Research on Sediment Deposition Characteristics and the Vegetation Restoration of Ecological Riverbanks in the Deep Waterway Regulation Scheme of Yangtze River

Abstract

In order to explore the sediment deposition characteristic of ecological riverbanks associated with vegetation restoration in the deep waterway regulation scheme of Yangtze River, two kinds of typical ecological riverbanks and a traditional riprap riverbank (TRR) in engineering areas were investigated. The vegetation community characteristics, sediment aggregate compositions, nutrient contents, total organic carbon (TOC), sediment microbial biomass carbon (MBC), sediment microbial biomass nitrogen (MBN), and sediment microbial biomass phosphorus (MBP) were determined. The results indicated that the ecological restoration effect of the lattice gabion ballasted vegetation mat riverbank (LGBVR) was best, followed by the mesh grid riverbank (MGR), and that of the TRR was relatively poor. In different ecological riverbanks, the sediment aggregated compositions were not significantly varied. The sediment contents of NH₄⁺-N, available phosphorus (AP), and TOC in ecological riverbank areas were relatively higher than those of the TRR. In the LGBVR, the sediment contents of MBC were relatively higher than those of the others. The sediment deposition characteristics and ecological restoration effects in the study area should be monitored for a long time.

1. Introduction

As an important ecotone between terrestrial and aquatic ecosystems, a riparian zone can act as an indispensable corridor for substance, energy, and biological information flows [1,2,3]. At the same time, riparian zones have been strongly impacted and modified by artificial projects and natural flow regime alternations [4]. Conventionally, rock and concrete materials have been widely applicated in riverbank protection projects. This emphasizes the structural stability and safety of flood control. However, these hard riverbanks and stream canalizations have negative effects on riparian ecosystems, ecological integrity, habitat connectivity, and esthetic value [5,6,7,8]. As effective methods for riparian protection and restoration, the ecological riverbank concepts have been widely accepted and applied in developed countries, such as countries in the European Union, the USA, and Japan, for years [9,10,11]. Ecological riverbank development is a promising tool to maintain structure stability, safety, and riparian ecological restoration [12]. Recently, ecological riverbanks have played significant roles in improving surface water quality, aquatic ecosystem health, and biodiversity protection in the lower Yangtze River basin [13], and they have become the main development trend in waterway regulation projects on ecological construction [12,14,15]. In general, for different rivers and navigation conditions, ecological riverbank development must conform to engineering stability and facilitate the growth of plants, organisms’ movement, surface and groundwater flow, riparian ecosystem protection, and ecological restoration effects [16,17].

The Yangtze River waterway has unique advantages and great development potential in navigation development, and it is known as the “golden waterway”, which is the main framework for building a comprehensive three-dimensional transportation corridor in the Yangtze River Economic Belt. In recent years, new riverbank structures such as concrete grids, steel wire mesh gabions, and geogrids have been adopted to the waterway regulation scheme in Yangtze River [15,18]. During the construction of the 12.5 m deep waterway regulation scheme of Yangtze River below Nanjing, a variety of ecological riverbank forms were developed and applied in some river sections. In this study, two kinds of typical ecological riverbanks and a riprap riverbank in the engineering areas were selected and surveyed in 2018. The vegetation restoration and sediment deposition characteristics associated with ecological riverbanks were investigated in this study. The objectives of this study were as follows: (1) to assess vegetation restoration and sediment deposition characteristics in different ecological riverbanks; (2) to explore the ecological effects and influence factors associated with ecological riverbank restoration; (3) to provide practices for the structural design of ecological riverbanks in the waterway regulation scheme of Yangtze River.

2. Materials and Methods

2.1. Study Area

As the national key engineer of the 12th and 13th five-year plans in China, the 12.5 m deep waterway regulation scheme of Yangtze River below Nanjing started in June 2015 and was completed in June 2017. With the river reach length of 227 km, it includes the Fujiangsha waterway, Kouanzhi waterway, Hechangzhou waterway, and Yizheng waterway [19]. Affected by tidal flow at varying degrees, the Yangtze River is located at a fluvial plain in this section near to the estuary. In this study area of Yangtze River, the hydrodynamic and sediment movements are complex, which are affected by the current to varying degrees [20].

In this study, the traditional riprap riverbank (TRR) region (Figure 1a) and two typical ecological riverbanks, including a lattice gabion ballasted vegetation-mat riverbank (LGBVR) region (Figure 1b) and a mesh grid riverbank (MGR) region (Figure 1c), were selected as study areas to explore the vegetation restoration associated with the sediment deposition characteristic in the dry season in November of 2018.

2.2. Sampling Surveys

At the spatial scale for the MGR region, 3 survey transects were set perpendicular to the river flow direction (Figure 1c). On each transect, 3 sampling sites were set at equal distances along the transverse gradient of the riparian zone. There were 12 sampling sites (E1–E12) for the MGR region. As for the LGBVR region, 2 gabions with 2 sampling sites (L1 and L2) were randomly selected (Figure 1b). As for the TRR region, 3 survey transects were also set perpendicular to the river flow direction (Figure 1a). On each transect, 3 sampling sites were set at equal distances along the transverse gradient. There were 9 sampling sites (S1–S9) for the TRR region. In addition, two control sampling sites (CSs) (D1 and D2) were randomly selected and set outside the influence range of the riverbank engineering areas (Figure 1a,c).

According to the terrestrial vegetation community survey method [21], in each sampling site, the quadrat size was 1 m × 1 m and the plants’ Latin names and numbers were recorded. At the same time, the aboveground parts of plant biomasses were obtained by a collection method, and the biomasses (g·m⁻²) were determined in the laboratory (by drying in a 75 °C oven for 48 h). At each sampling site, surface sediment samples (0–5 cm) were collected with an inner diameter of 5 cm of soil drilling, and 3 samples were taken from each sampling site and mixed into 1 sediment sample.

2.3. Analysis

In this study, the pH values of sediment were measured using an acidity meter (the water to sediment ratio was 2.5:1) (PHS-3C, Sanxin, China). The values of sediment electrical conductivity (EC) were measured using a conductivity meter (the water to sediment ratio was 5:1) (DDS-11A, Leici, China). The contents of ammonia nitrogen (NH₄⁺-N), nitrate nitrogen (NO₃⁻-N), available phosphorus (AP), and available potassium (AK) in sediment were measured using flow analyzers (AA3, Bran+Luebbe, German). The contents of total nitrogen (TN) were determined via the addition of the ammonia nitrogen (NH₄⁺-N) and nitrate nitrogen (NO₃⁻-N) contents. The contents of sediment microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), and microbial biomass phosphorus (MBP) were measured by the chloroform fumigation–K₂SO₄ extraction method [22]. By using the wet sieving method [23], the sediment aggregate could be classified and calculated as large macroaggregates (>2.00 mm), small macroaggregates (2.00–0.25 mm), microaggregates (0.25–0.053 mm), and silt + clay fractions (<0.053 mm). The separated sediment aggregates were air-dried and the sediment total organic carbon (TOC) contents were measured. The sediment TOC contents were the mean values of the sediment aggregate TOC contents. The sediment organic carbon (TOC) contents were measured by the K₂Cr₂O₇-FeSO₄ oxidation method [24].

2.4. Biodiversity Index

The biodiversity index can reflect the species composition, structural stability, and complexity of a vegetation community [25]. These were calculated as follows:

Margalef index (d) [26]:

$$d=\left(S-1\right)/lnN$$

(1)

Shannon–Weaver index (H′) [27]:

$$H^{\prime }=-{\displaystyle \sum }_{i=1}^n{P}_{i}ln{P}_i$$

(2)

Pielou index (J) [28]:

$$J=H^{\prime }/ln\left(S\right)$$

(3)

where P_i is the proportion of the number of plant species (i) individuals to the total number of individuals in each sampling site, n is the total number of plant species in each sampling site, and N is the total number of all plant species individuals in each sampling site. S is the classification taxa number of plant species in each sampling site.

2.5. Data Analysis

SPSS Statistical 22.0 was performed for one-way ANOVA test and Pearson correlation analysis. To analyze the sediment characteristic differences on the ecological riverbanks, a one-way ANOVA test was employed, and the significance was tested using the LSD method (α = 0.05). Pearson correlation analysis was employed to analyze the vegetation community and sediment characteristics of the ecological riverbanks.

3. Results

3.1. Vegetation Community Characteristics

Regarding different ecological riverbanks,

Table 1. Vegetation community characteristics in different ecological riverbanks. Mesh grid riverbank (MGR); lattice gabion ballasted vegetation-mat riverbank (LGBVR); traditional riprap riverbank (TRR); control sampling sites (CSs). Lowercase letters in same rows indicate a significant difference at the 0.05 level by one-way ANOVA test; the same applies below.

Types of Riverbank	MGR	LGBVR	TRR	CSs
	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)
Numbers of sampling sites	12	2	9	2
Numbers of species	8	6	6	12
Margalef index (d)	0.37 ± 0.14 a	0.65 ± 0.00 a	0.47 ± 0.04 a	0.45 ± 0.01 a
Shannon–Weaver index (H’)	0.66 ± 0.24 a	1.08 ± 0.17 a	0.92 ± 0.07 a	0.48 ± 0.01 a
Pielou index (J)	0.82 ± 0.08 a	0.78 ± 0.12 a	0.80 ± 0.07 a	0.44 ± 0.01 b
Biomass (g·m−2)	74.53 ± 27.08 b	1225.01 ± 200.33 a	11.47 ± 4.79 b	368.34 ± 25.48 ab
Latin names of dominant species	Kalimeris indica, Cardamine hirsute	Phragmites australis, Cardamine hirsute	Veronica anagallis-aquatica	Phragmites australis, Achyranthes aspera, Polygonum hydropiper, etc.

shows that the MGR region had eight plant species, the Pielou index (J) was at its highest, and the dominant species were Kalimeris indica and Cardamine hirsute. The LGBVR region had six plant species, it had the highest Margalef index (d) and Shannon-Weaver index (H′), and the dominant species were Phragmites australis and Cardamine hirsute. The TRR region had six plant species and the dominant species was Veronica anagallis-aquatica. The LGBVR region had the highest biomass due to large and tall phragmite communities, followed by the MGR region, and the TRR region had lowest biomass. In addition, the CSs region of the natural riparian riverbank had 12 plant species and a relatively higher biomass. The one-way ANOVA test also indicated that the biomass of the LGBVR region was significantly different compared to the MGR and TRR regions. The biomass was not significantly different between the LGBVR and CSs regions. For biodiversity indices, the Pielou index (J) of the CSs region was significantly different compared to that of the MGR, LGBVR, and TRR region.

3.2. Sediment Aggregate Compositions

Table 2. Sediment aggregate compositions in different ecological riverbanks. Lower case letters in same rows are same to Table 1.

Types of Riverbank	MGR	LGBVR	TRR	CSs
	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)
Numbers of sampling sites	12	2	9	2
pH	7.52 ± 0.08 a	7.64 ± 0.11 a	7.53 ± 0.15 a	7.29 ± 0.25 a
EC(μs·cm−1)	109.1 ± 8.6 ab	151.7 ± 19.9 a	111.4 ± 8.2 ab	91.4 ± 1.0 b
>2.00 mm (%)	5.4 ± 0.3 a	4.8 ± 0.1 a	4.9 ± 0.3 a	5.5 ± 1.4 a
2.00–0.25 mm (%)	26.1 ± 0.5 a	27.4 ± 0.2 a	25.7 ± 0.4 a	27.8 ± 2.8 a
0.25–0.053 mm (%)	45.9 ± 0.9 a	45.7 ± 0.5 a	46.7 ± 0.8 a	45.7 ± 2.6 a
<0.053 mm (%)	22.7 ± 0.4 a	22.1 ± 0.8 a	22.7 ± 0.7 a	21.1 ± 1.2 a

shows that the LGBVR region had relatively higher values of pH and EC in sediments. The sediment aggregate compositions of large macroaggregates (>2.00 mm) and small macroaggregates (2.00–0.25 mm) had relatively higher values in the CSs region (33.22%), followed by the LGBVR region (32.23%). Otherwise, they had relatively lower values in the MGR region (31.41%) and TRR region (30.62%). The one-way ANOVA test also indicated that the EC values of sediment were significantly different between the LGBVR and CSs regions.

3.3. Sediment Nutrient Contents

Table 3. Sediment nutrient contents in different ecological riverbanks. Lower case letters in same rows are same to Table 1.

Types of Riverbank	MGR	LGBVR	TRR	CSs
	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)
Numbers of sampling sites	12	2	9	2
NH4+-N (mg·kg−1)	1.97 ± 0.12 b	2.71 ± 0.52 a	1.95 ± 0.15 b	2.04 ± 0.06 ab
NO3-N(mg·kg−1)	2.35 ± 0.18 a	3.04 ± 0.08 a	2.42 ± 0.30 a	3.29 ± 1.06 a
TN (mg·kg−1)	4.32 ± 0.19 b	5.75 ± 0.43 a	4.38 ± 0.30 b	5.33 ± 1.12 ab
AP (mg·kg−1)	18.13 ± 1.65 a	21.97 ± 3.41 a	18.93 ± 1.54 a	21.68 ± 2.39 a
AK (mg·kg−1)	315.37 ± 15.15 a	339.55 ± 41.33 a	341.74 ± 21.25 a	315.88 ± 40.68 a

shows that the LGBVR region had relatively higher values of NH₄⁺-N, TN, and AP content in sediments. The TRR region had relatively higher values of AK content. As for the LGBVR region, the one-way ANOVA test indicated that sediment TN contents were significantly different compared to those of the MGR (p = 0.032) and TRR (p = 0.042) regions. Its sediment NH₄⁺-N contents were significantly different compared to those of the MGR (p = 0.042) and TRR (p = 0.043) regions. Furthermore, the sediment NO₃⁻-N, AP, and AK contents displayed no difference among the riverbanks and the CSs region.

3.4. Sediment TOC Contents

Figure 2 shows that the MGR region had relatively higher mean values of TOC content in sediments. The LGBVR region had relatively lower mean values of TOC content. The one-way ANOVA test indicated that the sediment mean TOC content in the MGR region was significantly different compared to that in the TRR region (p = 0.026). Moreover, the sediment aggregate TOC contents were not significantly different among ecological riverbanks and the CSs region.

3.5. Sediment MBC, MBN, and MBP Contents

Table 4. Sediment MBC, MBN, and MBP contents in different ecological riverbanks. Lower case letters in same rows are same to Table 1.

Types of Riverbank	MGR	LGBVR	TRR	CSs
	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)	(Mean ± SE)
Numbers of sampling sites	12	2	9	2
MBC (mg·kg−1)	596.23 ± 27.56 b	803.67 ± 47.67 ac	671.16 ± 28.56 ab	704.90 ± 107.40 ab
MBN (mg·kg−1)	33.12 ± 1.53 a	32.88 ± 3.90 a	37.29 ± 1.59 a	39.16 ± 5.97 a
MBP (mg·kg−1)	6.58 ± 0.07 a	6.60 ± 0.20 a	6.52 ± 0.07 a	6.65 ± 0.15 a
MBC/TOC	0.010 ± 0.000 c	0.014 ± 0.001 a	0.012 ± 0.001 b	0.012 ± 0.001 abc
MBC/MBN	18.000 ± 0.000 b	24.970 ± 4.415 a	18.000 ± 0.000 b	18.000 ± 0.000 b

shows that the values of MBC content, MBC/TOC, and MBC/MBN were highest in the LGBVR-region sediment. The sediment in the CSs region had highest MBN and MBP contents. The sediments in the TRR region had relatively higher MBN contents. The sediments in the LGBVR region had relatively higher MBP contents. As for the LGBVR region, one-way ANOVA demonstrated that the sediment MBC content was significantly different compared to that in the MGR region (p = 0.009). The sediment value of MBC/TOC in the LGBVR region was significantly different compared to that in the MGR region (p = 0.001) and TRR region (p = 0.001). The sediment value of MBC/MBN in the LGBVR region was significantly different compared to that in the MGR region (p = 0.000) and TRR region (p = 0.000).

3.6. Correlation Analysis

Table 5. Pearson correlation coefficients between vegetation and sediment characteristics in ecological riverbanks. ** means significant correlation at 0.01 level (bilateral); * means significant correlation at 0.05 level (bilateral).

	N	Biomass	pH	EC	NH4+-N	NO3−-N	TN	AP	AK	TOC	MBC	MBN	MBP	MBC/TOC	MBC/MBN
N	1.000
Biomass	0.251	1.000
pH	0.406	−0.291	1.000
EC	−0.481	−0.308	−0.225	1.000
NH4+-N	0.067	−0.124	−0.064	−0.237	1.000
NO3--N	0.219	−0.145	0.231	−0.028	−0.038	1.000
TN	0.225	−0.192	0.165	−0.152	0.509	0.841 **	1.000
AP	−0.143	−0.034	−0.022	−0.078	−0.075	−0.020	−0.058	1.000
AK	−0.491	−0.383	−0.187	0.643 *	−0.273	0.047	−0.108	−0.185	1.000
TOC	0.001	0.340	−0.145	−0.310	0.054	0.007	0.035	−0.276	−0.487	1.000
MBC	0.401	0.480	−0.180	0.061	0.138	0.016	0.088	0.112	−0.356	−0.159	1.000
MBN	0.360	0.536 *	−0.245	0.027	−0.039	−0.017	−0.036	0.142	−0.323	−0.100	0.966 **	1.000
MBP	0.112	−0.203	0.372	−0.185	0.267	0.065	0.201	0.105	0.029	−0.253	−0.403	−0.509	1.000
MBC/TOC	0.367	0.289	−0.113	0.161	0.125	0.010	0.077	0.160	−0.143	−0.487	0.938 **	0.882 **	−0.270	1.000
MBC/MBN	0.234	−0.104	0.201	0.137	0.674 **	0.126	0.474	−0.089	−0.194	−0.251	0.333	0.079	0.301	0.402	1.000

shows the Pearson correlation coefficients among the vegetation and sediment characteristics in different ecological riverbanks. The sediment contents of NO₃⁻-N were significantly correlated with TN (r = 0.841, p < 0.01). The sediment contents of NH₄⁺-N were significantly correlated with MBC/MBN (r = 0.674, p < 0.01). The sediment contents of MBC were significantly correlated with MBN (r = 0.966, p < 0.01) and MBC/TOC (r = 0.938, p < 0.01). The sediment contents of MBN were significantly correlated with MBC/TOC (r = 0.882, p < 0.01). The biomasses were significantly correlated with MBN (r = 0.536, p < 0.05). The sediment EC values were significantly correlated with AK (r = 0.643, p < 0.05).

4. Discussion

4.1. Sediment Deposition Characteristics in Ecological Riverbanks

Worldwide, rivers have been tremendously affected and modified by artificial riverbanks for navigation channels [29,30]. The accumulation of sediment deposition and organic matter in riparian zones has improved the soil conditions that are the basis for vegetation restoration and biodiversity maintenance [31]. Traditionally, in Yangtze River waterway regulation schemes, riverbank constructions were mostly made of concrete, cemented rock, riprap, and wire gabions, which are not suitable for sedimental deposition and vegetation restoration [16,32,33]. Ecological riverbanks can prevent and remove large amounts of nutrients from upland areas [30]. In recent years, ecological riverbanks have been widely used to restore the riparian ecosystem, and various forms have been developed in China [18,33]. This study demonstrated that the LGBVR region had the best effects of sedimental deposition, nutrient accumulation, and biological fixed carbon. It also had relatively higher values of NH₄⁺-N, TN, AP, MBC contents, MBC/TOC, and MBC/MBN in sediments. MBC was regarded as bioactive carbon components in sediment and is sensitive to the ecological restoration effect [25]. The structure of the LGBVR was developed to facilitate sediment deposition and nutrient accumulation and does not affect the growth of plant roots. According to the survey, it had 10–15 cm of siltation in the gabion after the completion of 1 year, which made it especially suitable for the growth of perennial phragmites. In addition, the bottom layers of the MGR are geotextiles and a bamboo woven mat, and sediments are deposited among three-dimensional network components. Although it has certain advantages in promoting sedimental deposition, its structure is more suitable for the growth of annual shallow root herbaceous plants. Furthermore, the MGR region had relatively higher values of TOC content in sediments. For the TRR region, the sediments were concentrated between the stone crevices, and the sediment deposition and plant growth conditions were poor.

4.2. Vegetation Restoration and Sediment Deposition Characteristics in Ecological Riverbanks

The accumulation of sediment and organic matter improves soil conditions and was the material basis for vegetation restoration and biodiversity maintenance in riparian zones [31]. In riparian zones, ecological riverbanks can benefit sediment deposition, decrease the water flow velocity, and facilitate plant growth [34,35]. The unique structure of ecological riverbanks can promote plant colonization, growth, and sedimental deposition, but the vegetation restoration and sediment deposition effects are obviously different in different forms of ecological riverbanks. In the same climate zone, an ecological riverbank structure such as an eco-bag, an eco concrete tetrad-ball in Hang-jia-hu plain [13], or the planting of a concrete revetment in an island of the estuary of Yangtze River [8] can have a good vegetation restoration effect. The plant species composition and dominant species are important indicators for vegetation succession [36]. In this study, the LGBVR region had the highest biomass, Margalef index (d), and Shannon–Weaver index (H′), and its dominant species was Phragmites australis. Although the MGR has certain advantages in resisting water flow impact and promoting sedimental deposition, it is not suitable for the colonization and growth of perennial plants in the short term. In the frequently submerged area near the riverside, this study also revealed that the sediment deposition and vegetation restoration effect of the riprap riverbank were better because it was closer to the natural state. In the French and Swiss Alps region, riprap riverbanks had relatively poor vegetation restoration effects [37]. However, in sub-tropical areas, they had relatively good ecological restoration effects compared to a gabion revetment in Cuatien River, Vietnam [17]. For periodically submerged riverbank areas, the riprap riverbank’s sediment deposition and vegetation restoration effects were comparatively better than those achieved using the MGR and LGBVR. Therefore, along the transverse gradient of riparian zones, ecological riverbank development should adopt the mixed use of different structure forms in the deep waterway regulation scheme of Yangtze River [38,39]. In the lower reach of the Yangtze River, the deep waterway regulation scheme is affected by both runoff and tidal currents, and the river level variation and riverbed evolution are complex [15]. In the riparian zone survey, the natural habitat conditions varied with different plant communities, which was related to the water flow magnitude and sedimental deposition characteristics [40,41]. Due to the high structural stability and safety requirements, the ecological riverbanks were only studied in parts of river sections in this project. Furthermore, the design, construction, and maintenance of ecological riverbanks needs to be further evaluated.

5. Conclusions

In this study, the sediment deposition characteristics associated with the vegetation restoration of two kinds of typical ecological riverbanks, LGBVR and MGR, and TRR were investigated in the deep waterway regulation scheme of Yangtze River. For the LGBVR region, it had the highest biomass due to large and tall phragmite communities. It also had the highest Margalef index (d) and Shannon–Weaver index (H′) values, as well as the highest MBC content, MBC/TOC, and MBC/MBN in sediments. For the MGR region, it had the highest number of plant species (eight) and Pielou index (J) and a relatively higher biomass in sampling sites. It also had highest values of TOC contents in sediment. The TRR region had the lowest biomass. In general, the LGBVR region had the best vegetation restoration and sediment deposition effects, followed by the MGR region. The TRR region had a relatively poor ecological restoration effect. Furthermore, the sediment deposition and vegetation restoration effects in different forms of ecological riverbanks are still in the process of dynamic succession, and these ecological effects need to be studied for a long time.