Читать книгу Tropical Marine Ecology - Daniel M. Alongi - Страница 43

Tropical rivers of various sizes occupy a significant fraction of the world’s coastlines, but tidal rivers located in the wet tropics have the most impact on the geology and hydrology of the global ocean. 69.1% of the world’s river water (26 084 × 10³ km a⁻¹; Laruelle et al. 2013), laden with nearly 60% of the world sediment discharge (10 × 10⁹ tonnes a⁻¹), enters the tropical coastal ocean annually, mostly from the largest rivers (Table 4.1). The Amazon is the world’s largest river, but most tropical river water and sediment enters the global ocean from the Indo‐Pacific archipelago where high relief and rainfall produce high freshwater and sediment yields. Other major river/ocean boundary regions are in north‐eastern South America and west‐central Africa. The Amazon alone accounts for a disproportionate amount of the global flux, but the smaller mountainous rivers in Southeast Asia, ignored until recently, account for a greater proportion (43%) of the present sediment discharge estimates.

Tropical estuarine (349.4 × 10³ km²) and watershed (58 707 × 10³ km²) areas constitute 34.5% and 52.0% of the world’s totals, respectively (Laruelle et al. 2013). Tropical continental shelf area (11 094 × 10³ km²) and volume (720 576 km³) constitute 36.6 and 18.7% of the world’s totals. The small percentage of shelf volume is due to the fact the tropical shelves are on average narrower and shallower than shelves of higher latitude (Laruelle et al. 2013).

The dissolved loads of wet tropical rivers constitute about 65% of the world’s total (Huang et al. 2012). The proportion of water and sediment discharged from tropical rivers are a likely underestimate as many small‐ and medium‐sized tropical rivers remain ungauged (Latrubesse et al. 2005).

The relative importance of small mountainous rivers to the coastal ocean is exemplified on the islands of New Guinea and Timor. On the island of New Guinea, the ten largest rivers contribute only 35% to the island’s total river yield of 1.7 × 10⁹ t of sediment to the adjacent coastal zone as discharge from roughly 240 smaller rivers make up the balance (Milliman 1995). There are no large rivers on the much smaller island of Timor and the island discharges much smaller amounts of water (170 km³ a⁻¹) and sediment (133 × 10⁶ t a⁻¹), but area‐specific rates, including carbon and dissolved and particulate nutrients, are much higher than in New Guinea due to very high rates of deforestation and land degradation (Alongi et al. 2013). Borneo is a special case where despite moderate relief the sediment yield is high (even under rainforest), probably because of continuous uplift.

TABLE 4.1 Estimates of water (km3 a−1) and suspended sediment (109 tonnes a−1) discharge from gauged tropical rivers. Rivers are ranked by water discharge.

Source: Alongi (1990), Milliman and Farnsworth (2011) and Liu et al. (2020). © John Wiley & Sons.

River	Country	Water discharge	Sediment yield
Amazon	Brazil	6300	1200
Zaire	Zaire	1300	43
Orinoco	Venezuela	1100	210
Brahmaputra	Bangladesh	630	540
Mekong	Vietnam	550	110
Ganges	Bangladesh	490	520
Ayeyarwady	Burma	430	360
Tocantins	Brazil	370	75
Pearl	China	300	69
Zambesi	Mozambique	220	20
Salween	Burma	210	180
Fly	New Guinea	170	120
Niger	Nigeria	160	40
Magdalena	Colombia	140	140
Hungho	Vietnam	123	130
Song Hong	Vietnam	120	50
Rajang	Borneo	110	30
Purari	New Guinea	105	105
Sao Francisco	Brazil	97	1
Godavari	India	92	47
San Juan	Colombia	82	16
Atrato	Colombia	76	11
Sepik	New Guinea	72	45
Essequibo	Guyana	70	4.5
Usumacinta	Mexico	67	6.2
Sanaga	Cameroon	65	6
Marowijne	French Guinea	57	1.4
Corantijn	Guyana	47	1.1
Mahanadi	India	47	16
Tigris‐Euphrates	Iraq	46	1
Doce	Brazil	45	10
Papaloapan	Mexico	44	6.9
Kikori	New Guinea	43	33
Sai Gon	Vietnam	42	3
Patia	Colombia	40	21
Volta	Ghana	40	1.6
Parnaiba	Brazil	40	3
Brahmani	India	36	7
Coco	Honduras	36	6.5
Rufiji	Tanzania	35	17
Tsiribihina	Madagascar	31	12
Araguari	Brazil	30	0.5
Chao Phraya	Thailand	30	3
Nile	Egypt	30	0.2
Grande de Matagalpa	Nicaragua	29	5.3
Oyapoc	French Guiana	28	0.5
Paraiba	Brazil	28	4
Mira	Colombia	27	9.7
Escondido	Nicaragua	26	4.7
Grijalva	Mexico	23	1.3
Jequitinhonha	Brazil	23	5
Narmada	India	23	15
Ca	Vietnam	22	4
Senegal	Senegal	22	3
Cauweri	India	21	0.4
Pampanga	Philippines	21	1.4
Prinza Polka	Nicaragua	21	3.6
Mangoky	Madagascar	20	10
Panuco	Mexico	19	6.6
Ikopa	Madagascar	19	15
San Juan	Nicaragua	18	4.9
Pahang	Malaysia	18	3
Kelantan	Malaysia	18	2.5
Ma	Vietnam	17	3
Mearim	Brazil	17	0.7
Bengawan Solo	Indonesia	15	19
Coppename	Suriname	15	0.4
Suriname	Suriname	14	0.3
Thu‐Bon	Vietnam	14	2
Gurupi	Brazil	14	10
Lempa	El Salvador	14	7
Mae Klong	Thailand	13	8.1
Balsas	Mexico	13	11
Cavaly	Liberia	13	5.3
Perak	Malaysia	12	0.9
Sinu	Colombia	12	4.2
Approuague	French Guiana	12	0.2
Mitchell	Australia	12	0.4
Porong	Indonesia	12	6.2
Krishna	India	12	1
Mana	French Guiana	12	0.1
Berbice	Guyana	11	0.2
Damodar	India	10	28
Sassandra	Ivory Coast	10	2.9
Grande de Terraba	Costa Rica	10	1.9
Cross	Nigeria	10	7.5
Subarnarekha	India	10	3
Indus	Pakistan	5	10
Limpopo	Mozambique	5	33

Regardless of size, a great variety of physical processes, ultimately driven by climate, make these tropical coastal margins unique compared to coastal settings of higher latitude. The climate of the equatorial region is characterised by high rates of rainfall, solar insolation, and temperature. By virtue of these characteristics and global position, Coriolis forces are small, and winds are dominated by easterly trade winds (Chapter 2). These physical forces, coupled with the enormous loads of freshwater and sediment draining from the land, produce extensive buoyant plumes, in some instances, extending beyond the shelf edge. Rapid rates of sediment accumulation and high rates of nutrient flux and primary productivity are but a few of the unique characteristics of these river‐dominated coastal systems.

Tropical rivers worldwide drain a variety of geologic/geomorphologic settings: (i) orogenic mountain belts, (ii) sedimentary and basaltic plateau/platforms, (iii) cratonic areas, (iv) lowland plains in sedimentary basins, and (v) mixed terrains (Latrubesse et al. 2005). These types of rivers show high but variable peak discharges during the rainy season and a period of low flow during the dry season; some rivers show two flood peaks during the year, a main one and a secondary flood peak.

Tropical rivers exhibit a variety of channel forms and, consequently, a variety of different delta and mouth morphologies (Latrubesse et al. 2005). In most cases, rivers morph from one form to another over time so they are difficult to classify. Two main settings are rivers that discharge onto a tectonically active margin and those that discharge onto a passive margin (Leithold et al. 2016). Active margins are narrow and passive margins are wide.

Exactly how and where the discharged sediment deposits onto the adjacent shelf are not well understood, although it appears that sediments being discharged onto a wide, passive margin deposit in shallow waters often proximal to the river mouths, while a significant, if highly variable, proportion of sediments discharging onto narrow, active margins are transported to the adjacent shelf slope and deep sea (Wright and Nittrouer 1995; Leithold et al. 2016). The fate of sediment seaward of river mouths involves five stages: (i) supply via plumes, (ii) initial deposition, (iii) resuspension and transport by marine hydrographic processes, (iv) sediment that comes back on shore from far away and/or via tidal pumping, and (v) long‐term net accumulation (Figure 4.9). These processes vary greatly with river regime and coastal physics. The Amazon plume extends along the NW portion of the coast and far seaward. And although tide range is large and mixing processes are relatively intense, the enormous volume of outflow results in the effluent filling the entire water column beyond the mouth, before ascending above the seawater. It continues to expand as a thick (5–10 m) buoyant plume which reaches more than 300 km offshore and about 1000 km to the NW entrained by the North Brazil Current (NBC). However, the plume is highly variable over time and space due to wind forcing and tidal variations in bottom drag and vertical mixing. There is also seasonality in discharge, with maximum discharge during May–June; peak sediment discharge precedes peak water discharge by about a month or more, so the volume‐specific rates of discharge vary considerably. Strong tidal currents and waves generated by the easterly trade winds dominate the Amazon shelf resulting in variable spatial and temporal distribution of sediments on the shelf. Intense reworking of sediment on the inner shelf allows only temporary deposition. Once resuspended, the NBC carries sediment far to the NW. Despite high alongshore current flux, erosion occurs along the shore although erosion–deposition episodes depend on the strength of the NBC. At least 50% of sediment accumulation takes place on the mid‐shelf (depth 30–50 m) seaward and NW of the mouth. Remaining sediment is probably stored in the tidal reaches of the lower river.

FIGURE 4.9 Major stages in sediment dispersal of river sediments in the coastal ocean.

Source: Wright and Nittrouer (1995), figure 3, p. 503. © Springer Nature Switzerland AG.

The Purari River system on New Guinea is different, being much smaller (Table 4.1) and having a mountainous watershed (Wright and Nittrouer 1995). The Purari delta is heavily vegetated by mangroves and crossed by an intricate network of interconnected channels which trap most of the river sediment load. Saltwater intrusion is prevented by large shallow and mobile sand banks within and outside the river mouth. Fine sediments are carried onto the inner Gulf of Papua shelf as muddy, low‐salinity plumes that are broken up by the coastal oceanographic regime which is dominated by onshore‐directed southeast trade winds for most of the year. Thus, much of the sediment remains trapped relatively close inshore as a turbid band and is advected alongshore. Plumes enter tidal channels on flood tides supporting the extensive mangroves within the delta. Some sediment is transported directly offshore especially during summer when the trade winds are weak, and rainfall is at its peak.

In the Ayeyarwady River system, little modern sediment accumulates on the shelf immediately off the delta. Instead, a major mud wedge with a distal depocentre of up to 60 m thickness has been deposited seaward of the adjacent Gulf of Martaban, extending to about 130 m into the Martaban Depression (Liu et al. 2020). However, no river‐derived sediment has been found in the adjacent deep Martaban Canyon. There is a mud drape wrapping around the narrow western Myanmar shelf in the eastern Bay of Bengal. Unlike other large river systems in Asia, such as the Yangtze and Mekong, there is a bidirectional transport and depositional pattern controlled by local currents and tides, and seasonally varying monsoonal winds and waves.

In the Ganges–Brahmaputra River system, approximately one‐third of total sediment discharge is sequestered within the flood plain and delta plain (Rahman et al. 2018). The remaining load appears to be apportioned between the accumulating subaqueous delta and the deep‐sea Bengal fan via a nearshore canyon. The roughly equal partitioning of sediment among the flood plain, shelf, and deep sea reflects the respective influence of an inland subsiding tectonic basin, a wide shelf, and a deeply incised canyon system (Rahman et al. 2018).

Plumes of other large tropical river systems may be dispersed laterally due to local coastal currents and hydrography. For instance, the typical seasonal orientation of the Zaire River plume is northward for most of the year, except during February–March when the plume has a large westward extension onto the narrow shelf (Denamiel et al. 2013). The northward extension of the plume is explained by a buoyancy‐driven upstream coastal flow and the combined influences of the ambient ocean currents and the wind. During February–March, the surface ocean circulation drives the westward expansion of the plume and the presence of the deep Congo canyon increases the intrusion of seawater into the river mouth.

Off the Mekong delta, a similar lateral plume occurs throughout most of the year, with a net deposition SW of the river mouth down the Ca Mau peninsula (Szczuciński et al. 2013). In summer, a large amount of fluvial sediment is deposited near the Mekong River mouth, but in the following winter, strong mixing and coastal currents lead to resuspension and south‐westward dispersal of previously deposited sediment. Strong wave mixing and downwelling‐favourable coastal current associated with the more energetic NE monsoon in the winter are the main factors controlling post‐depositional dispersal to the SW.

For tropical river plumes, coastal hydrography plays an important role in governing how the discharged sediment is dispersed onto the adjacent shelf (Wright and Nittrouer 1995; Hetland and Hsu 2014). Oceanic processes that resuspend and transport sediment act in concert with maximum plume outflow, causing sediments to be dispersed farther from the delta mouth. Sediments can be dispersed over relatively wide areas that extend considerable distances from the delta; some rivers do not have a subaerial delta or do not protrude much beyond the regional coastline. Alongshore dispersal of sediments takes place primarily on the inner shelf but near the delta mouth the depth is too shallow and energetic for fine‐grain sediments to be deposited, except temporarily (Hetland and Hsu 2014). Much of the material transported alongshore becomes sequestered within tidal currents, and off the Amazon, muds that are transported alongshore ultimately accumulate, forming expensive, accumulating mud banks hundreds of km to the NW of the mouth. Despite a high level of instability, these mud banks are colonised by mangroves. However, the Amazon is characterised by accumulations of sediments as subaqueous delta deposits at relatively large distances on the mid‐shelf. A large fraction of Amazon sediments reaches the mid‐shelf due to the energetic currents and waves that sustain sediments in suspension until they reach relatively deep water (Wright and Friedrichs 2006). Much of the observed diversity of sediment dispersal and accumulation is attributable to variations in coastal energy regimes and to the temporal sequencing of river discharge relative to oceanographic transport processes. Sediment transport in the southeast section of the Amazon coastal zone is greatly affected by tidal asymmetry, seasonal variation of the wind and wave regime, and river discharge (Gomes et al. 2020). Climate and geological configuration have resulted in numerous estuaries and large‐mangrove‐lined coastal plains that partially divided the estuarine basins, which are connected by tidal channels within the Amazon delta. Convergence of transported sediment occurs within a channel connecting the estuaries, resulting in mud retention and further delivery to the mangrove‐covered plains, with a net flux of suspended sediments between estuaries. The connectivity between estuaries via channels is a key process to redistribute muddy sediments along this coastal sector, which helps to explain the evolution and maintenance of the relatively homogenous and widespread progradation of mangroves along the coast.

Mechanisms that dominate the short‐term spreading and mixing of riverine sediment may differ from the mechanisms that determine the longer‐term dispersal of sediment. Sediment records from the South China Sea show that strong monsoons are associated with intensified reworking of pre‐existing floodplain sediment over millennial timescales (Clift 2020). Strong monsoons result in deposition of more altered material that is also delivered at higher rates than during drier periods. Millennial‐scale changes in monsoon strength result in changes in the weathering regime but not fast enough to account for the changes seen in the sediments preserved in Asian deltas; instead, monsoon‐modulated recycling dominates. Over longer time periods (>10⁶ year) strengthening of the monsoon is linked to faster bedrock erosion and increased sediment flux to the ocean.

An additional complication is the fact that most tropical river systems are heavily affected by humans; few, if any, tropical rivers are pristine. Human disturbances such as the construction of dams and deforestation can greatly impact water and sediment discharge. Increased greenhouse gas emissions are projected to impact twenty‐first century precipitation distribution, altering riverine water and sediment discharge. Modelling indicates that increasing global warming will lead to more extreme changes and greater rates of increasing or decreasing changes in fluvial discharge (Moragoda and Cohen 2020). At the end of the twenty‐first century under all IPCC climate change scenarios, mean global river discharge will increase by 2–11% relative to the 1950–2005 period, while global suspended sediment flux will increase by 11–16% under pristine conditions. Combining the effects of climate change with natural and anthropogenic impacts, tropical rivers have and will continue to be affected greatly in future. For example, natural and human‐induced factors have greatly altered the discharge of the Patía River of Colombia (Restrepo and Kettner 2012). In 1972, the river flow was diverted to an adjacent river resulting in several environmental changes, such as coastal retreat along the abandoned delta, the formation of barrier islands with exposed peat soils in the surf zone, abandonment of former active distributions in the southern delta plain with associated closing of inlets and formation of ebb tidal inlets, breaching events on the barrier islands, and accretion on the northern delta plain.

Tropical deltas are sensitive to human encroachment due to regional water management, global sea‐level rise, and climatic extremes (Shearman et al. 2013; Darby et al. 2020). Vertical change within a delta is a function of the change in delta surface elevation relative to sea‐level (Darby et al. 2020), derived as the sum of the rates of natural compaction and anthropogenic subsidence, eustatic sea‐level change, rate of crustal deformation due to local geodynamics, and the rate of surface aggradation (Figure 4.10). A significant fraction of the world’s deltas is subsiding and at risk of imminent drowning, having significant impacts on changes in mangrove area (Darby et al. 2020). In the Ayeyarwady delta, the loss of mangrove habitats and its conversion to cultivation has led to increased salinity intrusion, coastline retreat, and increased flood risk (Kroon et al. 2015). Trends in other rivers of the Asia‐Pacific region indicate deforestation and subsequent destabilisation of coastline (Shearman et al. 2013). Overall, Shearman et al. (2013) observed a net contraction of 76 km², but trends varied among different river systems. Further, some systems such as the Ganges–Brahmaputra are naturally subsiding, resulting in a high‐risk situation in relation to sea‐level rise. Thus, most tropical river systems are at moderate to high risk of anthropogenic change. With increasing rates of sea‐level rise and more intense cyclones, tropical river systems will increasingly undergo environmental and ecological change into the foreseeable future. A model to estimate such future impacts on the Mekong delta (Bussi et al. 2021) indicates that climate change will play a secondary role compared to dams; planned dams will reduce suspended sediment fluxes to the delta by up to 50% over the next two decades.

FIGURE 4.10 Idealised scheme of the factors and processes contributing to vertical changes within river deltas in the face of relative sea‐level rise (RSLR). Most of the world’s deltas currently have low rates of natural sediment supply and high rates of eustatic sea‐level rise (E = 1–6 mm a⁻¹) and often higher rates of human‐induced subsidence (S_A < 250 mm a⁻¹), meaning that many deltas are subsiding or in danger of drowning as sediment accretion (A_N = 0–5 mm a⁻¹) is the only factor that can offset relative sea‐level rise. Natural compaction (C_N = 0–5 mm a^–1) and crustal deformation (G = 0–3 mm a^–1) are also important factors.

Source: Darby et al. (2020), figure 5.1, p. 105. Licensed under CC BY 4.0. © Springer Nature Switzerland AG.

The geomorphology of continental margins influences the biosphere by helping to mediate genetic connectivity of populations during sea‐level change (Dolby et al. 2020). Combining genetic data, geographical information systems‐based estuarine habitat modelling, and paleobiologic and recent effects of sea‐level change on evolution, Dolby et al. (2020) tested the relation between overall shelf area and species richness using data of 1721 fish species. They found an 82% global reduction of estuarine habitat abundance at low‐stand relative to high‐stand periods and found that large habitats change in size much more than small habitats. Narrow continental margins have significantly less habitat at high‐stand and low‐stand than wide margins and narrow margins significantly associate with active settings, effectively linking tectonic setting to habitat abundance. Narrow margins host greater species richness. Dolby et al. (2020) offer three possible explanations for this finding. First, physical isolation imposed by narrow margins may facilitate the formation of new species over time. Second, the size stability of small habitats, which disproportionately occur on narrow margins, may increase and retain species extirpated in the more variable habitats on wide margins. Third, the smaller habitats on narrow margins may facilitate greater species richness through greater habitat heterogeneity. The concept of narrow shelf margins as a ‘diversification pump’ is in opposition to previous paleontological information that generally argued areal restrictions were unimportant and/or regressions led to extinctions. However, these results support the idea that the complex and peculiar relation between habitat and sea‐level change depends on the inherent geomorphic properties of the coastline. These results remain to be tested but do illustrate the crucial role of the geosphere in marine ecological processes.