILLIAMS and LENTON, 1975; MCLACHLAN, 1979). Still more c.omplex is L. Chad, with two main basins, the drrper at great.er distance from the main inflow. This fact, and t.he development at low level of aquatic vegetation as obst.acle to flow between basins, led

? 2 282 L

J

BOL

E 280 n

to isolation of the deeper northern basin during the low levels of 1973 and final drying out in a period (1975-6) when wat,er was being replenished in t,he more accessible basin (CAR~ZOUZE, DURAND and LÉVI~QUE, 1983; sec Fig. 42). In a third region, of a north-east dune-archipelago, isolated depression pools developed and dried out during low lake level. Yet another, more linear pattern is represented by river floodplains (WELCOWIE, 1979), annually cove- red by overspill of water from the river channel. Such water cari replenish previously isolated or dried-out. lagoons or pans, as on the ‘ interna1 delta ’ of the Middle Niger (BLANC et al., 1955) and the Pon- golo floodplain in northern Natal (HEEG and BREEN, 1982; sec Fig. 44) or augment others in permanent connection to the river (e.g. Shambe lagoon, Sudd swamps, Sudan : RZOSKA, 1974). Such floodplains merge with the littoral regions of elongate reservoirs or ’ river-lakes ’ subject to seasonal drawdown.

The flooding-drying cycle, as expressed by water level change with time, has a common asymmetry. Wat.er influx with rising level is typically more abrupt than the later subiidence conditioned by evaporation and reduced run-off. An example from the floodplain of the Senegal River is illustrated by THOM~SON (1985, Fig. 3.3). The periodicity may be seasonal-annual, long-t,erm and inter-annual, or a combination of both as illustrated in the records from lakes Chad and Chilwa (Figs. 42, 43). These lakes also demonstrate that the seasonal amplitude of environmental oondit.ions tends to be great.er when the seasonal cycle occurs at- a low-level and low-capacity st.age of a long-t,erm oscillation (L. Chad : CARMOUZE et al., 1983; L. Chilwa : KALK et al., 1979) or after a particularly dry summer (South African highveld pan : ROGERS et al., 1989).

Water charact,eristics which tend to be especially

3 g m ül h

278 -

m Malamfatori

E, 276 ,,,I, ,,I,,,,,,,, ,I,,,.,,,,, ,,,,,,I,,I, ,.,,, ,,,. ,,,,I., ,,, ,,,,,,,,,,,

1971 1972 1973 1974 1975 1976 1977 Water New drying up of Water retum VA”r” the northern basin 09.01.77

FIG. 12. - Tim+variat.ion in water levels of L. Chad during a drought. phase, 1971-7, showing divergence between those in the strutheast.ern archipelago at Bol and t,he northern basin at Malamfatori. Redrawn from CARMOUZE and LEMOALLE (1983).

Évolution du niveau du lac Tchad au cours de l’installation d’une période de sécheresse en 1971-77. Divers bassins s’individualisent : l’archipel du sud-est 0 Bol, et la cuvette nord du lac, suivie à Malamfatori. Redessink d’après CARMOLIZE ef LEMOALLE, 1983.

AFRICAIN SHALLOW LAKES 131

10 000 25

ccl 3 6000 2.0

x E .5 r s 6000 1.5 5

z 4z

8 4000 1.0 $

2 2000 a5

0 0

1965 1966 1967 1966 1969 1970 1971 1972

FIG. 43. - Lake Chilwa : the relat,ionship between changes in seasonal maxima and minima of wat.er-depth at. one st.ation before, during, and after a drought phase (0) and those of ionic concentration reflected by conductivity (at 20 OC) (0). From

MCLACHLAN (1979). Lac Chihva : relation entre maximums et minimums de profondeur (s) et (s) de conductivitG (à 20 OC) avant, pendant ef après

une phase de sécheresse. D’après MCLACHLAN, 1979.

influenred by the lowr-level stages of a flooding- drying cycle include t.emperat.ure, turbidity, water circulation, salinit,y, and dissolved oxygen aontent-. Thus, in lake Chilwa, the last phase of open water before desicçation was marked by elevated daytime temperature, high turbidity due to dense algal growth, redxed horizontal circulation of water, rai- srd salinity and phosphate concentrat.ion, and oxygen depletron at a short distance below the surface. Most. of these conditions occurred in the low level 1973 stage of L. Chad, although here macrophyte growth rather than algal growth often predomina- ted. For 1972, BENECH ut al. (1976) describe how a t.ornado at. Bol disturbed sediment and displaçed water-masses ; concentraCons of oxygen fell below 5 ‘$& saturation, and a heavy Ash-kil1 resulted.

A cycle of ionic c»nt.ent nornially accompanies a flooding-drying cycle. This is illust,rat.ed Fig. 43 for a station in L. Chilwa, where water level and conductivity are inversely relat.rtl (MCLACHLAN et al., 1972; MICLACHLAN. 1979) . Parallrl data exist for L. Nakuru (VAHEBCHI, lDY?), 1,. Chad (CARXPOUZE, (:HANTRATNE ami LEMOALLE, 1983), and the POngOlO pans (HEEG and BREEN, 1982; sec: Fig. 44). Never-

Flow > 120 m3 s-1

HU ----l Datum

1600

at Makhani’s Bridge

1975 I 1976

FIG. 44. - Mhlolo Pan, Pongolo floodplain : conductivity changes, indicative of concent.rat.ion and flushing, in relation to pan water-level and river flow and conductivity. Adapted from ITEEG and BREEN (1982).

Mhlolo Pan, dans la plaine d’inondation du Pongolo: les variations de conductivité indiqueni les périodes de dilution et de concentration, en relation avec le niveau du lac, la conductivité et le débit du fleuve. D’après HEEG et BREEN, 1982.

Rev. hydrobiol. trop. 25 (2) : 87-144 (1992).

.T. F. TALLING

9.0

6.0

I I I

0 2 4 6

Alkalinity (meq 1-l)

FIG. 15. - Srasonal succession (arrowed) of combinations of pH and alkalinity in 6 shallow pools (pans) of the Wankie National Park, Zimbabwe. Mrasnrements made in the wet season from November (N), December (D), January (J) to April are connected by

t.hick arrows, and those from April, September, to November by thin arrows. Adapted from WEIR (1968). Évolution. au cours des saisons, de la relation alcalinitt’ - pH dans 6 mares peu profondes (pans) du Pare Nafional W’ankie au Zimbabwe. Les mesures en saison des pluies de novembre, décembre ef janvier à avril sont reliées par des Pèches épaisses; les mesures

d’avril, septembre à novembre sont reliées par des /lèches fines. AdaptE de WEIR, 1968.

thrless t herr art: bath spatial and ion-specific compli- cations. In L. Chilwa the renewal of river and swamp inflow mag generate a ring of fresher water that is at first distinct. from more central remnant saline Lvatrr. In t.he low-level L. Chad of 1973-5, particularly high salinit,y developed in the isolated northern basin, from which later influx of low-sa1init.y watrr from the main southern inflow, the Cha;i Hiver, waa obstructed by exposed sediment and rnacrophytP barrirrs. At. another extreme of size, t.he salinity, alka1init.y and conductivity of rainpools t.fhnd to increase during prolonged isolation and contraction (RZkKA. 1961, 1984), as do those of pans studietl in t,he Pongolo floodplain (HEEG and GREEN, 1982) and in Zimbabwe (WEIR, 1968; see Fig. 45). Thr latt,er show an element of ‘ chemical hysteresis ’ in thrb cycle, with pH values usually rela- tivelp deprrssçd during the expansion and re-dilution phase, prrsumably by Cc32 accumulation after decomposit.ic:~Ii.

However, thr sensitivity of salinity to lake volume mag be offset or ’ buffered ’ by several factors, including direct on-lake rainfall, the separation of margi-

Hw. hydrobiol. frop. 2.~ (2) : X7-1& (1992).

na1 salt dep0sit.s outside the main water mass and their aerial deflation by wind, their re-cont,act to t,he water-mass at timrs of advancing level, and geoche- mica1 sedimentary transformations or ‘ reverse weathering ‘. Thesr processes are documented for L. Chad in CARMOUZE, DIJRAND and LÉVBQUE (1983).

The seasonal or longer-terni salinity cycle cari involve disproportionate changes in some consti- t,uents ; some may be represented in the sedimentary histor? (e.g. carbonate-rich horizons). One mecha- nism IS t.he precipit,at,ion of divalent cations, Ca”+ and Mg’~L+, from waler of increasing alkalinity and pH, a procrss demonstrat.ed from syst,ematic obser- vation and experiment. on L. Chad (CHANTRAINE, 1978 ; GAC, 19%); CARMOUZE, 1983 ; see Fig. 26) and L. Guiers (&GELS and GAC, 1983). Another is the uptake and subsequent. release of nutrient. elements (N, P, K) during associat.ed growth cycles of plants, especially macrophytes, as in L. Chilwa (HOWARD- WILLIAMS and LENTON, 1975) and the Pongolo floodplain (HEEG and BREEN, 1952; ROGERS and BREEN, 19%). The areal stocks of these nutrients in dense


16 19~;

-2 gm

19

10

P

0:

FIG. 46. - Stocks per unit area of t.he elements K, N and P in dense stands of reedswamp, submerged macrophytes and phytoplankton of varied dry weight (D. W.). Based, respectively, on data in GAUDET (1977), HOWARD-WILLIAMS and LENTON (1975). HOWARD- WILLIAMS and JUNK

J 1977), HOWARD-WILLIAMS (1977), ROGERS and BREEN (1980), and VINER (1977c) with BURGIS et al. (1973). alues from L. Chilwa, Swartvlei and the Pongolo Pan are at or near seasonal maxima.

Stocks de K, N et P ef de poids sec par unité de surface dans des peuplements denses de roseaux, de maerophgtes immergés ou de phytoplancton. Les données utilisées sont, respectivement, de GAUDET (1977), HOWARD-~~~~~~~~~~~ et LENTON (1975), HOWARD-WIL- LIAMS ei JLINK (1977), H~~ARD- WILLIAMS (1977), ROGERS et BREEN (1980), et VINER (1977c) avec BURGIS et- al. (1973). Les valeurs

pour les lacs Chilma, Swartvlei et Pongolo Pan sont proches ou égales au maximum saisonnier.

macrophyte st,ands are high and typically exceed those in most phytoplankton communities ; comparative examples appear in, Fig. 46. A nitrate-flush cari follow t,he re-wett,ing of soi1 and sediment, as noted for a Typha swamp at L. Chilwa (HOWARD- WILLIAMS, 1972, 1979), and re-flooding cari induce nutrient release from grasses and the dung of large herbivores (S. M. MCLACHLAN, 1971). During such initial re-tlooding, the mass-release of available solutes to a limit.ed volume of water often produces a transient peak of tot,al ionic concent.ration, as at L. Chilwa (MCLACHLAN, 1979) and.elsewhere (McLA- CHLAN, 1974).

Biologically, the expansion phase of the cycle cari be viewed as aquatic recovery followed by terrestrial stress (e.g. on the grass Cynodon dactylon : FURNESS and BREEN, 1982, 1986), and the contraction phase

as terrestrial recovery followed by aquatic stress. The c.omponents of aquatic stress may include supra- optimal salinity (e.g. for the macr0phyt.e Potamoge- forz crispns in t.he Pongolo pans : HEEG and BREEN, 1982; mortality of a cichlid fish in L. Chilwa : MOR- GAN, 1972, 1979) and adversely limiting light attenuation (e.g. for algae in L. Chilwa : Moss and MOSS, 1969) as wrll as the ultimate deficiency of water itself.

7. SYNOPSIS : DISTINCTIVE ENVIRONMENTAL CONTROLS IN SHALLOW AFRICAN WATER-BODIES

Drawing on t.he previous extended ac.count, it may be useful to single out distinctive features of environmental control in shallow African wat.er-bodies.

Reu, hydrobiol. trop. 26 (2) : X7-144 (1992).

134 J. F. TALLING

i. Storasr rapacit.y in the water-column prr unit area is hmited. 111 consrquenc7~ an area-related input or loss I~I~- more readily induce large relative changr~s of concentration. Examplrs are changes of heat Lntent (temperature) in relation to solar input ctir evaporation loss ; of oxygen cent ent , including complete anoxia, in relation to photosynthetic gain or respiratory ronsumpt.ion ; and of water content. ii. Thr shallow water-c*olumn is relatively susceptible to wirlc-l-inducecl turbulence and convective mising. It. t.ht~rrforr usually lacks any persistent t.rniprratllre-c.ler~sit‘; stratifirat-ion, except. in some rstreme cases of salinity-layrring. However, a diel trrrlprraturr-tleI1sit.y stratification normally develops in relat.ion to t hr diurnal input of short.-wave solar radiation. It is often associated with »t.her interacting dirl c*yc*les (r.g. of 02, COQ, phytoplankton rc,distril,iitiun). iii. Interactions involving underwat.er sediment,s are often enhanced. Thus superficial sediments are oft;en disturbed EiIld resuspended by wind-induced carrent-3 ; thry may be sufficiently well-illuminated as to support brnthic growth of algae and macrophytrs. Their ratio to thr water-mass, physico-che- mica1 reactivity, and hydrodynamic accessibility ma y lead them tu dorninat-6% chemical rec.ycling - unless a clrnsr end buoyant. plankton transfers the prrponderancc of such prof,essing t.o the water- c\r)lumn. The form of thr water-b&n is relativrly sensitive to srdiment generation and ac.cumulation. iv. Shallow water-bodies provide the conditions under which most saline lakes have developed. These include a favourable rat-io of cumulative evaporat,ion to lakr volu~~w and the possible location in a spring- fcd aupplp of rec~ycled salinp reserves. Where such reserves are ahsent, and srepage-out and possibly ’ reverse wrat hering ’ of sediments appreciable, a shallow water of 0111-y moderate salinity is cnmpa- tible with the absence of surface out.flow (lakes Chad, Naivasha). Y. The development ia favourrd of high areal stocks

of biomass (and incorporated nutrirnt, elements) in swarnp macrophytes and of higher biomass concentrations pclr unit, volume. in phytoplankton. The former plants are environmentally important for local Spat*ial gradients, shelt.er. niche-diversification, and nutrient pat.hways - inc.luding net upward transfer as a ‘nutrient, purnp ‘. The latter often produce, by photosynt het.ic activity, large diel fluctuations in concentrations and layering of 02 and COT, with elevat,ed pH. However such -ac.t.ivity is count.ered, as regards areal product.ivit.y, by increased light at.te- nuation due to phytoplankton and bac.kground pig- ments. Thus rnany product.ive shallow lakes are optically deep. vi. There is increased 1iahilit.y in African climat.es, wit,h periodic rainfall and high insolation, to large horizontal excursions of water level in a flooding- drying cycle. These bave numerous correlates in a tirne-sequence of physical. chemical, and biological conditions.

1 am indebtr,d to other memhers of t,hP Working Group on African Shallow Lakes and Wetlands. and especially Christ.ian I.tivfiyus, .Iacques LEMMOAL.IX and Francoise Gosse, for much valuahle information and criticism: to Trevor FURNASS for assistance with test figures; and to Julie WATERHOUSE for typing t.he manuscript.

For permission t.o reproduce diagrams from trt,her publications 1 um grateful t.« : The Hoyal Society (Fig. 3, from Proc. R. Soc. (R) 184, p. 637). E. Schweizerbart’sche Verlagsbuch- handlung (Figs 33 and 17, from Arch. Hydrohiol. 77, p. 78 and 92, p. 511 ; Figs 2 and 20, f rom Arch. Hydrobiol. Beih. Ergehn. Limnol. 33, p. 65a and p. 655; Figs 18 and 39, from Verh. int. l’erein. theor. angern. Limnol. 17, p. 1000 and 20, p. 1147). Hlackwrll S&ntific Publications (Fig. 12, from J. appl. Ecof. 9, p. 749; Fig. 40, from Freshwafer Biol. 14. p. 605). and Klu- wer Acadrnlic Publishws (Figs 37, 10, 36 ond 22 ; from Hydro- biologia 33, p. 95; 52, p. 186; 81, p. 75; 99, p. 69; Fig. 13 from Monographiae biologicae 35, p. 67). Names of the authors concerned are givrn in the Iegends of t.he present Figures, and the titles of their works in t.he Heferencrs.


REFERENCES

AKHURST (E. G. J.) and BREEN (C. M.), 1988. - Ionie content as a factor influencing turbidity in t.wo floodplain lakes after a tlood. Hydrobiologia, 160 : 19-31.

ALEEM (A. A.), 1959a. - A taxonomie and paleoecological invest,igation of t,he diatom-flora of the extinct Fayoum Lake (Upper Egypt). Bull. Fac. Sci. Alexandria Univ., 2 : 99-138.

ALEEM (A. A.), 1959b. - A taxonomie and paleoecological investigation of the diatom flora of the extinct Fayoum Lake. Part II. Distribution and ecology. Bull. Fac. Sci. Alexandria Univ., 2 : 217-244.

ALEEM (A. A.) and SAMAAN (A. A.), 1969. - Productivity of Lake Mariut, Egypt.. Part 1. Physical and chemical aspects. Inf. Revue ges. Hydrobiol., 54 : 313-355.

ALLANSON (B. R.), 1979. - The physico-chemical limnology of Lake Sibaya. In : Lake Sibaya, Allanson (B. R.) (ed.). Monographiae biol., 36. Dr W. Junk Publ., The Hague : 42-74.

ALLANSON (B. R.) and HOWARD-WILLIAMS (C.), 1984. - A contribut.ion to the physicochemical limnology of Swartvlei. .Arch. Hydrobiol., 99 : 133-159.

ALLANSON (B. R.) and VAN WYK (J. D.), 1969. -An introduction to the physics and chemistry of some lakes in northern Zululand. Trans. R. Soc. S. Afr., 38 : 217-240.

ARAD (A.) and MORTON (W. H.), 1969. - Minera1 Springs and saline lakes of the Western Rift Valley, Uganda. Geo- chim. Cosmochim. Acfa, 33 : 1169-1181.

ASHTON (P. J.), 1977. - Factors affecting the growth and development of Azolla filiculoides Lam. Proc. Second Nafional Weeds Conf. Souih Africa, Sfellenbosch Univ., 2-4 Feb. 1977. Balkema, Cape Town and Rotterdam, p. 249-268.

ASHTON (P. J.), 1979. - Nitrogen fixation in a nit.rogen-limited impoundment. J. Waf. Poli. Control Fed., 51 : 570- 579.

ASHTON (P. J.), 1981. - Nitrogen fixation and the nitrogen budget of a eutrophic impoundment. Wafer Res., 15 : 823-833.

ASHTON (P. J.), 1985. - Nitrogen transformations and the nit.rogen budget of a hypertrophie impoundment (Hart- beespoort. Dam, South Africa). J. Limnof. Soc. Sfh Afr,, 11 (1), 32-42.

ASHTON (P. J.) and SCHOEMAN (F. R.), 1983. - Limnological studies on t.he Pretoria Salt Pan, a hypersaline maar

Rev. hydrobiol. trop. 25 (2) : BY-l&& (1992).

lake. 1 : Morphomet.ric, physical and chemical feat.ures. liydrobioIo$ia, 99 : 61-73.

ASHTON (P. .J.) and SCHOE~IAN (F. R.), 1988. -Thermal stratification and the stabi1it.y of meromixis in t.he Pretoria Salt Pan, Sout.h Africa. Hydrobiologia. 158 : 253-265.

BAKER (B. H.), 1958. - Geology of the Magadi area. Geol. Surv. Kenya, Rep. no 42. 81 p.

BANOUR (M. W.), 1979. - The salt regime of Lake Edku (Egypt) before and after t.he const.ruction of Asswan’s High Dam. Arch. Hydrobiol., 85, 392-399.

BANOUB (M. W.), 1983. - Comparative study of the nutrient? salts of Lake Edku (Egypt) before and after the construction of Asswan’s high dam in 1958 and 1969. Arch. Hydrobiol., 99 : 106-I 17.

BANOUR (M. W.) and WAHBY (S. D.), 1961. - Further cont.ribution t,o the chemist.ry and hydrography of the Nozha Hydrodrome near Alexandria. Alex. Inst. Hydrobiol. Fish., Notes and Mernoirs, 61 : 42 p.

BA~TER (R. M.), WOOD (R. B.) and ~ROSSER (M. V.), 1973. - The probable occurrence of hydroxylamine in the water of an Et.hi»pian lake. Limnol. Oceanogr., 18 : 470-472.

BEADLE (L. C.), 1932a. - Scient.ific resu1t.s of the Cambridge Expedition to the East. African Lakes, 1930-l. 3. Obser- vations on the bionomics of some East. African swamps. J. Linn. Soc. (ZOO~)., 38 : 135-155.

BEADLE (L. C.), 393213. - Scientific results of the Cambridge Expedit.ion to t.he East. African Lakes. 1930-I. 4. The waters of some East. African lakes in relation to their fauna and Bora. J. Linn. Soc. Lond. (ZOO~.), 38 : 157- 211.

BEADLE (L. C.), 1943. - An ecological survey of some inland saline wat,ers of Algeria. J. Linn. Soc. (ZOO~.), 41 : 218- 242.

BEADI.E (L. C.), 1981. - The inland waters of tropical Africa. An int,roduc.tion t.o t%ropical limnology. 2nd Edn. Long- man, London, p. 475.

BEAUCHA~XP (R. S. A.), 1953. - Sulphates in African inland waters. Nature, Lond., 171 : 769-771.

BEAUCHAILIP (R. S. A.), 1958. - Ut.ilizing the nat.ural resources of Lake Vict.oria for the bpnefit of fisheries and agri- culture. Nature, Lond., 181 : 1634-1636.

BENECH (V.), LEI\IOALLE (J.) and QUENSIÈRE (J.), 1976. - Mort.alités de poissons et conditions de milieu dans Ie

136 .T. F. TALLING

RERG (A.). lR61. - Hole 6cologique des eaux de la Cuv&e cllngolaise sur la crtr~ssance de la jacinthe d'eau (/?Si& hornio rrassipes (Mart.) SOI~S). ~llern. .-Icad. R. C;c;. oufre-mer, Sri. naf., 8, ns. 12 (3) : 120 p.

filSHAl (H. bI.), If>fy>. Thr water characteristirs of the Nile in the SU&I> with a note OIL thr effect of Eichhornia crtlssipes ~II thr hydrohiology of thr Xile. Hydrohiolo- qicr. 10 : Xlï-382.

RLACK (J. Y.). 1953. -TOP distsihntion of solar radiation Oder t.ho earth’s wrfare. .-1rch. Jlef. Geophys. Bioklint., 7 : lt~o-lt‘cn.

BI,4Nl.. (hf.). hc~r;:r (.J.l and r\CIRENTON (F.), 1955. - RPcherchrs hydrohiologiqurs dans le hassin du Moyen Niger. Bull. Insf. Fraqais .Ifr. Noire (.a), 17 : BYO-746.

Howxar\-e~ (,A. P.), 1902. - Srosonal hydrological changes affecting the erolofry of Luaka Lagoon, Lake Rang- wn~lu : some preliminary observations. Hydrobiology antl Inland Fishrrics. Symposium on Prohlems of Major Lakes (Lusaka, 18th-Z4t.h August. 1960). Landon : C:l::TA/cSh L,\KES (60) Cri(b). p. I5+[11].

Howv\r&r<e~ (A. P.). 19%. - Thr physicochemical limnology of the RIwrnda Hiver mouth. Lake Kariba. .-lrch. Hydro-

hiol.. ïï : f.X-108.

RUGRN~I (F. IV. H.). 19ïU. - GopIler ion dist.ribution in the +urface waters of Iakes George and Idd hmin. H,ydro- biologie, ts-i : $15.

RIIC;ENYI (F. M’. H.), 19412. - (:oppPr pollution studies in Lakes George and Edward, Irganda : t-he distrihubion of copper, cadmium and iron in the water and sediments. Enuiron. Pollrtf., Scr. R, 3 : 129-138.

BURGIS (M. J.j, 1978. - The Lake George ecosystem. 1.erh. inf. 1.erein. fheor. cmgrul. Limnol., 20 : 1139-I 152.

BuIW;Is w. J.), D.4HLINGTON (J. P. E. c.), DUNN (1. G.), GNF (G. G.). Gvc~~3.4~~4 (J. J.) and ïVfd;OWAN (L. hl.), 1973. - The hiomass and distrihut.ion of organisms in Lake George. Uganda. Proc. R. Soc. B, 1% : 271-298.

C.~I..ION (.\. G.), 1987. - A recent.ly landlocked brarkish-water lagoon of Lake Tangaoyika : physical and chemical c.haract.eristics, and spatio-t.rmporal dist.rihution of phytoplankttsn. Hytlrohiolqin, 153 : 55-70.

(:-\RXISI-~E (J. P.). l!j/-. -- Circulation @néraIe des eaux dans le lac Tchad. J-wh. int. [-erein. fheor. ang~ro. Limnol., 19 : fI-8-635.

~:.~RM~~LI~F: (J. P.). 1976. - La [email protected]«n hydrogéochimique du lat. Tchatl. 7’rnu. cf dot. &sfom., 11” 58 : 418 p.

I~:~RV~IIVE (J. P.). 1983. - Hgdrochrmical regulation of the Iake. In : I,clkr Ch&. Kcology and producfivify of a shal- loro fropicul t~cosystenl. Carmouze (J. P.). Durand (.J. R.) üIJd 1.6~6que IC.) (eds.j, J)r !Y. Junk Pnhl., The I-lagur : 95- 123.

CARM~~~E (.J. P.), Apu~s~ (E.), ARCE (Cl.) and QUINTANILLA (.J.), 1983. - LP bilan énergétique du lac Titicaca. Reu. Hydrobiol. trop., 16 : 135-144.

CARMOUZF: (J. P.). C:HANTRAINE (J. M.) and LEMOALLE (J.), 1983. - Physical and chemical characteristics of the waters. In : Lulie Chad. EcoIogy and producfiuity of a shallou~ fropical ecosysfem, Carmouze (J. P.), Durand (J. K.) and L+vEqur (Cl.) (eds), Dr W. .Junk Publ., The Hague : 65-94.

CARIIOUZE (J. P.). DURAND (.J. H.) and LÉ~ÊQuF. (C.) (eds), 1983. - Lake Chad. Ecology and producfivify of a shnl- low tropical ecosysfem. Dr Y%‘. Junk Publ., The Hague. 576 pp.

CARMOUZE (J. P.), F~TIUS (G.) and LÉ~ÈQuE (C.), 1978. - Influence qualitative des macrophytes sur la regulat.ion hydrochimique du lac Tchad: Cah. ORSTOM St+. Hydrohiol., l? : 65-69.

CARMOZIZE (.J. P.) and LEMOALLE (J.), 1983. - The lacustrine environment. In : Lake Chad. Ecology and producfivify of o shallom tropical erosysftm, Carmouze (.J. P.), Durand (J. R.) and L;~+ue (C.) (rds), Dr IV. Junk Puhl., The IIague : 27-63.

(I~RTER (ci. s.). 19%. - The papyrus suramps of L!qanda. Hef- fers. Cambridge. 25 p.

CHANTRAINE (J.-M.). 19%. - fivolution hydrochimique du lac Tchad de septembre 1973 5 septembre 1975 au cours d’une phasr dtx décrue. Cah. ORSTOM, S&. Hydrobiol., 12 : n-17.

~~GELS (F. X.) and C;&c (J. Y.), 1982. - Le lac de Guiers : fonctionnement. bilans hydriques. Évolution d’une nappe d’eau libre en zone sahélienne (Sénégal). L’ah. OFLSTOdf, St+. Géol.. 7 : 41-60.

COGELS (F. X.) and GAC (J. kV.), 1983. - Circulation et salinit.8 des eaux du lac de Guiers. Problème de dbveloppement, et modele de gestion. Orstom, Cent.re de Dakar-Hann. 14 p., ndtigr.

DAMAS (H.), 1954. --. Étude limnologique de quelques lacs ruandais. Il. Êt.ude t.hermique et. chimique. Mem. Insf. Roy. colon. Belge, Sect. Sci. naf. mrd. Bruxelles. 24 : l-116.

DE KIMPE (P.), 1964. - Cont.ribution à l’étude hydrobiologique du Luapula-Morro. .-In& Mus. r. Afr. cent., Ser. 8 vol. Sri. Zool., nu 128 : 238 p.

DENNY (P.). 1978. - Nyumha ya Mungu reservoir, Tanzania. The general features. B;ol. J. Linn. Soc., 10 : 5-28.

DENNY (P.) (Rd.), 1985a. - The ecology and management of African metland vegrfafion. Dr W. .Junk Publ., Dor- drecht, 344 p.

DENNY (P.). IYSbb. - The struct.ure and functioning of Afri- cari euhydrophyt,e communities. The floating-leaved and submerged vepetation. In : The ecology and management of African toefland vegefafion, Denny (P.) (ed.). Dr W. Junk Puhl., Dordrerht : 125-151.


D~NNY (P.), HARMAN (J.), ABRAHAMSSON (J.) and BRYCESON (I.), 1978. - Limnological and phytoplankt.on studies on Nyumba ya Mungu reservoir, Tanzania. Biol. J. Linn. Soc., 10 : 29-48.

DILLON (P. J.) and RIGLER (F. H.), 1974. - The phosphorus- chlorophyll relat,ionship in lakes. Limnol. Oceanogr., 19 : 767-773.

DUBOIS (J. Th.), 1955. - Étude hydrobiologique d’un lac Afri- cain d’alt.it.ude : Le Lungwe. Ét,ude physique et chimique. Bull. Cenfr. belge éfude ef de documenfafion des Eaux, 110 53 : 79-82.

DUBOIS (J. Th.), 1959. - Note sur la chimie des eaux du lac Tumba. Bull. Séances Acad. R. SC. oufre-mer, 5 : 1321- 1333.

DUFOUR (P.)? 1982a. - Les front.ières naturelles et humaines du système lagunaire Ebrié. Incidences sur I’hydrocli- mat. Hydrobiologia, 94 : 105-120.

DUFOUR (P.), 1982b. - Influence des conditions de milieu sur la biodégradation des matières organiques dans une lagune tropicale. Oceanologia Acfa, 5 : 355-363.

DUFOUR (P.), 198%. - Modéles semi-empiriques de la production phytoplanctoniyue en milieu lagunaire tropical (C&e d’ivoire). Acfa Owologia, Oecol. gener., 3 : 223-239.

DUFOUR (P.), CREMOUX (.J. L.) and SLEPOUKHA (M.), 1981.- Contràle nutritif de la biomasse du seston dans une lagune tropicale de Côte d’ivoire. 1. Étude méthodolo- gique et. premiers résultats. J. erp. mar. Biol. Ecol., 51 : 246-267.

DUFOUR (P.) and DURAND (J. R.), 1982. - Production végé- tale des lagunes ivoiriennes. Reo. Hydrobiol. trop., 15 : 209-230.

DUFOUR (P.) and LEMASSON (L.), 1985. -Le régime nutritif de la lagune tropicale l?brié (Côte d’ivoire). Oceanogr. trop. 20 (1) : 41-69.

DUFOUR (P.), LEMASSON (L.) and CREMOUX (J. L.), 1981. - ControIe nutritif de la biomasse du sest.on dans une lagune t,ropicale de COte d’ivoire. II. Variations géo- graphiques et saisonnieres. J. exp. mar. Biol. Ecol., 51 : 269-284.

DUFOUR (P.) and SLEPOUKHA (M.), 1981. - fitude de la ferti- lité d’une lagune tropicale au moyen de tests biologiques sur populations phytoplanctoniques naturelles. Reo. Hydrobiol. trop., 14 : 103-114.

DUMONT (I-I. J.), in press. - African shallow lakes and wetlands : t.he regulation of species and communities. Rev. Hydrobiol. trop.

DUMONT (H. J.), PENSAERT (J.) and VAN DE VELDE (I.), 1981. - The crustacean zooplankton of Mali (West Africa). Hydrobiologia, 80 : 161-187.

DURAND (J. R.) and CHANTRAINE (J. M.), 1982. - L’environnement climatique des lagunes ivoiriennes. Rev. Hydro- biol. trop., 15 : 85-114.

Rev. hgdrobiol. trop. 25 (2) : 87-M (1992).

ECXRRS (K. A.) and TETZLAFF. (G.), 1978. - A simple mode1 for descrihing the heat balance of a shallow lake with applicat.ion to L.ake Chad. Boundary-Layer Mefeorol., 15 : 205-214.

EL-HEHYAWI (M. L. El-S.), 197ï. - Some aspects of the chemistry nf Lake Manzalah wat.er. Bull. Insf. Oceanogr. Fish., Alexandria, 7 : l-30.

EL-&GHRABY (A. M.) and DOWIDAR (N. M.), 1969. - The occurrence of marine plankt.on organisms in Lake Qarun (Egypt). Rapp. P.-v. Neun. Commn inf. Explor. scient. Mer Mediferr., 19 : 849-851.

ELSTER (H. J.) and VCJLLENWEIUEH (R. A.), 1961. - Beitrage zur Limnologie Agyptens. .Irch. Hydrobiol., 57 : 241- 343.

EL-WAKEEL (S. Ii.), Mo~cos (S. A.) and hlahlis (A. M.), 1970. - Thr major rations in Lake Maryut, waters. Hydro- biologiu, 36 : 263-271.

EL-WAKEEL (S. K.) antl WAHRY (S. D.), 1970. - Hydrography and chenlistry of Lake Manzalah, Egypt.. Arch. Hydro- biol., 67 : 1 ï3-‘200.

ENTE (B.), 1976. -- Lake Nasser and Lake Nnbia. In : The Nile, biology nf an ancienf riaer, Rzoska (J.), (ed.), Dr W. .Junk Pul)l.. The Hague : 271-298.

ENTE (B.), 1978. - Sedimentation prorrsses above the Aswan High Dam in Lake Nasser - Nuhia (Egypt - Sudan). Terh. inf. 1.erein. fheor. nngezo. Limnol.. 80 : 3667-1671.

EUGSTER (FI. P.), 1967. - Hydrons sodium silicat.es from Lake Magadi, Kenya : precursors of beddrd chert. Science, 1V.Y., 157 : llT7-1180.

EUGSTER (H. P.), IRTO. - Chemistry and origin of the brinrs of Lake Magadi. In : Mineralogy and chemistry of non- marine eoaporifes, Morgan (R. A.) (ed.). Alineral. Soc. .4mer., Spec. Pap. no 3 : 213-235.

EWBTER (H. P.) and JONES (B. F.), 1968. - Gels composed of sodium-aluminum silicate, Lake Magadi, Kenya. Science, N.>-., 161 : 160-163.

EVANS (J. IT.), 1961. - A phytoplankton multi-sampler and its use in Lake Vict.oria. Nnfzzre, Land., 191 : 53-5.5.

FISH (G. R.), 1957. - A seiche movement and its effect on the hydrology of Lake Vict.oria. Coloninl Office Fisheries Pzzbl., Land., 10 : l-68.

FRYER (Ci.) and TALLING (J. F.). 1986. - Africa : t.he FBA connection. Hep. Frrshzoaf. biol. :lss., no 54 : 97-1‘22.

FURNESS (II. D.) and RRREN (C. M.), 1982. - Dec:omposit.ion of Cynodon dacfylon (L.) Pers. in t.he seasonally flooded areas of the Pnngolo river floodplain : patt.ern and significance of dry mat.ter and nutrient loss. Hydrobiolo- gia, 97 : 119-126.

FLJRNESS (H. D.) and BREEN (C. M.), 1986. -An assessment. of the importance of aater stress in seasonally flooded C. dacfylon on th? Pongolo River floodplain in the for-

138 J. F. TALLING

mulation of guidelinrs for controlled tlooding. Hydro- biologia. 137 : 175-184.

G4c (J. j‘.), l!lHO. - Gbochimie du bassin du lac Tchad. Bilan de l’alt~raticm de I’erosion et. de la sAdimentat.ion. Trczo. rf dot. Orsfom. 123 : 251 p.

GAL.L.AHL)~ (Y.), 19’78. - 4svmmetry and anomalies of cirrula- tion and vertical mising in the branching of a lagoon est-unry. Hydrodynamics of estuaries and fjords, Nihoul (J. C.) (ed.), Elsevier : 197-206.

GANF (G. G.), lY7-l~. - Phytoplankton biomass and clist,ribu- tien in a shallow eutrophic lake (Lake George, Uganda). Oecok+. 16 : Q-2Q.

G~NF (G. G.j, 197lh. - Diurnal mixing and t.he v-ertical tiist.ri- buticm of phytoplankton in a shallow equatorial lake (Lake George, Uganda). .I. Ecol., 62 : 61 I-629.

G~NF (G. C;.). 19ïlc. - Incident snlar irradiante and underwater light penet.ration as factors controlling the chlorophyll o content. of a shallow equatorial lake (Lake George, LIgarrda). .I. Eco/.. 62 : 593-609.

C.;~NF (C;. G.1. 1971~1. - Rates of oxvaen unt.ake bv the plank- .I v

tonic rorr1munit.y of a shallow equat.orial“ lake *(Lake George, Uganda). Orcologia, 15 : 17-32.

GANF (G. G.). 1975. - Photosynthetic product.ion and irra- diancr - photosynthesis relationships of the phytoplankton from a shallow equatorial lake (Lake George, Llyanda). Oecologia, 18 : 165-183.

GANF (G. G.) snd BLA~K,~ (P.), 1974. - Oxygen upt.ake, ammonia and phosphate excretion by zooplankt,on of a shalIow equatorial lake (Lake George, Uganda). Limnol. Ocranogr., 10, : 313-326.

(.i 4NF (G. G.) arltl HORNE (A. .J.), 1975. - Diurnal stratillca- t.ion, photosynthesia and nit.rogen-fixation in a shallow equatorial lakr (Lake George, Uganda). Frrshtvat. Biol., 5 : 1:3-3!t.

C;4NF lG. Ci.) and MILRURN (T. R.), 1971. - A conduct.imet.ric methtrd for thr determinat.inn of t.ot.al inorganic and partirniat.e organic carbon frart.ions in freshwater. .4rrh. Hydrobiol., ti!) : I-13.

GANF (G. G.) and VINER (A. B.), 1973. - Ecological st.abi1it.y in a shallow equatorial lakr (Lake George, Uganda). Pro<. R. Soc. B, 184 : 321-316.

GA~~E (F.), l!jïla. - Les diatombes des sédiments holocènes du bassin du Lac Afrrra (Giulietti) (Afar Septentrional, Ethiopie). Essai dr reronst.it,ution de l’évolution du milieu. Int. Heu. ges. Hydrobiol., 59: 96-122.

Gtsan (F.), 1971b. - Les diatombes holocènes du bassin infé- rieur de I’.Aouache (dépression des Danakil, Ethiopie). Leur signification paléoecologique. Inf. Rev. ges. Hydro- biol.. 59 : 12%146.

GASSE (F.), 1975. - L’ivolrrfion des lacs de l“,lfar Ceniral (Éfhiopie ef T.F..I..I.) du Plio-Pléistocène à I’Actuel. Thrsis, LTniversité de Paris VI, Paris. 3 vols, 387 p.

GASSE (F.). 1977. - Evolution of an intert.ropical African lake from 70000 BP t.o present : lake Abhé (Et.hiopia and T.A.F. 1.). Natare, Lond., 265 : 42-45.

GASSE (F.) and DELIRRIAS (G.), 1976. - Les lacs de I’Afar cent.ral (Ethiopie et T.F.A.I.) au Pléist.ocène Superieur. In : Paleolimnology of Lake Bima and fhe Japanese Pleis- focene, Horie (S.) (ed.), 4 : p. 529-575.

GASSE (F.), TALLING (J. F.) and KILHAM (P.), 1983. - Diatom assemblages in East Africa : classiflcat.ion, distribution and ecology. Reu. Hydrobiol. trop., 16 : 3-34.

GAUDET (J. J.), 1976. - Nutrient relationships in the detritus of a tropical swamp. Arch. Hydrobiol., 78 : 213-239.

GACJDET (J.), 1977. - Uptake, accumulat.ion and loss of nutrients by papyrus in tropical swamps. Ecology, 58 : 415-122.

GAUDET (J. J.), 1978. - Effect. of a tropical swamp on water qua1it.y. l’erh. inf. Iérein. fheor. angem. Limnol., 20 : 2202-2206.

GAUDET (J. J.), 1979. - Seasonal changes of nutrients in a tropical swamp : North Swamp, Lake Naivasha, Kenya. J. Ecol.. 67 : 953-981.

GAUDET (J. J.) and MELACK (.J. M.), 1981. - Major ion chemis- t.ry in a tropical African lake basin. Freshwaf. Biol., 11 : 309-333.

GAUDET (J. J.) and MUTHURI (F. M.), 1981a. - Nutrient regeneration in shallow tropical lake wat,er. Verh. inf. Verein. theor. angew. Limnol., 21 : 725-729.

GAUDET (J. J.) and MIITHURI (F. M.), 1981h. - Nutrient relationships in shallow water in an African lake, Lake Nai- vasha. Oecoloyia, 49 : 109-118.

GOLTERMAN (H. L.) AND KOUWE (F.), 1980. - Chemical budgets and nutrient pat.hways. In : The fanctioning of freshroater ecosysfems, Le Cren (E. D.) and Lowe- McGonnell (H. H.) (eds). Internat.ional Biological Pro- gramme, 22. Gambridge Univ. Press : 85-140.

GONFJANTINI (R.), Rn~sr (S.), FERRARA (G.)and PANJCHI (C.), 1973. - Isotopic composit.ion of waters from the Dana- kil depression (Ethiopia). Earth Planef. Sci. Leff., 18 : 13-21.

GORGY (S.), 1959. - The use of energy equations in the cal- culation of the rate of evaporation from Lake Qarun. Notes & Mem., Flydrobiol. Dept. Cairo, 42 : 26.

GREENWOOL~ (P. H.), 1976. - Lake George, Uganda. Phil. Trans. R. Soc. Lond. B, 274 : 375-391.

GREENW»OD (P. H.) and I.uN~ (J. W. G.) (eds.), 1973. - A discussion on t.he biology of an equatorial lake : Lake George, Llganda. Proc. R. Soc. Lond. B, 184 : 227-346.

GRJFFJTHS (J. F.) (ed.), 1972. - Climates of Africa. Elsevier. Amsterdam.

GROBBELAAR (J. LT.), 1985. - Carbon flow in the pelagic zone of a shallow t.urbid impoundment, Wuras Dam. Areh. Hydrobiol., 103 : I-24.

Hro. hydrohiol. frop. 25 (2) : 8i-144 (LY92).


GROBBELAAR (J. U.) and TOERIEN (D. F.), 1985. - Carbon flow in a small t.urbid man-made impoundment.. Hydrobiolo- gia, 121 : 237-247.

GROVE (A. T.), 1972. - The dissolved and solid load carried by some West African rivers : Senegal, Niger, Benue and Shari. J. Hydrol., 16 : 277-300.

GUEST (N. J.) and STEVENS (J. A.), 1951. - Lake Natron, its Springs, rivers, brines and visible saline reserves. Geol. Surv. Tanganyika, Minera1 Resources Pamphlet, no 58 : 21 p.

HABERYAN (K. A.), 1985. - The role of copepod fecal pel1et.s in the deposition of diatoms in Lake Tanganyika. Lim- nol. Oceanogr., 30 : 1010-1023.

HARE (L.) and CARTER (J. C. H.), 1984. - Diel and seasonal physico-chemical fluctuations in a small natural West African lake. Freshwaf. Biol., 14 : 597-610.

HAWORTH (E. Y.), 1977. - The sediments of Lake George (Uganda). V. The diatom assemblages in relation to the ecological history. Arch. Hydrobiol., 80 : 200-215.

HECKY (R. E.) and KILHAM (P.), 1973. - Diatoms in alkaline saline lakes : ecology and geochemical implications. Limnol. Oceanogr., 18 : 53-71.

HEEG (J.) and BREEN (C. M.), 1982. - Man and the Pongolo floodplain. Soufh African National Scientific Pro- grammes Report, no 56 : 117 p.

HEEG (J.), BREEN (C. M.), COLVIN (P. M.), FURNESS (H. D.) and MUSIL (C. F.), 1978. - On the dissolved solids of the Pongolo floodplain pans. J. limnol. Soc. Souih. Africa, 4 : 59-64.

HEMENS (J.), SISIPSON (D. E.) and WARICK (R. J.), 1977. - Nitrogen and phosphorus input t,o the Midmar Dam, Natal. Wafer SA, 3 : 193-201.

IIESSE (P. R.), 1958. - Fixation of sulphur in the muds of Lake Victoria. Hydrobiologia, 11 : 171-181.

HOLDSHIP (S. A.), 1976. - The paleolimnology of Lake Manyara, Tanzania. A diat.om analysis of a 56 meter sediment tore. PhD thesis, Duke Univ., 121 p.

HORNE (A. J.) and VINER (A. B.), 1971. - Nitrogen fixation and its signiflcance in tropical Lake George, Uganda. Nature, Lond., 23’2 : 417-418.

HOWARD-WILLIAMS (C.), 1972. - Limnological studies in an African swamp : seasonal and spatial changes in the swamps of Lake Chilwa, Malawi. Arch. Hydrobiol., 70 : 379-391.

H~~ARD-WILLIAMS (C.), 1977. - The distribution of nut.rients in Swartvlei, a Southern Cape coastal lake. Water SA, 3 : 213-217.

H~~ARD-WILLIAMS (C.), 1979. - The aquatic environment : II. Chemical and physical characteristics of the Lake Chilwa swamps. In : Lake Chihva, ed. Kalk (M.), McLa- chlan (A. J.) and Howard-Williams (C.). Monographia biol. 35. Dr W. Junk Publ., The Hague : 79-90.

Reo. hydrobiol. trop. 25 (2) : 87-144 (1992).

HOWARD-WII.LIAX~S (C.), 1981 - St.udies on t,he abi1it.y of a Pofamogeton pecfinatus community t.o remove dissolved nit.rogen and phosphorus compounds from lake water. J. appl. Ecol., 18 : 610-637.

HOWARD-WILLIAI~IS (C.), 1985. - Cycling and retention of nitrogen and phosphorus in wetlands : a theoretical and applied perspective. Freshwaf. Riof., 15 : 391-431.

HOWARD-WILLIAIGS (C.) and ALLANSON (B. R.), 1981a. - An integrated st.udy on lit.toral and pelagic primary production in a Southern African coastal lake. Arch. Hydrobiol., 92 : 507-53-l.

HOWARD-WILLIADIS (C.) and ALLANSON (B. R.), 1981b. - Phosphorus cycling in a dense Pofamogefon pecfinafus L. bed. Oecologia, 49 : 56-66.

H~~ARD-WILLIAMS (C.) and DAVIES (B. R.), 1979. - The rates of dry matter and nutrient loss from decomposing Pofa- mogefon pecfinafus in a brackish south-temperate coas- ta1 lake. Freshmat. Biol.. 9 : 13-21.

HOWARD-WILIATNS (C.) and GANF (G. G.), 1981. - Shallow wat.ers. In : The ecology and ufilizafion of African inland waters, Symoens (J. J.), Burgis (M.) and Gaudet (J. ,J.) (eds.), United Nations Environment Programme : 103- 113.

HOWARD-WILLIA~~S (C.) and G.~UDET (J. J.), 1985. - The st.ructure and functioning of African swamps. In : The ecology and management of African wefland vegefafion, Denny (P.) (rd.), Dr W. Junk Publ., Dordrecht : 153- 175.

H~~ARD-WILLIAMS (C.) and JUNK (W. J.), 1977. - The chemical composit.ion of Central Amazonian aquatic macrophgtes with special reference to their role in t.he ecosystem. Arch. Hydrobiol.. 79 : 446-464.

HOWARD-WII.LIAIS ((1.) and LENTON (G. M.), 1975. - The role of the litt.oral zone in t.he functioning of a shallow tropical lake ecosystem. Freshwaf. Biol., 5 : 445-459.

H~~ARD-WILLIAMS (cl.) and H~~ARD-WILLIAMS (W.), 1978. - Nutrient leaching from t.he swamp vegetation of Lake Chilwa, a shallow African lake. Aquat. Bof., 4 : 257-267.

HUTCHINSON (G. E.), 196ï. - .4 freafise on limnology. Vol. I. Geography, physics and chemisfry. John Wiley, New York and London. 1015 p.

ILTIS (A.), 3969. - Phytoplanct.on des eaux natronées du Kanem (Tchad). II. Les mares temporaires. Cah. ORS TOM, ser. Hydrobiol., 3 : k-19.

IMHOFF (J. F.), SAHL (H. Ci.). SOLIMAN (G. S. H.) and TRÜPER (II. G.), 1979. - The Wadi Natrun : chemical composi- t.ion and microhial mass developments in alkaline brines of eutrophic desert lakes. Geomicrobiol. J., 1 : 219-234.

ISHAK (M. M.) and ABDEL-MAI.EK (S. A.). 1980. - Some ecological aspects of Lake Qarun, Fayoum, Egypt,. Part. 1. Physico-chemical environment,. Hydrobiologia, 74 : 173- 178.

140 J. F. TALLING

JOHN (D. M.), 1986. - The inland waters of tropical West Xfrica. Ergebn. Limnof., 33 : l-244.

JOII’E~ (El. F.), EI.IC;BTER (H. P.) and RETTIG (S. L.), 3977. - ffydrochrmistry of t.he Lake Magadi basin, Kenya. Cieo- chim. cosmochim. .4cfu, dl : 53-72.

KALFF (J.), 1983. - Phosphorus limitat.ion in some tropical -\frican lakes. Hydrobioloyiu, 100 : 101-l 12.

K4r..h. I&I.), ~.~o~AH»-~vrLLI4hïS (c.) and MCLACHLAN (A. J.) (eds.), 1979. - Luke Chilwu. Studies of change in a tropi- cc11 rcosysfwn. Mono$r. Biologicue 35. Dr W. Junk Publ., Dordrecht., 462 p.

GLLQVWT (T.) and MEADOWS (B. S.), 1978. - The toxic effect. of copper on algae and rot,ifers from a soda lake (Lake Nakuru, East Africa). IVufer Res., 12 : 771-775.

K~MP-NIELPEN (L.), .J$RGENSEN (L. A.) and J~RGENSEN (S. E.), 1980. - A sediment-wat.er exchange mode1 for lakes in t.he LJpper Nile Basin. In : Proyress in ecologicul engineering und management by mufhemafical modelling. Edititms Crbcdoc. Li&ge : 557-581.

KEN»ALL (R. L.j, 1969. - An ecoloaical history of t.he Lake Victoria hasin. Ecol. 12ionoyr.,-J9 : 121-176.

KILHAM (P.), 1971. - A hypothesis concerning silica and the freshwater plankt,onic diatoms. Limnol. Oceunoyr., 16 : IO-18.

KlL.H4M (FI.). 1973. - Biogeochemisfry of Africun lukes und riorrx. PhD t.hrsis. Ilniv. Michigan, 212 p.

I~ILH~W (P.), 198-. - The effect. of hippopotamuses on potas- sium and phosphatr ion concent.rations in an African lake. .Im. :IJidl. Nuf.. 108 : 402-205.

KILHAM (P.) and HECKY (R. E.), 1973. - Fluoride : geochemical and ecolo~ical significance in East. African waters and sedimentS. Limnof. Oceunogr., 18 : 932-945.

KOE~AN (J. H.), F’ENNINGS (J. H.). DE GOEIJ (J. J. M.), TJIOE (P. S.), ~LIN~O (P. hl.) .~ND HOPCRAFT (J.), 1972. - A preliminary survey of t.he possible contaminat.ion of Lake Nakuru in Kenya wit.h some metals and chlorina- trd hydrocarhon pest.icides. .J. uppl. Ecol., 9 : 411-416.

LEMASSON (L.) and PAGÈS (J.), 1982. - Apports de phosphore et d’azote par la pluie en zone tropicale (Chte d’ivoire). Rro. Hydrohiol. trop., 15 : 9-14.

LEMASSON CL.). PAGÈS (J.) & GREMOUX (J. L.), 1982. - Échanges d’Cléments nutrit.ifs dissous entre l’eau et le &liment dans une lagune t.ropicale saumat.re. Oceanoyr. trop., 17 : -lF>-58.

LEMCIALLE (J.), 197Ya. - Iran hearing oolites and the present. conditions of iron sediment,at.ion in Lake Chad (Africa). In : Ores in sedimenfs, AMSTUTZ (G. C.) and BERNARD (A. J.j (eds); Springer. Berlin.

LEMOALLE (J.), 19731,. - Azote et phosphore dans les eaux de pluies H Fort-Lama (1970). Cah. ORSTOM, se%. Hydrol., 9 : 614x3.

~+CI. hydrobiol. lrop 2.5 (2) : 87-I-M (1992).

LEMOALLE (J.), 19T8. - Relations silice-diat,omées dans le lac Tchad. Cah. ORSTOM, Sér. Hydrobiol., 12: 137-141.

LEMOALLE (J.), 1979. - Biomasse et product.ion phytoplancto- niqurs du lac Tchad (1968-1976). Relations avec les condit.ions du milieu. Orst.om, Paris. 311 p., mulfigr.

LEMOALLE (J.), 1983a. - Le lac Icheul. Éléments de I’hydro- climat. en 1981-82. Publ. frimesfr. Jnsf. naf. xi. fechn. oceunoyr. pèche Sulummbo, Rapp. Docum. 1, 13-33.

LEMOALLE (J.), 1983b. - Phytoplankton production. In : Lake Chad. Ecology and productivity of a shallow tropical ecosystem, Carmouze (J. P.), Durand (J. R.) and L&&~ue (C.) (eds). Monographiae biol., 53. Dr W. Junk Publ., The Hagne : 357-384.

LÉVÈQUE (C.), 1972. - Mollusques bent.hiques du lac Tchad : écologie. product.ion et. bilans énergét,iques. ThBse, Uniu. Paris VI. 295 p. (cited after Carmouze, 1983).

LIWNGSTONE (D. A.j, 1963. - Chapter G. Chemical composi- t,ion of rivers and lakes. Data of Geochemistry, Fleischer (M.) (ed.). Sixth Edition. Geol. Surv. Professional Paper 440-G. Waahingt.on, vr1+6d p.

LIVINGSTONE (D. A.) and MELACK (J. M.). 1984. - Some lakes of subsaharan ‘ifrica. In : Ecosysfems of fhe world. 23. Lakes and resrrrroirs, Taub (F. B.) (ed.), Elsevier, Ams- terdam : 467-497.

LOFFLER (H.), 1964. - The limnology of tropical high-moun- t,ain lakes. Iv&. inf. lerein. theor. unyew. Limnol., 15 : 176-193.

LOFFLER (H.), 1968. - Tropical high mountain lakes. Collo- quium Geoyraphicum, 9 : 67-76.

MACINTYRE (S.), 1981. - Stratification und mizing in shullow tropical Africun lakes. PhD Thesis, Duke Univ., 338 p.

MACTNTYRE (S.) and MELACK (.J. M.), 1982. - Meromixis in an equatorial hfrican soda lake. Limno/. Oceanogr., 27 : 595-609.

MCLACHLAN (A. J.), 1979. - The aquat.ic environment : 1. Chemical and physical characterist.ics of Lake Chilwa. In : Lake Chilmu. Studies of change in u tropical ecosysfem, Kalk (M.). Mclachlan (A. J.) and Howard-Wil- liams (Cl.) (eds). Monographiae biol., 35. Dr W. Junk Publ.. The Hague : 59-78.

MAGLIONE (C.), 1968. - Présence de gay1ussit.e et de trona dans les ’ natronières ’ du Kanem (pourtour nord-est du Lac Trhad). Bull. Soc. fr. Minéral. Cristallogr., 91 : 388- 395.

MAGLIONE (G.), 1969. - Premières donnees sur le régime hydrogéochimique des lacs permanents du Kanem (Tchad). Cuh. ORSTOM, sér. Hydrobiol., 3 : 121-141.

M.~GLIONE (G.), 1976. - Géochimie des évaporites et. silicates néoformé en milieu continental confiné. Les dépressions int,erdunaires du Tchad. Trao. et dot. Orstom, no 50 : 335 1’.


MAHLIS (A. M.), MORCOS (S. A.) and EL-WAKEEL (S. K.), 1970. - The major anions in Lake Maryut waters. Hydro- biologia, 36 : 275-294.

MARTIN~ (M.). 1969. - La geochimica del lago Giulietti (Etio- pia). Rend. Soc. Miner. Pefrol. ifal., 25 : 65-78.

MCDOUGAL (T.), MORTON (W. H.) and WILLIAMS (111. A. J.), 1975. - Age and rates of denudation of Trap Series basalts at Blue Nile Gorge, Ethiopia. Nature, Lond., 254 : 207-209.

MCLACHLAN (A. J.), 1974. - Development of some lake ecosystems in tropical Africa, with special reference to t.he invert.ebrat.es. Biol. Reo., 49 : 365-397.

MCLACHLAN (A. J.), 1979. - The aquatic. environment : 1. Chemical and physical characteristics of Lake Chilwa. In : Lake Chilwa. Studies of change in a tropical ecosystem, Kalk (M.), McLachlan (A.J.) and Howard-Williams (C.) (eds), Monographiae biol., 35. Dr W. Junk Publ., The Hague : 59-78.

MCLACHLAN (A. J.) and MCLACHLAN (S. M.), 1976. - Develop- ment of the mud habitat during t,he filling of two new lakes. Freshmaf. Biol., 6 : 59-67.

MCLACHLAN (A.J.), MORGAN (P. R.), HOWARD-WILLIAMS(~.), MCLACHLAN (S. M.) and BOURN (D.), 1972. - Aspects of the recovery of a saline African lake following a dry period. Arch. Hydrobiol., 70 : 325-340.

MCLACHLAN (S. M.), 1970. - The influence of lake level fluctuations and the thermocline on wat,er chemistry in t,wo gradually shelving areas in Lake Kariba, Central Africa. Arch. Hydrobiol., 66 : 499-510.

MCLACHLAN (S. M.), 1971. - The rate of nutrient release from grass and dung following immersion in lake water. Hydrobiologia, 37 : 521-530.

MELACK (J. RI.), 1978. - Morphometric, physical and chemical features of the volcanic crater lakes of western Uganda. Arch. Hydrobiol., 84 : 430-453.

MELACK (J. M.), 1979a. - Photosynthesis and growth of Spi- rulina plafensis (Cyanophyta) in an equatorial lake (Lake Simbi, Kenya). Limnol. Oceanogr., 24: 753-760.

MELACK (J. M.), 1979b. - Phot.osynthet.ic rates in four tropical African fresh waters. Freshwaf. Biol., 9: 555-571.

MELACK (J. M.), 1981. - Photosynthet.ic activity of phytoplankton in tropical African soda lakes. Hydrobiologia, 81 : 71-85.

MELACK (J. M.), 1983. - Large, deep Salt, lakes : a comparative limnological analysis. Hydrobiologia, 105 : 223230.

MELACK (J. Ri.) and KILHAM (P.), 1972. - Lake Mahega : a mesothermic, sulphato-chloride lake in western Uganda. Afr. J. trop. Hydrobiol. Fish., 2 : 141-150.

MELACK (J. M.) and KILHAM (P.), 1974. - Photosynthetic rat.es of phytoplankton in East African alkaline, saline lakes. Limnol. Oceanogr., 19 : 743-755.

R~U. hydrobiol. trop. 25 (2) : U-144 (1992).

MELACK (J. M.), KILHAI\I (P.) and FISHER (T. R.), 1982. - Response of phyt.oplankton to experimental fertiliza- t.ion witb ammonium and ph0sphat.e in an African soda lake. Oecologia, 52: 321-326.

MELACK (J. M.) and MAC~NTYRE (S.), in press. - Phosphorus concentrations, supply and limitation in tropical Afri- cari iakes and rivers. In Tiessen (H.) (ed.), Phosphorus cycles in terrestrial and aquat.ic ecosystems. Regional Workshop 4 : Africa. SCOPE and UNEP, Nairobi 1991.

MESHAL (A. H.) and MORCOS (S. A.), 1980. - The relation between chlorinit.y, conduct,ivity ratio and salinity in the wat.er of Lake Qarun, Egypt. J. Cons. perm. inf. Exfilor. Mer, 39 : 20-25.

MESHAL (A. H.) and MORCOS (S. A.), 1981. - Evaporation from Lake Qarun and its water budget. Bull. Inst. Ocea- nogr. Fish. Caire. 7 : 474-453.

MONNIN (C.) and SCHOTT (J.), 1984. - Determination of t.he solubilitS; products of sodium carb0nat.e minerals and an application to tronn deposition in Lake Magadi (Kenya). Geochim. cosmochim. Acta, 48 : 571-581.

MONTEITH (J. L.), 1972. - Solar radiation and productivity in t.ropical ecosyst.ems. J. appl. Ecol., 9 : 747-766.

MORANDINI (G.), 1940. - ftlissione di sfudio a1 Lago Tana. Volume Terzo. Richerche limnologiche. Parte Prima. Geo- grafia @ca. Rome : Reale Accademia d’Italia, 319 p.

MOREAU (J.), 1982. - Le lac Thotry, lac plat hypersale (Mada- gascar). Écologie et peuplement piscicole. Reu. Hydro- biol. trop., 15 : 71-80.

MORGAN (P. R.), 1972. - Causes of mortality in t.he endemic Tilapia of Lake Chilwa (Malawi). Hydrobiofogia, 40 : 101-119.

MORGAN (P. R.), 1979. - Thr effects of t.he receding and rising lake on the fishes of Lake Chilwa. In : Lake Chilwa. Studies of change in a tropical ecosystem. Kalk (M.), McLachlan (A. J.) and Howard-Williams (C.), Monogra- phiae biologicae, 35 : 193-202. Junk Pub]., The Hague.

MORIARTY (D.-J. W.), DAHLINUTON (J.P. E.C.), DUNN (I.C.), MORIARTY (C. M.) and TELVIN (M. P.), 1973. - Feeding and prazing in Lake George, Land: B, l& : 299-319.

Uganda. Proc. R. Soc.

Moss (B.) and Mass (J.), 1969. - Aspects of the limnology of an endorheic African lake (L. Chilwa, Malawi). Ecology, 60 : 109-118.

MUSIL (C.). BORNMAN (C. H.) and GRUNOW (J. O.), 1976. - Some observed int.er-relat.ionships between the caver of aquat.ic vegetatinn and various physical properties of t,he wat.er medium. .I. S. .4fr. Bof., 42 : 157-169.

PA&S (J.), LEMASSON (L.) and DUFOUR (P.), 1979. -Éléments nutritifs et production primaire dans les lagunes de Cote d’ivoire. Cycle annuel. Arch. Sri. Cenfre Rech. Océanogr. .Ibidjan, 5 : I-60.

PAGES (J.), LEMASSON (L.) and DUFOUR (P.). 1981. - Primary production meaPurement in a hrackish tropical lagoon.

142 .J. F. TALLING

Effect of light, as studied at some stations by the 1% mrthod. Rro. Hydrobiol. trop., 14 : 3-15.

F’ETEHS (R. H.) and &~AC~NT’CRF: (S.), 1976. - Orthophosphate turnaver in East Xfrican lakes. Oecologia, 25 : 313-319.

Prrwmr. (1~. H.), 1975. - Drtermination of hydroxytamine in Ethiopian rivers and lakes. Mikrochim. Acfa, 2 : 425 I-9.

F’R»WBE ((;. A.) and TALLING (J. F.), 1958. - The seasonal growth and succession of plankton algae in the White Nile. Limnol. Oceunogr., 3 : 223-238.

HIC:FIAH~~~ON (J. L.), HARVEY (T. J.) and HOLDSHIP (S. A.), 1!1ï8. - Diatoms in t.he historv of shallow East African lakrs. Pol&ie .lrchwm Hydrohiol., 25 : 341-353.

RICAAHDSON (J. L.) and RI(:HARDSON (A. E.), 1971. - Hist.ory of ;~II .African rift. lake and its climatic implications. Ed. Mfmogr.. 32 : 199-534.

ROBARTR (R. D.), 1979. - Underwater light. penetrat.ion, chlo- rophgll a and primary product.ion-in a tropical African Iake (Lake MtrIiwaine, Hhodesia). Arch. Hydrobiol.. 86 : 123-41-l.

ROR4RTS (R. D.) arid ALL.ANL;ON (B. R.), 1977. - Meromixis in the lake-like upper reaches of a Sout.h African estuary. .Irch. Hydrobiol., HO : 531-5-10.

R~INSON (A.), I!i& - Nntes on diurnal and seasonal changes in t.emperature and oxygen regimrs in Lake Chad. Rep. fd. F’iah. Sera., Nigeria, Lake Chad Res. Sfn., 196667 : 26-3 1.

RODHE (H.), MIWOLWE (E.) and SCIEDERLUND (R.), 1981. - Chemiral composition of prrcipitat.ion in East. Africa. Icenya J. Sci. Technol. Ser. .A, 2 : 3-11.

ROLEHR (K. H.j antl BRWN (C;. hi.), 1980. - Growth and reproduction of Potumogefon crispus in a Sout.h African Inkr. .I. Ecol.. 68 : 501-571.

ROGERS (K. H.) and BREEN (C. M.), 1982. - Decomposition of Potumogefon crispus L. : the efferts of drying on t.he pattern of mass and nutrient toss. Aquaf. Bof.. 12 : l-12.

HOI;EH~ IG. K.), EI.LERY (PV. N.). WINTERNITZ (N. L.) and I)OH’LIEIER (R.), 1989. - Physical, chemical and biotic reip~~nses to tiecreasing water dept.h in a highveld pan following wet and dry summers. Soufhern Afr. J. uquuf. Sri., 15 : 67-96.

Roou (P. M.) and KERR (J. D.), 1985. - Temperat.ures expe- rienced hy thr float-ing weed Suloinia molesta Mitchell and their prrdiction from meteorological data. .4quaf. Bof.. 16 : 91-103.

Rz8s~4 (J.). 195%. - Observations on t.ropical rainpools and grneral remarks on temporary wat.ers. Hydrobiologia, 1; : PGP2XIi.

Rzosx~ (J.). 1973. - The LJpper Nilr swamps, a tropical wettand study. Freshwuf. Biol., 4 : I-30.

R~DSKA (J.), (ed.), 1976. - Thr Nile, hiology of an ancient

river. Monogr. Riol., 29. Dr W. Junk Publ., The Hague, 417 p.

RZOSKA (J.), 1983. - Temporary waters. In: The Sahara, Cloudsley-Thompson (J.) (ed.). Key Environments - The Sahara. Pergamon Press.

RZOSKA (J.), 1985. - Wat.er quality and hydrobiology of the Niger. - In : The Niger and ifs neighbours, Grove (A. T.) (ed.), Balkema Publ., Net.herlands : 77-99.

SAAD (M. A. H.), 1973. - Distribution of phosphat.es in Lake Mariut., a hsavily polluted lake in Egypt.. lt’at. Air & Soif Polluf., 2 : 515-522 ta< .

SAAD (M. A. H.), 1971;. - Some limnological investigat.ions of Lake Edku, Egypt. .Jrch. Hydrobiol., 77 : 411-80.

SAaD (M. A. H.), 1!17X. - Distribution of phosphat.e, nitrit,e and si1icat.e in Lake Edku, Egypt. FTerh. inf. Serein. fheor. ungew. Limnol., 20 : 1124-l 130.

SAAD (M. A. H.). 1980. - Eutrophication of Lake Mariut., a heavity polImed lake in Egypt.. Jn : Agrochemical resi- due-biofu infrracfions in soi1 und uquafic ecosysfems : 153-163. Vienna. L~EA.

SAMAAN (A. A.) and -\RDALLAH (R. R.), 1981. - Effect. of wat.er pollutiorr on the development. of periphyton in Lake Mariut., Egypt.. Bull. Insf. Oceanogr. Fish. Cuira. 7 : 527-537.

SERRUYA (C.) (ed.), 1978. - Lake Kinneret. Monographiae biol., 32. Dr W’. Junk, Publ. The Hague, 501 p.

SERVANT (M.) and SERVANT (S.), 1983. - The lacustrine environment and its evolut.ion. In : Lake Chad. Ecology and productivity of a shallow t.ropical ecosyst.em, Car- mouze (J. P.), Durand (J. R.) and Lévkque (C.) (eds). Monogruphiue hiol., 53. Dr W. Junk Publ.. The Hague : 1 I-26.

SERVANT-VILDARY (S.), 1978. - Étude des Diat.omées et paléolimnologie du bassin Tchadien au Cénozoique superieur. (2 vols). Trurr. et dot. Orsfom, n0 Y4 (1). 329 p.

SHAHEER’ (A. H.) and YOSEF (S. F.), 1978. - The effect. of the cessation of Nile flood on the hydrographie features of Lake Manzala, Egypt.. .irrh. Hydrobiol., Y4 : 339-367.

Smit.hsonian Meteorological Tables, 1951. - Sixth Revised Edn. Washingt,on.

SY~IOENS (J. J.), I!X%. - La minéralisation des eaux nat.u- relies. In : Explor. Hydrobiol. du Bassin du Lac Bang- coeolo ef du Luapuln, Symoens (J. J.) (ed.), Bruxelles, 2 : l-199.

SYWENS (J. J.), BURGIS (M.) AND GAUDET (J. .J.) (eds), 1981. - The ecology and utilization of African inland waters. United Nat.ions Environmental Programme, Reporfs and Proceedings Ser.. 1, Nairobi. 191 p.

TALLING (J. F.), 195ïa. - The tongitudinat succession of wat.er charact.srist.ics in t.he Whit,e Nile. Hydrobiologia, 11 : 73-89.

AFRICAN SHALLOW LAKE8 143

TALLING (J. F.), 1957b. - Diurnal changes and photosynthesis in some tropical Proc. R. Soc. Lond. B, 147 : 57-83.

of stratification African waters.

TALLIN~ (J. F.), 1963. - Grigin of st.ratification in an African Rift lake. Limnol. Oceanogr., 8 : 68-78.

TALLING (J. F.), 1965. - The photosynthetic activity of phy- t.oplankton in East. African lakes. Int. Reu. ges. Hydro- biol., 50 : l-32.

TALLING (J. F.), 1966. - The annual cycle of stratification and phyt.oplankton growt.h in Lake Victoria (East Africa). Inf. Reo. ges. Hydrobiol., 51 : 545-621.

TALLING (J. F.), 1969. - The incidence of vertical mixing, and some biological and chemical consequences, in tropic.al African lakes. Verh. inf. Serein. fheor. angew. Limnol., 17 : 998-1012.

TALLING (J. F.), 1970. - Generalized and specialized features of phytoplankton as a form of photosynthetic caver. In : Predicfion and measuremenf of phofosynfhefic pro- ducfioity. Centre for Agricultural Publishing & Docu- ment.at.ion, Wageningen, Netherlands : 431-445.

TALLING (J. F.). 1976a. - Water characterist.ics. In : The Nile, biology of an ancient river, Rzoska (J.) (ed.). illonogra- phiae biol., 29. Dr W. Junk, The Hague : 357-384.

TALLING (J. F.), 1976b. - Phytoplankton : composition, development and productivity. In : The Nile, biology of an ancient. river, Rzoska (J.) (ed.). Monographiae biol., 29. Dr W. Junk, The Hague : 385-402.

TALLING (J. F.), 1980. - Some problems of aquatic environ- n1ent.s in Egypt. from a general viewpoint of Nile ecology. Waf. Supply Mgmf, 4 : 13-20.

TALLING (J. F.), 1981. - Phytoplankt.on. The condit.ions for high concent,rations of biomass. In : Ecology and ut.ili- zat,ion of African inland waters, Symoens (J. J.), Burgis (M.) & Gaudet (J. J.) (eds). Unifed Mations Enuironmenf Programme, Reports and Proc. Series, 1, Nairobi : 41-45.

TAILING (J. F.), 1986. - The seasona1it.y of phytoplankton in African lakes. Hydrobiologia, 138 : 139-160.

TALLING (J. F.), 1987. - The phyt.oplankton of Lake Victoria (East. Africa). In : Proceedings of the symposium on t.he phycology of large lakes of the world, Munawar (M.) (ed.). Arch. Hydrobiol., Beih. Ergebn. Limnol.. 25 : 229- 256.

TALLING (J. F.), 1990. - Diel and seasonal energy iransfer, st.orage and stratification in Africa,n reservoirs and lakes. Arch. Hydrobiol., Beih. Ergebri. -Limnol., 33 : 651- 660.

TALLING (J. F.) and RZOSKA (J.), 1967. - The development of plankt.on in relation to hydrological regime in t.he Blue Nile. J. Ecol., 55 : 637-662.

TALLING (J. F.) and TALLING (1. B.), 1965. - The chemical composit.ion of African lake waters. Inf. Reo. ges. Hydro- biol., 50 : 421-463.

Heu. hydrobiol. lrop. k’s (2) : X7-194 (1992).

TALLING (J. F.), Woon (R. R.), PROSSEH (M. V.) and BA~TER (R. M.), 1973. - The upper limit. of photosynthetic productivity by pbytoplankton : evidence from Ethiopian soda lakes. Freshwaf. Biol., 3 : 53-76.

TAYLOR (D. I.), 1982. - The effert of a major macrophyte regreasion upon primary production in the littoral of Swart.vlei. Arch. Hydrobiol., 96 : 345-353.

TAYLOR (M.) and AQUISE (E.), 1984. - A alimatological energy budget of Lake Titicaca (Peru/Bolivia). Verh. inf. Serein. fheor. angcw. Limnol., 22 : 1246-1251.

THOMP~ON (B. W.j, 1965. - The climat,e of Africa. Oxford Univ. Press.

THO~IPSON (K.), 1985. - Emergent. p1ant.s of permanent. and seasonally-flooded wetland. In : ThP ecology and management of African metland uegefafion, Denny (P.) (ed.). Dr W. Junk Pub]., Dordrecht : 43-107.

THOWPSON (K.) and HAMILTON (A. C.), 1983. - Peatlands and swamps of the African continent. In : Ecosysfems of fhe world. JB. Mires : swamp, bog, fen and moor. Regional sfudies, Gore (A. J. P.) (ed.). Elsevier, Amsterdam : 331- 373.

THOHNTON (.J.A.), 1986. - Nutrients in African lake ecosys- t.ems : do we know ah’? J. Limnol. Soc. Sfhn Afr., 12 : o-21.

THORNTON (J. A.) and WAL~~SLEY (R. D.), 1982. - Applicabi- 1it.y of phosphorus budget models t.o southern African mari--made lakes. Hydrobiologia. 89 : 237-245.

TUITE (C. H.), 1981. - Standing trop densities and distribu- t.ion of Spirrrlina and bent.hir diatoms in East African alkaline saline lakea. Freshroaf. Biol., 11 : 315-360.

TWINCH (A. J.) and BREEN (C. M.), 1981. - A study of nitrogen and phosphorus fluxes in rnriched isolation columns. Hydrobiologia. 77 : 49-60.

TWINCH (A. J.) and BREEN (C. M.), 1982. - A comparison of nutrient. availability measured by chemical analysis and calculat.rd from bioassag yields. Hydrobiologia, 94 : 247-255.

TWINCH (A. J.) and BREEN (C. M.), 1984. - A study of phosphorus partitioning and exrhange in a South African impoundmrnt. JVaf. Res.. 18 : 775-782.

VARESCHI (E.), 198‘2. - The rcology of Lake Nakuru (Kenya). III. Ahiotic factors and primary production. Oecologiu, 55 : 81-101.

VARLET (F.), 1978. - Le regime de la lagune Ebrie (COt.e d’ivoire). Traits physiques essentiels. Trao. ef dot. Ors- fom. 83 : 164 p.

VINER (A. B.). 1969. - The chrmist,ry of the water of Lake George, Uganda. b-erh. inf. Irein. fheor. angew. Lim- nol., lï : 289-296.

VINER (A. B.), 1973. - Rrsponses of a mixed phyt.oplankton populat.ion to nutrient. enrirhments of ammonia and

1-L-I J. F. TALLING

phosphate. and some associated acological implications. Proc. R. Soc. Lond. B, 183 : 351-370.

\‘IMXH (A. R.). 1975a. - The su~)ply of minerals to tropical rivers and lakes (Uganda). In : An introduction fo fand- wczfer interactions. Olson (.I.) (ed.). Springer-Verlag : ?%‘&l.

YINER (.A. Pl.), 1975h. - The sediment.s of Lake George (Uganda). 1. Redos potentials, oxygen consumption and carhon dioside out.put. .-lrch. Hydrobiol., 76 : lSl- I!Cï.

VINEH (.A. B.). 1!175c. - Thr srdin1ent.s of Lake George, Ilganda. 11. Release of ammonium and ph0sphat.e from ;II; undisturhed mud surface. .Irch. Hydrobiol., 76 : 36% 378.

VINER (.A. I?I.), 1975d. - The sediments of Lake George (l’ganda). TII. The uptake of phosphate. .4rch. Hydro- hiol., ïR : 3!I3110.

VINi?R (.-\. El.), 1975~. - Non-hiological factors affecting phos- JJhat? recychng in the wat.er of a tropical emrophic lake. T-erh. inf. V’erein. fheor. angeto. Limnol., 19: 1404- 1115.

VINER (.A. ES.), IDîTa. - The sediments of Lake George (Uganda). 11’. Vertical distribution of chemical faatures in relation to ecological history and nut.rient recycling. .1rrh. Hydrobiol.. HO : M-69.

\ÍNFR 1.4. B.), 19ïïb. - The influence of sediments upon nntrient rxchanges in t.ropical Iskes. In : Interactions bcfmcen scdimrnfs und fresh mater, Golterman (H. L.) (rd.). .Junk & PUDGC, Thr Hague : 216-215.

1 INFH (4. H.). 19ïïc. - Rrlationships of nitrogen and phosphorus to a tropical phytoplankt.on population. Hydro- biologie. 5? : 185-l !16.

VIN~R (A. I3.j. 1982. - Nitrogen fixation and denit.rification in srdiments ~rf two Kenyan lakes. Biofropicu, 13 : 91-98.

YINRR (.A. H.). HRBEN (G. M.). GOLTERMAN (H. L.) and THORN- TON (J. A.), 1981. - Nmrient budgets. In : The ecofogy and ufilisafion of .-lfrican inland waters, Syrnoens (J. J.), Hurgis (hl.) & Baudet. (.J. J.) (eds.), Unit.ed Nations F:nvironment Programme, Nairobi : 137-148.

VINER (-\. R.) and SMITH (1. R.), 1973. - Geographical, hist.o- ncal antl physical aspect.s of Lake George. Proc. R. Soc. Lond. B, 18-I : 23%%O.

VISSER (S.), 1961. Ghemical composition of rain water in Kampala, I.lganda, and its relat.ion t.o meteorological and topographical condit.ions. .I. geophys. RU., 66 : 3’159-3765.

VIS~ER (5. A.), 1!174. - Variations in t.he composition of waters from lakes and rivers in East and West Africa. .Ifr. J. frop. Hydrobiol. Fiah., 3 : 62-83.

VIWER (S. A.) ani1 VILL.F.NEUVE (J. P.), 1975. - Similarities an11 differences in the chemical composition of wat.ers from iVest. Central and East. africa. I’erh. inf. Serein. fhcor. nngem. Limnot.. 19 : 1416-1125.

WAHRY (S. D.) and ABDEL-M• NEIM (M. A.), 1979. - The pro- hlem of phosphorus in the eutrophic Lake Maryut, (Egypt). Esfuarine Coasfal Alar. Ski., 9 : 615-622.

WAHRY (S. D.), KINAWY (S. M.) and EL-TARBACH (T. 1.). 1978. - Ghemical charact.erist,ics of Lake Maryut, a polluted lake sout.h of Alexandria, Egypt. Esfnarine Coasfnf Mnr. Sci. 7 17-28. < 3

WALLING (D. E.), 19x4. - The sediment yields of African rivers. In : Chulltwges in .4frican hydrology and mater resources (ed. Walling (D. E.), Foster (S. S. D.) and Wur- zel (P.)) : 265283. Wallingford. IAHS Press.

WALMSLEY (R. D.). RUTTY (M.), VAN DER PIF.PEN (H.) and GR~ELER (D.). 1980. - Light penetration and the int,er- relationships het,ween optical parameters in a t.urhid suht.ropical impoundment~. Hydrobiologia, 70 : 145-157.

WANDIGA (S. O.), 1’181. - The concentrat.ions of zinc, copper, lead, manganese, nickel and fluoride in rivers and Iakes of Kenya. Sinrf : Ethiop. J. Si., 3 : 67-74.

WEIR (J. S.), 196Y. Seasonal variation in alka1init.y in pans in Çent.ral Africa. Hydrobiologia, 32 : 69-80.

\VIZIR (J. S.), 1969. - Studies on Central African pans III. Fauna and physico-chemical environment. of some ephemeral ~m:~ls. Hydrobiologia, 33 : 93-116.

WEISSER (P.), 1978. - A conceptual mode1 of siltation system in shallow lakes with Moral veget,at.ion. J. limnol. Soc. Soufh. Afr.. 4 : 145-119.

WELCOMME (R. L.), 1979. - Fisheries erology of floodplain rioers. Longman. 317 p.

WOOD (R. El.), BASTER (R. IV~.) and ~ROSSER (M. V.), 1984. - Seasonal and comparat.ive aspects of chemical strat.ification in some tropical crater lakes, Et.hiopia. Freshwat. Biol., 1-I : ?r)1-r)73. << c

WOOD (R. B.), ~ROSSER (M. V.) and BA~TER (R. M.), 1976. - The seasonal pat.t.ern of thermal characteristics of four of the Bishoftu crat.er lakes, Et.hiopia. Freshwaf. Biol., 6 : 519-530.

Woon (R. B.), PRnssEn (I+l. V.) and BASTER (R. M.), 1979. - Opt.ical choracterist.ics of the Rift Valley lakes, Ethio- pia. Sinef : Efhiop. J. Sri., 1 : 73-85.

WO~D (R. B.) and TALLIN~ (J. F.), 1988. - Chemical and algal relationships in a salinity series of Et.hiopian inland waters. Hydrobiologia. 158 : 29-67.

WORTHINGTON (E. R.), 1930. - Observations on the t.empera- turc, hydrogen-ion concent.ration, and other physica1 condit.ions of the Victoria and Albert Nyanzas. Inf. Reu. ges. Hydrobiol. Qdrogr., 21 : 328-357.

YOUNG (J. O.), 1!:175. - Srasonal and diurnal changes in the water temperatura of a temperate pond (England) and a tropical pond (Kenya). Hydrobiologia, 47 : 513-526.

YURETICH (R. F.) and CERLING (T. E.), 1933. - Hydrogeoche- mistry of Lake Turkana, Kenya : mass balance and minera1 reactions in an alkaline lake. Geochim. cosmochim. .4cfa, -17 : 1099-l 169.

Documents

Environmental vegulation in African shallow lakes and wetlands