Assessment of the Lower Danube Sediment Regime and Morphology for the Identification of Habitats for Critically Endangered Sturgeon ^†

Abstract

The investigation of sediment morphology and sediment regime is crucial for the initial stages of the hydromorphodynamic modeling of the Lower Danube basin. This helps in identifying significant habitats and potential obstacles that may disrupt the migration patterns of anadromous sturgeon species in the Lower Danube. This investigation involved the analysis of 10 samples, which were collected in equal quantities from specific places where hydrotechnical activities were conducted, specifically the Bala and Caleia branches. The sediment samples were analyzed to determine their morphological and structural characteristics through granulometric assessments. Additionally, three significant parameters, namely, the standard deviation, symmetry index (skewness index), and flattening index (kurtosis index), were used for further characterization.

1. Introduction

The morphology of the substrate is a vital aspect in comprehending the specific needs of anadromous sturgeon species for their wintering, feeding, and particularly spawning habitats [1,2]. Considering that the three species can only reproduce in the Lower Danube [3], it is crucial to analyze the morphology and sediment regime. These data are necessary for further complex hydromorphodynamic modeling, which can give a greater awareness of the potential risks of migration route disruption [4].

Over time, a number of hydrotechnical works were undertaken to maintain navigation along the Danube and assure the uninterrupted flow of economic activities. Between 2011 and 2014, a project was constructed near the county of Braila, specifically, on a subsidiary branch of the Danube known as the Caleia branch, with the primary objective being to reallocate the flow of water from the secondary channel to the main branch through the installation of a bottom sill. The bottom sill and consequent redistribution of water flow resulted in hydrodynamic changes due to the higher velocity of the water current [5,6]. These alterations can have an impact on the migration of anadromous sturgeon species that move upstream from the Black Sea to reproduce [7,8,9].

On the other hand, sediments play an important role in defining the degree of water pollution and can provide a better understanding of how different pollutants are dispersed in the Danube basin [10,11]. The quality of the input data utilized is a crucial aspect in maintaining a high level of confidence in the output of numerical modeling methods. To guarantee this, the National Institute for Research and Development in Environmental Protection (INCDPM) conducts in situ measurements for the parameters required by hydromorphodynamic models. These measurements include both input data and the parameters necessary for numerical simulations, calibration, and validation processes. As granulometric analysis methods we mention the following [12]:

• Sieving using sieves (grain diameter d > 2.00 mm);
• In situ sieving method (0.05–0.08 (0.063) < d< 2.00 mm);
• Sedimentation method (d < 0.05–0.08 (0.063) mm).

The granulometric analysis provides correlations between the mass percentage of solid particles below a given dimension and the corresponding particle diameter, represented through specific graphs, such as the following:

• The frequency curve, which is a graphical depiction of a string that consists of pairs of particle size classes and their corresponding fractions—a histogram representing the simple frequency distribution in percentage [%] and the proportion of involvement for each distinct size class.
• The cumulative frequency refers to the total accumulation of frequencies in a dataset. The cumulative percentage involvement refers to the total proportion of each particle size class up to a specific diameter value.

Furthermore, the granulometric curves were characterized by the following parameters: the median, the graphical mean, the standard deviation, the skewness index, and the kurtosis index. In this study, the structure and composition of sediment, along with the patterns of sediment movement, that are crucial factors in developing numerical models that are key habitats for anadromous sturgeon species in the Lower Danube, were analyzed.

2. Materials and Methods

2.1. Sampling Campaigns

In order to conduct sediment transport modeling, it is essential to determine the sedimentological properties of the river bed within the modeling area. In order to achieve this objective, INCDPM conducted in situ sampling campaigns, specifically targeting submerged obstacles on the Bala and Caleia branch areas (Figure 1), due to their potential impact on anadromous sturgeon migration upstream from the Black Sea. During these campaigns, a total of 10 sediment samples were collected, 5 for each branch, at different depths.

In order to establish the sediment databases, samples were assigned unique identification parameters in the format PS_location_sample_number: PS_Bala_1–PS_Bala_5 and PS_Caleia_1–Ps_Caleia_5, respectively.

2.2. Granulometric Assessment

The preliminary evaluation of silt involves performing a particle size analysis, which is a technical procedure that separates and categorizes the particles in a sediment sample based on their size, indicated as a percentage distribution. This process establishes the geometric properties of the particles, which are categorized into granulometric fractions that represent clusters of solid fragments. These fractions are measured as units of mass or as a percentage of the total mass of the dry sample. The particles are defined by their dimensions, which fall within specific size ranges defined by upper and lower limits.

The INCDPM campaigns obtained representative samples from various depths in the examined areas. The granulometric analysis of these samples was conducted using the sieving methodology. The STAS 1913/5-85 standard was considered in order to ascertain the grain size. Each sample was weighed at 300 g. The laboratory sample was then dried at a temperature of 105 °C, until it reached a consistent mass (the difference between two consecutive weighing was no more than 0.0002 g). The moisture content of the samples was then calculated.

The sample to be processed was poured over the sieve with the largest mesh of the set used, the action being carried out manually (Figure 2a) or by using a machine (Figure 2b), until the granular fractions separated.

In order to verify the completion of the sieving process, each sieve is vigorously shaken over a sheet of paper for a duration of 1 min. The material that has successfully gone through the shaking sieve should not exceed 1% of the total mass of the material on that sieve. The residue left on the paper is then combined with the material on the subsequent sieve. The weight of the granular fractions that remain on each sieve is measured.

The most widely used grain size scale in sedimentology is the Udden–Wentworth granular scale [13], with particle diameters expressed in millimetres and unchanged ratios between class boundaries. The PHI scale was introduced by Krumbein [14] and represents the logarithmic transformation of the Udden–Wentworth geometric scale, the phi diameter being calculated as the binary logarithm (log base 2) of the particle diameter (D) in mm.

ϕphi = log2 D

(1)

This enables the analysis of sediment size distributions throughout a broader range of diameter values and facilitates a more concentrated examination of smaller particles.

Additionally, the granulometric curves were characterized by the following parameters:

• The median, also known as D50, is the diameter value that splits the particle size distribution into two equal sections, with 50% of the particles having a diameter smaller than or equal to this value. Additional significant values are the size values corresponding to the selected cumulative frequencies equidistant from D50: D16 and D84, D5 and D95, and D25 and D75.
• The graphical mean of a distribution refers to the arithmetic mean of frequencies that are equidistant from the 50th percentile.

$${\mathrm{M}}_{\mathrm{F}\mathrm{o}\mathrm{l}\mathrm{k}-\mathrm{W}\mathrm{a}\mathrm{r}\mathrm{d}}={\displaystyle \frac{\mathrm{D}16+\mathrm{D}50+\mathrm{D}84}{3}}$$

(2)

• Standard deviation corresponds to the dispersion of size values within the distribution around the mean of 50%.

$${\sigma }_{\mathrm{F}\mathrm{o}\mathrm{l}\mathrm{k}-\mathrm{W}\mathrm{a}\mathrm{r}\mathrm{d}}={\displaystyle \frac{\mathrm{D}84-\mathrm{D}16}{4}}+{\displaystyle \frac{\mathrm{D}95-\mathrm{D}5}{6.6}}$$

(3)

• The skewness index, also known as the mean squared deviation, quantifies the extent to which values are dispersed in a frequency distribution.

$$\mathrm{S}\mathrm{K}={\displaystyle \frac{\mathrm{D}16+\mathrm{D}84-2\times \mathrm{D}50}{2\times \left(\mathrm{D}84-\mathrm{D}16\right)}}+{\displaystyle \frac{\mathrm{D}5+\mathrm{D}95-2\times \mathrm{D}50}{2\times \left(\mathrm{D}95-\mathrm{D}5\right)}}$$

(4)

• The kurtosis index quantifies the degree to which values in a frequency distribution are concentrated in a certain region.

$${\mathrm{K}}_{\mathrm{G}}={\displaystyle \frac{\mathrm{D}95-\mathrm{D}5}{2.44\times \left(\mathrm{D}75-\mathrm{D}25\right)}}$$

(5)

3. Results and Discussion

The percentage distribution of the granular fractions for the five samples collected from the bottom sills area of the Bala branch are presented in

Table 1. Sediment sample granulometry—Bala branch.

Sample	Sieve Diameter [mm]	2	1	0.5	0.25	0.16	0.09	<0.09	Total
PS_Bala_1	Sieve debris [g]	0.000	0.000	0.410	36.300	202.099	45.601	3.027	287.437
	Gravimetric percentage [%]	0.000	0.000	0.143	12.629	70.311	15.865	1.053	100.000
PS_Bala_2	Sieve debris [g]	0.000	0.000	0.445	16.800	170.232	51.250	5.575	244.302
	Gravimetric percentage [%]	0.000	0.000	0.182	6.877	69.681	20.978	2.282	100.000
PS_Bala_3	Sieve debris [g]	0.000	3.494	10.695	20.528	77.361	84.371	33.278	229.727
	Gravimetric percentage [%]	0.000	1.521	4.656	8.936	33.675	36.727	14.486	100.000
PS_Bala_4	Sieve debris [g]	0.000	16.020	17.066	32.880	58.466	67.134	27.154	218.720
	Gravimetric percentage [%]	0.000	7.324	7.803	15.033	26.731	30.694	12.415	100.000
PS_Bala_5	Sieve debris [g]	0.000	6.160	42.221	43.202	39.342	29.983	18.565	179.473
	Gravimetric percentage [%]	0.000	3.432	23.525	24.072	21.921	16.706	10.344	100.000

. The same results for the Caleia branch can be found in

Table 2. Sediment samples granulometry—Caleia branch.

Sample	Sieve Diameter [mm]	2	1	0.5	0.25	0.16	0.09	<0.09	Total
PS_Caleia_1	Sieve debris [g]	5.804	8.570	5.430	16.336	26.329	98.050	42.474	202.993
	Gravimetric percentage [%]	2.859	4.222	2.675	8.048	12.970	48.302	20.924	100.000
PS_Caleia_2	Sieve debris [g]	5.806	11.937	6.274	9.249	14.010	85.940	66.399	199.615
	Gravimetric percentage [%]	2.909	5.980	3.143	4.633	7.019	43.053	33.264	100.000
PS_Caleia_3	Sieve debris [g]	6.225	11.220	5.640	2.933	14.639	117.176	42.572	200.405
	Gravimetric percentage [%]	3.106	5.599	2.814	1.464	7.305	58.470	21.243	100.000
PS_Caleia_4	Sieve debris [g]	6.059	10.039	7.598	41.958	54.362	69.538	29.371	218.925
	Gravimetric percentage [%]	2.768	4.586	3.471	19.165	24.831	31.763	13.416	100.000
PS_Caleia_5	Sieve debris [g]	4.015	18.340	14.690	21.061	16.759	65.102	60.015	199.982
	Gravimetric percentage [%]	2.008	9.171	7.346	10.531	8.380	32.554	30.010	100.000

.

Figure 3 present the histograms, corresponding to simple frequency curves for the Bala branch samples. Figure 4 shows the cumulative frequency curves for the data collected from the Bala branch area.

Regarding the samples taken from the submerged obstacle area in the Caleia branch, the histograms (simple frequency curves) are shown in Figure 5 and the cumulative frequency curves are shown in Figure 6.

Using the characteristic values for particle diameter obtained from the examination of sediment samples, we calculated the indicators to characterize the 10 sediment samples. The results can be found in

Table 3. Granulometric curve parameters—Bala branch samples.

Sample	Standard Deviation	Skewness Index	Kurtosis Index
PS_Bala_1	0.042	1.078	0.810
PS_Bala_2	0.052	0.171	1.112
PS_Bala_3	0.119	0.342	2.054
PS_Bala_4	0.251	0.641	2.055
PS_Bala_5	0.276	0.535	0.839

and

Table 4. Granulometric curve parameters—Caleia branch samples.

Sample	Standard Deviation	Skewness Index	Kurtosis Index
PS_Caleia_1	0.274	0.724	5.670
PS_Caleia_2	0.298	0.717	10.022
PS_Caleia_3	0.278	0.609	11.023
PS_Caleia_4	0.305	0.675	2.987
PS_Caleia_5	0.398	0.838	2.517

. The diagnostic analysis was conducted on the gathered sediment samples to determine the classes of values for the cumulative curve parameters. The results of this study are depicted using the color coding presented in Figure 7.

4. Conclusions

The structure and composition of sediment, as well as the patterns of sediment movement, are essential factors in the development of numerical models that study the important habitats for the anadromous sturgeon species in the Lower Danube, and also help identify areas where migration routes may be at risk of disruption, especially due to hydrotechnical works, such as the ones on the Bala and Caleia branches.

In this context, upon analyzing the cumulative curve representation of the samples collected from the Bala branch area, it was determined that the primary grain size fraction is the fine–medium sand class. Furthermore, following the assessment of the cumulative curves of the samples taken from the Caleia branch area, it was found that the primary grain size fraction is the fine–medium sand class. It was observed that the range of values is wider compared to the samples from the Bala branch area, where the coarse sand fraction is more prominent.

It was demonstrated that, with respect to the standard deviation, 9 out of 10 samples exhibit “Excellent sorting”, while PS_Caleia_5 is characterized by “Good sorting”. When it comes to the symmetry index (skewness), 9 out of 10 samples show a “Positive asymmetry”, while PS_Bala_1 exhibits a “Strongly positive asymmetry”.

The parameter that differentiates the samples the most is the flattening index (kurtosis). Even though all the samples from the Caleia branch are characterized as “Strong flattening”, when it comes to the Bala branch, two out of the five samples exhibit a sharp character, while the remaining three samples show a flattening character.

These data regarding the morphology and regime of sediments in two areas of interest for the anadromous sturgeon migration in the Lower Danube will be employed in further studies regarding the numerical modeling of the river basin in order to better understand its morphology and to identify important habitats and potential disruptions in migration routes.