The Impact of Socio-Economic Factors on Sediment Load: A Case Study of the Yanhe River Watershed

Abstract

Under the influence of climate change and human activities, sediment load in rivers has changed significantly, which has a profound impact on the stability of ecosystems and the sustainable development of human beings. Taking the Yanhe River watershed as a case, this paper expounds the dynamic relationship among the Grain for Green Project, social and economic development, population migration, and sediment transport. The variability of sediment load was detected by Pettitt test, the double cumulative curve method, and the regression analysis method, and the effects of climate and human activities on sediment load were quantitatively analyzed. The results showed that 1) from 1956 to 2016, the precipitation of Yanhe River watershed rose slightly in the past 10 years, but the sediment load decreased significantly; 1996 was identified as the catastrophic year of the study period, when the contribution of climate change and human activity to reduced sediment load was 14.1% and 85.9%, respectively. 2) The Grain for Green Project increased the vegetation coverage of the study area from 40.6% to 78.5%. 3) The proportion of agricultural GDP in total GDP decreased from 52.26% to 7.3%, and the proportion of agricultural GDP was positively correlated with sediment transport and cultivated land area (p < 0.01). 4) Population migration resulted in the urbanization rate reaching 40.23%, and the urbanization rate is negatively correlated with sediment load and cultivated land area (p < 0.01), while the cultivated land area is positively correlated with sediment load (p < 0.01). The decrease of cultivated land area makes the sediment load gradually decrease. Therefore, socio-economic factors promote the sustainable development of the river basin.

1. Introduction

A watershed is a relatively independent and naturally formed ecosystem and also a significant place for human habitation and socio-economic development [1]. In recent decades, the impact of climate change and human activities on the global ecological environment has become more and more obvious. In particular, it has affected the spatial and temporal distribution of surface water resources and various factors in the water cycle (runoff and sediment load), which has led to the worldwide concern. About 50% of the world’s rivers show a significant decline in river sediment load [2,3,4,5,6], affecting the structure, processes, and functions of societies and ecosystems [7,8,9]. The decrease of sediment load is due to the change of underlying surface caused by human activities, which leads to significant changes in sediment production mechanisms and sediment transport processes [10]. Human activities are restricted by policies and socio-economic factors. Therefore, understanding and balancing the relationship between policies, socio-economic factors, and sediment load is critical to the sustainable management of the basin.

River sediment transport is one of the major problems facing many countries [11,12,13]. Sediment deposition in excess supply will lead to bed aggradation, so as to raise the flood water level [14,15]. In rivers, reduced sediment load can improve water quality, aquatic habitats and reduce flooding issues [16,17]. However, reduced sediment loads may also cause problems, such as channel erosion and erosion of the delta coastline when the critical shear stress threshold is exceeded and reduced nutrient inputs to aquatic and riparian ecosystems [18,19]. Therefore, the decrease of sediment discharge and its influence on river and ocean system has become a global issue [20,21,22,23].

The Yellow River is commonly held to be the birthplace of Chinese civilization, and the Yellow River is one of the rivers with the highest sand content in the world. In recent decades, in order to effectively control soil erosion and restore ecosystems, the government has adopted various water-saving measures and policies—such as building silt dams [24,25], reservoirs [26], and terraces [21]—and other ecological measures and the policy of returning farmland to forests [27,28]. These measures have altered the process of sediment supply in natural watersheds and the geomorphology of rivers, thereby affecting the connectivity of runoff, soil erosion, water, and sediment transport and drastically reducing sediment load in the Yellow River watershed [29]. The amount of sediment transported by the Yellow River has been restored to almost 600 Anno Domini before human activity [30,31]. Clarifying the impact of various measures on the amount of sediment load is of great significance for future soil and water conservation work [32]. Hu et al. [33] found that human activities accounted for an average of 47% and 81% of the changes in runoff and sediment load, and the rest were due to climate change. Li et al. [34] found that silt dams have an important role in controlling sediment transport during heavy rains by using a multivariate mixed model and watershed comparative analysis. Yang et al. [35] selected eight basins on the Loess Plateau and found that the increase in vegetation coverage since 1996 has been the main factor for sediment reduction, accounting for 47.7% of total sediment reduction. Climate change and terrace construction account for 9.1% and 18.6% of sediment emissions, respectively. However, current studies mainly focus on water and sediment changes, water and soil conservation, and rarely take socio-economic factors as a starting point to explore their impact on sediment transport.

Different from natural factors (temperature and precipitation), socio-economic factors mainly change natural conditions by controlling human activities and have an impact on sediment load. Increased rural population may exacerbate soil erosion, which may affect sediment load [36]. The policy of returning farmland to forest, rapid economic development and industrial structure adjustment have profoundly changed the land use types in the basin [37,38], effectively controlled soil erosion, reduced soil erosion and improved ecological environment [39]. In this paper, Yanhe River watershed is selected as an example. Based on the meteorological data, runoff and sediment data and socio-economic data from the Yanhe River watershed are used to explore the relationship between the policy of returning farmland to forest, social and economic factors and sediment load, so as to clarify the relationship between social development and sediment load and promote the sustainable development of the watershed.

2. Data and Methods

2.1. Study Area

This paper presents a socio-hydrologic analysis centered on the Yanhe River watershed, which is located in the middle of the Loess Plateau (Figure 1). The Yanhe River watershed is one of the main sources of sediment in the Yellow River, with an area of 7725 km², most of which is covered by loess hilly-gully areas with serious soil erosion. It experiences a semi-arid continental climate with the catchment-wide average annual precipitation is 520 mm, of which the majority occurs in the summer months of June–September. Since the implementation of the policy of the Grain for Green in 1997, the ecological environment of the Yanhe River watershed has been greatly improved. The average annual runoff is 1.82 × 10⁸ m³, and the average annual sediment transport is 26.2 million tons. By 2016, the population in Yanhe River watershed reached 992,678, among whom 593,337 were agricultural people, accounting for 59.8% of the total population.

2.2. Data

The Yellow River Conservancy Commission (YRCC) provided annually observed runoff discharge, and sediment load data at the Ganguyi station (1956–2016). Meteorological data from six weather stations in the Yanhe River watershed from 1956 to 2016 were obtained from the China National Meteorological Information Center (http://cdc.cma.gov.cn/home.do). The inverse distance weight interpolation (IDW) method for the weather station was applied to obtain the watershed climate information. Land use data was provided by the Chinese Academy of Sciences Resource and Environment Science Data Center. Vegetation coverage and DEM were provided by Geospatial Data Cloud site, Computer Network Information Center, Chinese Academy of Sciences. (http://www.gscloud.cn). Vegetation coverage is extracted from theGIMMS (glaobal inventory modelling and mapping studies) NDVI (normalized difference vegetation index) and MODIS (moderate-resolution imaging spectroradiometer) NDVI data sets (http://globalchange.nsdc.cn). The vegetation coverage in 1980 was linearly calculated based on the vegetation coverage in 1981–2016. Data on cultivated area, human population size and economic data, which are supported by the Yanhe River watershed, have come from several statistics yearbooks, including the Shannxi Statistical Yearbook (http://cnki.netdata.cn) and the Yanan Statistical Yearbook (http://dfz.shaanxi.gov.cn).

2.3. Method

2.3.1. Change-Point Analysis

Identifying change points is one of the important analysis methods to study the impact of climate change and human activities on runoff and sediment. This study uses the non-parametric method proposed by Pettitt (1979) [40], which is widely used in the detection of mutation points in hydrological and climate records. This method detects a significant change in the mean of a time series when the exact time of the change is unknown [41].The test uses a version of the Mann–Whitney statistic Ut,N, which verifies whether two samples x1,…, xt and xt + 1,…, xN are from the same population. The test statistic Ut,N is expressed by

$$U_{t,N}=U_{t-1,N}+{\displaystyle {\sum }_{j=1}^N\mathrm{sgn}(x_t-x_j)(\mathrm{t}=2,\dots ,N)}$$

(1)

where

$$\left\{\begin{array}{l}\mathrm{s}\mathrm{g}\mathrm{n}({\mathrm{x}}_{\mathrm{i}}-{\mathrm{x}}_{\mathrm{j}})=1\mathrm{i}\mathrm{f}({\mathrm{x}}_{\mathrm{i}}-{\mathrm{x}}_{\mathrm{j}})>0\\ \mathrm{s}\mathrm{g}\mathrm{n}({\mathrm{x}}_{\mathrm{i}}-{\mathrm{x}}_{\mathrm{j}})=0\mathrm{i}\mathrm{f}({\mathrm{x}}_{\mathrm{i}}-{\mathrm{x}}_{\mathrm{j}})=0\\ \mathrm{s}\mathrm{g}\mathrm{n}({\mathrm{x}}_{\mathrm{i}}-{\mathrm{x}}_{\mathrm{j}})=-1\mathrm{i}\mathrm{f}({\mathrm{x}}_{\mathrm{i}}-{\mathrm{x}}_{\mathrm{j}})<0\end{array}\right.$$

(2)

The test statistic counts the number of times a member of the first sample exceeds that of the second. The null hypothesis of Pettitt’s test is the absence of a changing point. Its statistic k(t) and the associated probabilities used in significance testing are expressed as (Pettitt, 1979) [40]

k(t) = Max1 ≤ t ≤ N|Ut,N|,

(3)

and

$$\mathrm{P}\cong 2\mathrm{e}\mathrm{x}\mathrm{p}\left\{-6\left[\mathrm{k}{(\mathrm{t})}^2\right]/\left({\mathrm{N}}^3+{\mathrm{N}}^2\right)\right\}$$

(4)

If P < 0.05, a significant change point exists. Therefore, the time series was divided into two parts at the location of the change point.

2.3.2. Climate Change and Human Impact Identification Methods

The double cumulative curve is a method widely used to study the consistency and long-term trends of hydrometeorological time series and is used to quantitatively evaluate the impact of climate change or human activities on runoff and sediment [42]. If the proportional relationship between the two cumulants remains the same during the same period, the double accumulation curve is represented by a straight line [43]. The abrupt change in the slope of the curve may be caused by soil and water conservation measures, vegetation restoration or destruction, and climate change [44]. In this paper, the mutation point is determined by the Pettitt test [40], the period before the mutation point is determined as the reference period, and the period after the mutation point is the research period. The relationship curve between rainfall and sediment in two periods was established, and a linear regression equation was established according to the base period, thereby reconstructing the annual sediment load not affected by human activities during the study period [45]. The following equations can be used to calculate the contribution of climate change and human activities to sediment load:

$$\Delta {\overline{\mathrm{S}}}_{total}={\overline{\mathrm{S}}}_{ao}-{\overline{\mathrm{S}}}_{po}$$

(5)

$$\Delta {\overline{\mathrm{S}}}_{hum}={\overline{\mathrm{S}}}_{ao}-{\overline{\mathrm{S}}}_{ac}$$

(6)

$$\Delta {\overline{\mathrm{S}}}_{\mathrm{cli}}=\Delta {\overline{\mathrm{S}}}_{total}-\Delta {\overline{\mathrm{S}}}_{hum}$$

(7)

$${\mu }_{\mathrm{cli}}=\frac{\Delta {\overline{\mathrm{S}}}_{\mathrm{cli}}}{\Delta {\overline{\mathrm{S}}}_{total}}\times 100\%$$

(8)

$${\mu }_{\mathrm{hum}}=\frac{\Delta {\overline{\mathrm{S}}}_{\mathrm{hum}}}{\Delta {\overline{\mathrm{S}}}_{total}}\times 100\%$$

(9)

where $\Delta {\overline{\mathrm{S}}}_{total}$ is the total change of annual average sediment transport volume (tons), ${\overline{\mathrm{S}}}_{po}$ and ${\overline{\mathrm{S}}}_{ao}$ are the average annual sediment transport volume during the reference period and the research period (tons), respectively, and ${\overline{\mathrm{S}}}_{ac}$ is the average annual sediment transport volume calculated during the research period (tons). $\Delta {\overline{\mathrm{S}}}_{\mathrm{cli}}$ and $\Delta {\overline{\mathrm{S}}}_{hum}$ are changes in annual average sediment volume caused by climate change and human activities(tons). ${\mu }_{\mathrm{cli}}$ and ${\mu }_{\mathrm{hum}}$ are the contribution rate of climate change and human activities to sediment change [46].

3. Results

3.1. Changes in Runoff, Sediment Load, and Precipitation

Figure 2a shows the changes in annual precipitation in Yanhe River watershed from 1956 to 2016. It can be seen from the figure that the overall annual change in Yanhe River watershed precipitation is not large. The maximum annual precipitation is 832 mm (1996), and the minimum annual precipitation is 315 mm (1997). As can be seen from

Table 1. The mean of precipitation, runoff, and sediment in the Yanhe River watershed.

Years	Precipitation (mm)			Runoff (108m3)			Sediment (108t)
	Mean	Cv	K	Mean	Cv	K	Mean	Cv	K
1956–1965	561	0.24	2.53	2.39	0.44	3.72	0.6	0.82	36.4
1966–1975	524	0.17	1.54	2.19	0.28	2.65	0.55	0.48	5.3
1976–1985	544	0.31	1.5	2.18	0.29	2.63	0.37	0.99	10
1986–1995	494	0.37	1.65	2.21	0.22	2.04	0.47	0.42	5.8
1996–2005	465	0.42	2.08	1.72	0.32	2.43	0.31	0.85	15.3
2006–2016	535	0.18	1.54	1.43	0.44	3.52	0.05	1.26	24
1956–2016	520	0.2	2.64	2.05	0.38	5.28	0.384	0.93	363.6

, the average annual precipitation of Yanhe River watershed is 520 mm, the extreme value ratio (maximum value/minimum value) is 2.64, and the coefficient of variation is 0.2. The average annual precipitation in each period is 561 mm, 524 mm, 544 mm, 494 mm, 465 mm, and 535 mm. Compared with other periods, the average annual precipitation in 1996–2005 is significantly reduced. In the past decade, precipitation in the Yanhe River watershed has been gradually increasing.

According to the annual runoff data observed at the Ganguyi Hydrological Station (Figure 2b), the annual runoff of the entire basin showed a significant downward trend during 1956–2016. The annual average runoff in Yanhe River watershed is 2.05 × 10⁸ m³, the maximum runoff is 5.02 × 10⁸ m³ (1964), the minimum is 0.94 × 10⁸ m³ (2015), the annual coefficient of variation is 0.38, and the extreme ratio is 5.28. The annual average runoff decreased from 2.39 × 10⁸ m³ (1956–1966) to 1.43 × 10⁸ m³ (2006–2016). The average annual runoff from 2006 to 2016 was lower than the average annual runoff from other periods.

Figure 2c shows the change of annual sediment load in Yanhe River watershed. As can be seen from the figure, the annual sediment load in the river basin decreased significantly from 1956 to 2016. The average annual sediment load was 0.384 × 10⁸ t, the maximum annual sediment load was 1.818 × 10⁸ t (1964), and the minimum value was 0.005 × 10⁸ t (2015), the overall change was basically consistent with the runoff trend. However, the average annual sediment load from 1976 to 1985 decreased significantly to 0.37 × 10⁸ t, and the average annual sediment load from 2006 to 2016 decreased to 0.05 × 10⁸ t. The coefficient of variation of annual sediment load is 0.93, and the extreme ratio is as high as 363.6. Comparing precipitation and runoff, the trend of sediment load is more obvious.

3.2. Change-Point Analysis for Sediment Loads

In order to quantify the change in sediment loads before and after the sudden change, a double cumulative curve was drawn to represent the correlation between cumulative annual sediment loads and precipitation. According to the Pettitt test [40], 1996 was the abrupt change point of annual sediment load. As can be seen from Figure 3, the annual sediment load suddenly decreased around 1996. The annual sediment loads decreased from 0.52 × 10⁸ t∙yr⁻¹ in the reference period to 0.133 × 10⁸ t∙yr⁻¹ in the study period. A regression equation was established based on the cumulative precipitation and sediment loads before the change, and the contribution rate of climate change to sediment loads reduction during the study period was calculated to be only 14.11%, while the contribution of human activities to sediment loads reduction was 85.89%.

3.3. Change in Maximum Sediment Concentration

Runoff sediment concentration is an important parameter that must be measured in soil and water conservation monitoring. Figure 4 shows the change of maximum sediment concentration in Yanhe River watershed from 1963 to 2016, and the maximum sediment concentration generally occurs in flood season. It can be seen that before the change, the fluctuation of the maximum sediment concentration was stable, while after the change, the maximum sediment concentration decreased significantly, and the fluctuation was large. The maximum value was 1200 kg/m³ in 1963, and the minimum value was 75.9 kg/m³ in 2014. The average annual maximum sediment concentration decreased from 959.5 kg/m³ to 595.1 kg/m³.

3.4. Changes in Socio-Economic Factors

3.4.1. The Grain for Green Project

In recent centuries, unsustainable cultivation and grazing in extremely hilly areas have transformed most of the arable land in the loess plateau (up to 70%) into degraded land [47]. In order to address the serious land degradation problem, the Chinese government (since 1999) initiated a program the Grain for Green on the loess plateau, giving priority to action to address large-scale soil erosion [48]. Figure 5 shows the area of the Yanhe River watershed that has been converted to forest since 1999. It can be seen that the implementation process of “Grain for Green” in the Yanhe River watershed presents three stages—first slowly increasing, then accelerating, and finally approaching equilibrium. From 2000 to 2016, a total of 1675 km² of farmland in the Yanhe River watershed was converted into woodland, with an annual average conversion rate of 98.5 km². The implementation of the policy of returning farmland to forests has changed the land use types in Yanhe River watershed.

Table 2. The land use types from 2000 to 2015 in the Yanhe River watershed (km²).

Landuse Types	2000	2005	2010	2015
Arableland	3217	3009.5	2375	2360.5
Forestland	808.5	1018	1074.5	1072
Grassland	3376	3373	3934.5	3915.5
Water	25	22	24	26
Building	27	31	47	81
Unutilized land	2.5	2.5	1	1

shows land use types from 2000 to 2015 obtained by remote sensing image classification of the Yanhe River watershed. As can be seen from the table, the cultivated land area of Yanhe River watershed continuously decreased from 3217 km² to 2360.5 km² from 2000 to 2015. The area of woodland, grassland and urban construction land increased continuously, with an increase of 263.5 km² of woodland, 539.5 km² of grassland and 54 km² of urban construction land.

3.4.2. Economic Development and Industrial Restructuring

The Chinese government introduced a reform and opening-up policy in 1978 to promote rapid economic development in China [49]. This paper chooses the Yanhe River watershed from 1980 to 2016, such as industrial production value, population, urbanization, and other social indicators to represent the economic development of the river basin. Figure 6 is the distribution of Gross Domestic Product (GDP) in various industries in the Yanhe River watershed. It can be seen that the GDP of the Yanhe River watershed has continued to rise, from 0.02 × 10⁹ USD in 1980 to 6.89 × 10⁹ USD, of which industrial GDP has increased from 0.56 × 10⁷ USD rose to 3.17 × 10⁹ USD, the service industry increased from 0.38 × 10⁷ USD to 3.21 × 10⁹ USD, and the agriculture increased from 0.01×10⁹ USD to 0.51 × 10⁹ USD. Since entering the 21st century, with the rapid development of industry and service industry, the total GDP has increased from 0.17 × 10⁹ USD∙yr⁻¹ by 4.62 × 10⁹ USD∙yr⁻¹. The proportion of industrial GDP in the basin’s GDP increased from 27.9% to 46%, the proportion of the service industry increased from 19.7% to 46.6%, and the proportion of agricultural GDP in the basin’s GDP decreased from 52.26% to 7.4%.

3.4.3. Migration and Urbanization

The population of the Yanhe River watershed increased from 1980 to 2016 (Figure 7e). The total population increased from 5.4 × 10⁵ to 9.3 × 10⁵ in 1980, of which 9 × 10⁵ people lived in urban areas for a long time (about 91% of the total population), and 9 × 10⁴ people lived in rural areas. A large-scale population migration occurred in 1998, and a large number of rural people moved to cities. The increase in urban population led to the expansion of urban construction land and the acceleration of urbanization. Urbanization refers to the process by which rural populations converge to cities. The essence is the spatial transformation of the rural population, the integration of non-agricultural industries into cities and towns, and the transfer of agricultural labor to non-agricultural labor [50]. As of 2016, the non-agricultural population in the Yanhe River watershed totaled 3.9 × 10⁵ and the agricultural population was 5 × 10⁵. The urbanization rate increased from 17.2% to 40.23% (Figure 7f).

3.5. Relationship Between Socio-Economic Factors and Sediment Loads

According to the above analysis, the ratio of agricultural GDP to total GDP, urbanization rate, and cultivated land area and sediment loads were regressed, as shown in the Figure 8, Figure 9 and Figure 10. As can be seen from Figure 8, there is a significant positive correlation between sediment loads and the ratio of agricultural GDP to total GDP (p < 0.01, indicates a significant level of confidence at 0.01), and the correlation coefficient is 0.471. As the proportion of agricultural GDP decreases gradually, the sediment loads decrease gradually. Figure 9 shows the relationship between urbanization rate and sediment loads. It can be seen from the figure that the higher urbanization rate is, the lower sediment loads. There was a significant negative correlation between urbanization development and sediment loads (p < 0.01), and the correlation coefficient was 0.547. It can be seen from Figure 10 that the smaller the area of cultivated land, the less the amount of sand transported. Cultivated land area was positively correlated with sediment loads (p < 0.01), and the correlation coefficient was 0.43. The lower correlation coefficient was mainly due to the fact that the cultivated land area adopted in this paper was the common cultivated land area, not the main cultivated land area converted to forest or reduced.

4. Discussion

4.1. Effects of Precipitation and Runoff on Sediment Loads

The impact of climate change on soil erosion has drawn global attention, and precipitation is a major factor directly affecting runoff and further affecting soil erosion [51,52,53]. According to the above analysis, it can be seen that in the past ten years, the precipitation in the Yanhe River watershed showed an upward trend, but the runoff and sediment loads showed a slight decline and a significant decline, respectively. After 1996, the impact of climate change on sediment transport only accounted for 14.11%. The decrease of runoff plays a role in the decrease of sediment loads. Wang et al. [54] found that the sharp decrease in sediment loads in the loess plateau after 1979 (58%) was mainly due to the decrease in water yield (59%), and the engineering and vegetation measures led to the change of the surface, which was the reason for the 76% decrease in water volume. Among them, ecological restoration is considered to be the main reason for the decrease of runoff [55]. The increase of vegetation coverage allows more precipitation to be intercepted by vegetation canopy, and the intercepted water returns to the atmosphere quickly through evaporation [56]. Dense litter beneath the trees also favors precipitation infiltration. As the Yanhe River watershed is a typical hilly region, only a small part of rainfall can replenish groundwater, while the rest of water dissipates into the atmosphere through evapotranspiration during the dry season [57], resulting in continuous reduction of runoff. Therefore, the restoration of vegetation is the main cause of the decrease of sediment loads in Yanhe River watershed after 1996.

4.2. Effect of Socio-Economic Factors

4.2.1. Policy Effects

Environmental protection policy is an important tool to regulate the relationship between human beings and ecosystems, which can promote ecological restoration at different spatial and temporal scales. Policy and institutional factors have a mandatory impact on land use change. Since 1999, with the development of China’s economy, government departments have adopted administrative, legal, financial and various technical means to improve the ecological environment of the plateau and implemented such large-scale projects as the “Grain for Green Project” and “China’s ecological environment construction.” After the implementation of the policy of returning farmland to forests, the area of forests and grasslands in the eight tributaries of the loess plateau increased to 2208.8 km² (4.4%) and 2948.7 km² (4.4%), respectively [35]. Zuo et al. [58] found that the decrease of cultivated land (3.7%) and the expansion of forestland (14.7%) in the Huangfu Chuan watershed can reduce the sediment transport (40.6%). Since the implementation of the policy of Grain for Green, the vegetation coverage of Yanhe River watershed has improved on the whole. The area of cultivated land has decreased by 856.5 km², the forest land has increased by 263.5 km², and the grassland has increased by 539.5 km². As can be seen from Figure 7c, the vegetation coverage increased from 40.6% to 78.5%. The increase of vegetation coverage can enhance rainfall interception, reduce raindrop kinetic energy and diversion runoff, and reduce the incidence of runoff, peak discharge, and soil erosion in Yanhe River watershed by 50%, 64%, and 72%, respectively [59,60], reducing sediment loads from the root [61].

4.2.2. Effect of Economic Development and Industrial Structure Adjustment

The influence of socio-economic factors on the amount of sediment transport is mainly through human activities. In the primitive period or at a relatively low level of economic development, humans have less influence on the underlying surface and runoff, resulting in less soil erosion and less sediment loads; during the economic take-off and industrial expansion stages, large-scale infrastructure construction, and the exploitation of energy resources, the surface vegetation has been destroyed to a great extent, causing serious soil erosion and increasing sediment transport. In the development of economic development, the optimization of economic structure, and the regulation of systems, people have the conditions to use the fruits of economic development and technology to establish progressive methods to reduce soil erosion and artificially repair the environment, control erosion and sediment yield in the river basin, and reduce the amount of sediment transported [62,63]. Since the beginning of the 21st century, the government has issued a series of encouragement policies in urban construction, land management, population and labor mobility, and industrial layout that have promoted the development of the local economy. The Yanhe River watershed is rich in oil and natural gas, and the industry is relatively developed. At the same time, with the rise of the service industry, the demand for agriculture is getting lower. It can be seen from Figure 11 that, as the ratio of agricultural GDP to total GDP declines, the area of cultivated land is also decreasing (p < 0.01).

4.2.3. Effect of Population Migration and Urbanization

Economic development and industrial restructuring have provided employment opportunities for a large number of rural laborers, thereby changing the employment and land use patterns of rural residents. According to statistics from socio-economic data, the per capita income of cities in the Yanhe River watershed increased from $36.39 USD in 1980 to $4432.78 USD in 2016, while per capita income in rural areas increased from $9.68 USD to $1526.26 USD (Figure 7d). Although the per capita income is increasing simultaneously, it can be clearly seen that the growth rate of urban per capita income is significantly higher than that of rural per capita income. The gap in per capita income between urban and rural areas has attracted more and more rural laborers to move to cities. From Figure 12, we can see that there is a linear relationship between the per capita income gap and the number of immigrants. As the per capita income gap widens, the number of rural populations moving to cities also increases (p < 0.01). Population migration promotes the development of urbanization. Urbanization is a fundamental change in a country’s economic structure, social structure, production mode and lifestyle [64]. As can be seen from Figure 13, the higher the level of urbanization, the less cultivated land (p < 0.01). The reason is that in the transfer of rural labor force is mainly young adults, engaged in industry or service industry in the city to obtain greater benefits, the elderly generally continue to stay in the countryside for agricultural production. With convenient transportation and sound infrastructure, more and more young people choose to start their own businesses in cities rather than return to the countryside. As the elderly age, the declining labor force has been unable to carry out large-scale agricultural production, thus reducing the demand for arable land. With fewer and fewer people left behind in the countryside, arable land has been abandoned. Compared with cultivated land, grassland and shrubland have higher infiltration capacity compared to bare land, and the infiltration capacity of rainfall is increased to improve the water retention capacity of soil [65]. The restoration of natural vegetation in abandoned farmland will change the hydraulic characteristics of soil, thus changing the transport process of runoff and sediment [66,67]. The decrease of rural population leads to the decrease of unreasonable reclamation activities, which reduces the intensity of soil and water loss in the river basin and correspondingly reduces the amount of sediment transported.

5. Conclusions

This study analyzed the change of sediment loads in the Yanhe River watershed from 1956 to 2016 and analyzed the influence of socio-economic factors on sediment transport processes using regression analysis. The results showed that the amount of sediment transported in Yanhe River watershed decreased significantly, and 1996 was the abrupt change year of sediment loads. Sediment loads are positively correlated with the ratio of agricultural GDP to total GDP and cultivated land area and negatively correlated with urbanization rate. The sharp decline in sediment transport since 1996 is mainly due to the restoration of vegetation in the Yanhe River basin. The main measure to promote vegetation restoration was the conversion of Grain for Green Project implemented in 1999, which changed the land use type of the study area and increased the area of forest and grassland. Meanwhile, with the adjustment of economic development and industrial structure, the rapid development of industry and service industry, a large portion of the rural labor force migrated to cities, promoting the process of urbanization, gradually reducing agricultural demand, and thus reducing the destruction of arable land and promoting the restoration of vegetation. In short, the driving factors behind sediment reduction are the Grain for Green Project, economic development and industrial restructuring, population migration, and urbanization. Therefore, balancing the relationship between socio-economic development and ecological environment is conducive to sustainable development.