Home - Syllabus - Seminar Section
U4735x Environmental Science for Decision Makers
Lecture 5: River discharge by continent: S & SE Asia, Central & Eastern Russia, Africa & Australia; River management in China and Pakistan.
Jim Simpson
1 - South and East Asia rivers: Ganges-Brahmaputra, Yangtze, Huang Ho, Indus.
2 - Arctic Russia rivers: Yenesei, Lena, Ob.
3 - Africa rivers: Zaire, Niger, Zambezi, Nile.
4 - Australia rivers: Murray/Darling.
5 - River TDS vs Q (mm/yr): river basins in tropical and arid regions.
6 - "Average" world river chemical composition (table).
7 - Water policy issues for surface waters in South & East Asia.
8 - Map of China: locations of Yangtze and Yellow Rivers.
9 - Drainage basin of Yangtze River: population = 400 million.
10 - Map of Yangtze, including Three Gorges Dam site.
11 - Three Gorges Dam statistics table.
12 - Map of Yellow River & Sanmenzia Dam site.
13 - Map of Henan Province: Bangiao and Shimantan Dam failures (1975).
14 - Pakistan - main rivers and large dams.
15 - Pakistan - irrigation drainage network.
16 - Tarbela reservoir - plan view.
17 - Tarbela Dam - spillways and power tunnels.
18 - Tarbela reservoir sedimentation: 1974-95; schematic cross section.
19 - Indus River & Tarbela Dam statistics.
20 - Tarbela reservoir water level operations.
21 - Dams worldwide & sedimentation in Puerto Rico reservoirs.
22 - Large river sediment discharge by region (map).
23 - Large river table of total suspended solids.
24 - Export of TSS per unit area from continents: world average.
25 - Calculation of export of TSS per unit area from two rivers: Huang ho and St. Lawrence.
26 - Reservoir siltation time-scales: Nile, Indus, Yangtze.
27 - Global erosion rates of land (time-scale calculations).
28 - Global erosion rates of land: implications for dam construction.
The river systems of South and East Asia, Russia, Africa and Australia have had dramatic influence on the history of human habitation in those regions, and provide some of the most important constraints and opportunities for the future. Some illustration of those influences can be seen from mean annual water discharge characteristics of the largest rivers in these major continental areas. The water and suspended sediment discharge characteristics of the largest rivers account for many of the major environmental issues in each of these regions.
In terms of early human history and present rapid growth in population, the most important region of rivers lies in South and East Asia (Projection #1). The largest discharge to the ocean of this area, the combined flow of the Ganges and Brahmaputra (Q = 1000 km3/yr) is almost double that of the Mississippi River. These river basins support some of highest total populations and population densities on Earth. Those living in the delta region of Bangladesh are subjected to risk from flooding that is probably worse than that of any other large population. Both of these rivers carry huge quantities of suspended sediments, as do nearly all of the large rivers in the region. Thus they present special difficulties for water control efforts through construction of dams, in addition to their very large water discharge rates. The huge quantities of river discharge from this region are a reflection of a major phenomenon usually described as the Indian Monsoon. This zone of high precipitation is the result of intense solar heating of the midcontinent of Asia during summer months, which causes air to rise forming a low pressure center that leads to inflow of marine air from the south. As this moist air encounters the largest area of high elevation and mountain ranges on Earth, the Tibetan Plateau and Himalaya Mountains, it rises, cools and delivers enormous amounts of rain to the subcontinent. There are very high precipitation rates in much of India and Bangladesh during most years, usually in the months of June through August. For years in which the strength of the monsoon rains is weaker than average, there are often major difficulties of food supply because of reduced crop yields.
The valleys of the Yangtze (Q = 900 km3/yr) and the Huang Ho (Yellow River, Q = 50 km3/yr) have been centers of civilization in China throughout its history. They are also the sites of some of the largest and most controversial dam construction projects of the 20th and 21st centuries. The Indus River network (Q = 200 km3/yr), including the Punjab, key to agriculture in Pakistan and northwestern India, includes probably the largest contiguous area of irrigation diversion of river water for any basin in the world. Construction of huge dams in this region, especially since partitioning of colonial India into Pakistan, India and Bangladesh, has greatly altered the regional economies and set in motion a sequence of major environmental problems associated with the new reservoirs and heavily irrigated lands.
Examples of public policy issues for water management in South and East Asia include: 1) cross international border conflicts, such as Ganges diversions by India, just upstream of Bangladesh, during low discharge seasons; 2) overpumping of groundwaters, especially for irrigation in India and China, which will eventually lead to major dislocations in future crop production; 3) nonsustainable sediment accumulation problems behind major dams, especially on the Yellow, Indus and Yangtze rivers; 4) dominance of mega-scale river management infrastructure investments prevents capital from being directed to other purposes.
There are about a factor of four more suspended solids in rivers than dissolved ions, using a flux-weighted average to represent all river discharge to the ocean in terms of a single concentration value for particles and another value for dissolved materials. These particles include clay minerals produced by chemical weathering of rocks, as well as minerals such as quartz that undergo no significant alteration in soil environments, and are carried by rivers as residual detritus from continental rocks to the edge of the seas without having any appreciable impact on the chemistry of rivers. However these suspended particles can be of great importance to human uses of rivers, especially when large dams are constructed. Unlike the dissolved ions in rivers, which tend to stay in solution "forever", and can be increased only as water evaporates or decreased by chemical precipitation of new minerals such as calcite, the suspended load of rivers is only retained as long as turbulent flow of water continues in the channel. Once river water is impounded behind a dam, nearly all of the suspended load (typically greater than 90%) falls out of suspension and accumulates as bottom sediments. This simple, very well understood process, still is frequently ignored in the critical examination of huge reservoir construction projects. Hydroelectric and irrigation storage projects are often considered as "permanent" additions to the infrastructure of countries, despite the steady "aging" of reservoirs to the point at which their original design uses are no longer possible. Once a reservoir has an appreciable fraction of its usable storage filled with sediments, that volume is no longer available for delivering irrigation water and often the operation of turbines for generation of electricity is also no longer economically feasible. The critical issue of the length of time for such losses of reservoir storage volume is rarely given full scrutiny in the conception and design of large dam projects.
In Russia lies the largest group of Arctic rivers in the world (Projection #2). The Yenesei (Q = 600 km3/yr), Lena (Q = 500 km3/yr) and Ob (Q = 400 km3/yr) are all comparable in discharge to the Mississippi, yet lie in an extremely low population region with very low topography. The high Q values for these rivers result from the combination of very large drainage basins and low rates of ET typical of Arctic climates. They dominate the budget of fresh water delivered to the Arctic Ocean, which in turn is crucial to the dynamics of the sea ice which covers that ocean. There have been a number of serious proposals to divert large amounts of the water from some of these rivers to arid lands far to the south, providing irrigation water for crops such as wheat and cotton. The likely effects of such massive fresh water diversions on circulation of water and formation of sea ice in the Arctic Ocean and subsequent climatic evolution of high latitudes in the Northern Hemisphere is unknown, but a considerably community of scientists have concluded that such a perturbation of natural Arctic river discharges could involve major risk of large-scale climate change. Diversion of rivers from Russia and Kazakhstan at lower latitude for irrigation has already led to major problems for the Aral Sea, which is shrinking rapidly.
Examples of public policy issues for water management in central and eastern Russia include: 1) How do you balance large-scale climate risks such as loss of sea ice in the Arctic due to reduced freshwater inflows from Arctic rivers vs demands for more irrigation water for crop production at lower latitudes in Russia? 2) How do you balance environmental and economic damage of a shrinking Aral Sea against the demands for more irrigation water in that region? A similar, although much smaller scale problem involving the Salton Sea in southern California, will probably develop soon as Colorado River water is diverted away from irrigation of the Imperial Valley to provide domestic water supplies, especially for San Diego.
Africa is drained by some of the largest and most important rivers to human history (Projection #3). The Zaire (also known as the Congo) is the world's second highest discharge river (Q = 1300 km3/yr). This river basin receives very high P because of its location beneath the ITCZ, as is true for the Amazon. The second largest river in Africa, the Niger (Q = 230 km3/yr), is one of the most important resources to the large population in West Africa, including Nigeria, the highest population country on the continent. The Zambezi (Q = 150 km3/yr) is the site of one of the largest dams in the world (Kariba), constructed in the late 1950s. The Nile (Q = 80 km3/yr), covering the longest distance from headwaters to mouth of any river in the world (about 6600 kilometers), has been the locus of some of the most dramatic agricultural and cultural achievements of our species. Without this river, Egyptian civilization and all that it has meant to human history would have been impossible. To the first approximation, there is no significant rainfall in Egypt, except in the northernmost Delta adjacent to the Mediterranean Sea, yet Egypt developed one of the world's highest density urban civilizations as early as 2000 BC. Historians have argued that the organizational efforts required to adapt to and control the annual floods of the Nile were some of the most critical steps in our progression towards the modern world.
The Tigris and Euphrates (combined Q = 50 km3/yr), although not located in Africa, are included here with the rivers from that continent because of similarities in their historical importance to that of the Nile. They are also located in an area with extreme low rainfall, yet served as home to many critical developments in agriculture and other elements of civilization. They also provide important lessons about the cumulative effects of irrigation over extended periods on soil and water quality.
Examples of public policy issues for water management in Africa and the Middle East include: 1) How do you balance losses sustained by displaced populations from flood plains when large dams are constructed, such as Kariba Dam on the Zambezi? 2) When problems such as salinization of soils occur many decades after start of irrigation, how should those future costs be included in cost/benefit analysis? 3) Africa, especially south of the Sahara, has the lowest level of cumulative river engineering investment of any major continental area, and also includes a number of very poor countries. Should large-scale investment in dams be promoted in that region to stimulate economic growth, despite a number of known likely negative consequences.
The highest discharge rivers in Australia (Projection #4), the driest continent, are very small in water discharge rate compared with the largest rivers of other continents. The flow of the Murray/Darling River in SE Australia is about a factor of 40 lower than that of the Mississippi, the largest river in North America. The rivers in the extreme northeast of Australia, although comparable in discharge to that of the Murray/Darling, have little economic importance because they have extremely large seasonal variations in discharge due to a few months with high summer monsoon precipitation rates followed by many months of drought. In contrast, the basins of the Murray and Darling Rivers, receive rain from both summer monsoons and winter cyclonic storms, and comprise a region that is the heart of the agricultural economy of Australia. The river in North America which is most comparable in terms of Q, basin area, and relative economic importance of the Murray/Darling, per unit of river discharge volume, is probably the Colorado. However, the Murray/Darling is the only important large river system in Australia, while North America has a number of much larger rivers of great economic value. Thus the very low population supported in Australia (currently about 20 million) is strongly rooted in the availability of water, as well as historical reasons of great distances from large population regions such as East Asia and Europe. With very low rates of rainfall and river discharge, this continent appears unlikely to become a major center of human population in the foreseeable future.
Rivers are usually described as being "fresh water", in contrast to saline water such as that of the ocean, which has a total dissolved solids (TDS) concentration of about 35 grams per liter (parts per thousand) or alternatively 35,000 parts per million (ppm). There is much less dissolved material in rivers than found in the ocean or salty lakes, averaging about 100 ppm, or about a factor of 350 lower in concentration than sea water. There is remarkably little variation in TDS from one large river to another, especially compared to variations in suspended solids concentrations (soil particles), which range over many orders of magnitude. The total range of TDS among large world rivers is only a little more than one order of magnitude, with a tendency for rivers with lower mean discharge per unit area to be higher in TDS (Projection #5). Mean annual TDS for large rivers vs Q shows the contrast in amounts of dissolved material from different climatic zones. Thus large tropical rivers such as the Amazon and Zaire, have quite low TDS values, with the Amazon averaging less than 50 ppm. At the other extreme, the Colorado River, a representative arid zone river, has a "natural" TDS value of about 700 ppm, slightly more than one order of magnitude greater than the largest tropical rivers. Note that the Hudson River has a mean TDS value similar to the world "average" river, while the Nile and Murray rivers are a bit more salty than the average, but appreciably less than the Colorado. New York City drinking water, which is derived from tributaries of the Hudson River that drain the Catskill Mountains to the northwest of NYC, has total dissolved solids values comparable to those of the tropical rivers, among the lowest in the world.
River water, to the first approximation, is a dilute solution of calcium bicarbonate. The highest concentrations of calcium and bicarbonate observed in world rivers approach levels expected for saturation of water with calcite, a mineral entirely composed of calcium and carbonate ions. The composition of average river water also includes smaller amounts of chloride, sulfate, sodium and other ions (Projection #6). The dissolved ions in rivers are derived from a combination of processes, including weathering of primary igneous rocks such as basalts, and granites. In such cases, all of the calcium is derived from the rocks and all of the bicarbonate is derived from carbon dioxide respired by bacteria and other plants in the root zone of soils. Thus river water chemistry is produced by the activities of living organism on the minerals that form the rocks of the crust of the Earth, another example of the interaction of biological and geological processes. Other contributions to river water chemistry include weathering of limestone rocks, which are the result of precipitation of calcium carbonate minerals by marine microorganisms living in the ocean during earlier geological periods. After these marine fossil layers were buried to great depth, hardened by elevated pressure and temperature and then thrust up onto the continents by mountain building processes, they became available for weathering and return as dissolved ions in rivers to the ocean.
Dam construction in China and other areas of South and East Asia is likely to be a major focus of water resource policies for much of the coming century. Challenges of design, construction and operation of such facilities, especially when they are very large, will be complicated and difficult compared to other parts of the world. Some of key water management policy and environmental issues relevant to dam construction in this region include:
Population densities in S & SE Asia are among the highest in the world, especially in flood plains of major rivers, leading to major disruptions associated with reservoir construction.
Suspended particle loads of rivers in this region are extremely high, resulting in very short reservoir "lifetimes" relative to filling up significantly with sediments. Thus design and operation of reservoirs in this region probably should be altered substantially from practices in much of the rest of the world. Such changes would probably raise investment costs for such projects substantially.
The two most important river systems in China are the Yangtze and Yellow Rivers (Projection #8), both of which flow from the mountainous regions of western China towards the East China and Yellow Seas. The Three Gorges Dam site lies in the middle reach of the Yangtze. The Yangtze drainage basin has a very large resident population (about 400 million), and includes about 55% of the total area as crop lands (Projection #9). Note that in the USA, highest population densities are in coastal areas, far from the flood plains of our largest river systems. The global fraction of cropland for all inhabited continents is about 10% of total land area, far less than the drainage basin of the Yangtze River. Many large dams (17) are already present in the catchment. The Three Gorges Dam, upstream of Gezhouba Dam, will have a reservoir which extends upstream about 600 km (Projection #10).
Total electricity generating capacity for the Three Gorges Dam is designed for 18,000 MW (Projection #11), representing about 10% of the total generating capacity for the entire country in 1993. To generate a comparable amount of electricity by combustion of coal would require annual burning of about 40-50 million tons of coal. Thus construction of the dam should reduce production of green house gases appreciably.
Rapid siltation of the Sanmenxia Dam on the Yellow River (Projection #12) caused major economic costs several decades ago. Within only two years of initial construction (1960-62) that dam had to be completely altered, with electric generating capacity reduced from 1000 MW to only about 250 MW. Reservoir storage volume losses due to siltation amounted to about 5% per year, even after extensive alterations to permit enhanced sediment flushing.
The most catastrophic failure of dams in China, in terms of human life, occurred during August, 1975, when two major dams (Bangiao and Shimantan) in Henan province following a major typhoon cause massive flooding (Projection #13). When those two dams, and more than 50 smaller dams failed, the loss of life was huge (86,000 to 230,000). Note that this flood event is not included in the list discussed earlier of major floods reported for China (Munich Re). The history of previous catastrophic dam failures and siltation of reservoirs in China suggests that there are a number of serous potential problems likely to result from the Three Gorges Dam and other major dam construction projects in the region. Some of the kinds of problems likely to evolve on the Yangtze River can be illustrated by considering the history of dam construction on the Indus River.
Accumulation of sediments behind dams is a general problem associated with reservoirs on large rivers. The time-scale of loss of reservoir water storage volume is dependent upon the ratio of annual sediment transport to the location and the volume of the reservoir. The area of the world with the most serious difficulties from reservoir sedimentation is South Asia, due to the very large amounts of suspended particles carried by major rivers of this region, which result from erosion of high mountains and intense rainfall associated with the Indian Monsoon.
When colonial India was partitioned in the late 1940's into India and Pakistan (East and West), planning was initiated for a series of large dams in West Pakistan and India on the group of rivers that make up the Indus River system, including the Punjab in northwestern India. The most readily constructed dams were built first and finally in the late 1960's, a huge dam on the main stem of the Indus at the base of the foothills of the Himalaya was begun and eventually completed in 1975 (Projection #14). This project, known as Tarbela Dam, named for the small Pakistani village which formerly occupied the site, was the largest Earth-fill dam in the world at the time of completion. Storage water from Tarbela Reservoir became the largest component of irrigation water for the network of distribution canals which feed districts throughout the Indus Valley in Pakistan (Projection #15). Much of Pakistan is arid, and irrigation has been critical for agriculture for most of the history of human occupation, as in the Nile Valley of Egypt.
The surface of Tarbela Reservoir extends about 80 kilometers upstream of the Dam, and is confined in a narrow gorge at an elevation of about 500 meters above sea level at maximum water storage level (Projection #16). The natural discharge of the Indus River in southern Pakistan to the Arabian Sea, after inflow from the large tributary rivers of the Punjab, is quite large (Q = 240 km3/yr), sufficient to rank approximately 20th among all the rivers of the world in mean annual Q. During annual maximum river discharge (usually in late August or early September) following monsoon rains, the main stem of the Indus at the site of Tarbela Dam can carry about as much water as average discharge of the St. Lawrence River over Niagara Falls ( >104 cubic meters per second), presenting an extremely difficult engineering challenge for construction of a dam. The total volume of the reservoir following construction was about 14 km3, equivalent to 1 to 2 months of average Q of the Indus at the site, or about 25% of average annual releases of water from the High Dam at Aswan on the Nile. Compared to the volume of Lake Nasser (170 km3), the reservoir behind Tarbela Dam has a relatively small volume, due to the steep topography of the foothills of the Himalaya. Tarbela Reservoir has a total volume about 35% of that for the Three Gorges Dam, on a river much lower in annual Q than the Yangtze River.
The main embankment dam at Tarbela has a crest about 140 meters above the bed of the Indus River, with two huge spillways on the eastern shore to permit passage of the flood pulse of the Indus when the combined capacities of all of the hydroelectric generating tunnels on the western side (right abutment) are not sufficient to transmit all of the flood flow (Projection #17). The scale of construction required and design features necessary to manage the extremely large flood discharge at the end of each monsoon season are very impressive engineering achievements. The total electricity generation capacity installed at Tarbela is approximately 3800 megawatts, sufficient to provide about 40% of demand for the entire country during the late 1980's. The maximum installed capacity for hydroelectricity generation at Tarbela is appreciably greater than that for the Aswan High Dam in Egypt, due to both the higher elevation drop across the dam axis at Tarbela and the larger discharge of the Indus during the south Asian summer monsoon compared to the season of highest irrigation water releases from Lake Nasser through the turbines at the Aswan High Dam.
Within the first two decades of operation, the reservoir had accumulated a very large "delta" of sediments (Projection #18), since more than 90% of all the suspended particles are trapped behind the dam as water from the turbulent river slows down to form Tarbela Reservoir. In the years following 1974, the downstream margin of the delta of sediments reached closer and closer to the water intake structures for the turbines to generate electricity. At low water, once the sediment delta reaches the intake structures, the sand-laden reservoir water will rush down the power tunnels at high velocity and rapidly damage the turbine blades beyond repair. Thus, as sediments approach the water intake location, the only feasible short-term solution is to not decrease the reservoir level to the low elevation that was originally designed for operations (1300 feet), referred to as the "minimum pool". This change has already begun, resulting in a major decrease (by about one third) in the amount of available irrigation water downstream.
The cross section of sediment distribution in Tarbela reservoir through 1995 is presented in a schematic drawing to illustrate terminology about sediment types, and the general tendency of coarse-grained particles (sand) to deposit near the upstream end of the reservoir and fine particles (clay) near the downstream end. This actually refers only to deposition patterns when the water level is relatively high. During months when reservoir levels are low after water releases to meet irrigation demand, the river current is able to erode some of the upstream sediments and move them downstream towards the dam. Based on detailed surveys of sediments in 1990, about 14% of total storage capacity had been lost, or about 1% per year between 1974 and 1990.
Discharge of suspended solids to the location of Tarbela Dam on the Indus River is about 210 million tons (metric) per year, about the same as the current discharge of the Mississippi to the Gulf of Mexico, in an annual flux of water only about 1/7 of that for the Mississippi at its mouth; clearly the Indus carries a very large load of suspended sediments. The installed electricity generating capacity at Tarbela Dam is about 3800 MW (Projection #19), about one third of the electricity capacity for the entire country of Pakistan in the early 1990's. Annual loss of reservoir volume to sediment accumulation is about 150 million m3, or almost 50% greater than the volume of the main embankment dam itself. An order of magnitude estimate of the maximum "lifetime" of the project (before filling completely with sediments) can be made by dividing the original reservoir volume [14.3 km3] by annual loss of volume to sedimentation (1974-92) [0.154 km3] = 93 years. Actual "lifetime" without major modifications of operating practices and new construction is much shorter. In practice, only two decades after construction major problems had developed.
Although the original economic justification for Tarbela had much of the economic benefit assigned to irrigation water storage, the supply of electricity in Pakistan is now so close to being overwhelmed by demand that extending the usable life of the reservoir for electricity generation by raising the minimum pool has become the highest priority. Thus within the first two decades of operations, Tarbela Dam had collected sufficient sediments from the Indus behind its embankment to cause about one third of the formerly usable irrigation storage capacity ("live" storage) to be lost to the farmers downstream. Raising the minimum pool to permit another few decades of electricity generation to be sustained will delay facing loss of most of the current economic value of Tarbela Dam from electricity production, but within the first two or three decades of the 21st century, Pakistan will be confronted with the loss of its largest source of hydroelectricity in addition to its most important irrigation water storage facility. It is important to note that major disruptions to reservoir operations can occur far sooner than the time required for the entire reservoir volume to be filled with sediments. In the case of Tarbela, substantial reduction in the effective volume of irrigation water storage (Projection #20) has occurred before one quarter of the total storage volume was lost to sediments, only two decades after construction of the dam was completed. Smaller dams in the region have already had their reservoirs become completely filled with silt, and have essentially lost all of their economic value to the country, but there has not previously been such a dramatic short-term loss of high-cost infrastructure in Pakistan (or most other countries for that matter).
Control of surface waters in the 20th century frequently involved construction of dams, many of which are huge. Rivers carry particles in suspension due to turbulent water motion. Most of these particles settle to the bottom when the still waters of a reservoir behind a dam are reached, forming sediments that will eventually completely fill the storage volume. The length of time required for sediment accumulation is dependent upon the initial reservoir volume and the rate of influx of suspended particles, which varies by many orders of magnitude for different river basins. By the mid 1980s there were approximately 45,000 large dams (height >15 meters) world-wide, nearly all constructed in the 20th century, a very high fraction of which are filling with sediments and have no effective strategy for flushing accumulated sediments downstream past the reservoir accumulation areas (Projection #21). In areas of relatively high soil erosion and weathering, such as Puerto Rico, construction of new reservoir storage volume has barely kept pace with loss of water storage capacity to sediment accumulation (Morris, G.L., 1995), and most of the best available dam sites have now been exploited.
Earlier, the pattern of dissolved ion concentrations in major rivers was outlined, and the very narrow range of observed mean annual values (a total range of only about one order of magnitude) had a global mean of about 100 parts per million. For suspended particles, the range of mean annual values is much greater (by orders of magnitude), with some rivers, such as the Ganges/Brahmaputra, Amazon, Huang Ho (Yellow), Yangtze, Mississippi, Indus, Nile and Colorado, having extremely large mean annual transports of sediments (all greater than 100 million tons per year). Other rivers, such as the St. Lawrence and Columbia, are much more favorable for hydroelectric development because of lower transport of suspended particles.
Rivers in southern Asia that are influenced by large seasonal fluctuations of monsoon precipitation and draining high mountains that can rapidly erode, such as the Huang Ho and Indus, have two to three orders of magnitude higher concentrations of suspended particles per unit of river runoff (mm/yr) than rivers draining basins with more uniform seasonal precipitation and much less topographic relief. As an example, the Hudson River carries about a factor of 50 less suspended particles per unit volume of water than the Indus River.
The largest rivers of south and east Asia, together carry more than half of all of the suspended particles that are discharged from continents of the world into the ocean (Projection #22). Some areas with high net export of TSS to the ocean in S and SE Asia have particle yields in excess of 1000 T/km2/yr, while a number of river basins draining mountain areas have average particle yields in excess of 100 T/km2/yr. The largest rivers of south and east Asia, plus islands of the western Pacific such as New Zealand, Indonesia, New Guinea and Taiwan, together carry more than 70% of all of the suspended particles discharged from global land areas into the ocean. As a result, construction of dams almost anywhere in this region is more difficult than for any other area in the world, from the viewpoint of development of infrastructure which is likely to have a long service life.
Ranked in decreasing order of total particle yield, the largest river discharges are (Projection #23): Ganges + Brahmaputra (1670 million T/yr), Huang Ho (1080 million T/yr), Amazon (900 million T/yr) and Yangtze (480 million T/yr). If we compare the amounts of suspended sediment discharge from some large rivers around the world, the variability is clearly enormous. Arctic rivers in Russia, such as the Yenesei and Lena discharge very little sediment to the ocean, while the large rivers draining the Himalaya export huge amounts of particles.
As a frame of reference for considering export of suspended particles by rivers to the ocean, we can derive a flux-weighted average for all continental areas by multiplying total river discharge (Q), which is about one million m3/sec times 400 ppm (average TSS in global rivers), and divide by the total ice-free area of continents (about 1.3 x 108 km2): net result is about 100 T/km2/yr for export of TSS by rivers to the ocean (Projection #24).
Similar calculations can be done for two end-member rivers (Projection #25) which represent extremes of suspended particle export to the ocean; the Huang Ho in China, with TSS concentrations of >20,000 ppm, has a net export of about 1400 T/km2/yr, about 14 times the global "mean" for continents. In contrast, the St Lawrence River in N America, which has most of the particles from its catchment trapped by the Great Lakes before they can be discharged to the ocean, has a TSS concentration of only 9 ppm, and a net export of only about 4 T/km2/yr, a factor of more than 300 less than for the Huang Ho. Thus export of TSS to the ocean has a global mean concentration for all land area of about 100 T/km2/yr, but the distribution of export is extremely variable and large areas contribute very little TSS to the ocean, while others have huge fluxes.
The Three Gorges Dam, currently under construction on the Yangtze River in China, will very likely encounter siltation problems on a scale at least as great as for any major reservoir in the world (Projection #26). The time-scale for accumulating sufficient quantities of sediments to cause major perturbations in reservoir management practices will probably be similar to that observed at Tarbela Dam, which has been only about two decades.
Erosion of the land surface and transport of sediments to the oceans can occur at rates that present serious problems for management of surface waters in some river systems. Order of magnitude estimates of global erosion rates can be derived from considering the amount of suspended particles carried by the "average" river, which is about 400 parts per million (ppm). Using this value and the total discharge of world rivers, the amount of suspended particles carried by all rivers to the ocean is about 1.6x1010 Ton per year (Projection #27). Comparing this rate with the total mass of continent about sea level, assuming a mean land elevation of about 1 km, (3.3 x 1017 Tons), the time required to erode all land area down to sea level would be about 20 million years. Using other estimates for the mean elevation of continents (840 meters), and a mean erosion rate of 60 meters per million years, the time to erode all land area at the current rate would be about 14 million years (Projection #28). The difference between these two estimates is not significant, relative to the uncertainties associated with such an estimate. These calculations do illustrate that without active mountain building due primarily to plate motions that cause collisions of major segments of Earth's crust, erosion would reduce all exposed land area to near sea level quite rapidly compared to the history of the Earth. In areas near high mountains, the rates of erosion can be extremely high.
Leopold, Luna B., Sediment problems at the Three Gorges Dam, Appendix B, pp 194-199, The River Dragon has Come!, The Three Gorges Dam and the Fate of China's Yangtze River and Its People, Dai Qing, M. E Sharpe, Armonk, NY, 1998.
Qing, Dai and Lawrence R. Sullivan, The Three Gorges Dam and China's Energy Dilemma, J. International Affairs 53, No 1, Fall, 1999, pp 53-71, Columbia University, NY.
A Brief Introduction to the Three Gorges Project on the Yangtze River, February 1999, PRC Government Publication, 30pp.
Updated
September 15, 2003
webmaster.