Land degradation and soil erosion
Nowadays, with the growing population, the demand for agriculture production is increasing too. Large-scale conversion of forests to agriculture lands is a consequence of increased agriculture production demand, leading to increased soil erosion. High soil erosion must cause high depositions of sediment in rivers and lakes, and this is one of the major reasons for floods and water pollutions (Yang et al., 2003).
Wilkinson and McElroy (2007) indicated that subaerial erosion as a result of human activity, primarily through agricultural practices, had resulted in a sharp increase in net rates of continental denudation; and present farmland denudation is proceeding at a rate of around 75 Gt yr-1 and is largely confined to the lower elevations of Earth’s land surface. Increased soil erosion led to increased sediment fluxes in most of the rivers across the globe (Gupta et al., 2012). Syvitski et al. (2005) addressed that humans had increased the inland sediment transport by the global rivers through soil erosion by 2.3 ± 0.6 Gt yr-1. Yang et al. (2003) pointed out that nearly 33% of the world’s arable land was lost to erosion, with loss continuing at a rate of more than 10 million ha yr-1.
Asia probably has suffered more from soil erosion compared to other continents due to its geomorphology, climate factors and the human activities (Dregne, 1992; Ananda and Herath, 2003; Lal, 2003; Yang et al., 2003). That natural erosion is primarily confined to drainage headwaters, and around 83% of the global river sediment flux is derived from the highest 10% of Earth’s surface (Wilkinson and McElroy, 2007). The Himalayan-Tibetan Plateau is the birthplace of many important rivers including the Indus, the Ganges, the Brahmaputra, the Irrawaddy, the Salween, the Mekong, the Red, the Yangtze and the Yellow rivers, and this area has been recognized as a great SS contributor (Milliman and Syvitski, 1992; Ludwig and Probst, 1998). From Figure 1-6 we can see that the Himalayan-Tibetan Plateau is in high erosion. Active tectonic movements (such as earthquake, debris flow and landslide), steep slopes, freeze-thaw and weathering erosions are the main issues in the riverhead high-elevation areas. Active human activities inside these basins (such as deforestation, agriculture, urbanization, road construction and mining) accelerate the soil erosion processes.
Yang et al. (2003) found that Southeast Asia had the most serious soil erosion problems and hot spots were close mountainous areas located in the tectonic zones and dense croplands of the high population regions where both natural geomorphology and human activity are major factors for inducing soil erosion; and there was an increasing trend found in Asian, and the regions with the largest increases were in the tropic rainforest regions (Southeast Asia), such as Thailand and the lower Mekong basin (Yang et al., 2003). In the Mekong River basin, the soil erosion in the 1980s was 9.6 t ha yr-1 and predicted to reach 13.0 t ha yr-1 in the 2090s (Yang et al., 2003).
In Vietnam, more than 40% of its steeply sloping lands (62% of the country) suffer severe erosion (Dregne, 1992). From previous studies, the annual soil losses in the Red River basin in Vietnam’s part ranged from 0.9 to 174 t ha yr-1 (Podwojewski et al., 2008; Nguyen et al., 2011; Mai et al., 2013; Tuan et al., 2014). In the area of the Red River basin in China’s part, Gu (2016) found that the average annual soil erosion was 18.4 t ha yr-1 (136 Mt yr-1) in 2000 while it was 18.7 t ha yr-1 (138 Mt yr-1) in 2010; severe soil erosion area which was only less than 10% of the total erosion area, however, contributed 57% to 65% of the total erosion amount; farmland was the hot spot of soil erosion, followed by grassland and forest; slope above 15° and elevation between 1000-2000 m a.s.l. were the hot spots of erosion.
Sediment export by rivers
Global sediment flux to the oceans was estimated from 12.6 to 18.5 Gt yr-1, and Asia exported the most sediments (4.8 Gt yr-1) among continents (Syvitski et al., 2005; Gordeev, 2006; Syvitski and Kettner, 2011). High sediment loads is a common feature in many Asian basins due to the pronounced topography of the region, especially the basins originating from the Himalayan-Tibetan Plateau, such as the Mekong, the Red, the Yangtze and the Yellow rivers (Milliman and Syvitski, 1992; Ludwig and Probst, 1998; Evans et al., 2012).
Investigation of global value and the current trend in sediment exports has some constraints and uncertainties (Walling and Fang, 2003; Cohen et al., 2013). Firstly, lack of sediment data in many rivers, especially in the rivers in developing and underdeveloped countries, can cause an underestimation on the global sediment exports. Even the sediment flux data is available, but the measurement only considers suspended sediment flux, not the bed load transport. Secondly, analysis of annual sediment flux temporal trends requires records of enough length data. Long-term sediment monitoring programmes, however, are rare in many areas of the world.
Dam is a key tool for people to exploit the water resource (water supply, electricity generation and flood control) in many areas of the world, especially in the areas with intensive population and agricultural activities (Schmutz and Moog, 2018). Lehner et al. (2011b) estimated that there were 6862 records of reservoirs and their associated dams and about 2.8 million impoundments larger than 0.1 ha worldwide. However, dam constructions can induce associated impacts such as interruption of river continuity, siltation of river bed and clogging of interstitial, downstream river bed incision and downstream flow and water quality alteration (Figure 1-7) (Schmutz and Moog, 2018).
The evidence afforded by the sample of the world’s rivers indicates that reservoir construction is probably the most important influence on land-ocean sediment fluxes (Walling and Fang, 2003). Early, Vörösmarty et al. (1997) estimated that more than 40% of global river discharge was intercepted by the large impoundments and that an around 70% proportion of this discharge maintains a theoretical sediment trapping efficiency in excess of 50%; for regulated drainage basins the global, discharge-weighted residence time change was 0.16 years, representing a 30% potential sediment trapping. More recently, Vörösmarty et al. (2003) indicated that greater than 50% of basin-scale sediment flux in regulated basins was potentially trapped in artificial impoundments, with a discharge-weighted sediment trapping due to large reservoirs (≥ 0.5 km3) of 30%, and an additional contribution of 23% from smaller reservoirs.
Syvitski et al. (2005) addressed that fluvial sediment loads, over 100 Gt of sediment, including carbon (around 1 to 3%), had been sequestered behind reservoirs. With more dam are going to operate, sediment trapping loads and percentage might increase.
Figure 1-7 Dam construction interrupts the river continuity (such as fish migration, sediment and nutrient transport) and alters the downstream flow and sediment regimes.
Dam and weir constructions can reduce downstream gravel supply and therefore lead to armors, thus increase intensify flushing out heterogeneous sorted sediments. (Hauer et al., 2018). As a consequence, the resultant deficits in bed-load transport may lead to continuous riverbed incision with the risk of channel avulsion and riverbed breakthrough during single flood events. Dam constructions may cause significant alterations of the sediment regime (such as grain size distribution) based on the storage of water and the capture of sediment by dams which cause profound downstream changes in the natural patterns of the hydrologic variation and sediment transport (Hauer et al., 2018). Dam constructions not only retain the sediment, breaking the sediment continuum, but also alter the dynamics of sediment transport processes. Large amounts of retained suspended load in the reservoirs are released in a short period of time during flushing, mostly in conjunction with flood events, resulting in a surplus of sediments in downstream river sections. Consequently, high loads of mostly fine sediments cause high concentrations of turbidity and can be responsible for losses and mortality of aquatic organisms (Espa et al., 2015; Hauer et al., 2018).
Asian rivers were estimated to export 4.8 Gt yr-1 sediment to the oceans (Syvitski and Kettner, 2011). However, Africa and Asia showed the largest reduction in sediment flux to the coast in rivers (such as the Nile, Orange, Niger, and Zambezi in Africa and the Yangtze, Indus, and Yellow in Asia), and 31% of the total sediment load retained in reservoirs were indicated in Asia and 25% in Africa (Syvitski et al., 2005).
Since the 1980s, the sediment load of the Yellow River has dropped markedly to <50% of this earlier value in response to reduced precipitation, increased water abstraction and improved sediment control practices in the Loess region of the Middle Yellow River (Walling and Fang, 2003). Mean sediment flux of the Yangtze River decreased by 71% between 1950-1968 and the post-Three Gorge Dam decade, about half of which occurred prior to the pre-Three Gorge Dam decade; approximately 30% of the total decline and 65% of the decline since 2003 can be attributed to the Three Gorge Dam, 5% and 14% of these declines to precipitation change, and the remaining to other dams and soil conservation within the drainage basin (Yang et al., 2015). During 2007-2013, the fluvial sediment supply from the Pearl River to the coast showed a massive 71% decrease compared to period 1954-1979 when the human influences were not significant (Ranasinghe et al., 2019). In the Mekong River basin, the sediment trap efficiency was predicted to increase to 78-81% with all the planned reservoirs being built, and the potential annual sediment trap would be 70-73 Mt (Kummu et al., 2010). In the Ren River basin, due to the construction of the biggest dam, Hoa Binh dam, sediment flux drastically decreased from 100-160 Mt yr-1 to around 40 Mt yr-1 during 1997-2004, and the mean annual sediment trapping efficiency of Hoa Binh dam was 88% (Vinh et al., 2014). The annual sediment flux of the Ganga River was estimated from 262 to 390 Mt yr-1 during 2004-2010 (Rice, 2007; Lupker et al., 2011); and the annual sediment flux of the Brahmaputra River was 387 Mt yr-1 in 2006 (Rice, 2007), however, Rahman et al. (2018) found that these two major river systems were following a declining trend, and sediment load was decreasing at a rate of 4-10 Mt yr-1.
Asian rivers export high sediment fluxes to the oceans especially the Himalayan-Tibetan Plateau originating rivers (Milliman and Syvitski, 1992; Ludwig and Probst, 1998; Evans et al., 2012). Most estimations of sediment flux were calculated based on a monthly or an annual scale. However, most sediment export happens during flood events linked to daily discharge variations. Therefore, it would be necessary to study the sediment fluxes in this region on a daily time scale in order to precisely estimate the export of suspended sediment and its associated nutrients and contaminants, and also to understand the transport response to the flood events.
Fluvial Organic Carbon
Carbon (C) cycling is a cornerstone of ecosystem biogeochemistry as it is a critical element for all biota cellular processes (Kroeze et al., 2012). Carbon can be divided into inorganic carbon (IC) and organic carbon (OC). The most essential pools for carbon in an aquatic environment include dissolved inorganic carbon (DIC), dissolved organic carbon (DOC) and particulate organic carbon (POC) (Göltenboth and Lehmusluoto, 2006).
Inorganic carbon (IC) in the form of DIC is available in water in three forms: CO2, HCO3−, CO32−, and their availability depends on the pH values of the water (Göltenboth and Lehmusluoto, 2006); in the atmosphere, IC is primarily in the form of CO2 (Dodds and Whiles, 2010; Cole, 2013).
Organic carbon (OC) is the base of organic compounds which contain carbon atoms bonded to hydrogen atoms and possibly other elements such as nitrogen or phosphorous. OC provides the materials and energy for metabolism within the ecosystem. Soluble compounds including soil humic substances, polysaccharides, polypeptides and some colloidal materials comprise the DOC; living and dead micro-organisms and carbon in suspended sediments are isolated as particulate organic carbon (POC) (Schlesinger and Melack, 1981).
The OC transport by rivers to the ocean is important to coastal heterotrophic organisms, even though riverine OC represents only a small fraction (0.9%) of net global terrestrial primary production (Zhao and Running, 2010; Huang et al., 2012). Before reaching the ocean, C from land transit through the continuum formed by soil, groundwater, riparian zones, rivers, lakes, estuaries and coastal marine areas, combined with contaminants.
Sources of organic carbon
Riverine OC mainly comes from three sources: the allochthonous source, which is based on terrestrial origins, such as weathering from rocks, leaching from soil and the decomposed products like the tissue of plants on land; the autochthonous source, which derives from primary production within the river itself, such as from algae and phytoplankton. Anthropogenic influences from agricultural, domestic and industrial activities can also be regarded as an allochthonous source (Hope et al., 1994; Huang et al., 2012).
The DOC is a mixture of substances. Besides allochthonous inputs, leaching from the soil is the main sources for DOC while the uptake by bacteria is the most important output (Le, 2005). The POC is mainly composed of the substances bound in the organism and the detritus. The main source for POC is the primary production. POC can be transformed by secretion, excretion and autolysis into DOC and be derived from DOC by physico-chemical and biological processes (Göltenboth and Lehmusluoto, 2006).
Fluvial DOC and POC concentrations are mainly related to the soil organic carbon (Aitkenhead and McDowell, 2000; Huang et al., 2012; Li et al., 2017; Fabre et al., 2019). Soil organic carbon depends on land management and land use, and it can enter the river by washed out by rain and by leaching (Escolano et al., 2018). Soil resources in many Asian countries are being overexploited, degraded, and irreversibly lost due to inappropriate land management practices, industrial activities, and land use changes that lead to soil sealing, erosion, contamination, and loss of organic carbon, which subsequently increase the OC exports to the oceans.
Dissolved and particulate organic carbon exports by rivers
The riverine OC cycle and budget have been paid attention and studied in recent decades. Schlesinger and Melack (1981) used two ways to estimate the global OC flux: the first was by an inventory and extrapolation of data on loss of carbon per unit volume of river discharge from 12 intermediate and large rivers, and they found an OC export of 0.37 Gt yr-1; the second was using measurements of the fluvial loss of organic carbon per unit area of land in various ecosystem types, and they found the OC flux was 0.41 Gt yr-1. Ludwig and Probst (1996) utilized a database of mean annual DOC and POC fluxes of 32 rivers, respectively, and other ecological factors to calculate DOC and POC fluxes, and found a global annual OC flux of 0.38 Gt yr-1 (DOC of 0.21 Gt yr-1 and POC of 0.17 Gt yr-1). Aitkenhead and McDowell (2000) examined the relationship between DOC flux and soil C:N ratio on a biome basis; and by using their C:N model, they estimated the total export of riverine DOC from land to the oceans to be 0.36 Gt yr-1. Schlünz and Schneider (2000) re-estimated the modern global riverine OC flux and gave the value of 0.43 Gt yr-1, of which 0.18 Gt yr-1 was transported by Asian rivers. More recently, Li et al. (2017) re-evaluated the riverine global OC flux to 0.48 Gt yr-1, of which 0.24 Gt yr-1 was DOC and 0.24 Gt yr-1 was POC, and Asian rivers exported more DOC and POC than other continents. As presented in Section 1.3.2, Himalayan rivers export high quantity of sediments to the oceans, and it has also been recognized to be a great source of OC (Aucour et al., 2006; Galy et al., 2007).
The humid tropical climate is associated with the highest carbon yield (Ludwig and Probst, 1996b; Aitkenhead and McDowell, 2000; Huang et al., 2012; Carvalhais et al., 2014). The tropical region (30°N–30°S) covers around 43% of the world’s land but contributes 66% of global outflow, 73% of sediment load and over 61% of terrestrial net primary production (Milliman and Syvitski, 1992; Syvitski et al., 2005). Therefore, tropical rivers are critical to total global fluvial organic carbon flux. Aitkenhead and McDowell (2000) found the riverine DOC flux from tropical regions were 0.15-0.23 Gt yr-1. Huang et al. (2012) used published and unpublished data, considering 175 tropical rivers, estimated the fluvial carbon fluxes and found that these tropical rivers delivered approximately 0.27 Gt yr-1 organic carbon to the estuaries, of which 0.14 Gt was DOC and 0.13 Gt was POC. They found that rivers in the equatorial region between 3°N and 6°S produced high DOC; and the type of soil was a main influencing factor: the pattern of DOC distribution was similar to the distribution of soil OC density. They also pointed out that rivers in mainland Asia have the highest specific export rates in terms of DOC and POC. The tropical rivers in Asia exported 0.05 Gt yr-1 of DOC and 0.06 Mt yr-1 POC to the oceans (Huang et al., 2012).
From above we can see that Asian rivers export high OC fluxes to the oceans and the tropical region is high-yield of OC. Most estimations were calculated based on a monthly or an annual scale. However, most OC export happens during flood events linked to daily discharge variations. Hardly any study estimated OC fluxes on a daily time step. Therefore, the DOC and POC exported by Asian tropical rivers take a large portion of the whole Asian rivers OC exports, and it would be necessary to study the OC fluxes in this region at a daily time scale.
Riverine DOC concentration and fluxes are mainly influenced by basin soil OC, 49 hydrogeology and climate conditions. Soil OC condition is related to land cover and land management. High soil OC and riverine DOC concentration are found in the basins with the land cover of the forest, peatland, wetland and agriculture (Schlesinger and Melack, 1981; Bishop and Pettersson, 1996; Coynel et al., 2005; Billett et al., 2010; Ritson et al., 2019). Forest, peatland and wetland are rich in organic matters and active decomposition processes while farmland is with some organic fertilizer and is eroded by agricultural practices. Hydrogeology conditions such as drainage intensity and basin slopes and climate such as rainfall density can cause different erosion and leaching processes which are essential to DOC concentration and fluxes (Ludwig, 1997; Huang et al., 2012).
The main factors that govern riverine POC are suspended sediment (SS), phytoplankton, soil OC (Ludwig, 1997; Huang et al., 2012; Fabre et al., 2019). Some studies indicate an inverse relationship between %POC (which is the percentage of POC concentration in suspended sediment concentration) and the SS (Ludwig, 1997; Dang et al., 2013a; Fabre et al., 2019). POC concentration is high when the SS concentration is low, and autochthonous OC produced by phytoplankton is the main contribution; when POC concentration is low and SS concentration is high, the mineral matter, erosion soil and sedimentary rock are a major source (Ludwig, 1997; Dang et al., 2013a).
Besides the factors mentioned above, human interferences and climate changes have affected the OC transfer and fluxes (Hope et al., 1994; Seitzinger et al., 2010; Escolano et al., 2018). For example, agricultural activities have enhanced the soil erosion which consequently has increased POC input (Ludwig and Probst, 1996b; Ciais et al., 2008; Huang et al., 2012). Dam constructions have sequestered the suspended sediment which consequently decreased POC export, and over 100 Gt of sediment and 1 to 3 Gt of carbon were sequestered in reservoirs constructed largely within the past 50 years (Syvitski et al., 2005); and dam regulation on discharge also affect the DOC transport dynamic and fluxes (Hope et al., 1994; Seitzinger et al., 2010; Hu et al., 2015; Wu et al., 2015; Li and Bush, 2015; Li et al., 2015; Liu et al., 2015, 2019a; Shi et al., 2016; Xia et al., 2016; Huang et al., 2017; Le et al., 2018; Park et al., 2018). Climate variability has been verified to affect OC fluxes (Tian et al., 2013; Wu et al., 2015). Wu et al. (2015) indicated that climate change had influences on POC due to the variations of discharge and sediment load. Tian et al. (2013) compared the DOC in different climate zones in USA and found that temperature was a key variable for DOC export and climate warming would have a greater impact on riverine DOC yields in cooler climate zones than on those in warmer climate zones.
The Red River basin is a basin crossing subtropical and tropical climate zones and shared among China, Laos and Vietnam, combining different land uses and affected by human activities such as intensive agriculture and dam implements. Previous studies, both on using sampling data and modelling, have investigated that human activities have impacts on hydrology and SS (Le et al., 2007; Lu et al., 2015; Vinh et al., 2014; Wang et al., 2011; Wei et al., 2019, submitted); these studies especially found a strong retention of SS caused by dams. Hence, it would be interesting to understand the OC processes and quantify OC fluxes in this basin at the interannual scale with a daily time step.
General Approaches and Tools
For assessing the hydrological cycle and suspended sediment, the general method and tools are in-situ field measurements, empirical and simple equations, remote sensing techniques, geographic information systems, and numerical simulations. However, field-collecting data at large spatial and temporal scales is expensive, and often impracticable in some remote areas and underdeveloped regions. Empirical or/and simple equations, such as sediment rating curves are sometimes applied to quantify the sediment flux (SF) (Asselman, 2000; Syvitski et al., 2000; Achite and Ouillon, 2007; Zhang et al., 2018). However, a sediment rating curve requires discharge (Q) as an input, which might not be available for remote and underdeveloped regions, and its parameters can vary a lot among a big drainage basin. Therefore, this method might neither be the best choice for calculating the SF on a daily scale nor in a large basin.
Numerical models combined with other techniques (such as remote sensing) can fill the gap in sediment dynamic measurements (Syvitski et al., 2005; Wilkinson et al., 2009) and offer insight into future and past trends in response to environmental and human changes, such as land use change and climate change. In addition, simulations can be carried out at a large spatial scale and at a daily time scale to quantify, analyze and forecast water resources and quality. In particular, it can realistically represent the spatial variability of the basin, which will provide a global view of the whole basin. Many physically-based hydrological models had been used (Daniel et al., 2011; Islam, 2011; Devia et al., 2015; Fu et al., 2019), such as MIKESHE (Graham and Butts, 2005), HSPF (Bicknell et al., 1997) and Soil and Water Assessment Tool (SWAT) (Arnold et al., 1998). Among these models, SWAT has been proved to obtain good hydrological predictions with a little direct calibration in many different basins around the world (Gassman et al., 2007, 2014; Devia et al., 2015; Fu et al., 2019), and more applications can be found in SWAT literature database: https://www.card.iastate.edu/swat_articles/.
Although SWAT has been applied to many Asian basins, and also to subtropical or/and tropical areas, most of them were at a scale of 77 to 105,000 km2 (Gassman et al., 2007; Bannwarth et al., 2015; Lweendo et al., 2017; Li et al., 2018; Shrestha et al., 2018; Tan et al., 2019). Tan et al. (2019) summarized that a total of 126 articles related to application of SWAT on studying water-related issues (such as climate change, land use change, best management practices, water quality and hydropower) in Southeast Asia, half of which were in Vietnam and Thailand, and the performances of the SWAT model were generally above satisfactory.
The Red River is a typical Asian river system, combining different land uses, affected by human activities such as intensive dam implementations and agriculture (Le et al., 2007; Nguyen et al., 2011). Recent studies of hydrology and suspended sediment in this basin mainly used data from gauge stations or sampling to do statistical analysis (He et al., 2007; Dang et al., 2010; Lu et al., 2015; Le et al., 2017a), or use modelling to perform simulations at a local scale (Ngo et al., 2015) or in the delta part (Vinh et al., 2014) at a monthly scale; few studies analyzed fluxes at daily scale, but only on a short period (Le et al., 2007), in the delta (Luu et al., 2010). Hiep et al. (2018) used SWAT to simulate the discharge at daily scale for four stations (Lao Cai, Yen Bai, Son Tay and Hanoi) on the main Red River during 2005-2009, however, they did not simulate the Da and Lo rivers, nor the suspended sediment. Both Q and SSC can vary greatly from day to day; therefore, it would be more favourable to calculate flux at a daily time step. Also, water quality monitoring is usually carried out during some specific days in a month, and outputs from a model at daily scale can be practical and useful for further studies. In addition, different scenarios of global changes can be considered to help researchers or government administrators to compare different possibilities and set up long-term management plans.
Table of contents :
1. CHAPTER Ⅰ: Scientific Context and Objectives
1.1. Water Regime
1.1.1. Water resources worldwide
1.1.2. Water resources and consumptions in Asia
1.1.3. Water quality in Asia (especially in China and Vietnam)
1.2. Hydrologic Cycling
1.2.1. Global hydrologic cycling
1.2.2. Hydrologic cycling at a basin scale
1.3. Sediment Fluxes
1.3.1. Land degradation and soil erosion
1.3.2. Sediment export by rivers
1.4. Fluvial Organic Carbon
1.4.1. Sources of organic carbon
1.4.2. Dissolved and particulate organic carbon exports by rivers
1.4.3. Influence factors
1.5. General Approaches and Tools
1.6. Objectives of this Study
2. CHAPTER Ⅱ: Materials and Methods Global Theme of the Materials and Method
2.1. Study Area
2.1.1. General information
2.1.2. Climatic characteristics
2.1.3. Hydrological characteristics and water resources
2.1.4. Land use and cover
2.1.5. Basin social economy and human activities
2.2.1. Discharge and suspended sediment concentration dataset
2.2.2. Dissolved and particulate organic carbon dataset
2.3. General Introduction of the Modelling Approach
2.3.1. SWAT general introduction
2.3.2. SWAT application
2.3.3. Hydrological modelling component in SWAT
2.3.4. Sediment modelling component in SWAT
2.4. Modelling Setup for Hydrology and Suspended Sediment
2.4.1. Modelling inputs for SWAT
2.4.2. Dam implemented in SWAT
2.5. Fluvial Organic Carbon Computation
2.5.1. Dissolved organic carbon
2.5.2. Particulate organic carbon
2.6. Calibration processes
2.6.1. Discharge and Suspended Sediment Concentration calibration
2.6.2. Dissolved and particulate organic carbon parameters calibration
2.7. Simulation Performance
2.7.1. The coefficient of determination (R2)
2.7.2. The Nash–Sutcliffe efficiency (NSE)
2.7.3. The Percent bias (PBIAS)
2.7.4. Dissolved and particulate organic carbon validation
2.8. Scenarios and Output analysis
2.8.1. Scenarios implementations by SWAT
2.8.2. Identification of the influencing factors for soil erosion
2.8.3. Scenarios outputs for dissolved and particulate organic carbon
2.8.4. Analysis of the relationships between the parameters for calculating the dissolved and particulate organic carbon and the physical characteristic of eachsub-basin
3. CHAPTER Ⅲ: Modelling Discharge and Suspended Sediment Concentration
3.1. Scientific Context and Objectives
3.2. Materials and Methods
3.3. Main Results and Discussions
3.4. Conclusion and Perspectives
3.5. Full Article Published in Water
4. CHAPTER Ⅳ: Assessing the Sediment Fluxes and Soil Erosion
4.1. Scientific Context and Objectives
4.2. Materials and Methods
4.3. Main Results and Discussions
4.4. Conclusion and Perspectives
4.5. Full Article Submitted to Hydrological Processes
5. CHAPTER Ⅴ: Assessing Fluvial Organic Carbon Concentration and Fluxes
5.1. Scientific Context and Objectives
5.2. Materials and Methods
5.3. Main Results and Discussions
5.4. Conclusion and Perspectives
5.5. Full Article
6. CHAPTER Ⅵ: General Discussion
6.1. Water Regime
6.1.1. Hydrological cycle and water yield
6.1.2. Impacts of dams on discharge
6.1.3. Impacts of climate variability on discharge
6.2. Suspended Sediment
6.2.1. Sediment export
6.2.2. Soil erosion
6.2.3. Impacts of climate variability on suspended sediment
6.2.4. Impacts of dams on suspended sediment
6.3. Organic Carbon
6.3.1. Dissolved organic carbon export
6.3.2. Particulate organic carbon export
6.3.3. Total organic carbon export and evolution
7. CHAPTER Ⅶ: Conclusions and Perspectives
7.1. General Conclusions
7.1.1. Water regime
7.1.2. Suspended sediment
7.1.3. Organic carbon
7.1.4. Simple relationships proposed from this study
Conclusions Générales et Perspectives