SCOPE 42 - Biogeochemistry of Major World Rivers

  5.1 INTRODUCTION

Over 66% of South America's land mass, with a continental expanse of 77.8 x 10⁶ km² (12% of the Earth's surface), is drained by six large fluvial systems which deliver about 8000 km³/year of fresh water at the Atlantic Ocean. A remarkable feature is that about 58% of the total South American water discharge is supplied by one single river, the Amazon, which roughly occupies 40% of the continental area. This is the largest fluvial system on Earth and it is responsible¾by sheer volume¾for the significance of South America as the most important contributor to the world's runoff (Baumgartner and Reichel 1975). On the other hand, the South American continent contributes 13% of the total suspended sediments delivered by all rivers to the oceans, and this is mainly accomplished by three of the world's largest rivers: the Amazon, the Orinoco and the Paraná. They are rated first, third and ninth in terms of water discharge, and first, eighteenth and fifth in terms of drainage area (Milliman and Meade 1983).

Due to their importance, and to the relative lack of knowledge of the biogeochemistry of South American river systems¾partially in a pristine state but undergoing intense man-induced changes¾the SCOPE/UNEP Carbon Project placed special interest on the monitoring of these rivers, mainly to establish the transfer of carbon from land to sea, thus fulfilling a great need at present for more data.

 Of the six most important South American river basins¾Amazon, Orinoco, Paraná, São Francisco, Magdalena and Uruguay¾the first three were studied within the framework of the SCOPE/UNEP Project and their main biogeochemical aspects were analysed in a number of papers published in the Project's volumes (Degens 1982a; Degens et al. 1983, 1985, 1987). The remaining rivers have been subjected to less intense surveys. These results are a significant addition to the wealth of information reported during the late 70s in a number of specialized journals and in scientific papers¾which has dominantly focused on sedimentological/hydrological, hydrochemical and biological aspects. The SCOPE/UNEP Project, however, was a timely venture which has helped to overcome lack of precision in the published data (Degens 1982b).

Although this chapter deals with South American rivers, special attention was given to two case studies: the Paraná and the Orinoco Rivers. The Amazon has been considered elsewhere (Richey et al. , this volume, Chapter 3).

5.2 ENVIRONMENTAL CONDITIONS

South America exhibits ample variation in its geology, climate, relief and resulting biota. These variables interact in elaborate ways to imprint distinctive features in each major river system.

Clearly, geology/relief is of dominant importance and its main elements can be briefly described as follows: (a) the long¾over 7200 km¾and folded mountainous chain of the Andes of Tertiary orogeny, the continental back-bone along its western margin; (b) the Brazilian and Guyana Precambrian shields, with their metamorphic and igneous components; (c) the lowlands, with 20% of the total continental area, extending from the mouth of the Orinoco to Bahia Blanca, in Argentina. Outstanding features are the thick sediment sequences of the three largest drainage systems and the loess- mantled Chaco-Pampa plains; and (d) the Patagonian plateau. Furthermore, the central and eastern parts consist of a series of plateaus and extensive depressions filled with alluvia. Thus, the margins of the continents are often higher than the interior, a characteristic that produces profound effects on the pattern of drainage. A detailed description of the continent's geology is given by FAO-UNESCO (1971).

Due to the fact that the South American continent extends over 67° of latitude, it is not surprising that it presents a great variety of climate although 80% lies within the tropical zone (Schwerdtfeger 1976).

Three factors (i.e. circulation of air over the continent, ocean currents and altitude) have strong influences on the climate of South America, determining several well-marked climatic regions. According to Koeppen's classification, four major types of climates occur in the continent: tropical rain (A), arid (B), temperate (C) and polar (E) (Eidt 1968).

 The rivers concerned are concentrated in the tropical-subtropical zone, where annual atmospheric precipitations of 1000-2000 mm/year are common and certain regions have a rainfall which even exceeds 4000 mm/year. Aside from some Andean tributaries, the contribution from meltwater to annual runoff is relatively scarce.

The large extension of the South American continent and the presence of numerous and varied mountain ranges make possible a great variety of vegetation types. According to Scholten (FAO-UNESCO, 1971) the natural plant cover of South America may be divided into ten main ecological units, of which the forest vegetation is dominant. The most important regions are:

1. Tropical: wet evergreen forests (tropical forests) exist if there is no prolonged dry period and if the mean annual rainfall exceeds a minimum of 1600 mm. These forests are formed by a large number of woody species of which most have evergreen foliage and cover almost the whole Amazon basin, parts of the Guyanese and Brazilian Shields (highlands), the Pacific coast of Colombia and northern Ecuador.

2. Tropical seasonal forests: the climate is characterized by a prolonged seasonal dry season. This region comprises the belt around the savannahs of the 'llanos del Orinoco', parts of the Amazon drainage basin in Brazil and Bolivia, eastern and southern Brazil, east of Paraguay, and the Misiones Province in Argentina.

3. Tropical savannahs: the term corresponds to any grassland, with or without trees, natural or man-made. The climate in general is characterized by a tropical seasonal regime of rainy summers and dry winters, which occupies a vast area in central Brazil, the 'llanos' of the Orinoco River, part of the Guyana highlands, the 'Gran Pajonal' in Peru, the alluvial 'Pampas de Majos' in Santa Cruz (Bolivia), and the 'Pantanal' (Upper Paraguay basin).

4. Andean mountain forests of the tropics: these forests stand on the humid slopes of the Andean ranges from the coasts of Venezuela and the Santa Marta Massif in Colombia to the Tucumán area in northwestern Argentina.

5. Temperate forests: they occur in central and southern Chile and in the uplands of southern Brazil.

6. Temperate natural grasslands (pampas): the climate is warm temperate to subtropical, and precipitation varies from 500 mm/year (Bahía Blanca) to 1400 mm/year (Río Grande do Sul). The vegetation consists exclusively of herbaceous plants, dominated by species of Stipa. They are largely distributed in southern Brazil, Uruguay and Argentina.

7. Semi-arid formations: small trees and often thorny shrubs (prosopis, acacias, mimosas) coyer regions with alternate dry and relatively wet seasons. The precipitation is below 1000 mm/year. These regions include the coastal part of Venezuela, northern Brazil and the Chaco region (southern Bolivia, western Paraguay and northern Argentina).
   The vegetation of South America has been extensively treated by Hueck (1966), Weber (1968), Hueck and Seibert (1972) and Walter (1984), among others.

Following the previous description, it is evident that South America holds some of the largest phytomass pools on Earth, with vast areas considered as substantial carbon sources. Modeling efforts (Lieth and Esser 1985) attribute to them accumulation values between 1000 and over 2500 g C/m², for the timespan 1860-1981. Clearly, the Amazonian rain forest is the most important, with links which affect the transfer of carbon from the continent to the sea via the system's tributaries. In fact, all three major river systems (Amazon, Orinoco and Paraná) are affected by deforestation and rapid changes in the use of the land, thus accelerating the transfer of particulate (POM) and dissolved (DOM) organic matter, from the river basins to the sea.

5.3 DRAINAGE

Figure 5.1 depicts the vast drainage network of the South American continent. Approximately 90% of the continental water runs towards the Atlantic Ocean, and only one major riverine system¾the Magdalena¾pours its discharge towards the northern side. There are no major rivers on the western seaboard, although the large glacial lakes of southwestern Patagonia (Nahuel Huapi, Buenos Aires, Pueyrredón, Argentino, etc.), and the Titicaca and Poopo lakes in Bolivia are significant fresh water masses with insufficiently known biogeochemical characteristics. Lastly, the closed basins or drainage-less areas amount to less than 4.5% of the continental expanse.

The most important drainage system is the Amazon River which, with a length of 6577 km, is second only to the Nile. The volume of water supplied by the Amazon to the ocean (mean discharge of 5500 km³/year) represents 15% of the total fresh water existing in the world (Salati 1985).

The Amazon drains about 33% of South America and has more than 1000 tributaries, seven of which are more than 1600 km long. Its hydrograph shows only one peak, with a minimum at the end of October or early November, and a maximum at the end of May or early June. At Tefé, the difference between the high and low water mark is 15.2 m and at Obidos it is 7.6 m (Sioli 1975).

The Paraná River is the second largest drainage system in South America, with a basin area that exceeds 2.6 x 10⁶ km², a length close to 4000 km and a total annual water discharge of 500 km³. A number of tributaries are members of the drainage net: from the mountainous Andean environment, the Bermejo and Pilcomayo Rivers; the Paraguay, from the tropical lowlands of Matto Grosso, and the Iguazú, originating in the rainy slopes of Serra do Mar; and finally, the Salado and Carcarañá rivers which drain the flat Chaco- Pampa loess-mantled plains.

Figure 5.1 Map showing South America's major river basins

The headwaters are located in the Serra dos Preneos (Brazil), close to 15°30' S. From the headwaters, down to the confluence with the Paraguay River¾what is known as the Upper Paraná ¾the river has a relatively clearly defined channel with a restricted floodplain.

The Paraguay River, whose headwaters are located close to 14°20' S, has a very large¾about 80 000 km²¾marshy feature, known as 'El Gran Pantanal'. This little known area, about 250 km wide, probably has a significant impact on the river's hydrochemistry, as it has on its hydrology. The Bermejo River, Paraguay's main tributary, exhibits a pronounced slope in the headwaters (1.12 m/km) which contributes to a heavy sediment load during the rainy season (February-March).

In the 1200 km which remain to the outfall¾the middle course¾the river develops a floodplain valley with a width which varies between 10 and 40 km, easily flooded along most of its right margin.

The Paraná, before joining the Uruguay River to give birth to the Río de la Plata, forms an ample inner delta, 320 km long and over 60 km wide. This delta, like the floodplain in the middle course, is a highly complex geomorphological feature with numerous channels and streams, ponds, ox-bows, lakes and backwaters, separated by levees and bars of variable extension and height. The process of inundation of the floodplain is periodic in nature, and has a large impact¾as will be seen¾on the system's biogeochemistry.

At medium to low water levels, about 10% of the flood valley is occupied by the main channel, and roughly 2-5% by secondary channels and streams of variable size (lotic environment type). Numerous ponds and marshy areas occupy between 20% and 50% of the valley's total area. These proportions, which are similar to those determined for the Amazon (Junk 1985) and for other large rivers with developed floodplains, clearly indicate the enormous area of contact between water and land, thus affecting a number of biogeochemical aspects.

The Orinoco, 2150 km long, is the third largest drainage basin covering about 1.0 x 10⁶ km². In terms of discharge it ranks third in the world, after the Amazon and Zaire rivers (Paolini et al. 1987), with a mean annual discharge of 1100 km³/year. The annual hydrograph reaches the maximum during mid-August and early September, and the minimum during the months of February and March. By means of the Casiquiare channel, the Orinoco is linked to the Amazon system through the Río Negro.

The Orinoco drainage system drains the northern part of South America and its watershed is developed within Venezuela and Colombia. Born in Sierra Parima at an elevation of 1074 m, it receives more than 2000 rivers on its way to the Atlantic. Its major tributaries can be divided as: (a) right bank tributaries, whose headwaters are located in the Guyana Shield. They are of acidic nature, black in color and poor in electrolytes. Tropical rain forest covers most of these old erosional surfaces  (Precambrian) with the exception of some areas in the north and southeast, where savannah-type vegetation predominates; (b) left bank tributaries, originating in the Andes or in the 'llanos', rich in electrolytes and in suspended sediments. The headwaters of some of these tributaries are located in shallow-marine and brackish water shales, sandstones and carbonates of the Andes. Others originate in the plains, a region with Quaternary fluvio-lacustrine sediments. The vegetation is savannah-type with gallery forests along the river banks.

The São Francisco is another one of the large rivers of South America. It is the third longest river system of Brazil (2900 km) and its basin has 631 000 km². São Francisco's headwaters are on the eastern slopes of the Serra de Canastra in southwestern Minas Gerais State. It flows northward for more than 1600 km across the State of Bahia before curving eastward to the Atlantic. Its average discharge is about 120 km³/year, with maximum peak discharge in February and minimum values in December (Paredes et al. 1983). The river is an important waterway to the interior of eastern Brazil, and three large dams (Tres Marías, Sobradinho and Paulo Alfonso) are located along its course for hydroelectric production, urban and industrial water supply and irrigation of agricultural crops.

The Magdalena River in Colombia drains an area of 257 000 km² with a total length of 1316 km. Its main tributaries are the Cauca, the Sogamoso, the César and the San Jorge. The latter covers approximately 50% of the watershed. The regime of discharge is bimodal, due to the distribution of precipitation, with low values in January-February (about 4000 m³/s). The river discharge increases from March onwards and reaches peaks of 10 000 m³/s in November-December (Ducharne 1975). The mean discharge is 6800 m³/s (215 km³/year) at its mouth in the Caribbean Sea. One important fact on the Magdalena is that about 80% of Colombia's population is living on its watershed, leading to a demographic density of 54 inhabitants/km² which is relatively high when compared to 0.24 inhabitants/km² in Amazônia. Consequently, problems of water pollution have appeared, especially in the lower course (Serruya and Pollingher 1984).

The Uruguay River joins the Paraná to form the Río de la Plata. Its basin is rated sixth among the large rivers of South America and its annual discharge of 145 km³ is regulated by Salto Grande, a large dam whose reservoir holds some 5 km³ of water .

The drainage to the Pacific represents 5.6% of South America's fresh waters, due to the proximity of the Andes to the Pacific coast and the scarcity of rains falling in the region comprised from southern Ecuador to northern Chile. Consequently, the rivers are very short and only a few of them have water throughout the year, as the Guayas in Ecuador, the Santa (110m³/s) in Peru, and the Aconcagua (79 m³/s), the Mapo (102 m³/s), the Maule (250 m³/s) and the Bio-Bio (350 m³/s) in Chile.

Seasonal floodplains are important in most South American rivers. Around the Caribbean Sea, floodplains originate from internal or coastal deltas, which contain large numbers of permanent or semipermanent lakes, locally termed 'ciénagas'. The largest of these plains is the one of the Magdalena River of which the internal delta¾with the San Jorge and Cauca Rivers¾ extends over 20 000 km² of savannah. The Atrato River, also in Colombia, covers 5300 km² and the Catatumbo River in Venezuela, 5000 km².

Several vast areas subjected to sheet flooding are found on the continent. The largest is the already mentioned Gran Pantanal in the Paraguay River basin, the shallow interconnecting net of lakes of which extends over 100 000 km². The 'llanos' in Venezuela and Colombia, between the Apure and Meta rivers (about 70 000 km²) are also subjected annually to shallow sheet flooding.

The Amazon's floodplain, 'várzea', varies from 20 to 100 km in width. It is limited by the slopes of the 'terra firme' and covers more than 50 000 km². The soils developed on the 'várzea' are rich in nutrients and highly productive, being fertilized yearly by flooding water (Salati 1985). Along the Paraguay River¾below the 'Pantanal'¾and the Paraná, down to its mouth, the floodplain covers an area of over 25 000 km². The coastal floodplains of the Orinoco and the Amazon have been estimated at 20 000 and 25 000 km², respectively.

The water bodies (lakes and lagoons) developed on floodplains are among the most important aquatic habitats in the tropics (Welcomme 1979). Junk (1985) pointed out that the floodplain is an important carbon source to the main channel and Novoa et al. (1984) and Novoa (1982) underlined the fact that the floodplain in the Orinoco is a highly productive area for fisheries. 

Table 5.1 General features of the South American rivers

Table 5.1 summarizes the most important features of the largest South American rivers.

5.4 BIOGEOCHEMICAL ASPECTS OF THE PARANÁ AND ORINOCO RIVERS

Two of the large rivers in South America, the Paraná and the Orinoco, have been selected as case studies to reflect the rates of continental transport of carbon to the sea, as well as the nature of the organic matter derived from autochthonous and allochthonous sources. All the aspects that have just been examined, but mostly the morphological characteristics of the rivers, their hydrology and the processes that take place within their basins, are the driving variables determining the above-mentioned carbon dynamics.

5.4.1 THE PARANÁ RIVER

Hydrochemical, sedimentological and ecological features of the Paraná have been reported during the last decades in several papers with international reach (e.g. Depetris and Griffin 1968; Bonetto 1975; Depetris 1976; Kempe 1982).

Table 5.2 summarizes the main characteristics of Paraná's hydrochemistry, as determined within the framework of the SCOPE/UNEP Carbon Project.

Suspended sediments in the Paraná are transported at a mean rate of about 80 x 10⁶ t/year. There is conclusive evidence, however, that this rate could fluctuate between 40 X 10⁶ t/year and over 100 x 10⁶ t/year, according to the prevailing hydrological conditions of the system. The above-cited mean value-based upon depth¾integrated sampling¾represents a sediment concentration of 170 mg/l for a mean discharge of 15 000 m³/s.

Most of Paraná's suspended load is supplied by the Bermejo River, the main Andean tributary of the Paraguay, whose mean sediment concentration is about 4500 mg/l (Drago and Amsler 1981).

Over 45 samples collected in the period from March 1981 to November 1984 allowed the calculation of a mean POC concentration of 4.5% (S_x = 5.1). The POC discharge-weighted mean for the same period was 2.1 mg/l, while individual determinations showed a trend to increase with increasing sediment concentrations and a moderate dilution with increasing discharges.

The allochthonous fraction (organic debris from plants) of POC probably plays a dominant role over autochthonous materials (fresh water plankton) when initial overbank conditions are reached. If, however, the river reaches higher water levels, mean POC values decrease as does the mass transport rate, and planktonic material becomes most likely the dominant fraction of POC.

Table 5.2 Paraná River basic hydrochemical data

During the period of sampling, the 1982 El Niño/Southern Oscillation (ENSO) event triggered flooding conditions of exceptional magnitude, allowing the possibility of assessing its impact on the Paraná River system. The collected data showed a POC discharge-weighted mean of 3.5 mg/l during pre-ENSO conditions (16.3 x 10⁵ t/year) which decreased to 1.3 mg/l (9.3 x 10⁵ t/year) during the ENSO-affected hydrology. Conversely, DOC increased its discharge-weighted mean from 6.1 up to 10.2 mg/l and more than doubled the transport rate, from 2.8 up to 7.5 x 10⁶ t/year .

Clearly, DOC is far more significant than POC in the Paraná. DOC/POC is 4.4; under normal hydrologic conditions the ratio is about 2, and it can reach 8 or more during the highest water stages. Although there is no clear-cut relationship with discharge (Figure 5.2), DOC concentrations are high during flooding, probably due to the contribution of organics leached from soils present in the drainage basin and also to the influence of material decomposed in the floodplain. Further, DOC concentrations form hysteresis loops¾clockwise and counterclockwise¾if plotted against discharge values, showing the dependence of the states of the system on its previous history (Depetris and Cascante 1985).

As mentioned before, the Paraná has an ample floodplain (about 25 000 km²) where organic matter is produced and temporarily stored. Decomposition and remineralization of POC occurs in this environment, substantially increasing the DOC content of ponds, ox-bows, etc., and thus raising DOC concentrations when these waters are introduced into the mainstream during overbank stages. Thus controlled by the system's hydrology, DOC can reach high values (20 mg/l or more) in a pulse-like fashion (Figure 5.2). This mechanism has been described for systems with well-developed floodplains and is valid for many of the world's rivers ( e.g. Ittekkot and Arain 1986).

Figure 5.2 Variation of DOC (0-50 mg/l), POC (0-15 mg/l) and gauge height (0-7 m) in a station located about 600 km upstream from the mouth in the Paraná River . Sequence starts in March 1981 and ends in November 1984

Data published by lttekkot et al. (1982) on dissolved carbohydrates (DCHO)¾215¾681 µg/l ¾and amino acids (DAA)¾157¾842µg/l¾in Paraná's waters, allow some additional comments on the biogeochemical processes taking place within river systems. Due to its significance as dominant constituents of all forms of organic matter, glucose is the most abundant sugar followed by variable proportions of fructose, mannose, galactose, etc.

As with other rivers (e.g. Ganges, Indus, Orinoco and Brahmaputra) (Ittekkot et al. 1982), DCHO concentrations increase with increasing discharge. In the Paraná, this is even more evident with DAA concentrations (Depetris and Lenardón 1982, 1983), when a clear overbank stage is reached, with discharges of 17 000 m³/s or higher (Figure 5.3). Clearly, the floodplain is the likely source of DCHO and DAA under these conditions of early overbank flooding. The percentage of total DCHO-carbon and DAA-carbon (Ittekkot et al. 1982) fluctuated between 0.6% and 8.0%, with a moderate tendency to increase the contribution with increasing discharge.

The Paraná River transports about 10.2 x 10⁶ t/year of total carbon. About 88% is accounted for by total dissolved carbon (organic plus inorganic).

Figure 5.3 Third-degree polynomial fitted to dissolved amino acids (DAA) data from the Paraná River. Initial overbank stage coincides with the minimum of curve. Analytical data from Ittekkot et al. (1982)

5.4.2 THE ORINOCO RIVER

According to Meade et al. (1983) and Meade (pers. comm.) the left bank tributaries contribute about 95% of the total suspended sediment load of the Orinoco River, although the water discharged is supplied in approximately equal amounts by the Guyana Shield and the Andes-'llanos'.

The chemistry of the Orinoco has been poorly documented (Edwards and Thornes 1970; MOP 1972) but during the last six years the research efforts on its watershed have increased noticeably (Németh et al. 1982; Escandon 1983; Paolini et al. 1983, 1987; Sanchez et al. 1985; Weibezahn, 1985; Lewis et al. 1986; Contreras, 1988).

Tables 5.3 and 5.4 summarize the main characteristics of both types of tributaries. The left-bank tributaries show high values of pH, conductivity, total cations, alkalinity, suspended load and turbidity, when they are compared with the right bank tributaries. Outstanding exceptions are the Cinaruco, Vichada, Inirida and Capanaparo, whose headwaters are located in the 'llanos' where oxisols and quartzipsamments dominate. Clearly, the natural environment (lithology, vegetation and soils) is dominant in determining the water chemistry of these fresh waters (Vegas-Vilarrubia and Paolini 1985).

Although the presented data correspond only to the rainy season, the seasonal variations of the physical and chemical parameters for the Orinoco main stem and some tributaries have been studied by several authors (Németh et al. 1982; Paolini, 1986; Paolini et al. 1983,1987; Weibezahn 1985; Lewis et al. 1986; Contreras 1988). All the works have shown that the concentration of most chemical species is clearly related to Orinoco's discharge, obtaining the highest values during low discharge, i.e., a classical dilution effect.

Table 5.3 Left-bank tributaries of the Orinoco (Plains and Andes)

 Table 5.4 Right-bank tributaries of the Orinoco (Guyana Shield)

Table 5.5 shows the concentrations and mass transport rates calculated by Paolini et al. (1987) for the Orinoco River at Ciudad Bolivar. Clearly, Orinoco's waters are diluted in their dissolved components, but the basin yield is significant if one takes into account the mass of water delivered to the ocean (1135 km³/year). The dissolved load is 30.5 x 10⁶ t/year, this value is eight times lower than that of the Amazon (Meybeck 1976) and similar to the Paraná River (Depetris 1976).

Table 5.5 Orinoco River basic hydrochemical data from 1983 to 1984

With respect to dissolved carbon (DOC and DIC), a sharp contrast is evident between left and right bank tributaries. In the left bank, inorganic carbon predominates (i.e. high alkalinity) while in the right bank organic carbon is more important (i.e. high DOC and color values). Although it is not known yet how much DOC is in the form of humic substances in the Guyana Shield tributaries, it seems probable¾according to spectroscopic and molecular properties¾that humic substances of the fulvic acid type (low molecular weight) are dominant (Vegas et al. 1988).

Figure 5.4 shows the variation of DOC, POC and TSS with discharge in the Orinoco River during the period 1981-82. Clearly, there is a good correlation between POC and total suspended solids (TSS), indicating that a common process controls their concentrations. POC varied between 1.5% and 6% of the suspended fraction; the lowest corresponded to the rising water stage in May while the maximum was registered during the low water stage in February, following a scheme similar to the one pointed out by Ittekkot et al. (1986). 

Figure 5.4 Variation of DOC (0-12mg/l), POC (0-12mg/l), TSS (0-200mg/l) and discharge (0-70 000m³/s) in the Orinoco River (Ciudad Bolivar). Sequence starts in February 1981 and ends in June 1982 

During the same period, DOC showed a more blurred image, with the maximum concentration (10.8 mg/l) in July, one month previous to maximum discharge . 

Ittekkot et al. (1982) published data on dissolved carbohydrates (DCHO)¾103¾970 µg/l¾and amino acids (DAA)¾65¾284 µg/l. Total DCHO concentrations follow the discharge values (r = 0.73, 90% significance) suggesting the floodplain as the source when overbank conditions are reached. Conversely, DAA do not show correlations with fluctuating discharge. 

Arabinose, one of the most abundant sugars in the Orinoco samples, is closely related to discharge (r = 0.85,95% significance) as well as non-protein DAA (e.g. g-amino butyric acid, .b-alanine), suggesting a microbial reworking of organic debris, possibly within the floodplain. Total DCHO concentrations in the Caroni, the most important black water tributary¾124¾656 µg/l¾are close to those determined in the Orinoco.

The fraction of DOC accounted for by carbon from DAA and DCHO fluctuated between 1.9% and 9.7% , with a tendency to reach higher contributions to DOC when discharges are also high.

For the total carbon transport, Paolini et al. (1983, 1987) calculated 8.8 x 10⁶ t/year and 7.7 x 10⁶ t/year for 1981-82 and 1983-84 respectively. About 75% is accounted for by total dissolved carbon (organic plus inorganic).

5.5 CONCLUDING REMARKS

The results obtained in the Paraná and Orinoco Rivers within the framework of the SCOPE/UNEP Project, 'Transport of Carbon and Minerals in Major World Rivers', now allow the projection of a clearer image of the biogeochemical processes taking place in the rivers concerned.

Table 5.6 summarizes data collected in both systems. It is quite clear that, although the rivers exhibit comparable discharge-weighted means and transport rates, the total organic carbon yield of the Orinoco is more than twice that of the Paraná. This difference is surely reflecting the relative importance of the tropical environment type in the Orinoco.

In connection with the general mechanisms governing the concentrations of carbon in both riverine systems, it is possible to conclude that:

1. the POC concentrations are often positively correlated with TSS, particularly during the rising stages; increasing discharges trigger a dilution effect;

2. DOC/POC ratios are always higher than 1, and often higher than 2 or 3, a distinctive characteristic for the rivers in South America;

3. floodplains are a clear source of POC and DOC, exporting dissolved and particulate organic constituents during the rising, peak and receding water stages, following a scheme similar to the one proposed by lttekkot et al. (1986) for the Ganges-Brahmaputra and Indus;

4. DOC concentrations exhibit an increasing trend with increasing discharge; flooding of exceptional magnitude¾such as the ENSO-triggered one in the Paraná¾can double or triple the DOC flux of a large river system;

5. although many more data are needed, the available information on DCHO and DAA suggest some reworking of organic matter, possibly within the extensive floodplains. Further, additional research should seek explanations for differences or similarities in DCHO and DAA composition among diverse rivers (e.g. DCHO in Figure 5.5);

6.  the labile, metabolizable fraction of POC was investigated for most of the world's rivers (Degens and lttekkot 1985). The Orinoco and Paraná showed high values for amino acids, higher than 20% , while the Orinoco also exhibited one of the highest total sugar contributions, more than 10% . The rivers of South America have the world's highest percentage of metabolizable carbon in the POC fraction;

7. although more data are needed, particularly in some of South America's rivers (e.g, São Francisco and Magdalena), much is being done to fill up the gap. The Uruguay River, for example, has been monitored in the Salto Grande dam during 1986 (Mañosa, pers. comm.) and now information exists on the variability of some of its biogeochemical parameters: DOC, 1.6-8.2 mg/l; POC, 0.36-1.15 mg/l; DIC, 1.6-4.7 mg/l; PN, 0.07-0.9 mg/l; estimated DOC transport rate, 0.5 x 10⁶ t/year; estimated POC transport rate, 0.12 x 10⁶t/year);

8. from the present worldwide flux estimate of 0.53-0.82 x 10⁹ t/year for TOC (Degens and lttekkot 1985), the rivers of South America contribute 8-13% (Kempe 1985). This is considered a minimum estimate, since most studies have been performed with samples collected from the surface, without the benefit of depth-integrated sampling and bedload determinations.

Table 5.6 Basic biogeochemical data for the Paraná and Orinoco Rivers

Figure 5.5 Relative proportions of dissolved carbohydrates (DCHO) in the Paraná, Orinoco and Caroni Rivers. RHA = rhamnose; RIB = ribose; MAN = mannose; FRU = fructose; ARA = arabinose; FUC = fucose; GAL = galactose; XYL = xylose; GLC = glucose. Basic data from Ittekkot et al. (1982)

REFERENCES

Baumgartner, A. and Reichel, E. (1975) The World Water Balance, R. Oldenbourg, Munich, Vienna, 179pp.

Bonetto, A. A. (1975) Hydrologic regime of the Paraná River and its influence on ecosystems. In: Hasler, A. D. (Ed.) Coupling of Land and Water Systems, Springer, pp.175-97.

Cascante, E., Giombi, N. and Depetris, P. J. (1985). Abundances and fluxes of inorganic particulate and dissolved phases in the Paraná River (Argentina). In: Degens, E. T., Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/ UNEP Sonderbd. 58, pp. 305-10.

Contreras, M. (1988) Hidroquímica de los ríos Cojedes, Acarigua, Portuguesa, Morador y Guanare. M.Sc. Thesis CEA-IVIC, Caracas, Venezuela.

Degens, E. T. (Ed.) (1982a) Transport of Carbon and Minerals in Major World Rivers, Pt. 1. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 52, 766pp.

Degens, E. T. (1982b) Riverine carbon¾An overview. In: Degens, E. T. (Ed.) Transport of Carbon and Minerals in Major World Rivers, Pt. 1. Mitt. Geol-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 52, pp. 1-12.

Degens, E. T. and Ittekkot, V. (1985) Particulate organic carbon¾An overview. In: Degens, E. T., Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/ UNEP Sonderbd. 58, pp. 7-27.

Degens, E. T., Kempe, S. and Soliman, S. (Eds) (1983) Transport of Carbon and Minerals in Major World Rivers, Pt. 2. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 55, 535pp.

Degens, E. T., Kempe, S. and Herrera, R. (Eds) (1985) Transport of Carbon and Minerals in Major World Rivers, Pt.3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 58, 645pp.

Degens, E. T., Kempe, S. and Gall Weibin (Eds) (1987) Transport of Carbon and Minerals in Majõr World Rivers, Pt. 4. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 64, 512pp.

Depetris, P. J. (1976) Hydrochemistry of the Paraná River. Limnol. Oceanogr. 21(5), 736-9.

Depetris, P. J. and Cascante E (1985) Carbon transport in the Paraná River. In: Degens, E. T. Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/ UNEP Sonderbd. 58, pp. 209-304.

Depetris, P. J. and Griffin, J. J. (1968) Suspended load in the Río de la Plata Drainage Basin. Sedimentology 11, 53-60.

Depetris, P. J. and Lenardón, A. M. (1982) Particulate and dissolved phases in the Paraná River. In: Degens, E. T. (Ed. ) Transport of Carbon and Minerals in Major World Rivers, Pt.1. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 52, pp. 385-95.

Depetris, P. J. and Lenardón, A. M. (1983) A second report on: Particulate and dissolved phases in the Paraná River. In: Degens, E. T., Kempe, S. and Soliman, H. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt.2. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 55, pp. 167-81.

Drago, E. C. and Amsler, M. L. (1981) Sedimentos suspendidos en el tramo medio del río Paraná: Variaciones temporales e influencia de los principales tributarios. Rev. Asoc. Cienc Nat. Litoral 12, 28-43.

Ducharne, A. (1975) Informe técnico de biología pesquera (Limnologia). Proyecto para el Desarrollo de la Pesca Continental. INDERENA-FAO Publ. 4 DP/COL 71552/4. Bogotá, Colombia.

Edwards, A. M. and Thornes, J. B. (1970) Observations on the dissolved solids of the Casiquiare and upper Río Negro, April-June 1968. Amazoniana (Kiel) 2(3),245- 56.

Eidt, R. C. (1968) The climatology of South America. In: Fittkau, E. J., Illies, J., Klinge, H., Schwabe, G. H. and Sioli, H. (Eds) Biogeography and Ecology of South America, Junk, The Hague.

Escandon ,M. (1983) Caracterización geoquímica de las aguas en la cuenca del río Orinoco (Tramo Río Arauca-río Caroní). B.Sc. Thesis. Universidad Central de Venezuela, Faculty of Sciences, 206pp.

FAO-UNESCO (1971) Soil Map of the World 1: 5,000,000, Vol. IV, South America, UNESCO, Paris, 193pp. 

Furch, K. (1984) Water chemistry of the Amazon basin: The distribution of chemical elements among freshwaters. In: Sioli, H. (Ed.) The Amazon. Limnology and Landscape Ecology of a Mighty Tropical River and its Basin, W. Junk, Dordrecht, pp. 167-99.

Hueck, K. (1966) Die Wälder Südamerikas Ökologie, Zusammensetzung und wirt- schaftliche Bedeutung, G. Fischer Verlag, Stuttgart.

Hueck, K. and Seibert, P. (1972) Vegetationskarte von Südamerika, G. Fischer Verlag, Stuttgart. 70pp.

Ittekkot, V. and Arain, R. (1986) Nature of particulate organic matter in the river Indus, Pakistan. Geochim. Cosmochim. Acta 50, 1643-53.

Ittekkot, V. , Spitzy, A. and Lammerz, U. (1982) Data on dissolved carbohydrates and amino acids in world rivers: A documentation. In: Degens, E. T. (Ed. ) Transport of Carbon and Minerals in Major World Rivers, Pt. 1. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 52, pp. 575-84.

Ittekkot, V., Safiullah, S. and Arain, R. (1986) Nature of organic matter in rivers with deep sea connections: the Ganges-Brahmaputra and Indus. The Science of Total Environment 58, 93-107.

Junk, W. J. (1985) The Amazon floodplain-A sink or source for organic carbon? In: Degens, E. T. Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/ UNEP Sonderbd. 58, pp. 267-83.

Kempe, S. (1982) Long-term records of CO₂ pressure fluctuations in fresh waters. In: Degens, E. T. (Ed.) Transport of Carbon and Minerals in Major World Rivers, Pt. 1. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 52, pp. 91-332.

Kempe, S. (1985) Compilation of carbon and nutrient discharge from major world rivers. In: Degens, E. T., Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt.3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 58, pp. 29-32.

Lewis, W. , Saunders, J. F. , Levine, S. N. and Weibezahn, F. (1986) Organic carbon in the Caura River, Venezuela. Limnol. Oceanogr. 31(3), 653-6.

Lieth, H. and Esser, G. (1985) The attempt to simulate the global carbon flux from 1860 to 1981 using the Osnabrück Biosphere Model. In: Degens, E. T., Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd, 58, pp. 137-44.

Meade, R. H., Nordin, C. F., pérez-Hernández, D., Mejía, A. and Paredes Godoy, J. H. (1983) Sediment and Water Discharge in Río Orinoco, Venezuela and Colombia. Proceedings of the Second 1nternational Symposium on River Sedimentation, Water Resources and Electric Power Press, Beijing, pp. 1134-44.

Meybeck, M. (1976) Total minerals dissolved transport by world major rivers. Hydrol. Scient. Bull. 21, 265-84.

Milliman, J. D. and Meade, R. H. (1983) World-wide delivery of river sediments to the oceans. J. Geol. 91(1), 1-21.

MOP (1972) Mediciones en ríos grandes. Ministerío de Obras Públicas, Dirección General de Recursos Hidráulicos. Caracas Venezuela, 9Opp.

Nemeth, A., Paolini, J. and Herrera, R. (1982) Carbon transport in the Orinoco River: Preliminary results. In: Degens, E. T. (Ed.) Transport of Carbon and Minerals in Major World Rivers, Pt.1. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 52, pp. 357-64.

Novoa, D. (1982) Los Recursos Pesqueros del Río Orinoco y su Explotación, Editorial Arte, Caracas.

Novoa, D., Ramos, F. and Cartaya, E. (1984) Las pesquerías artesanales del río Orinoco sector Caicara-Cabruta. Parte 1. Mem. Soc. Cienc. Nat. La Salle 4(121), 295-7.

Paolini, J. (1986) Transporte de carbono y minerales en el río Caroni. Interciencia 11(6), 295-7.

Paolini, J., Herrera, R. and Nemeth, A. (1983) Hydrochemistry of the Orinoco and Caroni rivers. In: Degens, E. T., Kempe, S. and Soliman, H. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 2. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderb. 55, pp. 223-36.

Paolini, J. , Hevia, R. and Herrera, R. (1987) Transport of carbon and minerals in the Orinoco and Caroni rivers during the years 1983-84. In: Degens, E. T. , Kempe, S. and Gan Weibin (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 4. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 64, pp. 325-38.

Paredes, J. F., Paim, A. J., Da Costa-Doria, E. M. and Rocha, W. L. (1983) São Francisco River: Hydrological studies in the dammed lake of Sobradinho. In: Degens, E. T., Kempe, S. and Soliman, H. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 2. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/ UNEP Sonderbd. 55, pp. 193-202.

Salati, E. (1985) The climatology and hydrology of Amazônia. In: Prance, G. T. and Lovejoy, T. E. (Eds) Key Environment Amazônia, Pergamon Press, Oxford, pp. 18-48.

Sánchez, J. C., Pena, R. and Núñez, I. (1985) Estudio de línea base del río Orinoco, INTEVEP S.A. , Informes Técnicos, Los Teques, 432pp.

Schwerdtfeger, W. (1976) Climates of Central and South America. World Survey of Climatology, Vol. 12, Elsevier, Amsterdam, 532pp.

Serruya, C. and Pollingher, U. (1984) Lakes of the Warm Belt, Cambridge University Press, Cambridge, 553pp.

Sioli, H. (1975) Amazon tributaries and drainage basin. In: Hasler, A. D. (Ed.) Coupling of Land and Water Systems. Ecol. Studies, Vol.10, Springer-Verlag, Berlin, pp. 199-213.

Vegas-Vilarrubia, T. and Paolini, J. (1985) Some physical and chemical characteristics of black water rivers in Venezuela, a preliminary report. In: Degens, E. T., Kempe, S. and Herrera, R. (Eds) Transport of Carbon and Minerals in Major World Rivers, Pt. 3. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonderbd. 58, pp. 191-6.

Vegas-Vilarrubia, T., Paolini, J. and García-Miragaya, J. (1988) Differentiation of some Venezuelan black water rivers based upon physicochemical properties of their humic substances. Biogeochemistry 6,59-77.

Walter, H. (1984) Vegetation of the Earth, 3rd edn, Springer-Verlag, Berlin.

Weber, H. (1968) Zur natürlichen Vegetationsgliederung von Südamerika. In: Fitt- kau, E. J., Illies, J., Klinge, H., Schwabe, G. H. and Sioli, H. (Eds) Biogeography and Ecology of South America, Junk, The Hague, pp. 475-517.

Weibezahn, F. (1985) Concentraciones de especies químicas disueltas y transporte de sólidos suspendidos en el Alto y Medio Orinoco y sus variaciones estacionales (Feb. 1984-Feb. 1985). Convenio MARN-PDVSA, Caracas 223pp.

Welcomme, R. L. (1979) Fisheries Ecology of the Floodplain Rivers, Longman, London, 297pp.