Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China

Quaternary International xxx (2015) 1e11

Contents lists available at ScienceDirect

Quaternary International journal homepage: www.elsevier.com/locate/quaint

Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China Jian Hua Gao a, *, Jianjun Jia b, Albert J. Kettner c, Fei Xing c, Ya Ping Wang a, Jun Li d, Fenglong Bai d, Xinqing Zou a, Shu Gao a a

Ministry of Education Key Laboratory for Coast and Island Development, Nanjing University, Nanjing 210093, China Key Laboratory of Submarine Geo-Sciences, Second Institute of Oceanography, State Oceanic Administration, Hangzhou 310012, China CSDMS Integration Facility, INSTAAR, University of Colorado, Boulder, CO 80309-0545, USA d Qingdao Institute of Marine Geology, Qingdao 266071, China b c

a r t i c l e i n f o

a b s t r a c t

Article history: Available online xxx

Reservoir interception has significantly affected the fluvial sediment budget as well as the sedimentary processes of the entire Changjiang catchment. To evaluate the impact of reservoirs, we analyze the combined effects of 1037 large and medium-sized reservoirs on the fluvial flux in general, and more specifically on the sedimentary processes in the middle and lower reaches. Results indicate that reservoir emplacement in the Changjiang catchment currently reduces the sediment load towards the East China Sea by 453 Mt y
1. Estimates at Yichang station show that the sediment discharge would exceed 555 Mt y
1, if there were no reservoirs involved. It is expected that in the near future, more dams will be constructed. The entire reach of the Changjiang River can be divided at Yichang station into two distinctly characterised reaches with regard to sedimentation, where the upper reach exhibits mostly siltation (over 589 Mt y
1 of sediment deposition), and the lower reach is affected by erosion (sediment loss, including sand extraction, exceeding 112 Mt y
1). As a consequence, the sediment flux to the sea will further decrease to 100 Mt y
1. Due to human interference, the upstream sediment load reduced and caused significant changes in the erosion/deposition pattern of the middle and lower reaches, which together altered the terrestrial sediment input to the sea. Before 2003, the upstream reaches were the dominant sediment source. After 2003, the sediment contribution of the middle and lower reaches became more important, and its sediment contribution will further increase to 78% of the total sediment load reaching the sea, after completion of the cascade reservoirs at the Jinsha Tributary. Hence, the middle and lower reaches are converting from a sediment sink to a major sediment source. © 2015 Elsevier Ltd and INQUA. All rights reserved.

Keywords: Reservoir emplacement Water and sediment discharge Sediment interception Changjiang River

1. Introduction Under natural conditions, temporal variations in fluvial water and sediment discharges are influenced mainly by changes in climate that indirectly can alter the land cover (Farnsworth and Milliman, 2003). During the last century, human activities such as freshwater extraction, sand mining of riverbeds, land use changes in river catchments and damming of rivers were intensified at an increasing rate, altering most global river systems (Syvitski, 2003,

* Corresponding author. E-mail address: [email protected] (J.H. Gao).

2011; Wang et al., 2008a). Reservoir construction for e.g. flood control, storage of freshwater for irrigation, navigation, and generation of hydroelectric power, has the most significant anthropogenic impact on riverine fluxes on a global scale. By reducing the connectivity of rivers, watersheds become more fragmented, resulting in changes in the hydrological processes, downstream river channel morphology, and the depositional and erosional processes of estuary and subaqueous deltas (Milliman, 1997; Gao et al., 2011; Gupta et al., 2012). Currently, there are more than 45,000 large and 800,000 medium to small reservoirs in the world (World Commission of Dams, 2005), which intercept ~25e30% of the global fluvial sediment € ro €smarty et al., 2003). This reduction in sediment towards the (Vo

http://dx.doi.org/10.1016/j.quaint.2015.03.015 1040-6182/© 2015 Elsevier Ltd and INQUA. All rights reserved.

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

2

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

oceans has its impact on river and coastal geomorphology as well as threatening fragile ecosystems (Walling, 2006; Li et al., 2007; Gao and Wang, 2008). The reduction in fluvial sediment load caused by reservoir emplacements has become a topic of global concern (Syvitski et al., 2005). Since 1950, China has built more than half of the world's large reservoirs, mainly for hydro-electrical purposes (Fuggle and Smith, 2000). Most of these are located in the Changjiang catchment (GMWR, 2009). The Changjiang River is 6370 km long and is globally ranked as the fifth river system in terms of water discharge, and ranks fourth when considering sediment load (Milliman and Farnsworth, 2011). There are over 50,000 reservoirs in the Changjiang catchment (Gao, 2006, 2007) and, therefore, the Changjiang River has become one of the highest impacted river systems in the world (Wang et al., 2011; Yang et al., 2011). The Three Gorges Dam (TGD) became operational in 2003 and is currently the largest concrete gravity dam in the world. Due to its high capacity of sediment retention, the impacts of the TGD on the middle and lower reaches of the Changjiang River and its estuary became a focus for recent studies (e.g., Yang et al., 2002; Dai et al., 2008; Wang et al., 2013). However, few attempts have been made to quantify the overall impact of all reservoirs on the sediment flux. One might argue that a large reservoir such as the TGD has a significant influence on the total suspended sediment load. However, smaller reservoirs are often located closer to the sediment source areas, and as a consequence, they could collectively play a similar or even more important role as compared to large reservoirs (Shi et al., 2012). Hu et al. (2011) reported already a significant reduction in the annual sediment flux towards the sea before the emplacement of the TGD. Neglecting these small reservoirs may result in misinterpreting the quantity of sediment trapped by large dams, inducing bias in evaluating human's impacts on sediment discharge of the Changjiang River. A proposed cascade reservoirs development plan for the upper stream of the Changjiang River has recently been approved, so water and sediment discharge will be subjected to even further adjustments in the coming decade (Q.S. Chen et al., 2008). To better predict the impact of future emplacement of dams, it is of importance to integrate all dams of the Changjiang catchment, and systematically analyze the relation between quantity of reservoir volume versus the sediment load reduction, in combination with the downstream effects, resulting in changes in freshwater and sediment discharge of the middle and lower reaches and to the estuary.

sediment load without the influence of dam interception, and evaluate the contribution of dam interception to the reduction of sediment load; (3) investigate changes in sedimentary processes of the Changjiang River basin due to sediment reduction; and (4) predict future changes of water and sediment discharge of the Changjiang catchment, after the completion of the upstream cascade reservoirs emplacements, and explore the downstream impact of these reservoirs on the channel and the Changjiang estuary. 2. Regional setting The Changjiang catchment covers an area of ~1.80  106 km2. The upper reach of the river, from the hinterland to the Yichang gauging station (Fig. 1), is the main sediment source of the entire catchment (Shi, 2008). Its middle reach extends from Yichang to Hukou gauging station and the section between Hukou and Datong gauging station defines the lower reach of the river (Fig. 1). No large tributaries join the lower reach, and downstream of Datong gauging station, the river is influenced by tides. Therefore, Datong station is the last station of the Changjiang River, and its records are generally used to represent the riverine flux into the East China Sea. The geomorphology is characterized by mountains and hills in the upstream area and by extensive fluvial plains with numerous lakes in the downstream area. The middle reach of the Changjiang River is a meandering river system, which evolves to a more braided system in the lower section of the river (Yin et al., 2007). The Changjiang River upstream of Datong gauging station includes seven major tributary basins: Jinsha, Min, Jialing, Wu, and Han rivers, together with two lakes: Dongting and Poyang Lakes (Fig. 1; Table 1). Dongting Lake is the second largest freshwater lake of China, and joins the main stream from the south, at Chenglingji gauging station (Du et al., 2001). The surface area of Dongting Lake decreased from 4350 km2 in 1949e2623 km2 in1995 due to siltation and land reclamation (BCRS, 2000). Poyang Lake is the largest freshwater lake of China, located at the junction of the south bank of the Changjiang River. Poyang Lake is a throughput type of lake; receiving runoff from five smaller tributaries: Gan, Fu, Xin, Rao and Xiu rivers and passing the freshwater through to the Changjiang River after regulation at Hukou. The contributed area to Poyang Lake is 16.2  104 km2, accounting for 9% of the Changjiang River drainage area (Shankman et al., 2006).

Table 1 Characteristics of the seven major tributary basins upstream of Datong station. River systems

Hydrological station

Lengtha (km)

Areaa (104 km2)

JinshaRiver Min River JialingRiver Wu River Han River Dongting Lake Poyang Lake Total

Pingshan Gaochang Beibei Wulong Huangzhuang e e e

2316 1062 1119 1018 1532 e e e

34 16.09 16 8.7 17.43 26.28 16.22 e

Discharge (km3 y
1)

Sediment load (Mt y
1)

I

II

IeII

I

II

IeII

143.0 84.93 65.32 48.56 46.51 166.28 100.6 655.19

146.48 77.97 58.66 44.08 44.41 162.32 102.60 636.53


3.48 6.95 6.66 4.47 2.1 3.96
2.0 18.67

232 45 108 24 39 26 17 491

164 31 26 8.0 8.0 10 6.0 253

68 14 82 16 31 16 11 238

I and II denotes the period of 1956e2010 and 2001e2010, respectively. Sediment load at Dongting Lake system represents the total load of its four tributaries: Xiang, Zi, Yuan and Li rivers. Similar for the sediment load of Poyang Lake system, which represents the total sediment load of its tributaries: Gan, Fu, Xin, Rao, and Xiu Rivers. a Represented data originating from Lin et al. (2010).

In this study, we aim to: (1) systematically analyze variation of reservoir storage capacity of the entire Changjiang catchment and its tributaries, and its effect on changes in downstream sediment load; (2) reconstruct the inter-annual variation of upstream

Recently, the water and sediment discharges of the Changjiang catchment have been significantly modified (Yan et al., 2011) due to the impact of climate variation, land cover changes, and human activities (Chen et al., 2001). A considerable number of reservoirs

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

3

Fig. 1. Sketch map of Changjiang catchment: a) location of the hydrologic stations for the Changjiang catchment, and b) the distribution of large ( ) and medium sized ( ) reservoirs in the Changjiang catchment. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

have been constructed in the Changjiang basin since the 1950s; by the end of the late 1980s, 1931 dams with a total water storage capacity of 20.5 km3 were built in the upstream region (BCRS, 2002). Since 2003, reservoir construction has been further accelerated; particular in the upstream tributary (Jinsha) where many cascade reservoirs are currently under construction or just finished (Shi and Du, 2009). The Changjiang catchment was once rich in forest resources. However, due to rapid economic development and population increase, forest area was significantly reduced (Lu and Higgitt, 2000). The total forest coverage in the Changjiang catchment decreased from 30% in the early 1930s to 20% in the early 1950s (Xu, 2000). By further utilizing waste land, the forest coverage was further reduced to ~10% in the early 1960s (Xu, 2009). The rapid decline in

forest cover led to soil erosion increase. The affected area that suffered from soil and water lose increased from 36  104 km2 in the 1950s to 62.2  104 km2 in the mid-1980s, accounting for 34.6% of the total area of the Changjiang catchment (BCRS, 2007). Presently, the lower Jinsha and Jialing Rivers, the middle reach of Min River, upstream of Wu River, the TGD area and upstream of Han River are the most serious deteriorated areas of the Changjiang catchment that suffer from soil and water losses (Xiong et al., 2009). At the beginning of 1989, a large-scale soil conservation campaign was implemented for the high sediment yielding regions of the upper Changjiang basin. In many small river valleys, serious erosion problems were tackled in a comprehensive way, concerning mountain slopes, rivers, farmland, and forests. Subsequently, the area of deteriorated land decreased to 53.1  104 km2 in the late

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

4

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

1990s, equaling to 29.5% of the total area of the Changjiang catchment (BCRS, 2007). 3. Data and methods 3.1. Freshwater and sediment discharge data We collected annual discharge and sediment load of 24 hydrological stations distributed in the main reach and 7 in the tributaries upstream of Datong station. The dataset covers a 57-year period (1953e2010) for three of the hydrological stations located at the main river (Yichang, Hankou and Datong station), and 54-years of data (1956e2010) for the other stations (Fig. 1a). The gauging stations of the four upstream tributaries are: Pingshan station for the Jinsha River, Gaochang station for the Min River, Beibei station for the Jialing River and Wulong station for the Wu River. Of the 24 hydrological stations, 17 are located in the tributaries of the middle and lower reaches of Changjiang River, of which the Huangzhuang station is located at Han River, and all of the other stations are situated in the drainage basins that flow towards Dongting and Poyang Lake. Ten hydrological stations belong to the Dongting Lake system: four stations, one for each of the four tributaries entering Dongting Lake, namely, Xiangtan station (Xiang River), Taojiang station (Zi River), Taoyuan station (Yuan River) and Shimen Station (Li River); 5 stations, one for each of the five different entrances where the Changjiang River flows into Dongting Lake, namely, Mituoshi station, Xinjiangkou station, Shadaoguan station, Ouchi (Kang) station and Ouchi (Guan) station; and Chenglingji station where the water flows from Dongting Lake into the Changjiang main river channel. Of the six hydrological stations of the Poyang Lake system, five are monitoring the 5 tributaries, namely, Waizhou station (Gan River), Lijiadu station (Fu River), Meigang station (Xin River), Wanjiabu station (Xiu River) and Hushan station (Rao River), and the Hukou station is monitoring the Poyang Lake flow towards the Changjiang River. Annual water and sediment discharge data of 1956e2001 were obtained from the Changjiang Water Resources Commission (CWRC), and those from 2002 to 2010 were obtained from the Bulletin of China River Sediment published by the Ministry of Water Resources, China (BCRS, 2002e2010; available at: http:// www.mwr.gov.cn/zwzc/hygb/zghlnsgb/). The recorded data were obtained from water samples collected throughout the water column at a fixed gauging section. Water discharge data were collected averaging velocity data obtained at various locations across a river cross-section in accordance to the Chinese national standards (Dai et al., 2008). 3.2. Dam and reservoir data According to the Ministry of Water Resources of China (MWRC), large reservoirs are defined as reservoirs with a storage capacity > 0.1 km3, medium sized reservoirs have a storage capacity between 0.01 and 0.1 km3, and small reservoirs can store < 0.01km3water. Here, the reservoir storage capacity index (RSCI) is defined as the ratio of the reservoir storage capacity and the annual average water discharge of the contributed catchment. Data on reservoir construction during 1949e2001 and 2002e2007 were obtained from the MWRC (2001) and the annual reports published by the MWRC (http://www.mwr.gov.cn/zwzc/ hygb/slbgb/), respectively. There are 1,037 large and medium sized reservoirs located upstream of Datong station (Fig. 1b), and the database includes information of reservoir storage capacity, construction and impoundment time. Although the quantity of large and medium sized reservoirs is less than that of small reservoirs, their total reservoir storage

capacity exceeds the total storage capacity of small sized reservoirs by far. In numbers, only 4.3% of all reservoirs in China are large and medium sized reservoirs, but these have a total storage capacity of 90.9% of the total reservoir storage capacity (GMWR, 2009). Therefore, it is valid to assume that only considering the impact of large and medium sized reservoirs over time will provide a good representation of changes over time of the total storage capacity for the Changjiang catchment. 3.3. Analytical method The ManneKendall test (M
K test) is a statistical method to determine the significance of monotonic trends in hydrometeorological time serials (Mann, 1945; Kendall, 1955). We used a modified M
K method to test variations of water discharge/ sediment load data (Hamed and Rao, 1998): Xt ¼ (x1, x2, x3, …, xn), the accumulative number mi of samples that xi > xj (1  j  i) is calculated. The normally distributed statistic dk can be expressed as:

dk ¼

k X

mi

2kn

(1)

i¼1

Under the null hypothesis of no trend, the mean and variance of the normally distributed statistic dk is as follows:

E½dk  ¼

kðk
1Þ 4

Var½dk  ¼

(2)

kðk
1Þð2k þ 5Þ 72

2kn

(3)

Based on the above assumption, the normalized variable statistical parameter UFk is defined as:

dk
E½dk  ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi UFk ¼ p var½dk 

k ¼ 1; 2; 3……n

(4)

where UFk is the forward sequence, and follows the standard normal distribution. In a two-sided test for trend, the null hypothesis should thus be accepted if jUFj UF1-a/2 (here a ¼ 5%) at the level of significance. A positive value of UF designates an “increasing trend”; likewise, a negative denotes “decreasing trend”. The backward sequence UBk is acquired by using the same equation but with the retrograde sample. The C values calculated with progressive and retrograde series are named C1 and C2, respectively. The intersection point of the two lines, C1 and C2 (k ¼ 1, 2....n) provides the point in time of the beginning of a developing trends within a time series. Assuming a normal distribution with a significant level of P ¼ 0.05, a Mann-Kendal statistics C > 1.96 indicates a significant increasing trend; while a C lower than
1.96 indicates a significant decreasing trend. 4. Results and discussions 4.1. Changes in water discharge and sediment load The variations of water discharge and sediment load during 1953e2010 at Yichang, Hankou and Datong station are shown in Fig. 2. Their M
K trends indicate that (Fig. 3): water discharge at the three stations does fluctuate but fail to pass the confidence level of the 95% test. The discharge at all 3 stations has a high variation but no significant increasing or decreasing trend. However, the sediment load variation at Yichang, Hankou and Datong station show reducing trends at 1985, and pass the confidence level of the 95% test at 1996, 1996 and 1989, respectively. Furthermore, we

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

5

Fig. 2. Annual discharge (red lines) and sediment load (grey bars) for Yichang, Hankou and Datong stations. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. The M
K trends of sediment load and water discharge, and cumulative water discharge and sediment load for Yichang (a), Hankou (b) and Datong (c) station.

analyzed the cumulative water discharge and sediment load variations over time for the three stations. Changes in the cumulative water e sediment load diagrams reveal changes in sediment concentration. Results show that three stepwise reduction stage periods were observed, namely, 1953e1989, 1990e2002, and 2003e2010 (Fig. 3). This change in trend is not only similar to the pattern revealed by the M
K test, but also reflects the impact of TGD emplacement (2003). During 1953e1989, the annual average sediment loads at Yichang, Hankou and Datong stations were 526, 425, and 461 Mt,

respectively. These were reduced to 390, 318 and 332 Mt respectively, during 1990e2002. During these two periods, the annual sediment discharge at Yichang station was significantly higher than at Hankou and Datong stations. In response to the impoundment of the TGD, the sediment loads observed at Yichang, Hankou, and Datong stations during 2003e2010 further decreased to 55, 118 and 152 Mt, respectively. There is a sharp reduction in sediment load at Yichang station. The total sediment load of Jinsha River, Jialing River, Min River, and Wu River reveals similar trends as observed at Yichang station

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

6

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

Fig. 4. Annual changes in water (red line) and sediment discharge (grey bars) of: a) the totals of Jinsha, Min, Jialing and Wu River, b) Han River (Huangzhuang station), c) the Dongting Lake, and d) Poyang Lake systems. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(Fig. 4a). Researchers suggest that the sediment load reduction at Yichang station was mainly caused by the TGD (Kuang et al., 2013). The sediment supply at Yichang station, representing the four tributaries, decreased to its lowest level ever recorded during 2003e2010. Thus, at least part of the decrease in sediment load at Yichang station can be explained by a reduction in sediment supply of the upstream tributaries. This sediment reduction is most likely more pronounced due to the impact of the TGD. The water and sediment discharge of the three tributaries of the middle and lower reaches also indicate reduction trends (Fig. 4b, c, d).

4.2. Temporal changes in reservoir storage capacity and sediment load 4.2.1. Reservoirs storage capacity over time Total Reservoir Storage Capacity (TRSC) upstream of Datong station shows an increasing trend from 1953 to 2010 (Fig. 5). By the end of 2010, the TRSC of large and medium sized reservoirs is 142.6 km3 and has a RSCI of 16.0% (Table 2). The TRSC upstream of Yichang station indicates that there is a slow increasing trend from 1953 to 1989 (to 12.1 km3) after which it accelerates from 1990 to 2002 (to 25.7 km3). Due to the emplacement of TGD in 2003, the TRSC upstream of Yichang rapidly increased to 65.0 km3, and has a RSCI of 15.3%.

During the last few decades, the cumulative water and sediment discharge relation upstream of Yichang and Datong stations changed, indicating that there is, proportional to water, less sediment transported. In addition, the changes of RSCI and sediment load indicate that, the decrease of sediment load is highly related to the increase of RSCI (Fig. 5). The above two relationships reflect the impact dams have on sediment load.

4.2.2. Sediment load trapped by reservoirs The deposition rate of reservoirs in here is defined as the annual percentage reduction of reservoir storage capacity due to sediment retention. The average reservoir deposition rate upstream Changjiang River is 0.65% for large sized reservoirs, 0.39% for medium sized reservoirs, and 0.90% for small sized reservoirs (BCRS, 2000). According to H.B. Li et al. (2011), the bulk density of deposited sediment in reservoirs upstream the Changjiang River is 1.23 t m
3. When applying this bulk density to large and medium sized reservoirs that are situated in the Jinsha, Min, Jialing, and Wu rivers, those fluvial systems trap 60 Mt y
1, 9 Mt y
1, 45 Mt y
1 and 35 Mt y
1, respectively. Sediment load intercepted by reservoirs upstream the Changjiang River also show that (Fig. 6): interception of sediments was less before 1980, then gradually increased to 124 Mt y
1 in 1989 and rapidly rose at the beginning of 1990, to208 Mt y
1 in 2002. The average sediment load interception of the TGD equals 146 Mt y
1

Table 2 Reservoir characteristics of the Changjiang catchment: number, storage capacity and RSCI values. Section

Number of reservoirs

Reservoir storage capacity (km3)

RSCI (%)

Upstream of Yichang station Yichang station e Datong station Upstream of Datong station Downstream of Datong station The total of the whole Changjing River

224 813 1037 95 1132.00

65.0 77.6 142.6 5.1 147.7

15.3 e 16.0 e e

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

7

Fig. 5. Increasing reservoir storage capacity (a) and the relationship between RSCI and sediment load (b) for Yichang and Datong station. The red dashed lines denote the average sediment load during different periods. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

after impoundment in 2003. Total sediment interception of reservoirs of the Changjiang River amount to 354 Mt y
1, of which the TGD alone and all the other reservoirs contributed 41.2% and 58.8%, respectively. The above estimate excludes the interception of a large number of small reservoirs. If we assume that the proportion of small sized reservoir storage capacity accounts for 0.9% of that of all reservoirs (GMWR, 2009), and annual deposition rate of these reservoir is also 0.9% (BCRS, 2000), and given the same bulk density (1.23 t m
3, Q. Li et al., 2011), the current sediment intercepted by small sized reservoirs would equal to 77 Mt y
1. In addition, there are a large number of ponds with storage capacity less than 10.00  104 m3 in

the upper stream of the Changjiang River, which form the main source for rural potable and irrigation water. According to CWRC (1997) there are 50.60  104 ponds with a total reservoir storage capacity of nearly 3.1 km3, trapping 0.06 km3 y
1 sediment. It is fair to assume that for a number of these ponds, e.g. those situated in small valleys, the sediment would not have reached the river regardless if there was a pond or not. However, assuming that the sediment delivery ratio to the river is 0.3 (Jing, 2009), then the contribution of these ponds to the downstream sediment load reduction is ~22 Mt y
1. The total sediment load trapped by all reservoirs and ponds upstream Changjiang River amount to 453 Mt y
1, exceeding the average sediment load passing Yichang station from 1953 to 2010 (430 Mt y-1). Notice that the TGD only intercept 32.2% of the total annual sediment that is trapped by all dams, and the quantity of sediment trapped by the other dams is 2.1 times of that of the TGD. Therefore, the influence of the TGD on the sediment load reduction of Changjiang River catchment should not be exaggerated. If no sediment was trapped, the average sediment load at Yichang station would have been 555 Mt y
1 without a significant increasing or decreasing trend over time (Fig. 7a); in addition, the sediment trapped by reservoirs show a highly inverse correlation to observed sediment loads at Yichang, Hankou and Datong stations (Fig. 7b), which further demonstrate that dam construction is the dominating factor resulting in sediment load reduction of Changjiang River catchment.

4.3. Changes in sedimentary processes

Fig. 6. Annual sediment interception of reservoirs and ponds of the Changjiang catchment. The annual sediment load trapped by the TGD during 2003e2010 was obtained from BCRS (2003e2010).

In order to analyzing changes in sedimentary process of different section of Changjiang main river, the sediment budget between Yichang-Hankou and Hankou-Yichang is calculated, according to following equations:

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

8

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

Fig. 7. The impact of reservoir emplacement on the sediment load: a) reconstructed sediment load purple dots and line) at Yichang station if there were not reservoirs in the upstream catchment. For comparison, the gray background shows the observed annual sediment load; b) relationships between sediment trapped in the reservoirs and the sediment load observed at Yichang, Hankou and Datong stations. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

SHankou ¼ DS þ SYichang þ SHan

(5)

SDatong ¼ DS þ SHankou þ SPoyang

(6)

where S is the sediment load of different hydrological station, DS is the quantity of deposited (
)/erosive (þ) sediment of the Changjiang main stream. The resulted indicated that, influenced by upstream sediment load reduction, the annual sediment deposition of the main river between Yichang and Hankou stations decreased from 52 Mt during 1956e1989 to 41 Mt during 1990e2002 (Fig. 8). After the emplacement of the TGD in 2003, this reach altered from a depositional to an erosive system, eroding on average annually 55 Mt. The main river channel between the stations Hankou and

Datong is dominated by erosion, 15 Mt y
1 during 1956e1989, 5 Mt y
1 during 1990e2002, and 21 Mt y
1 during 2003e2010. Due to upstream sediment load reduction, the contribution of sediment flowing from Changjiang River to Dongting Lake decreased from 95 Mt y
1 during 1956e1989 to 48 Mt y
1 during 1990e2002. However, Dongting Lake supplied 2 Mt y
1 sediment to Changjiang River after 2003e2010. Above changes suggest that Dongting Lake has been converting from a significant sink to a sediment source. Sediment record analyzes also indicate that the main river channel between Yichang and Datong station is a depositional system during the periods 1956e1989 and 1990e2002, which led to a sediment load decrease at Datong station of 37 Mt y
1and 36 Mt y
1, respectively. After 2003, sediment yield further reduced due to the impact of new emplaced reservoirs which started

Fig. 8. Fluvial sediment budget of different sections of the main river, sediment load trapped in reservoirs of Changjiang, as well as sediment load at Yichang, Hankou and Datong station during different periods.

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

channel erosion, contributing 76 Mt y
1of sediment at Datong station, which accounts for 48.0% of the averaged sediment load at Datong station (Fig. 8). According to river sand dredging practices in the middle and lower main river channel of Changjiang River issued by CWRC in 2003, the total sand extracted is just less than 34 Mt y
1 (Wang et al., 2008b). However, including less legal sand extractions, this would exceed 34 Mt y
1. Therefore, the actual sediment deficit of middle-lower channel is expected to be > 110 Mt y
1. 4.4. Estimated impact of future reservoir emplacements The Jinsha River is characterized not only for being the most important sediment source, but also for accommodating the largest hydropower bases upstream Changjiang River. According to the hydropower development plan, 12 large cascading reservoirs along the middle and lower Jinsha River will be built, with a total reservoir capacity of 81.0 km3 and RSCI of 56.6% (Feng, 1997), and the RSCI of Jinsha River will increase to 62.2%. Assuming a relative low RSCI of the catchment (less than 10%), we calculated the quantity of sediment trapped by its reservoirs. This method is not applicable if the RSCI will be >62.2% after completion of the cascade reservoirs. Feng et al. (2008) and Zhu (2000) demonstrated that the sediment load of Jinsha River might be reduced by 90% after construction of these reservoirs, with consideration of reservoir deposition, the current water and sediment condition, and future running pattern of cascade reservoirs. In addition, the Han River catchment is a valid analog for the Jinsha River catchment after emplacement of the cascade reservoirs. The RSCI of Han River catchment increased by 56.5%, and its sediment load decreased by 92.6%. Therefore, we are convinced that the sediment will be reduced by 90% is reliable, and the sediment load will be ~15 Mt y
1 (reduced by ~136 Mt y
1) after completion of the 12 cascading reservoirs. If the sediment discharge from the other three tributaries remains the same as of 2003e2010, the sediment load of the four rivers will become 82 Mt y
1, which is less than 62.4% of the average sediment discharge (218 Mt y
1) during this period. Therefore, the sediment trapped by reservoirs might increase to 589 Mt y
1 (Fig. 6). After 2003 (emplacement of the TGD), the relation between total sediment load of the four rivers and that at Yichang station significantly changed (Fig. 9a). Applying this relation, the average annual sediment load at Yichang station will only be 7 Mt y
1, which is only 12.7% of the present sediment load. If the function shown in Fig. 9b is used to further predict future sediment at

9

Hankou and Datong station after construction of the 12 cascading reservoirs, the average annual sediment load at the stations are70 Mt and 100 Mt, respectively. This is respectively 40.7% and 30.3% less than that of current observed load. Given the short period of observational data, and the deficiency of extreme events sample data, the predicted sediment discharge is not reflecting the actual annual sediment load in the future, but rather an average annual sediment load for the more typical years. Yang et al. (2007) also predicted future sediment flux at Datong station, and proposed that it is unlikely that the sediment flux to the rivers estuary will be less than100 Mt in the coming 50 years. However, given this study, it is likely that the sediment flux will become less than 100 Mt in the nearby future. X.Q. Chen et al. (2008) presented a method predicting the future sediment load at Datong station, their sediment predictions at Datong range from 101 Mt y
1 to 121 Mt y
1, which is in agreement with our predictions. We calculated sediment budget of different sections of Changjiang main river after completion of cascading reservoirs, using equations (5) and (6). Due to sediment load reduction in the upper part of the Changjiang River, the supply of sediment flux to the sea has changed its source from the upstream to the middle and lower reaches. After construction of the 12 cascade reservoirs, the predicted sediment recovery from Yichang to Hankou reach is 69 Mt y
1. Under the assumption that the Han River sediment load into Changjiang River remains 9 Mt y
1, 60 Mt y
1 of sediment will erode from the main riverbed and Dongting Lake will provide another 60 Mt y
1. Similarly, the predicted sediment supply from the section between Hankou-Datong is 30 Mt y
1. Assuming the sediment load from Poyang Lake remains 12 Mt y
1, then the riverbed will erode each year by18 Mt. Overall, the sediment load originating from riverbed erosion between the reach of Yichang and Datong, as well as the sediment contribution of Dongting Lake will reach 78 Mt y
1, accounting for 78% of the sediment load at Datong station. It is expected that in the nearby future, the entire reach of the Changjiang River will be divided at Yichang station into two distinctly characterised reaches with regard to sedimentation, with the upper reach exhibiting siltation of more than 589 Mt y
1 of sediment, and the lower reach experiencing erosion with a total sediment loss, including sand extraction, exceeding 112 Mt y
1. Currently, the reduction of sediment discharge to the sea, and its impacts on its estuary and the continental shelf has gained more and more attention (Syvitski et al., 2005; Gao et al., 2011). However, the river valleys also face environmental problems. For example,

Fig. 9. Sediment load at different gauge stations: a) The relationship between the sediment load of the total of the four upstream tributaries and Yichang station, and b) the correlation of sediment load at Yichang station to that of Hankou and Datong station.

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

10

J.H. Gao et al. / Quaternary International xxx (2015) 1e11

groundwater level is rising in some areas in the upper reaches, which may lead to land salinization of the riverbanks; the continuous erosion of middle and lower river channels will lower the water table of the riverbanks, causing instability of banks and embankment. Furthermore, the fragile ecosystem of both lakes and the Changjiang River will be challenged as a consequence of further modification of the sedimentary system. 5. Conclusions Although the water discharge of the Changjiang River does not show a significant change in trend, sediment load is stepwise decreasing over the last 50 years. The emplacements of reservoirs are testified to be the dominant factor leading to the sediment reduction towards the sea. Currently, the inclusion of all reservoirs and small ponds leads to a yearly sediment reduction rate of 453 Mt, among which 32.2% is caused by the TGD, and the remaining 67.8% is resulted by the other dams. As a result, the influences of the other dams on sediment load reduction of the Changjiang River should not be neglected. Due to reservoirs interception of sediments, the fluvial sediment budget of the whole Changjiang catchment has been greatly changed. The main river channel between Yichang and Hankou station, as well as Lake Dongting converted from a depositional to an eroding system. After the 12 cascading reservoirs of the Jinsha River will become operational, the Changjiang River will be divided at Yichang station: upstream the station will be characterized by sedimentation, more than 589 Mt of sediment will be deposited in its reservoirs and river channels; downstream the mid-lower reach mainly will undergo erosion, exceeding average sediment losses of 112 Mt y
1. The sediment flux towards the East China Sea will be further reduced to 100 Mt y
1. Geomorphologically, artificial modifications to the river valley result in readjustment of the catchment morphology, which will evolve towards a new equilibrium status. The source of terrestrial sediment to the East China Sea has been undergoing alteration, as a consequence of changes of system behavior of the Changjiang catchment. The sediment load of the Changjiang River entering the ocean was mainly contributed by the upper part of the catchment, whereas the sediment originating from the channel bed of the middle and lower reach has become more important as a sediment source. After 2003, 48.0% of the sediment to the sea originates from the middle and lower channel bed. After completion of the cascading reservoirs of the Jinsha River, the contribution of sediment to the sea, eroded of the middle and lower river channel will reach to 78%. The mid-lower river channels will completely convert from being a sediment sink to a source of sediment, entering the ocean. Acknowledgements The study was supported financially by National Basic Research Program of China (Grant No. 2013CB956503), and the Natural Science Foundation of China (Grant No.41376068 and 41476052). References BCRS (Bulletin of China River Sediment), 2000. Press of Ministry of Water Resources of the People's Republic of China. http://www.mwr.gov.cn/zwzc/hygb/zgstbcgb/. BCRS (Bulletin of China River Sediment), 2002. Press of Ministry of Water Resources of the People's Republic of China. http://www.mwr.gov.cn/zwzc/hygb/zgstbcgb/. BCRS (Bulletin of China River Sediment), 2007. Press of Ministry of Water Resources of the People's Republic of China. http://www.mwr.gov.cn/zwzc/hygb/zgstbcgb/. Chen, Q.S., Xu, Q.X., Chen, Z.F., 2008. Analysis on variation characteristics and influencing factors of runoff and sediment in the Wujiang River basin. Journal of Sediment Research 5, 43e48 (in Chinese with English abstract).

Chen, X.Q., Yan, Y.X., Fu, R.S., Dou, X.P., Zhang, E.F., 2008. Sediment transport from the Yangtze River, China, into the sea over the Post-Three Gorge Dam Period: a discussion. Quaternary International 186, 55e64. Chen, Z., Li, J., Shen, H., Wang, Z.H., 2001. Yangtze River of China: historical analysis of discharge variability and sediment flux. Geomorphology 41, 77e91. CWRC (Changjiang Water Resources Commission), 1997. Research in Three Gorge Project hydrology. Science and Technology Press of Hubei Province 10 (in Chinese). Dai, S.B., Lu, X.X., Yang, S.L., Cai, A.M., 2008. A preliminary estimate of human and natural contributions to the decline in sediment flux from the Yangtze River to the East China Sea. Quaternary International 186, 43e54. Du, Y., Cai, S., Zhang, X., Zhao, Y., 2001. Interpretation of the environmental change of Dongting Lake, middle reach of Yangtze River, China, by 210Pb measurement and satellite image analysis. Geomorphology 41, 171e181. Farnsworth, K.L., Milliman, J.D., 2003. Effects of climatic and anthropogenic change on small mountainous rivers: the Salinas River example. Global and Planetary Change 39, 53e64. Feng, J.L., 1997. An elementary introduction to the Stratagem and effect of the leading reservoir in Cascade. Yunnan Water Power 47 (2), 12e18 (in Chinese with English abstract). Feng, X.F., Yang, Q.Y., Zhang, O.Y., Wang, X.K., 2008. Sediment trap of Ertan Reservoir and its effect on sediment budget of Jinsha River. Journal of Sichuan University (Engineering Science Edition) 40 (6), 37e42 (in Chinese with English abstract). Fuggle, R., Smith, W., 2000. Large Dams in Water and Energy Resource Development in the People's Republic of China (PRC). Country Review Paper Prepared as an Input to the World Commission on Dams, Cape Town. http://www.dams.org. Gao, S., 2006. Catchment-coastal interaction in the Asia-Pacific region. In: Harvey, N. (Ed.), Global Change and Integrated Coastal Management: the AsianPacific Region. Springer, Dordrecht, pp. 67e92. Gao, S., 2007. The Three Gorges Dam: development and environmental issues. Macalester International 18, 146e171. Gao, S., Wang, Y.P., 2008. Changes in material fluxes from the Changjiang River and their implications on the adjoining continental shelf ecosystem. Continental Shelf Research 28, 1490e1500. Gao, S., Wang, Y.P., Gao, J.H., 2011. Sediment retention at the Changjiang subaqueous delta over a 57 year period, in response to catchment changes. Estuarine Coastal and Shelf Science 95, 29e38. GMWR (Gazette of the Ministry of Water Resources of China), 2009. Press of Ministry of Water Resources of the People's Republic of China. http://www. mwr.gov.cn/zwzc/hygb/slbgb/. Gupta, P., Kao, S.J., Dai, M., 2012. The role of mega dams in reducing sediment fluxes: a case study of large Asian rivers. Journal of Hydrology 464e465, 447e458. Hamed, K.H., Rao, A.R., 1998. A modified Mann-Kendall trend test for autocorrelated data. Journal of Hydrology 204, 182e196. Hu, B.Q., Wang, H.J., Yang, Z.S., Sun, X.X., 2011. Temporal and spatial variations of sediment rating curves in the Changjiang (Yangtze River) basin and their implications. Quaternary International 230, 34e43. Jing, K., 2009. Sediment delivery ratio in the Upper Yangtze River. Journal of Sediment Research 1, 53e59 (in Chinese with English abstract. Kendall, M.G., 1955. Rank Correlation Methods. Griffin, London. Kuang, C., Liu, X., Gu, J., Guo, Y., Huang, S., Liu, S., Yu, W., Huang, J., Sun, B., 2013. Numerical prediction of medium-term tidal flat evolution in the Yangtze Estuary: impacts of the Three Gorges project. Continental Shelf Research 52, 12e26. Li, Q., Yu, M., Lu, G., Cai, T., Bai, X., Xia, Z., 2011. Impacts of the Gezhouba and Three Gorges reservoirs on the sediment regime in the Yangtze River, China. Journal of Hydrology 403, 224e233. Li, H.B., Zhang, X.F., Xu, Q.X., 2011. Analysis and prediction of sediment trapped by large-scale reservoir group on upstream of Three Gorges Dam. Engineering Journal of Wuhan University 44 (4), 604e612. Li, L., Lu, X., Chen, Z., 2007. River channel change during the last 50 years in the middle Yangtze River, the Jianli reach. Geomorphology 85 (3), 185e196. Lin, Q.S., Huang, L., Yao, S.M., 2010. Analysis on variation of runoff and sediment load in main stream of upper Yangtze River in recent period. Yangtze River 41 (10), 5e8 (in Chinese with English abstract). Lu, X.X., Higgitt, D.L., 2000. Estimating erosion rates on sloping agricultural land in the Yangtze Three Gorges, China, from caesium-137 measurements. Catena 39, 33e51. Mann, H.B., 1945. Nonparametric tests against trend. Econometrica 13, 245e259. Milliman, J.D., 1997. Blessed dams or damned dams? Nature 388, 325e326. Milliman, J.D., Farnsworth, K.L., 2011. River Discharge to the Coastal Ocean: a Global Synthesis. Cambridge University Press, Cambridge. MWRC (Ministry of Water Resources of the People's Republic of China), 2001. The Code for China Reservoir Name. Chinese Water Conservancy and Hydroelectric Press. Shankman, D., Heim, B.D., Song, J., 2006. Flood frequency in China's Poyang Lake region: trends and teleconnections. International Journal of Climatology 26, 1255e1266. Shi, Z.H., Ai, L., Fang, N.F., Zhu, H.D., 2012. Modeling the impacts of integrated small watershed management on soil erosion and sediment delivery: a case study in the Three Gorges Area, China. Journal of Hydrology 438e439, 156e167. Shi, C.X., 2008. Scaling effects on sediment yield in the upper Yangtze River. Geographical Research 27 (4), 800e811 (in Chinese with English abstract).

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015

J.H. Gao et al. / Quaternary International xxx (2015) 1e11 Shi, C.X., Du, J., 2009. Causes for stepped changes of sediment load in the upper Yangtze River. Journal of Sediment Research 4, 17e24 (in Chinese with English abstract). Syvitski, J.P.M., 2003. Supply and flux of sediment along hydrological pathways: research for the 21st century. Global and Planetary Change 39, 1e11. € ro €marty, C., Kettner, A.J., Green, P., 2005. Impact of humans on Syvitski, J.P.M., Vo the flux of terrestrial sediment to the global coastal ocean. Science 308, 376e380. Syvitski, J.P.M., 2011. Global sediment fluxes to the Earth's coastal ocean. Applied Geochemistry 26, S373eS374. € ro €smarty, C.J., Meybeck, M., Fekete, B., Sharma, K., Green, P., Syvitski, J.P.M., 2003. Vo Anthropogenic sediment retention: major global impact from registered river impoundments. Global and Planetary Change 39, 169e190. Walling, D.E., 2006. Human impact on land-ocean sediment transfer by the world' rivers. Geomorphology 79, 192e216. Wang, Y., Dong, P., Oguchi, T., Chen, S., Shen, H., 2013. Long-term (1842e2006) morphological change and equilibrium state of the Changjiang (Yangtze) Estuary, China. Continental Shelf Research 56, 71e81. Wang, H.J., Yang, Z.S., Wang, Y., 2008a. Reconstruction of sediment flux from the Changjiang (Yangtze River) to the sea since the 1860s. Journal of Hydrology 349, 318e332. Wang, H.J., Saito, Y., Zhang, Y., Bi, N., Sun, X., Yang, Z., 2011. Recent changes of sediment flux to the western Pacific Ocean from major rivers in East and Southeast Asia. Earth Science Reviews 108, 80e100. Wang, Z.Y., Huang, W.D., He, Y.P., 2008b. Study on the sediment demand of the Yangtze River. Journal of Sediment Research 1, 26e34 (in Chinese with English abstract).

11

World Commission of Dams, 2005. Dams and Development: a New Framework for Decision-making World Commission of Dams (WDC). Earthscan Publications Ltd, London. Xiong, M., Xu, Q.X., Yuan, J., 2009. Analysis of multi-factors affecting sediment load in the Three Gorges Reservoir. Quaternary International 208, 76e84. Xu, J.X., 2000. Runoff and sediment variations in the upper reaches of Changjiang River and its tributaries due to deforestation. Journal of Hydraulic Engineering 1, 72e80 (in Chinese with English abstract). Xu, J.X., 2009. Plausible causes of temporal variation in suspended sediment concentration in the upper Changjiang River and major tributaries during the second half of the 20th century. Quaternary International 208, 85e92. Yan, Y., Wang, S., Chen, J., 2011. Spatial patterns of scale effect on specific sediment yield in the Yangtze River basin. Geomorphology 130, 29e39. Yang, S.L., Milliman, J.D., Li, P., Xu, K., 2011. 50,000 dams later: erosion of the Yangtze River and its delta. Global and Planetary Change 75, 14e20. Yang, S.L., Zhang, J., Dai, S.B., Li, M., Xu, X.J., 2007. Effect of deposition and erosion within the main river channel and large lakes on sediment delivery to the estuary of the Yangtze River. Journal of Geophysical Research 112, F02005. http:// dx.doi.org/10.1029/2006JF000484. Yang, S.L., Zhao, Q.Y., Belkin, I.M., 2002. Temporal variation in the sediment load of the Yangtze River and the influences of human activities. Journal of Hydrology 263, 56e71. Yin, H.F., Liu, G.R., Pi, J.G., Chen, G.J., Li, C.G., 2007. On the river-lake relationship of the middle Yangtze reaches. Geomorphology 85, 197e207. Zhu, J., 2000. Variation of sediment transportation in the Yangtze River and the way for its reduction. Journal of Hydroelectric Engineering 70, 38e48 (in Chinese with English abstract).

Please cite this article in press as: Gao, J.H., et al., Reservoir-induced changes to fluvial fluxes and their downstream impacts on sedimentary processes: The Changjiang (Yangtze) River, China, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.03.015