Soil erosion and sediment transport have been modeled at several
spatial and temporal scales, yet few models have been reported for
large river basins (e.g., drainage areas
Effective management of sediment in rivers is becoming increasingly important from an economic, environmental, and ecological perspective. A recent study of 145 major rivers with longer-term records of annual sediment loads showed that approximately 50 % of them experienced a statistically significant upward or downward seasonal trend (Walling and Fang, 2003). The majority of them showed declining sediment loads because of dams and other river control structures trapping sediment. Moreover, human activities such as deforestation and water diversion might cause sedimentation or erosion in coastal regions. Such sediment-related problems could be more serious in the future because of further dam construction, climate variability, and deforestation (Walling, 2011; Zarfl et al., 2014). Currently, river basins in Southeast Asia have serious soil erosion potential and excessive sedimentation. They are also experiencing dramatic land surface changes, such as forest clearing, reservoir construction, and hydropower construction and water diversion (Tacio, 1993), because of rapid population and economic growth in the region (Walling, 2009).
A wide range of models exists for simulating erosion and sediment transport. These models differ in terms of complexity, processes considered, and the data required for model calibration and model use (Roberto et al., 2012). In general, there is no “best” model for all applications. The most appropriate model will depend on intended use, spatial scale, and characteristics of the catchment being considered. The Universal Soil Loss Equation (USLE) (Wischmeier and Smith, 1978) and its revised version (RUSLE) (Renard and Freimund, 1994) are widely used as tools for empirical assessment of soil erosion. Both USLE and RUSLE account for sediment eroded from the catchment in the long term (e.g., for 20 years). In these empirical equations, however, the deposition of sediment is not considered to occur in the modeled area.
A number of process-based soil erosion and sediment transport models
have also been developed, but those applications are limited to
individual storm events and small (max. 2.6
Process-based models are based on the solutions of fundamental physical equations describing stream flow and sediment production in a river basin. They represent the physical processes observed in the real world, such as surface runoff, subsurface flow, ground flow, and evapotranspiration. Process-based models provide several major advantages over empirical and conceptual model, including capabilities for estimating spatial and temporal distributions of net soil loss (or gain, in the case of deposition) for an entire hillslope or for each point on a hillslope. Further, process-based models can estimate sediment simulation on a daily, monthly, or an average annual basis. Since these models are process-based, they can also be seasonally interpolated and extrapolated to some extent to a broad range of conditions, including some conditions that might be difficult to measure with field testing. Given the complexity of the relationships affecting sediment dynamics, it is important to develop a robust process-based model of sediment dynamics that can be used to predict the consequences of natural systems as well as human-induced environmental changes and impacts, especially in large catchments.
This study aimed to develop a process-based distributed model that can
simulate the sediment dynamic process at a large basin scale. The
feasibility of the model was confirmed in large catchments (i.e.,
The important processes of sediment dynamics (soil erosion, sediment transport, and deposition) were modeled and integrated with a process-based distributed hydrological model (DHM) (Fig. 1). In the sediment model, sediment dynamics on hillslopes and rivers were separately modeled and systematically linked each other. The sediment model was developed using FORTRAN to create a compatible link to the adopted distributed hydrological model. The runoff and river routing were incorporated within the sediment model. Hydrological and sediment-related processes were calculated on a daily time-step. The overall model was designed to target suspended sediment load (SSL), because suspended sediment (SS) is dominant portion of the transported sediment in many of the world's rivers (Ongley, 1996), and it is frequently assumed that the suspended load makes up about 90 % of the total load in the world (Milliman and Meade, 1983).
The distributed hydrological model used in this study is a geomorphology-based hydrological model (GBHM) developed by Yang et al. (2001). It solves the continuity, momentum, and energy equations using two modules: hillslope module and river routing module.
In the GBHM, the target watershed is divided into grids, and a digital elevation model (DEM) is used to determine the flow direction and accumulation pattern that creates the river network. Each subbasin is divided into a number of flow intervals. In the subbasins, flow intervals are defined as a function of distance from the subbasin's outlet. Lateral flow to the main stream estimated by accumulating runoff at each grid in one hillslope unit. This means that all hillslopes of a flow interval drain into the main stream in this model. The flow interval–hillslope system enabled the GBHM to realize a fast flow computation even in a large basin. The hillslope unit is viewed as a rectangular inclined plane with a defined length and unit width. The inclination angle givens by the corresponding surface slope.
In the hillslope model, each grid is divided into four layers: canopy, soil surface, unsaturated zone, and groundwater. Vegetation covered the surface soil and prevented direct rainfall onto the land. The deficit of canopy interception is calculated by vegetation coverage and leaf area index. The evapotranspiration module simulated the water volume that evaporated from the surface soil and transpirated from the canopy, where pan observation could also be used. In the module, Priestley–Taylor's method was applied for the canopy water storage, root zone, surface storage, and soil surface. In order to describe the unsaturated zone water flow, a vertical one-dimensional Richards equation is used with soil infiltration rate and soil moisture contents in the root zone. Saturated water flow and exchange with the river is described using basic mass balance equations and Darcy's Law. The simulation module of surface water flow estimated the infiltration excess and saturation excess discharging into the river system as lateral flow.
In the river routing system, the Pfafstetter numbering system is applied to track water flow efficiently from upper to downstream. The water routing on the river network is determined along the river stream using one-dimensional kinematic wave equations. Further details are described by Yang et al. (2001).
Soil detachment by raindrop impacts was estimated by Eq. (
The kinetic energy for direct rainfall
Soil detachments by flow and sediment deposition in rivers are
generally considered to occur simultaneously. Flow detachment or
deposition can be expressed by Eq. (
The inflow to a dam,
In the following case studies, the dam operation rule was applied, and
release was assumed to be equal to observed release from the Bhumibol
and Sirikit dams in the Chao Phraya River Basin. In contrast, in the
Mekong River Basin, the mean annual discharge from the Lancang
subbasin (2332.29
The Chao Phraya River Basin covers about one third of Thailand, which
is approximately 160 000
The climate in Thailand is strongly affected by the Southeast Asian
monsoon and characterized by distinct rainy and dry
seasons. Basically, the rainy season starts at the middle or end of
May and lasts until the middle of October. Annual precipitation in the
Chao Phraya River Basin varies between 1000 and 1500
Geographical information for the Chao Phraya River Basin (e.g.,
topography, soil type, and land use) was collected for the development
of a hydrological model. A DEM was obtained from the Shuttle Radar
Topography Mission (URL:
River discharge and SSL were calibrated and validated in combination
with dam operation for the period from 2001 to 2010 at four stream
gauges in the upper region (P73-Ping River, W3A-Wang River, Y37-Yom
River, and N13A-Nan River) and one stream gauge in the outlet (C2-Chao
Phraya River) (Fig. 2). Taking into account the availability of data,
the monthly river discharge and SSL data for 2001 were used for the
calibration model at all five stream gauges, whereas the observation
data from 2002–2010 were used for validation. For the parameter
calibration, a semi-automatic calibration method was used. It was the
Shuffled Complex Evolution (SCE) algorithm (Duan et al., 1992). It was
implemented in 2001 to identify suitable parameters. The dominant
factors affecting the hydrological process and soil erosion, such as
land use and soil characteristics, were considered for parameter
calibration, as listed in Table 1. As for parameters related to
sediment transport and soil erosion, the FAO global soil dataset was
used to consider spatial distribution of soil properties. The
parameters of sediment detachability from rain drop (
The monthly calibration for the hydrological and sediment process was
implemented with SCE in Chao Phraya in 2001 (Table 1). The parameters
for sediment (
Model evaluation revealed that the river discharge simulation
performed satisfactorily, as shown by NSE and
Regarding SSL, NSE for all stream gauges was larger than 0.5
except at C2 (Table 2). The simulation results captured the high peak
of SSL during rainy season, as shown in Fig. 3. Overall, the
performance of the model in the Chao Phraya River Basin indicated
sufficient accuracy for long-term simulation (Table 2). The results
from C2, located at the lowest reach, were not as good as in other
stream gauges. At C2, the results indicate underestimation, even
though the total simulated river discharge at the lower reach was
overestimated and supposedly resulted in higher simulated SSL in the
lower reach than in the upper reach. However, it appears the
simulation error may have been larger at the lower reach because of
accumulating uncertainty. The average total annual SSL was estimated
to be
It is inferred that the process of soil loss was strongly influenced
by rainfall intensity. This was clearly shown by the simulated SSL at
the Nan River (Fig. 3), where rainfall is higher (1341.8
The simulated SSC also shows good correlations with observed data at
all upper stream gauges, indicated by
The sensitivity of modeled SSL was also investigated for the
reasonable ranges of the input parameters. The target parameters for
this sensitivity analysis were soil detachability from rain drop
(
First, results were obtained by changing the detachability of soil
(
Second, we focused on
Third, soil cohesion (
Sensitivity analysis was conducted for three parameters to evaluate
the reliability of the model for simulating sediment
dynamics. Overall, the two input parameters (
The second study area was the Mekong River Basin, covering an area of
approximately 795 000
In this basin, acrisols were found to be the dominant soil type. These are tropical soils that have a high clay accumulation in a horizon and are extremely weathered and leached. Their characteristics include low fertility and ease of erosion if they are used for arable cultivation. The average textures of soils in the Mekong River Basin are 27.1 % sand, 30.4 % silt, and 42.5 % clay (Kyuma, 1976). The forest coverage in the Mekong River Basin is 30.5 %. The agricultural land coverage is 40.7 %. The rest of the areas are shrubland (17.2 %), urban (2.1 %), and water bodies (8.7 %) (MRC, 2000). This study examined the model outputs (river discharge, SSL, and SSC) at three hydrologic stations; 1-Chiang Sean, 2-Khong Chiam, and 3-Phnom Penh (Fig. 6).
The input data for the model include weather data, topography data,
soil properties and land cover. In this study, the GTOPO30 global DEM
data with a horizontal grid spacing of
Annual records of river discharge and SSC in the study were extracted
from the historical record published by the MRC (Mekong River
Commission, 2005). The historical record tabulated measurements of
river discharge, SSC, water quality, and other physical
characteristics at gauging stations located along the Mekong River
Basin and those of the river's tributaries. In this study, river
discharge and SSC records from the three targeted stations were
identified and used to calibrate and validate SSL simulation. The
stations were selected based on their relative locations and the
completeness of river discharge and sediment records at the
station. Unlike river discharge, which was measured daily, SSC was
monitored monthly. SSC samples were collected near the surface of the
river (0.3
The river discharge and SSL were simulated by considering an existing
dam in the Chinese section of the main stream (Manwan Dam). The model
simulated river discharge, SSL, and SSC for 10 years from 1991
to 2000, and three stream gauges along the main stream were adopted
for calibration and validation (Fig. 6). The daily river discharge and
sediment data for the period from 1991 to 1995 were used for
calibration, whereas the data from 1996 to 2000 were used for
validation. For the sediment particle size (d
The river discharge of the Mekong was well simulated at all three
stations (Table 3) (refer to Fig. B1 for hydrographs). The NSE
values for river discharge at Chiang Sean, Khong Chiam and Phnom Penh
were larger than 0.7 for calibration and validation from
1991 to 2000. The average correlation
For the sediment model, we calibrated the same parameters as in the
Chao Phraya River basin. In the Mekong River, wider ranges were
determined for
Figure 7 shows the simulated results of monthly SSL compared with
measurements at three gauging stations from 1991 to 2000. The simulated
results are in good agreement with observations, as summarized in
Table 3. NSE was larger than 0.6 for the upper, middle, and
lower stations in the calibration (1991–1995) and validation
(1996–2000) periods, except in the case of the validation period for
the upper station (Chiang Sean) (NSE
The simulated total annual SSL at Phnom Penh fluctuated over the
period 1991–2000 (average
The monthly SSC simulation at the stream gauging stations for the
period 1991–2000 is shown in Fig. 8. The correlation coefficient
A sensitivity analysis was applied for SSL in the Mekong River Basin
in the same manner as in the Chao Phraya River Basin. First, all the
parameters were set to calibrated values, which were
The results revealed that SSL increases slightly in September when
Regarding
The analysis of soil cohesion (
The three input parameters that describe soil erodibility have
a significant influence on the output of SSL. Generally, soil cohesion
(
In this study, a physically-based model of sediment transport
targeting a large basin scale was developed and coupled with
a distributed hydrological model. The model enables us to simulate
rainfall–runoff processes and sediment transport on hillslope and
within a river network. In its application to the Chao Phraya River
and Mekong River basins, the sediment dynamics (i.e., yield and
erosion) were reasonably simulated in hillslope areas. As it is
a grid-based model, it can identify locations of serious sediment
dynamics by a fine grid scale. Moreover, the present model
applications estimated soil cohesion (
However, the present model assumed a single SS size instead of a wide range of SS sizes, due to limited information in both case studies. Thus, insufficient modeling of SS size distribution might have limited the applicability of the sediment model in the case studies. Therefore, the model performance may be further improved by incorporating multi-size sediment particles into the model. Uncertainties in terms of model inputs, parameters and structure may also have influenced the simulation results. For example, the estimation of net sediment detachment (Eq. 9) could be improved by revising the equations. Currently, this equation (Eq. 9) assumed that the soil particles were detached (limitation to deposition) and limited by factors such as soil cohesion. Thus, this equation should be improved by considering the reasonable balance between erosion and deposition, especially for river basins. Sediment management in river basins is highly affected by both processes.
Nevertheless, the outputs from this model at the basin scale may provide useful information to developers, decision makers, and other stakeholders when planning and implementing appropriate basin-wide sediment management strategies, which can also be integrated with water resource management. The model could also be used also to project the anthropogenic impacts on sediment dynamics under different scenarios in large river basins.
S. Zuliziana and K. Tanuma made substantial contributions to model develop, simulations, data collection and analysis, and drafting and editing the manuscript. C. Yoshimura and O. C. Saavedra developed the research framework and models and helped edit the manuscript.
This research for Chao Phraya River was supported by Asian Core Program of Japan Society for the Promotion of Science (JSPS). Also, the part of the modeling work for Mekong River was supported by JSPS Core-to-Core Program (B. Asia–Africa Science Platforms) and Collaborative Research Program (CRA) of AUN/SEED-Net.
Model parameters calibrated for Chao Phraya River and Mekong River basins.
Model performance indicators for monthly river discharge, SSL, and
SSC in Chao Phraya River Basin from 2001 to 2010. NSE and
Model performance indicators for monthly river discharge, SSL, and SSC in Mekong River Basin from 1991 to 2000. NSE and
Structure of the distributed sediment model integrated with a process-based distributed hydrological model.
The target area of Chao Phraya River Basin.
Monthly suspended sediment load (SSL) at stream gauge stations in Chao Phraya River Basin for 2001–2010.
Average monthly suspended sediment concentration (SSC) at stream gauge stations in Chao Phraya River Basin for 2001–2010.
Sensitivity of suspended sediment load (SSL) at P73 in Chao
Phraya River Basin to
The target area and the modelled river network of Mekong River Basin.
Monthly suspended sediment load (SSL) at stream gauge stations in the Mekong River Basin for 1991–2000.
Monthly suspended sediment concentration (SSC) at stream gauge stations in Mekong River Basin for 1991–2000.
Sensitivity of suspended sediment load (SSL) at Khong Chiam
station in Mekong River Basin to
Monthly average river discharge at stream gauge stations from 2001 to 2010 at Chao Phraya River Basin.
Monthly average river discharge at stream gauge stations from 1991 to 2000 at Mekong River Basin.