*Article* **Distribution Characteristics and Potential Risks of Polycyclic Aromatic Hydrocarbon (PAH) Pollution at a Typical Industrial Legacy Site in Tianjin, North China**

**Chaocan Li 1, Xiaopeng Zhang 2,3,\*, Xuqin Wang 1, Xinbo Zhang 1, Shigang Liu 2,3, Ting Yuan 2,3, Weigui Qu 2,3 and Youjun Zhang 2,3**


**Abstract:** Polycyclic aromatic hydrocarbon (PAH) pollution in the soil of industrial legacy sites is a prominent problem when reusing urban land. To estimate the potential risks of PAHs, this study investigated 16 priority PAHs in the soil at different depths in a typical decommissioned industrial site in Tianjin. PAH concentrations were determined via gas chromatography-(tandem) quadrupole mass spectrometry. Incremental lifetime cancer risk (ILCR) assessment was applied to assess the potential risks to the population after land reconstruction. The total concentrations of PAHs in the soil at different depths ranged from 38.3 ng·g−<sup>1</sup> to 1782.5 ng·g−1, which were below the risk control standard for soil contamination of development land (GB 36600-2018). Low-ring (two-three ring) PAHs exhibit a dominant component, and the variations in PAH compositions were closely related to the former production units and soil properties. Compared to silty clay layers, PAHs tended to accumulate in the permeable miscellaneous fill layers. Incremental lifetime cancer risk assessment values associated with different exposure pathways for children, adolescents, and adults were calculated. The results showed potential carcinogenic risks for people of varying ages in this area, but they were still acceptable. In general, this legacy site can meet the demands of sustainable land development.

**Keywords:** PAHs pollution; risk assessment; industrial legacy sites; vertical distribution; land reuse

#### **1. Introduction**

China is currently undergoing an important restructuring of urban industrial structure and optimization of spatial layout [1]. Dramatic changes in land-use patterns have introduced a series of challenges, such as environmental pollution, climate change, and land deterioration, making soil resources increasingly strained [2,3]. With the pursuit of sustainable development, the remediation and reuse of polluted or degraded land have become increasingly important. So far, a large proportion of traditional industries have closed down or moved out of cities so that precious land resources can be reconstituted. However, most industrial legacy sites are polluted to varying degrees, making it impossible to reuse them directly, and polycyclic aromatic hydrocarbon (PAH) contamination is one of the major types of pollution [4,5]. PAHs are a type of semivolatile organic compound (SVOC) with low vapor pressure and hydrophobic property [6]. They are carcinogenic, teratogenic, and mutagenic contaminants, which can be easily bioconcentrated, can be transported over long distances [7–9], and have attracted great attention in terms of the global environment and public health [10,11]. PAHs originate from a wide range of sources, and the dominant contributing sources are the incomplete combustion of fossil fuels and

**Citation:** Li, C.; Zhang, X.; Wang, X.; Zhang, X.; Liu, S.; Yuan, T.; Qu, W.; Zhang, Y. Distribution Characteristics and Potential Risks of Polycyclic Aromatic Hydrocarbon (PAH) Pollution at a Typical Industrial Legacy Site in Tianjin, North China. *Land* **2022**, *11*, 1806. https://doi.org/ 10.3390/land11101806

Academic Editors: Jinyan Zhan, Xinqi Zheng, Shaikh Shamim Hasan and Wei Cheng

Received: 6 September 2022 Accepted: 11 October 2022 Published: 15 October 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

organic matter diagenesis. Therefore, PAHs are widely present in the environment, especially in areas with coal-combustion activities [12]. Due to their properties, more than 90% of released PAHs finally enter the soil (or sediment) through various pathways, making soil (or sediment) an important reservoir for PAHs [13,14]. After entering the soil, PAHs can be used as the sole carbon source and energy source for biological metabolism via soil microorganisms. If properly performed, the biodegradation of PAHs could represent a remediation strategy for petroleum-contaminated sites [15]. PAHs can combine with other organic matter for co-metabolism [16]. If PAHs are co-metabolized with organic maters and converted into phenols, quinones, and aromatic carboxylic acids via soil microorganisms and photochemical degradation, the conversion products are more toxic than the parent polycyclic aromatic hydrocarbons, resulting in a more severe threat to the environment [17]. Therefore, PAH pollution in the soil has always been a popular research topic.

Lately, there has been increasing research on PAH pollution in industrial legacy sites. Determining the characteristics of PAH in the soil will lay a solid foundation for soil remediation and land utilization [18]. Ma et al. [19] investigated the legacy sites left by relocating 26 chemical enterprises in Beijing and found that more than 30% of the sites were polluted by petroleum hydrocarbons and polycyclic aromatic hydrocarbons. Li et al. [20] studied PAH residues and the carcinogenic risks of dust samples at the surfaces of industrial legacy construction sites and in soil samples and found that the spatial distribution of PAHs in the dust was consistent with ∑16PAHs in the soil, and carcinogenic PAHs of industrial legacy sites should be regulated for regeneration. Cao et al. [9] investigated the contamination status of PAHs in the top soils of three industrial legacy sites (i.e., steelworks, a coking plant, and a gas station) and showed that the total concentration of 16 PAHs ranged from 371.1 ng·g−<sup>1</sup> to 4073.9 ng·g−1, and PAH pollution varied greatly among different types of enterprises. Most studies focus on the industrial legacy sites of petrochemical, coking, steel, and chemical industries [21,22], but few have focused on coal-fired power plants. PAHs are an integral part of the coal structure. During combustion processes, organic fragments are released through which cyclization or radical condensation reactions occur, leading to the formation of PAHs [23]. The characteristics and toxicity effects of parent PAHs and halogenated PAHs from active coal-fired power plants have been thoroughly studied [24,25]. However, the residual components of PAHs in the soil of a closed coal-fired power plant years after its abandonment remain unknown. Moreover, many studies have concentrated on PAH contamination in the surface soil of industrial legacy sites, but PAHs that remains in the soil for a long time may further contaminate groundwater [26,27]. Understanding the longitudinal pollution of PAHs from these sites remains an urgent matter.

Due to a high level of industrialization for several decades, several studies have reported a high level of PAH contamination in the sediment, water, and atmosphere of the Haihe River Basin in Tianjin [28,29]. Since coal-fired thermal power plants are a possible contributor to PAH pollution, in the present study, 7 drilling soil cores with a depth of 5 m from a typical former coal-fired thermal plant site, located in the human settlements in Tianjin, downstream of the Haihe River Basin, were collected for investigation. The objectives of this study were (1) to evaluate PAH contamination and longitudinal distribution at a former thermal power plant, (2) to identify the pollution contribution of different production units, and (3) to assess the risks posed by PAH residues at the study site to the environment and human health. The results will provide an environmentally relevant methodology and useful information for managing and remediating PAH contaminated sites. Additionally, this case study can provide data support for further exploration of land reuse and ecosystem response at decommissioned industrial sites.

#### **2. Materials and Methods**

#### *2.1. Study Area and Sampling*

The legacy site of the former thermal power plant is located in the downtown Hexi district in Tianjin, northeast and adjacent to the Haihe River. The plant was put into production after its completion in 2005, and after 10 years of production and operation, the unit in the plant was shut down in 2015. Since then, the land has been left vacant, with the original buildings on the site remaining intact until demolition beginning in October 2019. The historical production activities in this site were quite specific, mainly coal-fired power generation. The whole production process included the following steps: the combustion of coal in the boiler to generate heat, the heating of water into steam, the use of steam to drive steam turbine power generation, the power supplied to the power grid after being adjusted by the transformer, and the smoke and dust discharged through the bag filter and desulfurization device after generation.

#### *2.2. Sampling*

It has been preliminarily speculated that PAH contamination in soil is closely related to the different production processes, such as ground flushing, ash and slag stacking, and coal conveying [30]. Additionally, the daily maintenance of steam turbines, generators, and other equipment used in coal-fired power plants may also cause potential leakage. Based on the historical production processes at this site and the stationing conditions, a total of seven core sampling sites were set up to investigate the longitudinal characteristics of PAH in this legacy site, as shown in Figure 1. The drilling of the 7 core samples was executed before the demolition of the main structure, in case of disturbance.

**Figure 1.** Locations of core sampling sites at the thermal power plant legacy site, Tianjin, China.

The terrain where the site is located is low and flat. The stratigraphic conditions from top to bottom are as follows: (1) the artificial soil filling layer, with a thickness of 2.3–3.7 m; (2) the silty clay layer, with a thickness of 2.3–4.1 m. The buried depth of stable groundwater level ranges from approximately 1.8 to 2.4 m. According to the results of the geological surveys, the soil cores were sampled by Geoprobe (7822DT, USA) at a depth of 6 m with horizon as a base depth in December 2021. Each soil core was sampled at a different depth of 0.3 m, 0.6 m, 0.9 m, 1.2 m, 1.5 m, 2.4 m, 3 m, 4 m, and 5 m. Finally, a total of 70 samples (14 random, duplicated samples excluded) were collected and placed in polyethylene zipper bags and transported on ice to the laboratory. All the samples were naturally air-dried, ground, passed through 80-mesh sieves to remove non-soil materials such as plant roots and stones, and stored at −20 ◦C before analysis. A small amount was taken out from each sample to analyze the physical and chemical properties of the soil, as seen in Table S1 in Supplementary Materials. The lithology and location information of 7 core sites are listed in Table 1.


**Table 1.** Lithology and location information of 7 core samples sites.

#### *2.3. Sample Preparation and Analysis*

PAH congeners were analyzed following a method described in [31], with some modifications. Approximately 5 g of each soil sample spiked with surrogate standards (naphthalened8, acenaphthene-d10, phenanthrene-d10, chrysene-d12, and perylene-d12) was extracted with a mixture solvent of 100 mL acetone/n-hexane (V:V = 1:1) using an automatic Soxhlet extractor. Afterward, the extraction was solvent-exchanged to hexane and concentrated to approximately 1 mL by employing a rotary evaporator. The purification and fractionation processes were performed with a solid-phase extraction column (SPE, florisil, 1 g/6 mL). The column was first washed with 4 mL of n-hexane, and then 1 mL sample extract was transferred to the SPE column. A total of 10 mL of an eluent (a mixture solvent of dichloromethane and n-hexane) was added to the SPE column to achieve the elution of the target PAHs. The clean extract was then evaporated to 1 mL under a gentle stream of high-purity nitrogen and transferred to a 1.5 mL brown bottle before the instrument analysis.

The 16 priority PAHs designated by the United States Environmental Protection Agency (USEPA), naphthalene (Nap), acenaphthene (Acy), acenaphthene (Ace), fluorene (Fle), phenanthrene (Phe), anthracene (Ant), fluoranthene (FLu), pyrene (Pyr), benz[a]anthracene (BaA), chrysene (Chr), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), indeno [1,2,3-cd]pyrene (IcdP), dibenz[a, h]anthracene (DBA) and benzo[g, h, i]perylene (BghiP), were analyzed via gas chromatography-(tandem) quadrupole mass spectrometry (TQ8040, Shimadzu, Japan) in the electron-impacting (EI) mode. The chromatographic column was equipped with a DB-5ms fused silica capillary column (30 m × 0.25 mm × 0.25 μm, J&W Scientific, Folsom, CA, USA). The oven program for chromatographic column was set as follows: the initial temperature was set at 60 ◦C (for 1 min), then increased to 200 ◦C at a rate of 20 ◦C/min, followed by 1 min retention, and then lifted to 300 ◦C at a rate of 8 ◦C/min followed by 5 min retention, and finally to 310 ◦C at a rate of 5 ◦C/min. Helium was assigned as the carrier gas with a flow rate of 1.2 mL/min. The temperatures of ion source, transfer line, and quadrupole trap were maintained at 220 ◦C, 310 ◦C, and 150 ◦C, respectively. Multiple reaction monitoring (MRM) was utilized for quantitative analysis.

#### *2.4. Quality Assurance/Quality Control*

Besides the recovery of surrogate standard in every sample, procedure blanks, spiked blanks, spiked matrices, and randomly duplicated samples were also extracted and analyzed in the same way as the real soil samples in every batch of 8 samples. After the injection of each batch of samples, a calibration standard and a solvent blank were injected to check the background and stability of the instrument. The limit of detection (LOD) was quantified as the average of PAHs' concentration determined in blank samples summed with three times standard deviation values of each PAH compound (LOD = average ± 3 × SD) [32,33], and the limit of quantitation (LOQ), calculated as 3 times LOD, was 0.38~6.24 ng·g−1. Only a few PAH congeners with rather low concentrations were detected in the procedure blanks, excluding Nap. Since Nap has great volatility and a high background level, we calculated the other 15 PAHs, excluding naphthalene. The average recoveries obtained from the surrogate standard of d-acenaphthene, d-phenanthrene, d-chrysene, and d-perylene were 70.0 ± 9.2%, 84.4 ± 12.2%, 89.4 ± 11.3%, and 97.5 ± 14.2%, respectively. Recovery rates in the spiked matrices and blanks were 74.3%~118.32%, with a standard deviation of 8.5%–15.3%. The quantitation of PAHs was using an external calibration curve method with correlation coefficient (R2) ≥ 0.99.

#### *2.5. Lifetime Cancer Risk Assessment*

The USEPA standard [34,35] model of Lifetime Cancer Risk Assessment was used to assess the ILCR associated with PAH exposure in the soil of the legacy site. The calculation of ILCR for different population groups in terms of direct ingestion, dermal contact, and inhalation was as follows:

$$\text{ILCRs}\_{\text{Ingestion}} = \frac{CS \times \left(CSF\_{\text{Ingestion}} \times \sqrt[3]{\left(\frac{BW}{70}\right)}\right) \times IR\_{\text{soil}} \times EF \times ED}{BW \times AT \times 10^6} \tag{1}$$

$$\text{ILCRs}\_{\text{Derman}} = \frac{\text{CS} \times \left( \text{CSF}\_{\text{Derman}} \times \sqrt[3]{\left( \frac{\text{BW}}{70} \right)} \right) \times SA \times AF \times ABS \times EF \times ED}{BW \times AT \times 10^6} \tag{2}$$

$$\text{ILCRs}\_{Inhalation} = \frac{\text{CS} \times \left(\text{CSF}\_{Inhalation} \times \sqrt[3]{\left(\frac{\text{BW}}{70}\right)}\right) \times IR\_{air} \times EF \times ED}{BW \times AT \times PEF} \tag{3}$$

where *CS* is the PAH concentration in soil exposure in the topsoil of the study area (μg·kg<sup>−</sup>1) (namely PAH concentration in the surface soil), which was obtained by converting the concentrations of PAHs exposure according to toxic equivalents of BaP using the toxic equivalency factor [36]. The other parameters in the formula are listed in Table 2 [37,38].




**Table 2.** *Cont.*

The toxic equivalency quantities (TEQs) of PAHs were calculated first based on the following equation:

$$\text{TEQ} = \sum (\mathbb{C}\_{PAH} \times TEF) \tag{4}$$

where *CPAH* represents the concentration of individual PAHs.

#### **3. Results and Discussion**

#### *3.1. Levels of PAHs in the Legacy Site*

In this study, 70 samples from 7 drill holes were obtained from the former thermoelectric plant legacy site. The concentration of PAH compounds ranged from 38.3 ng·g−<sup>1</sup> to 1782.5 ng·g−<sup>1</sup> (dry weight, dw), with an average concentration of 542.2 ng·g−<sup>1</sup> dw (Nap was excluded). Table 3 presents the summary of the detection ranges of the individual PAHs. In total, the detection ratios of high-ring PAHs (5–6 rings) were significantly lower than those of low-and middle-ring PAHs (2–4 rings), which suggested an obvious source of incomplete combustion [8]. This is consistent with the expected characteristics of PAH pollution at this site. The maximum concentration of individual PAHs was anthracene (818.97 ng·g−1), followed by phenanthrene (712.91 ng·g−1). The standard deviation of all samples for each individual in this area was between 2.41 and 176.84, indicating that there are large differences between different samples at different locations or depths.

The lowest concentration of ∑15PAHs was detected at a depth of 5 m at site T1, the edge of the former plant, far away from the production area in the upwind direction. The highest concentration was found at 3 m depth (1782.5 ng·g<sup>−</sup>1) at T4, which was influenced by the temporal production process [32] as well as the vertical migration of PAHs [39]. The accumulated concentration of ∑15PAHs of all sectioned samples from each soil core was also calculated, and the highest cumulative concentration of ∑15PAHs was also detected at site T4. This again proved the speculation that PAHs originated from local contamination, which was closely related to the production process (boiler and oil pump unit). Seven kinds of carcinogenic PAHs, i.e., BaA, Chr, BbF, BkF, BaP, DBA, and IcdP, were calculated at levels ranging from 4.3 ng·g−<sup>1</sup> to 597.1 ng·g<sup>−</sup>1, with an average of 105.8 ng·g<sup>−</sup>1, which accounted for 19.5% of the average content of ∑PAHs. Despite the proportion not being particularly high, risks from PAHs at the industrial site still exist for groundwater and human health, even though the site has been abandoned for a long time.


**Table 3.** Statistical detection ranges of the individual PAHs in all soil samples.

The contents of PAHs in this thermal power plant legacy site were lower than those of the eastern coastal developed regions of China, which have a longer industrial history contributing to the pollution of PAHs in soil, such as the Pearl River Delta Region and the Yangtze River Delta Region. It was reported that the highest content of PAHs in soil samples collected from an e-waste recycling site in Guiyu reached 18,600 ng·g−<sup>1</sup> [40], and the highest ∑16PAH concentrations in soils from another e-waste recycling site in the Taizhou area reached 361,600 ng·g−<sup>1</sup> [41], which were tens to hundreds of times higher than the content of PAHs at this site. PAH contamination here was comparable to the residual levels of other industries in the Beijing–Tianjin–Hebei region, such as a cement factory (536.7 ng·g−1) [42], steel factories (1342 ng·g−1), and coking plants (735.3 ng·g−1) [9,38], which reflects the coordinated development and governance of the Beijing–Tianjin–Hebei region. In addition, the local PAH contamination was also compared with that of other coal-fired power plants in domestic and in the surrounding area. In this study, the PAH concentration in soil was comparable to that in Huainan City, a typical coal resource city in China (528.06 ng·g−1) [43]. However, it was lower than that in the surrounding surficial soil of Xuzhou, China (1089.69 ng·g−1) [25], and it was even lower in fly ash samples and bottom ash samples from an operating power plant in Anhui, China [23]. PAH contents here were also far lower than those in the soils of industrial heritage cities from other countries, such as Indonesia (11,720 ng·g<sup>−</sup>1) [44], Germany (15,879 ng·g<sup>−</sup>1) [45], South Africa (28,670 ng·g−1) [21], and France (181,000 ng·g−1) [46]. The relatively low PAH levels in the present study might be related to lower historical input and atmospheric deposition [43]. When compared to the background value around this region (336 ng·g−<sup>1</sup> on average) or compared to values from some countries that have always attached importance to cleaner production—for instance, Switzerland (225 ng·g−1) [47] and Japan (320 ng·g−1) [48]—PAH pollution in this legacy site was at a non-negligible level.

#### *3.2. Vertical Profiles of PAHs at the Legacy Site*

The vertical distribution of PAHs at different soil depths is illustrated in Figure 2. PAH contents at the surface soil samples (at a depth of 0.3 m) varied within a large range, from 124.4 ng·g−<sup>1</sup> to 1258.8 ng·g−1. The grading criteria of PAH contamination stipulate that PAH contamination levels can be divided into four grades: no contamination (∑PAH concentration < 200 ng·g−1, slight contamination (200 < <sup>∑</sup>PAHs concentration < 600 <sup>μ</sup>g/kg), moderate contamination (600 < ∑PAHs concentration < 1000 μg/kg), and severe contamination (∑PAHs concentration > 1000 μg/kg). PAH concentrations of over half of all the surface

soil samples in this study exceeded moderate and even severe contamination (site T4 and site T6), reflecting severe pollution of the current soil status [25,49].

**Figure 2.** Longitudinal distribution of PAH in each sampling core related to former production unit of the coal-fired power plant; the dividing line indicates the approximate boundary of soil lithology.

Moreover, the longitudinal distribution of ∑15PAHs showed a trend of first increasing, then decreasing, and then continuing to rise to the highest point, then declining, according to Figure 2. The first remarkable rise started at a depth of 0.9–1.2 m. The amounts of LMW PAHs were maintained at a high level, and the concentration of high-molecular-weight PAHs (HMW PAHs) exhibited a prominent increase (especially at T5, T6, and T7), which indicated that PAH accumulation corresponded with the release of the production process at that time; compared to low-molecular-weight PAHs (LMW PAHs), high-molecularweight PAHs are more likely to seep into the soil nearby [50]. As the soil depth deepened to below 1.8 m, the total concentration of PAHs dropped to the inflection point, then a second increase occurred. Apparently, the amounts of low-molecular-weight PAHs led to a second rise. This was because low-molecular-weight PAHs migrated downward more with groundwater seepage [51] and caused greater accumulation, while high-molecular-weight PAHs were enriched in the upper soil due to their low solubility and strong affinity to organic matter, which severely limited their vertical transport to deep soil [52]. A significant decrease in PAHs appeared in deeper soil at 3.0~5.0 m, which could be attributed to the change in soil lithology, as seen in Figure 3. A remarkable change took place in soil lithology from the miscellaneous fill layer to the silty clay layer at a depth of about 3.5 m. The poor permeability of the silty clay layer with high viscosity and low gravel makes it a barrier to the vertical migration of pollutants [53].

**Figure 3.** Vertical fractions of PAHs with different ring numbers in the soil of the legacy site.

The composition of PAHs at different depths of soil sampling points involved in this study is shown in Figure 3. In general, the 2–3 ring PAHs were the dominant contaminants out of all 15 PAHs at different soil depths, followed by the 4-ring PAHs. The composition of PAHs in topsoil (above 0.3 m) was not the same as that at other soil depths, with a high proportion of 5–6 ring PAHs (i.e., IcdP, DBA, and BghiP), which should be attributed to sources after shutting down the coal-fired plant, such as vehicular emission [44]. Besides this, PAHs with a low number of rings in the topsoil can also be easily revolatilized into the atmosphere [54]. Therefore, the content of PAHs in topsoil is unstable.

An obvious variation was revealed in the deep soil below the topsoil layer. The composition of PAHs in subsurface soil (at a depth of 0.3–0.9 m) samples mainly included 2–3 ring PAHs, whose proportion could reach more than 50%. Meanwhile, with the increase in depth (until a depth of about 1.8 m), the proportion of medium- and high-ring PAHs increased significantly. Additionally, as the depth of soil samples continued to increase, the proportion of low-ring PAHs increased again, and the proportion exceeded that of low-ring PAHs in surface soil. This regularity of PAH vertical composition manifested in the soil cores was partly due to the physicochemical properties of PAHs. Low-ring PAHs are of greater solubility and permeability than high-ring PAHs, and they are much more mobile to downward migration to deeper soil [51,55]. The composition of PAHs found in the mid-deep soil layers (1.2–3.0 m) was altered greatly compared to deep layers (4.0–5.0 m) and upper soil layers (0–0.9 m). The changes in organic matter in soil properties played a role in the variations, because soil organic matter content reaches its highest value in the soil surface layer and decreases significantly with increasing depth, which affects the adsorption and desorption processes in the soil to a great extent [56]. Accordingly, a better linear correlation between the TOC contents in soil samples and the ∑PAHs concentration—especially the HMW PAHs, as seen in Figure 4—verified the theory that soil organic matter (SOM) is usually responsible for the binding of PAHs in soil [57].

The PAH compositions at sampling sites T4 and T6 were clearly different from the others, with 4–6 ring PAHs dominant in the upper soil layers and the mid-deep soil layers. This may closely relate to the previous production status. These points were located at a downwind area of the boiler and oil pump units and transformer units. A major reason for the imparity of PAH was the impact of wind, and consequently the HMW-PAHs originated from the scattering and spreading of ashes and slags of process units deposited downwind [57,58].

**Figure 4.** The correlations of TOC contents with ∑PAHs (**a**), LMW-PAHs (**b**), and HMW-PAHs (**c**) and the correlation coefficient (R2) and significance level (p) are given.

#### *3.3. Influence of Different Process Units on PAH Pollution Status*

There was an obvious discrepancy in PAH characteristics in the soil at different sampling sites. The results were concordant with our hypothesis that different process units would lead to different degrees of PAH contamination. In the present study, the Kruskal–Wallis (K–W) nonparametric analysis was applied to investigate the discrepancy of PAH contamination within the regions of the four process units. As seen in Table 4, PAHs in soil samples from different regions of process units exhibit significant differences, because the *p* values for both ∑PAHs and HMW-PAHs were lower than 0.05 [59]. However, for LMW-PAHs, the data show non-significant differences, because LMW-PAHs are easily volatilized into the atmosphere and settle down over a period of time.

**Table 4.** Statistical data results of Kruskal–Wallis test, PAHs in soil from different regions of process units.


Figure 5 provides the contamination status of PAHs at key depths of each point, which was collected based on the location of the process unit in the legacy site. The soil at T4 and T5, distributed in the region of the boiler and oil pump, was heavily polluted with PAHs compared to soil in other process regions. According to the distribution of process units in Figure 5, the boiler room was the unit directly related to coal combustion. Due to coal-combustion production activities, the generated PAH first affected the nearby soil, resulting in a high cumulative PAH content in the surrounding soil. This was basically consistent with the results reported by Yang that the highest PAH content in soil was located in the heat-generating area [60]. Besides the area around the boiler and pump, PAH contents in the soils at sites T6 and T7 around the power-generation transformer were slightly higher than in other regions. This result may be related to the consumption or leakage of transformer oil during operation at that time; previous research has mentioned that aromatic hydrocarbons (polycyclic aromatic hydrocarbons and benzene series) make up more than 5% of common naphthenic transformer oil [61].

Due to the long-term operation of the transformer in this site, PAHs entered the surrounding environment with the phenomenon of "running, emitting, dripping, and leaking" and caused different degrees of pollution to the surrounding soil. The pollution content of PAHs at points T2 and T3 in the desulfurization area was lower than that in the soil at the other two regions mentioned above. Although large amounts of PAHs can be found in flue gases from thermal processes that involve incomplete combustion [62], the applied air pollution control devices had a significant effect on the removal of PAH in both particulate and gas phases [63]. The influence of natural factors such as monsoon climate

can also weaken the sedimentation of pollutants in the atmospheric environment [64]. Therefore, the pollution content in the soil around the chimneys, such as T2 and T3, was relatively low.

**Figure 5.** The contamination status of PAHs at key depths of each point collected based on the location of process unit in the legacy site.

Overall, process units contributed differently to PAH pollution in soil within the legacy area. The degree of different production units' PAH pollution ranked in the order of boiler process > power generation and transformation process > flue gas desulfurization process > boundary region (upwind direction).

Regarding the distribution of pollutants, there was no non-point source pollution of PAHs at this legacy site, which suggested that the soil pollution mainly comes from the production activities of the original plant. Additionally, the distribution of PAHs in the soil at different depths also reflected the obvious longitudinal migration of local pollution. Affected by the change of soil lithology, the migration depth accumulated to about 3.6 m underground at the demarcation of different soil lithologies, and the possibility that PAH continued to migrate slowly downward cannot be ruled out.

#### *3.4. Risk Assessment of Reconstructed Regional Population*

The PAH status of this site was compared with the values stipulated in the "Risk control standard for soil contamination of development land (GB 36600-2018)" implemented by China, as seen in Table 5. The controlled individual PAHs cited in the control standard above were all below the specified value, indicating that this legacy land would be classified as non-sensitive. The comparison between total PAHs in this site and the evaluation criteria proposed by Maliszewska [51] showed that 14.3% of all samples were considered heavily polluted (>1000 ng·g<sup>−</sup>1), 21% were considered moderately polluted (600–1000 ng·g<sup>−</sup>1), and 62% were considered mildly polluted (200–600 ng·g−1). What is particularly noteworthy is that all the samples were from different depths of just seven soil cores; thus, the total amount of PAHs accumulated in the longitudinal direction was a relatively high level, posing important risks to soil–plant system transportation [65] and even groundwater [26].

**Table 5.** Comparison between PAH in the target site and the "Risk control standard for soil contamination of development land (GB 36600-2018)".


Since the land met the criteria for reconstruction, it was necessary to evaluate the Incremental Lifetime Cancer Risk Assessment (ILCR) for the reconstruction regional population. The ILCRs of PAHs in the legacy site, classified for children, adolescents, and adults, were further calculated with Equations (1)–(3), and the results can be seen in Table 6. An ILCR of 10−<sup>6</sup> or less is considered a negligible risk, while ILCRs greater than 10−<sup>4</sup> indicate potentially high risk; ILCR values between 10−<sup>6</sup> and 10−<sup>4</sup> indicate an acceptable potential health risk [38,66]. In general, the carcinogenic risk values of direct ingestion, dermal contact, and inhalation for all populations estimated in this study ranged from 1.2 × <sup>10</sup>−<sup>8</sup> to 2.1 × <sup>10</sup><sup>−</sup>6, 1.2 × <sup>10</sup>−<sup>8</sup> to 6.2 × <sup>10</sup><sup>−</sup>6, and 4.2 × <sup>10</sup>−<sup>10</sup> to 1.5 × <sup>10</sup><sup>−</sup>7, respectively. Although none of the ILCR values exceeded 10−4, 28.6% of the soil samples exceeded the value of 10<sup>−</sup>6, indicating a potential risk to the local population at this site after reconstruction.

**Table 6.** Incremental lifetime cancer risks (ILCRs) for people via different exposure pathways.


There was no significant difference in the risks associated with different exposure routes between males and females. However, for different age groups (children, adolescents, and adults), the risks associated with the three exposure routes varied greatly. As shown in Figure 6, the risk of direct ingestion was greater than that of dermal contact and inhalation for children. Meanwhile, for adults and adolescents, the risk of exposure routes decreased in the following order: dermal contact > direct ingestion > inhalation. This is similar to other research in that differences existed in exposure pathways between children and adults [11]. Through all the health risk assessments, the results show potential risks in the legacy site if there is reconstruction, but the integrated lifetime cancer risks associated with exposure to soils with average PAH concentrations for different populations are acceptable.

**Figure 6.** Contributions of different exposure pathways for children, adolescents, and adults calculated by ILCR model.

#### **4. Conclusions**

The contaminated soil left after the relocation of industrial factories has attracted great attention. This study focused on the PAH contamination of soil in a typical thermal power plant legacy site in Tianjin, North China, aiming to determine the contamination status and vertical distribution of PAHs at the legacy site, and identify the intimate connection between different process units and the pollutant distribution. This study also provided a preliminary discussion on risk assessment after reconstructing a legacy site. Research on a single industrial site is not universal, but it can provide some methods and ideas for future research. In the thermal power plant legacy site examined, the total concentration of all PAHs ranged from 38.3 ng·g−<sup>1</sup> to 1782.5 ng·g−<sup>1</sup> (542.2 ng·g−<sup>1</sup> on average), which was at a comparable level to that from heavy industries in the Beijing–Tianjin–Hebei region. The 2–3 ring PAHs were the dominant contaminants among all individual PAHs at different soil depths at this site. PAH contents and constituents at different soil depths showed significant imparity since they were influenced by the physicochemical properties of PAHs as well as the variation of the soil lithology. Additionally, the poor permeability of the silty clay layer with high viscosity and low gravel makes it a barrier to pollutants' vertical migration. The distribution characteristics of PAHs in soil were also closely related to the production processes in the former factory. The degree of different production units in PAH pollution ranked in the order of boiler process > power generation and transformation process > flue gas desulfurization process. Health risks were assessed according to the incremental lifetime cancer risk assessment. In reality, there were potential carcinogenic risks for people of varying ages from this legacy site, but the values of all ILCRs were below 10<sup>−</sup>4, so the risks were still acceptable.

The present study is a reanalysis of pollution status and risk assessment of PAHs of the industrial legacy site, which has been abandoned and flagged as a residential area. The results of the current study provide an environmentally relevant methodology and useful information for managing and remediating industrial legacy sites. However, the contamination of other chemicals, such as heavy metals, volatile organic compounds, polychlorinated biphenyls and petroleum hydrocarbons, which are always associated with the operation of the coal-fired power plant, remain unexplored. Further studies should be focused on the pollution status of other chemicals, and the complex pollution

mechanism, providing a basis for assessing the possibility of remediating industrial legacy sites for reuse.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/xxx/s1, Table S1: Basic physicochemical properties and PAH concentration of collected soil samples.

**Author Contributions:** Conceptualization, X.Z. (Xiaopeng Zhang); methodology, C.L. and X.W.; writing—original draft preparation, C.L. and X.W.; writing—review and editing, C.L., X.Z. (Xinbo Zhang) and X.Z. (Xiaopeng Zhang) software, X.W.; data curation, C.L. and Y.Z.; investigation, S.L., T.Y. and W.Q.; resources, X.Z. (Xinbo Zhang) and W.Q.; project administration, X.Z. (Xiaopeng Zhang) and C.L.; funding acquisition, C.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** The research was financially supported by the Project of Tianjin Municipal Science and Technology Bureau of China, grant number 20YDTPJC00230; the Research Project of Tianjin Education Commission, grant number 2017KJ055; and the Project of Tianjin North China Geological Exploration Bureau, grant number HK2022-B3.

**Data Availability Statement:** All data are included in the manuscript, and additional data are available with the corresponding author and will be available upon request.

**Acknowledgments:** The authors are grateful to the editors and reviewers for their valuable comments and suggestions.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Impacts of Land Use on Pools and Indices of Soil Organic Carbon and Nitrogen in the Ghaggar Flood Plains of Arid India**

**Pravash Chandra Moharana 1, Roshan Lal Meena 2, Mahaveer Nogiya 2, Roomesh Kumar Jena 3, Gulshan Kumar Sharma 4,\*, Sonalika Sahoo 5, Prakash Kumar Jha 6, Kumari Aditi <sup>6</sup> and P. V. Vara Prasad 6,7**


**Abstract:** Changes in land use have several impacts on soil organic carbon (C) and nitrogen (N) cycling, both of which are important for soil stability and fertility. Initially, the study area was barren uncultivated desert land. During the late 1960s, the introduction of a canal in the arid region converted the barren deserts into cultivated land. The objectives of the present study were to evaluate the effects of various land use systems on temporal changes in soil organic C and N pools, and to evaluate the usefulness of different C and N management indices for suitable and sustainable land use systems under arid conditions. We quantified soil organic C and N pools in five different land uses of the Ghaggar flood plains, in hot, arid Rajasthan, India. The study focused on five land use systems: uncultivated, agroforestry, citrus orchard, rice–wheat, and forage crop. These land use systems are ≥20 years old. Our results showed that total organic carbon (TOC) was highest (7.20 g kg−1) in the forage crop and lowest in uncultivated land (3.10 g kg−1), and it decreased with depth. Across different land uses, the very labile carbon (VLC) fraction varied from 36.11 to 42.74% of TOC. In comparison to the uncultivated system, forage cropping, rice–wheat, citrus orchard, and agroforestry systems increased active carbon by 103%, 68.3%, 42.5%, and 30.6%, respectively. Changes in management and land use are more likely to affect the VLC. In soil under the forage crop, there was a considerable improvement in total N, labile N, and mineral N. Lability index of C (LIC), carbon management index (CMI), and TOC/clay indices were more sensitive to distinguishing land uses. The highest value of CMI was observed in the forage crop system followed by rice–wheat and agroforestry. In the long term, adoption of the forage crop increased soil quality in the hot, arid desert environment by enhancing CMI and VLC, which are the useful parameters for assessing the capacity of land use systems to promote soil quality.

**Keywords:** carbon and nitrogen pools; soil quality; carbon and nitrogen management index; land use; arid environment

#### **1. Introduction**

Hot, arid regions of India spanning across ~31.7 million hectares are characterized by a variety of landforms, soils, fauna, flora, and water resources as well as human activities [1,2]. As population and food demand continuously increase, these desert soils of hot, arid regions of India are being converted into arable lands, and more rapidly for the last 60 years. Desert

**Citation:** Moharana, P.C.; Meena, R.L.; Nogiya, M.; Jena, R.K.; Sharma, G.K.; Sahoo, S.; Jha, P.K.; Aditi, K.; Vara Prasad, P.V. Impacts of Land Use on Pools and Indices of Soil Organic Carbon and Nitrogen in the Ghaggar Flood Plains of Arid India. *Land* **2022**, *11*, 1180. https://doi.org/10.3390/ land11081180

Academic Editors: Shaikh Shamim Hasan, Jinyan Zhan, Xinqi Zheng and Wei Cheng

Received: 2 June 2022 Accepted: 25 July 2022 Published: 28 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

soils, however, have been harmed by increased wind erosion and salinity due to agricultural exploitation. There is an enormous amount of carbon (C) stored in desert ecosystems, and they store almost one-third of all terrestrial C (total C) [3,4], whereas 10% of the worldwide soil organic carbon (SOC) stock is found in arid and semiarid regions [5]. However, intensive cultivation, shrinking water resources, low biological productivity, severe erosion, and extreme climatic conditions in the arid regions of India have decreased the SOC [4,6]. As a result, identifying and implementing appropriate management techniques and land uses for arid regions to maintain or improve the SOC stock and recalcitrant or passive C pool are needed to enhance and sustain productivity while mitigating climate change.

Soil organic matter (SOM) is a critical component of soil quality and consequently a primary predictor of agricultural system sustainability [7]. Climate and management methods or cropping systems are the primary determinants of SOM maintenance in diverse land use systems. An important function of the SOM is to store nutrients, promote plant growth, and also sustain soil biodiversity, drive the nutrient cycle, maintain soil structural stability, increase infiltration of water, maintain porosity, and prevent erosion [8]. The dynamics of soil quality are determined by changes in SOM under crop cultivation. The primary constituents of SOM, SOC, and total nitrogen (TN) are strongly linked to a wide range of physical, chemical, and biological aspects of soil. Therefore, SOC and TN are used as important indicators of soil quality [9,10]. Since these labile forms of C and N are particularly sensitive to changes caused by agricultural management, they are employed to quantify SOM [11]. The total soil N content is the sum of all N pools in soil, most of which are organic in form and turn inorganic upon decomposition of SOM. For many arable crops, organic N mineralization is the primary process of N nutrition, and its potential in soil is regarded as a superior measure of fertility. Therefore, derived C and N indices such as carbon/N lability, carbon lability index, carbon pool index, and carbon management index (CMI)/nitrogen management index (NMI) may be used to analyze changes in SOM [7,12].

Knowledge of variations in SOC and TN under diverse land uses is required to understand the feasibility of applying conservation techniques to maintain production and safeguard the environment. CMI and NMI are good early indicators of whether or not a specific agricultural system is contributing to better soil quality. Land use changes can have a big influence on soil C storage. Agroforestry systems, diversified crop cycles, higher cropping intensity, and horticultural crops might all help to boost soil C sequestration [13]. However, very little information is available on these aspects for sandy desert soils of India. The current study examines the impact of diverse land uses on various soil organic C and N pools, as well as CMI and NMI. The objectives of this research were (a) to assess the effects of different land use patterns/systems on temporal variations in soil organic C and N pools in India's hot desert area; and (b) to evaluate the use of several C and N management indices as early indicators of overall C and N changes in various land uses in dry (arid) conditions. This knowledge would enable farmers to cultivate desert soils appropriately for long-term sustainability.

#### **2. Materials and Methods**

#### *2.1. Study Sites*

The study sites were the central state farm and central cattle breeding farm in the Suratgarh block of Sri Ganganagar district, Rajasthan, which lie between 29◦20 53 N to 29◦24 47 N latitude and 73◦30 0 E to 73◦37 38 E longitude and are situated at 171 m above mean sea level (Figure 1). The physiography was western plain–semiarid transitional plains, which constitute hot, arid sandy plains, and the agro-eco sub-region of the Ghaggar flood plains. The major soil series was Suratgarh soil series (fine, loamy, mixed (cal.) hyperthermic family of *Ustochreptic Haplocambids*). The dominant soils are deep to very deep. The soils are slightly alkaline (pHw of 8.31) and organic C and CaCO3 were 0.20 and 4.8%, respectively [14].

**Figure 1.** Study area and soil sampling locations in various land use systems of a hot, arid desert climate.

#### *2.2. Land Use Changes*

Initially, the study area was barren uncultivated desert land during the 1950s. In the late 1960s, the introduction of a canal in the arid region converted the barren deserts into cultivated land. The lands were brought under field as well as plantation crops and agroforestry trees since 1955. For this study, five different land use systems, namely (i) uncultivated, (ii) agroforestry, (iii) citrus orchard, (iv) rice–wheat system, and (v) forage crops were selected. All of the selected land uses were more than 20 years old, to examine the long-term impact of land uses on the buildup of SOC and N and their pools. Here, we compared SOC and N pools of different land uses with uncultivated land considering the initial soil condition with reference to climatic and topographic conditions (Table 1).

**Table 1.** Description of land use systems prevailed in the hot arid regions of India.


#### *2.3. Soil Sampling and Analyses*

During May 2015, three composite soil samples were taken using an auger at five intervals of 0–20, 20–40, 40–60, 60–80, and 80–100 cm from each land use type. Each sampling site had three plots, and as a result a total of 75 (5 land uses × 5 depths × 3 plots) composite samples were considered for laboratory analysis. Core samples were collected separately for determination of bulk density (BD).

The collected samples were analyzed for BD, pH, electrical conductivity (EC), cation exchange capacity (CEC), texture, and different pools of soil organic C and N following standard protocols. Soil BD was determined by core sampler (with known value) method [15]. Soil texture, pH, EC, and CEC were measured by Jackson's technique [16]. The rapid titration technique was used to examine calcium carbonate (CaCO3) [17]. Wet oxidation method was used to determine total organic C (TOC) in soil [18]. By treating the soil with 0.02 M KMnO4, oxidizable carbon (KMnO4-C) was calculated [19]. Particulate organic carbon (POC) was determined following the procedure as outlined by Camberdella and Elliot [20]. The difference between TOC and POC was used to determine mineral-associated organic carbon (MOC). Wet oxidation was used to estimate the oxidizable organic C (OOC) content of soil [21]. For the estimation of very labile C (VLC), labile C (LC), less labile C (LLC), and non-labile C (NLC), the modified Walkley and Black technique was used [22] with different concentrations of H2SO4 (5, 10, and 20 mL of concentrated (36.0 N) H2SO4 in the ratios of 0.5:1, 1:1, and 2:1). The amount of TN in the soil was assessed by digesting it with concentrated H2SO4 [23]. Keeney and Nelson's approach for determining inorganic N (NH4 +-N and NO3 —-N) was followed [24]. Organic N (Org-N) was calculated by deducting inorganic N from TN. The mineralizable N (labile N) was determined by the alkaline potassium permanganate (KMnO4-N) method [25].

#### *2.4. Soil Quality Indices*

Carbon management index (CMI) and nitrogen management index (NMI) were derived using the dynamics of SOC and N. The reference was an uncultivated soil near the experimental field; CMI and NMI were both set to 100.

CMI was calculated using the Blair et al. [7] mathematical methodologies, which are detailed below:

$$\text{CMI} = \text{CPI} \times \text{LIC} \times 100 \tag{1}$$

CPI is for C pool index, while LIC stands for C lability index. The following are the formulas for calculating the CPI and LIC:

$$\text{Carbon Pool Index (CPI)} = \frac{\text{Total C in sample (mg g}^{-1})}{\text{Total C in reference soil (mg g}^{-1})} \tag{2}$$

$$\text{Labbility Index of C (LIC)} = \frac{\text{Labbility of C in sample soil}}{\text{Labbility of C in reference soil}} \tag{3}$$

$$\text{Labbility of C (LC)} = \frac{\text{C in fraction oxidized by KMnO4} \left(\text{mg labile C g}^{-1} \text{ soil}\right)}{\text{C remaining unoccupiedized by KMnO4} \left(\text{mg labile C g}^{-1} \text{ soil}\right)} \tag{4}$$

The NMI was estimated using the techniques described by Gong et al. [26], which are identical to CMI [7]:

$$\text{NMI} = \text{NPI} \times \text{LIN} \times 100 \tag{5}$$

NPI stands for N pool index, while LIN stands for N lability index. The NPI and LIN are calculated using the following method:

$$\text{Nitrogen Pool Index (NPI)} = \frac{\text{Total N in sample (mg g}^{-1})}{\text{Total N in reference soil (mg g}^{-1})} \tag{6}$$

$$\text{Lability Index of N (LIN)} = \frac{\text{Lability of N in sample soil}}{\text{Lability of N in reference soil}} \tag{7}$$

$$\text{N in fraction oxidized by KMno} \\ \text{4 } \left( \text{mg labile N g}^{-1} \text{soil} \right)$$

$$\text{Labbility of N (LN)} = \frac{\text{N in fraction oxidized by KMno} \text{M} \left(\text{mg labile N g}^{-1} \text{soil}\right)}{\text{N remaining unoxidized by KMno} \left(\text{mg labile N g}^{-1} \text{soil}\right)} \quad \text{(8)}$$

C/N, POC/TOC, OOC/LBN, TOC/clay, C stratification ratio (CSR), and N stratification ratio (NSR) have all been proven to be good indicators for assessing soil quality (SQI) [27]. The ratio of TOC concentration to the TN concentration gave soil C/N ratio, and the other indices were derived by considering the same criteria. CSR and NSR were determined by comparing parameter values in the surface soil (0–20 cm) to those at a deeper depth [27,28].

#### *2.5. Carbon and Nitrogen Stock*

The SOC and N stock was calculated by multiplying their respective TOC and TN value with BD and depth of soil as:

$$\text{T SOC stock (Mg ha}^{-1}\text{)} = \text{TOC (or TN)} \text{ (g kg}^{-1}\text{)} \times \text{BD (Mg m}^{-3}\text{)} \times \text{Depth (m)} \times 10 \tag{9}$$

#### *2.6. Statistical Analysis*

Duncan's multiple range test (DMRT) at *p* < 0.05 was performed to find out specific differences between means of different soil depths as well as land use systems. Pearson's correlation matrix was used to assess the link between distinct pools of organic C and N and soil characteristics. A principal component analysis (PCA) was used to summarize the entire variance of the data for the examined depth (0–100 cm) data utilizing land use systems, which included all fractions of soil organic carbon and nitrogen as well as soil quality indicators (SQI). All these statistical analyses were performed by the R software version 3.6.2 [29]. The *prncomp*() function and *ggplot2* package of R were used for principal component analysis and graph preparation, respectively.

#### **3. Results**

#### *3.1. Effects of Land Use Systems on Soil Properties*

The mean BD varied from 1.47 (forage crop) to 1.52 Mg m−<sup>3</sup> (agroforestry) (Table 2). The mean BD, on the other hand, was not significantly altered depending on the land use scheme. The soil EC varied from 0.18 to 0.51 dS m−<sup>1</sup> across the land use systems, with forage crop soils having considerably lower soil EC (*p* < 0.05) than the other land use systems. However, there was no substantial change in soil pH and EC across soil depths within the land use system. The pattern of CEC became more uneven as soil depth increased across all land uses. The agroforestry system showed higher total CaCO3 compared to all land uses. CaCO3 concentration rose by 57.8%, 72.8%, 16.6%, and 1.11% in 0–20 cm soil depth in fodder crop, rice–wheat, citrus orchard, and agroforestry, respectively, over uncultivated soil. With respect to particle size fractions, i.e., sand and silt contents, which varied from 23.0 to 37.65% and 34.19 to 47.74%, respectively, soils under diverse land uses did not differ substantially (*p* < 0.05). The clay content ranged from 22.69 to 39.64% across various land uses and the mean clay content of the various land use systems did not differ much. However, with increasing depth there were significant changes in clay contents in forest and rice–wheat land use systems.


**Table 2.** In a hot, arid climate, the depth-wise distribution of soil attributes as influenced by various land use systems.

According to Duncan's multiple range test, values with different lower case (a–d) and upper case (A–D) superscript letters are significantly different (*p* < 0.05) between land uses for each soil depth and between soil depths for each land use, respectively, while mean values in a column with different lower case letters (w–z) are significantly different (*p* < 0.05). BD, bulk density; EC, electrical conductivity; CEC, cation exchange capacity.

#### *3.2. Effects of Land Use on TOC, POC, MOC, and KMnO4-C*

Although the content of TOC, POC, MOC, and KMnO4-C varied greatly amongst land uses, their order of magnitude remained stable throughout different depths (Table 3). The average TOCs for various land uses were varied in the order of forage crop (7.20 g kg−1) > rice–wheat (4.70 g kg−1) > citrus orchard (4.11 g kg−1) > agroforestry (3.54 g kg−1) > uncultivated (3.10 g kg<sup>−</sup>1). It was observed that different land uses significantly affected the MOC fraction. In uncultivated, agroforestry, citrus orchard, rice–wheat, and fodder crops, MOC varied from 1.10 to 2.81, 0.97 to 3.79, 1.76 to 3.67, 0.94 to 5.51, and 2.92 to 7.38 g kg<sup>−</sup>1, respectively, along the depth. In comparison to uncultivated land, the KMnO4 −C rose by 31.7 to 104.8% in various cultivated land uses.


**Table 3.** Depth-wise distribution of total organic carbon (TOC), particulate organic carbon (POC), mineral-associated organic carbon (MOC), and KMnO4 oxidizable carbon (KmnO4-C) as affected by different land use systems in a hot, arid environment.

According to Duncan's multiple range test, values with different lower case (a–d) and upper case (A–D) superscript letters are significantly different (*p* < 0.05) between land use for each soil depth and between soil depths for each land use, respectively, while mean values in a column with different lower-case letters (w–z) are significantly different (*p* < 0.05).

#### *3.3. Effects of Land Use on OOC and its Fractions*

The OOC and its fractions are extensively used in several agricultural sustainability or environmental quality monitoring programs. In the forage crop, rice–wheat, citrus orchard, and agroforestry systems, OOC buildup was 7.29, 5.95, 5.17, and 4.34 g kg−1, respectively, compared to 2.95 g kg−<sup>1</sup> in uncultivated soil (0–20 cm depth) (Table 4). The increases in OOC under the forage crop and rice–wheat was 116% and 74% greater over the uncultivated soil. The magnitude OOC under a gradient of oxidizing environments was as follows: under all land uses, NLC > LLC > LC > VLC. VLC concentrations in diverse land uses ranged from 0.19 to 1.27 g kg−<sup>1</sup> along the soil profile up to a depth of 100 cm. The LC and LLC concentrations of various land uses ranged from 0.46 to 2.60 g kg−<sup>1</sup> and 0.43 to 3.41 g kg−1, respectively. NLC concentration was found to be maximum (3.78 g kg−1) in 0–20 cm of the forage crop and minimum (0.92 g kg−1) in 80–100 cm of uncultivated land. In all land uses, the share of passive carbon pools (LLC and NLC) was higher than the active carbon pools (VLC and LC). There was no significant difference in OOC fractions with depth in agroforestry and uncultivated land.


**Table 4.** In a hot, arid environment, the depth-wise distribution of oxidizable organic C (OOC) and its fractions as impacted by different land use systems.

According to Duncan's multiple range test, values with different lower case (a–d) and upper case (A–D) superscript letters are significantly different (*p* < 0.05) between land use for each soil depth and between soil depths for each land use, respectively, while mean values in a column with different lower-case letters (w–z) are significantly different (*p* < 0.05). OOC, oxidizable organic C; VLC, very labile C; LC, labile C; LLC, less labile C; NLC, non-labile C; AC, active C; PC, passive C.

#### *3.4. Effects of Land Use on TN and its Fraction*

Higher accumulation of TN, Org-N, and KMnO4-N in the surface layers observed under all the land uses and different pools of lability showed a decreasing trend with increasing soil depth (Table 5). With respect to concentration of C fractions, the distribution of N fractions throughout depth in each land use followed a decreasing pattern. Average TN content followed the order: forage crop (488 mg kg−1) *>* rice–wheat (323 mg kg−1) *>* citrus orchard (316 mg kg−1) *>* agroforestry (244 mg kg−1) *>* uncultivated (254 mg kg−1). However, a similar pattern was observed in the distribution of TN and Org-N contents with respect to organic carbon distribution and was comparatively higher in the forage crop followed by the rice–wheat system than in other land uses. Significantly higher mean KMnO4-Nwas maintained up to 100 cm soil depth in the forage crop (50.1 mg kg−1) over the rice–wheat (47.9 mg kg−1) and uncultivated land (28.3 mg kg−1). All the land uses showed higher accumulation of mineral N in the 0–20 cm soil depth and then decreased with depth.


**Table 5.** Depth-wise distribution of total N and its fractions as impacted by different land use systems.

According to Duncan's multiple range test, values with different lower case (a–d) and upper case (A–D) superscript letters are significantly different (*p* < 0.05) between landscape for each soil layer and between soil layers for each land use, respectively, while mean values in a column with different lower-case letters (w–z) are significantly different (*p* < 0.05). TN, total N; KMnO4-N, KMnO4oxidizableN; Org-N, organic N; NH4-N, ammoniacal N; NO3-N, nitrate N.

#### *3.5. Carbon and Nitrogen Stock*

The SOC stock distribution revealed a diminishing trend with depth across all land uses (Figure 2). The forage crop showed a maximum SOC stock (26.36 Mg ha<sup>−</sup>1) at 0–20 cm soil depth. It was found to be highest in the rice–wheat system at soil depths of 20–40 and 40–60 cm, with 18.01 Mg ha−<sup>1</sup> and 12.59 Mg ha<sup>−</sup>1, respectively. SOC stock in the soil profile up to 100 cm depth was highest in fodder crops (52.74 Mg ha<sup>−</sup>1) and lowest in uncultivated land (22.92 Mg ha<sup>−</sup>1). TN stock followed a similar pattern as SOC stock across various land uses and depths. When compared to rice–wheat, citrus orchard, and agroforestry land use systems, forage crop land use systems had considerably larger TN stock (1.42 Mg ha−1) at the 0–20 cm depth. Rice–wheat (1.13 Mg ha−1) TN stock was also different from citrus orchard and agroforestry systems. Only in the 0–20 cm depth, the difference in TN stock between the fodder crop (0.78 Mg ha−1) and rice–wheat (0.25 Mg ha−1) was significant (*p* < 0.05). The TN stock did not differ between land use systems at 60–80 and 80–100 cm depths. All land uses showed no significant changes in N stock in the bottom soil layer.

**Figure 2.** (**a**,**b**) Soil organic carbon and nitrogen stock at different soil depths in different land use systems in hot, arid environment. UC, uncultivated; AF, agroforestry; CO, citrus orchard; RW, rice–wheat system; FC, forage crop.

#### *3.6. Relationship with Soil Properties and Pools of Soil C and N*

KMnO4-C displayed a negative and substantial connection with BD (r = −0.447b), pH (r = −0.691b), CaCO3 (r = −0.396b), and silt fractions (r = −0.290a) according to Pearson's correlation matrix (Table 6). However, it was shown to have a strong and positive relationship with CEC (r = 0.453b). TOC was significantly (*p* < 0.01) and inversely linked with BD (r = −0.421b), pH (r = −0.766b), and CaCO3 (r = −0.364b). The SOC and N fractions had a substantial and positive relationship with the silt fractions. CEC and KMnO4-N were shown to have a substantial correlation (r = 0.504b). BD, pH, CaCO3, and silt fractions all exhibited a negative and substantial relationship with N fractions. SOC and N fractions were not related to EC, sand, and clay.

**Table 6.** Correlation coefficient (r) between soil properties and various organic C and N pools in soils under different land use systems.



**Table 6.** *Cont*.

<sup>a</sup> Correlation is significant at the 0.05 probability level; <sup>b</sup> correlation is significant at the 0.01 probability level.

#### *3.7. Soil Quality Indices*

Under all land uses, the LI and CPI ranged from 0.92 to 1.18 and 1.00 to 2.54, respectively (Table 7). The ranking of the mean CMI under various land uses was as follows: forage crop (299) > rice–wheat (251) > citrus orchard (220) > agroforestry (169) > uncultivated land (147). Simple linear regression analysis revealed that OOC, KMnO4-C, VLC, and POC have strong linear correlations with CMI (Figure 3). In the 0–20 cm depth, a higher regression coefficient was found between CMI with KMnO4-C (R2 = 0.94) followed by OOC (R2 = 0.85), VLC (R<sup>2</sup> = 0.75), and AC (R2 = 0.73). Lower soil depths had a lower regression coefficient. Rice–wheat and agroforestry had much lower CMI than forage crop systems in this study. Because rice–wheat systems exhibited considerably lower rates of soil C rehabilitation than forage systems, these data suggest that forage systems provide better choices for C sequestration in soils in arid ecosystems than rice–wheat systems. The highest value of NMI was observed in rice–wheat followed by forage crop, agroforestry, and citrus orchard. The impacts of land use on soil NPI and NMI followed the same pattern as soil TN. The reference (uncultivated) value is 100. Values below 100 suggest that the system is deteriorating, while values over 100 show that the system is improving in terms of N. The highest NMI values were obtained in the forage crop land use (205 at 0–20 cm, 165 at 20–40 cm, 138 at 40–60 cm, 158 at 60–80 cm, and 157 at 80–100 cm soil depth). The correlation between NMI and KMnO4-N (R2 = 0.89) was stronger than the correlation between NMI and mineral N (R<sup>2</sup> = 0.60) (Figure 4). However, significantly higher NMI values were obtained from the continuous agricultural intensification compared to uncultivated soil.

**Table 7.** In a hot, arid environment, depth-wise distribution of carbon and nitrogen management indices as influenced by different land use systems.



**Table 7.** *Cont*.

According to Duncan's multiple range test, values with different lower case (a–d) and upper case (A–D) superscript letters are significantly different (*p* < 0.05) between land use for each soil depth and between soil depths for each land use, respectively, while mean values in a column with different lower case letters (w–z) are significantly different (*p* < 0.05). CPI, carbon pool index; LIC, lability index of carbon; CMI, carbon management index; NPI, nitrogen pool index; LIN, lability index of nitrogen; NMI, nitrogen management index.

The C/N ratio is a nutrient mineralization and immobilization indicator; a lower C/N ratio (<15:1) implies a higher mineralization rate. In the top 0–20 cm depth, the forage crop and rice–wheat systems showed significantly higher C/N ratios as compared to other land uses. In most land uses, C/N ratios declined from 0–20 cm to 20–40 cm depth, except for fodder crops, which exhibited a minor rise (Figure 5). Moreover, the C/N ratio in forage crops was considerably greater (*p* < 0.05) than in rice–wheat and agroforestry systems below 40 cm depth.

The C/N ratio in the research region was found to be greater above the standard range of 10:1 predicted in mineral soils. On the other hand, POC/TOC, OOC/LBN, and TOC/clay ratio showed differences between land use systems, with the highest values in the forage crop. Average CSR and NSR in the different land uses decreased in the following order: rice–wheat *>* forage crop *>* citrus orchard *>* agroforestry *>* uncultivated (Figure 6). As a result, the stratification ratio of C and N at lower depths was larger than in the top layers.

PCA is a more precise data selection approach of which variables or indices were more influential in differentiating land uses from the combined 0–100 cm data. The dimensionality of the data set in a PCA was defined by correlations and scatter plot matrices between variables, which selected variable candidates that may explain the variance in sensitivity indices for various fractions with respective pool sizes. The first two principal components (PCs) of the data set explained 83.9% and 8.32% of total variance, respectively (Table 8). The highly weighted variable in PC1 included TOC, OOC, POC, AC, TN, profile C, and N stock. In the PC2, variables of NLC, NH4-N, and NO3-N were found highly weighted. Regarding SQI, the PCA allowed a clearer differentiation of the land uses. The PC1 explained 32.0% of the variance where CMI, NPI, CPI, and CSR presented a positive and significant association (Table 9). The second PC explained 19.2% of the variance, where LIN, LIC, and POC/TOC ratio exhibited positive and significant associations in that component. Therefore, considering the mean value of SQI, it can be assumed that CMI, NPI, CPI, CSR, LIN, LIC, and POC/TOC ratio were the most sensitive indices for segregating land uses.

**Figure 3.** Relationship between carbon management index (CMI) with (**a**) oxidizable organic carbon (OOC), (**b**) KMnO4 oxidizable organic carbon (KMnO4-C), (**c**) very labile carbon (VLC), and (**d**) active carbon (AC) at different soil depths.

**Figure 4.** Relationship between nitrogen management index (NMI) with (**a**) KMnO4 oxidizable organic nitrogen (KMnO4-N) and (**b**) mineral N (NH4-N + NO3-N) at different soil depths.

**Figure 5.** Indicators of soil organic carbon and nitrogen (**a**) C/N ratio, (**b**) POC/TOC ratio, (**c**) OOC/LBN ratio and (**d**) TOC/Clay ratio at various soil depths in various land use systems. UC, uncultivated; AF, agroforestry; CO, citrus orchard; RW, rice–wheat system; FC, forage crop.

**Figure 6.** (**a**) Carbon stratification ratio (CSR) and (**b**) nitrogen stratification ratio (NSR) at different soil depths in different land use systems. UC, uncultivated; AF, agroforestry; CO, citrus orchard; RW, rice–wheat system; FC, forage crop.


**Table 8.** Principal component (PC) study of soil organic carbon and nitrogen pools in a hot, arid environment under various land use systems.


**Table 9.** In a hot, arid environment, principal component analysis of soil organic carbon and nitrogen indices under various land use systems.

The loading of each variable (arrows) and the scores of each land use (points) are shown in the PCA bi-plot (Figure 7). The length of the arrows and angle between them (cosine) approximates the variance and their correlations, respectively. The bi-plot between PC1 and PC2 has four quadrants. Our objective here is to establish some relation between the land use systems in different quadrants with the SOC and N fractions and their indices. The bi-plot showed an overlapping pattern while considering individual scores of each land use. For the TOC, NLC, MOC, PC, and TN, the forage crop was somewhat tilted to the right along the PC1 axis. Along the PC2 axis, rice–wheat scores were considerably biased toward greater negative values. The rice–wheat scores were clearly more impacted toward more positive values along the PC1 axis for the CMI, NPI, CPI, CSR, LIN, LIC, and POC/TOC ratios, according to SQI in the bi-plot.

**Figure 7.** Principal component analysis (PCA) bi-plot for all land use systems involving soil organic carbon (SOC) and N fractions (**a**) and indices (**b**).

#### **4. Discussion**

The soil characteristics along with C and N fractions varied greatly depending on the land use, but the order of magnitude remained similar throughout the depths. The difference in BD with soil depth was found to be substantial, with the lower depth layer having a greater BD than the topsoil layer, because of the overlying soil's weight, which produces compaction and a decrease in SOM content [11]. In all land uses, the pH and EC patterns were more erratic as depth increased. The impacts of land use on soil pH were not significant. In lower depths, there was no influence on EC. However, in the rice–wheat combination, a significant drop in EC was noted, which could be ascribed to the use of an irrigation source to leach off soluble salt [30]. Although there was an increase in clay and silt in the subsurface layers, along with a decline in sand content, the soils were primarily sandy [31]. Long-term irrigation under rice–wheat systems may have resulted in increased fine soil particles due to sediment movement by the canal [1,32].

The current study found that cultivating desert soil for 60 years enhanced TOC and its fractions under a variety of land uses. Due to the minimal vegetation found in desert soils, organic matter input into the soil is limited. However, differing land uses and soil layers were found to have a considerable impact on the KMnO4-C fraction [11,33]. In the surface depth of the forage crop, TOC and its fraction were much higher than in the lower depths. Overall, all land use systems and soil management approaches resulted in higher organic C buildup than uncultivated land. Land use changes can have a significant influence on SOC dynamics and carbon transport [34]. High TOC might be linked to high vegetative growth, fast root proliferation, organic matter breakdown, and subsequent organic matter retention in soil aggregates owing to clay complexes, as seen by the abundance of fine soil particles. The development of clay–organic complexes and soil aggregates in the arid region was likely facilitated by soil moisture resulting from alternating wet and dry conditions, accumulating the greatest amount of SOC. The decrease in TOC on uncultivated land is due to a drop in organic matter input and oxidation of SOC because of exposing soils to the blazing sun [35].

In terms of turnover time, the particulate organic matter pool is halfway between the active and passive organic matter pools (i.e., a slow pool) [20]. The primary sources of POC in this study were leftover root biomass, agricultural residues, leaf litter, and increased microbial biomass and plant debris. The various land uses investigated had a significant impact on the POC values. The high results under land uses were consistent with the findings of Kalambukattu et al. [36] that changes in land uses can lead to particle organic matter buildup. POC accounted for 37.7% (uncultivated) to 42% (citrus orchard) of the TOC across all land uses. In dry or cold climates, the POC reported a 50% greater level of SOC [37]. The lower POC to TOC ratios in our samples are most likely owing to the hot, dry environment, which favors biological decomposition of recent organic material inputs, resulting in less POC buildup [2]. The findings of Camberdella and Elliott [20] and Six et al. [38] demonstrated that soil disturbances such as tillage can lower POC levels.

Both OOC and TOC decreased with depth in all the land uses studied, probably due to a decrease in surface litter intake in lower soil layers [33,39,40]. These results are similar to those reported by Moharana et al. [30] for rice-based cropping systems in India's hot, dry region, where long-term farming increased the labile and recalcitrant fractions (LLC and NLC). Changes in land use were also particularly sensitive to the VLC and LC fractions of SOC [22]. This showed that monitoring the efficacy of various land uses in sustaining active C pools, which play a larger role in nitrogen cycling, is crucial. After 60 cm of soil depth, no significant difference in MOC and KMnO4-C concentrations was observed across all land uses. These findings corroborated those of Lal [41] and Gelaw et al. [42], who found that grazing field soils have greater SOC stock than agricultural soils due to more root biomass and residue returning to the surface.

Below 40 cm deep, a significant fall in the level of N fractions was seen for all land uses. The higher TN in soil cultivated with the forage crop might be attributed to the higher organic carbon, which came from the return of plant and root biomass as well as residues to the soil system [42,43]. Because of changes in SOM content and cultivation, Moharana et al. [2] found a substantial difference in KMnO4-N between barren and cultivated land. Mineral N concentrations in rice–wheat were similarly greater than in the citrus orchard and uncultivated land, showing that a higher rate of mineral fertilizer application in the rice–wheat system might boost N concentrations. Surface soil had higher KMnO4-N levels than subsurface soil, regardless of land use. This might be linked to the breakdown of root biomass in the surface layer, which releases nitrogen when organic matter is mineralized, re-leaving available nitrogen.

Despite the fact that pool sizes varied greatly among land use regimes, sensitivity indices for various fractions demonstrated that their susceptibility to change was comparable to total pools [26]. Due to different land use changes, no single pool could be employed as a sensitive indicator for SOC and N changes. VLC, LC, CMI, NPI, CPI, CSR, LIN, LIC, and POC/TOC ratio could be used as sensitive C and N indicators. The VLC was shown to be substantially more sensitive to management than the TOC. The LLC fraction, on the other hand, was far less affected by changes in land use than the TOC fraction. LBN (KMnO4-N) has a lower sensitivity than Org-N and TN, implying that it is ineffective as a sensitive indicator of land use changes. Westerhof et al. [44] indicated that the NMI was an excellent indication of N availability but not of total N. This was most likely owing to tillage's fast mineralization of labile organic materials. Labile N by KMnO4 is a quick and easy approach to assess the nitrogen status in soils.

#### **5. Conclusions**

Influence of land use and soil depth on variations in soil C and N fractions was investigated under arid conditions in India. The VLC, CMI, and NMI, among other soil quality indices, changed dramatically with land use. The VLC was substantially more responsive to changes in land use than the TOC. Forage crop and rice–wheat soils had greater TOC and TN than uncultivated soils, showing a large potential for adopting these methods to adsorb SOC and TN in these soils. The top 0–20 cm of the forage crop contained the majority of the SOC and TN. The sensitivity indices can be used to assess their utility and detect changes in SOC and N fractions caused by land use changes. NMI demonstrated to be a valuable indicator for analyzing changes in soil quality induced by rice–wheat land use because of the significant correlations between NMI and the OOC and N fractions. The study found that anthropogenic modifications of desert soils by changing to various land uses resulted in considerable improvements in C and N stock. In the arid region, therefore, integrating appropriate forage crops and agroforestry trees into agricultural fields and adopting restorative land uses can greatly influence the sequestration of both SOC and TN. Among the various land uses, forage crops, which have a larger biomass, have a higher TOC and CMI, and are considered the optimal systems for maintaining soil health in desert soil of India. The findings are particularly unique and useful for researchers, planners, and policymakers in desert ecosystems; nevertheless, such research can be improved in the future by considering climate, management, and socioeconomic factors of the region.

**Author Contributions:** Conceptualization, P.C.M.; writing—original draft preparation, P.C.M., R.K.J., G.K.S. and S.S.; supervision, P.C.M.; writing—review and editing, P.K.J., K.A., P.V.V.P., P.C.M., R.L.M. and M.N.; visualization, P.C.M. and P.K.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** The ICAR-National Bureau of Soil Survey and Land Use Planning, Nagpur, India financially supported the study.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** All data are included in the manuscript, and additional data are available with the corresponding author and will be available upon request.

**Acknowledgments:** For enabling this research to be completed successfully, the authors are very thankful to the Director of the ICAR-National Bureau of Soil Survey and Land Use Planning (NBSS and LUP), Nagpur, and to the Head, NBSS and LUP, Regional Center, Udaipur.

**Conflicts of Interest:** The authors declare that they do not have any competing or conflict of interest.

#### **References**


### *Article* **Quantitative Estimation of Saline-Soil Amelioration Using Remote-Sensing Indices in Arid Land for Better Management**

**Hesham M. Aboelsoud 1, Mohamed A. E. AbdelRahman 2,\*, Ahmed M. S. Kheir 1,3, Mona S. M. Eid 1, Khalil A. Ammar 3, Tamer H. Khalifa <sup>1</sup> and Antonio Scopa <sup>4</sup>**


**Abstract:** Soil salinity and sodicity are significant issues worldwide. In particular, they represent the most dominant types of degraded lands, especially in arid and semi-arid regions with minimal rainfall. Furthermore, in these areas, human activities mainly contribute to increasing the degree of soil salinity, especially in dry areas. This study developed a model for mapping soil salinity and sodicity using remote sensing and geographic information systems (GIS). It also provided salinity management techniques (leaching and gypsum requirements) to ameliorate soil and improve crop productivity. The model results showed a high correlation between the soil electrical conductivity (ECe) and remote-sensing spectral indices SIA, SI3, VSSI, and SI9 (R2 = 0.90, 0.89, 0.87, and 0.83), respectively. In contrast, it showed a low correlation between ECe and SI5 (R<sup>2</sup> = 0.21). The saltaffected soils in the study area cover about 56% of cultivated land, of which the spatial distribution of different soil salinity levels ranged from low soil salinity of 44% of the salinized cultivated land, moderate soil salinity of 27% of salinized cultivated land, high soil salinity of 29% of the salinized cultivated land, and extreme soil salinity of 1% of the salinized cultivated land. The leaching water requirement (LR) depths ranged from 0.1 to 0.30 m ha<sup>−</sup>1, while the gypsum requirement (GR) ranged from 0.1 to 9 ton ha<sup>−</sup>1.

**Keywords:** soil salinity; sodicity; GIS; RS; leaching and gypsum requirement

#### **1. Introduction**

Land degradation is one of the world's most severe environmental and socio-economic issues [1–3], occurring due to natural phenomena and anthropogenic factors that negatively impact land's ability to function effectively in an ecosystem which causes enormous challenges in achieving sustainable development goals [4–7]. Degraded lands could reach one-fifth of the total land in some countries [8,9]. Currently, salt-affected soil covers approximately 1.125 billion hectares, with anthropogenic activities affecting 76 million hectares. Soil salinity is a primary challenge to global food security and environmental sustainability. As climate change accelerates, the problem may soon spread to unaffected areas [10]. The high salinity levels could cause adverse effects on soil characteristics and plant physiology [11,12].

There is an urgent need to increase the area of agricultural land to meet the increasing demand for food due to the rapid population growth [13–15]. Therefore, one of the effective ways to raise the efficiency of the agricultural unit is land reclamation processes [16]. This helps in fixing one or more defects in the soils that hinder and/or reduce agriculture

**Citation:** Aboelsoud, H.M.; AbdelRahman, M.A.E.; Kheir, A.M.S.; Eid, M.S.M.; Ammar, K.A.; Khalifa, T.H.; Scopa, A. Quantitative Estimation of Saline-Soil Amelioration Using Remote-Sensing Indices in Arid Land for Better Management. *Land* **2022**, *11*, 1041. https://doi.org/10.3390/ land11071041

Academic Editors: Jinyan Zhan, Xinqi Zheng, Shaikh Shamim Hasan and Wei Cheng

Received: 31 May 2022 Accepted: 6 July 2022 Published: 8 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

productivity [17]. However, the traditional methods for reclamation operations are costly in time and effort [16]; therefore, looking to modern technologies to help in these calculations has become an urgent necessity.

Remote sensing and geographical information systems (GIS) are promising tools for assessing land degradation [18–22]. These tools can generate relevant maps and reliable spatial information to support decision making [23–25]. The earliest successful attempts to use remote sensing for the detection of salt-affected soils were preceded by Mougenot et al. [26]. The remote sensing and GIS datasets provide accurate information on large areas. Some satellite 'images' are low in cost, and the remote-sensing assessment tasks can be carried out in a shorter time than conventional fieldwork assessments [27–30]. The assessment results can be used to adequately manage soil and crops [31–34]. Using Landsat images allows for assessing the soil salinity features. The Landsat images are the best to capture soil salinity extent with different salinity levels [35–41]. In Egypt, several studies have also been conducted to map soil salinity using remote sensing and GIS datasets and showed reliable soil salinity results [42–57].

The prevalence of saline/sodic soils depends on two types of factors, namely, climatic factors and geomorphologic factors. Saline lands are found in regions with a continental climate or where droughts prevail, which leads to increased evaporation and salt accumulation [58–60]. Saline/sodic soils also spread in the lands of lakes, rivers, and sedimentary valleys, and abound in dry and semi-arid areas with little rain and high temperatures. All conditions are identical in the study area where the evaporation process accelerates the formation of salts and their ascent to the surface of the soil through its capillary property [31,32,61,62]. Saline soils have higher salt concentrations than usual, whereas sodic soils have higher concentrations of Na<sup>+</sup> than usual. Saline soils cause a chemical drought while sodic soils, conversely, cause waterlogging in soils [33]. Leaching is a vital soil management technique applied to salt-affected soil by adding supplemental irrigation water to remove salts from the root zone layers [26–28]. Understanding the hydraulic properties of the soil, water mobility, and salt dynamics are essential to correctly conduct the required leaching [63–65], while adding gypsum (CaSO4 × 2 H2O) to salt-affected soil is one of the oldest amendment methods. This method promotes the efficient replacement of Na+ by exchangeable Ca2+ leading to the improvement in the soil's physical–chemical and enzymatic properties [66–73].

In Egypt, salinity accumulation, sodicity, and waterlogging are the main form of land degradation. Soil salinity and sodicity seriously affect agriculture production, where saline/sodic soils occupy 46% of the total Nile Delta area [74]. Egypt's croplands are entirely irrigated due to the country's extremely low rainfall and high rates of evaporation. The primary cause of secondary soil salinization in Egypt is the extensive irrigation of agriculture in arid climate conditions [74,75]. Additionally, irrigation with contaminated water from the polluted drains led to increasing in some metals' concentration due to anthropogenic pollution through the spreading of contaminated dredged materials on agricultural fields [76]. The soil salinization problem in Egypt, caused by the reuse of irrigation drainage waters and limits on rice plantings due to the shortage of irrigation water, raises an urgent need for the agricultural productivity of the Nile Delta through, for instance, subsurface drainage in waterlogged lands, land leveling, and use of gypsum amendments [74]. Especially in Egypt, at least 20% of all irrigated areas are salt-affected, and other estimates put the figure as high as 50%. The northern part of the Nile Delta in Egypt contains a huge region of heavy clay soils with shallow open drainage, limited permeability, and low productivity.

In the North Delta region, particularly in Egypt's Kafr El-Sheikh Governorate, there is a serious lack of irrigation water supply. Farmlands at the end of irrigation canals must use available drainage water to compensate for the lack of source of water [76–79]. Therefore, this study aims to integrate remote sensing and GIS techniques to produce amelioration maps of leaching (LR) and gypsum (GR) requirements in the study area.

#### **2. Materials and Methods**

#### *2.1. Description of the Studied Area*

Location: The study area covers 373,191 km2, representing 28.1% of the total area of the Delta region and about 0.35% of the total area of Egypt. It is located in the Kafr El-Sheikh Governorate in the northern part of Egypt's Nile Delta. The latitude ranges from 31◦00 and 31◦ 15 in the east and 31◦00 and 31◦37 in the north, and an altitude of 9.14 m above sea level. It is bounded in the north by the Mediterranean Sea, the southern Gharbia Governorate, the eastern part of Dakahlia Governorate, and the western province of Bihaira. (Figure 1). In the north part of the area, Lake Burullus is located within the borders of the governorate with an area of 148,562 hectares. The lake is connected to the Mediterranean through the Burullus spur, which is 44m-wide.

**Figure 1.** Location map of the study area.

Landform: The region's topography has a diversity of natural life due to the diversity of environments and the diversity of the topography of the land. The natural environments in the province can be classified into three main types: agricultural and urban environments, coastal environments, and wetlands. Each of them is unique in its animal life, plants, and biodiversity [80,81].

Geology: The center and south of the province cover the sediments of the modern geological age (Holocene era), which are dark-brown formations composed of deposits of clay, clay, and sandy clay. These sediments are deposited over the ancient marine sediments (under delta formations) that date back to the Pleistocene era. They are yellow in color and consist of coarse and fine sand and pebbles consisting of quartz or igneous and metamorphic rocks. The northern coastal zone is a low, sandy coast consisting of soft, brittle sediments belonging to the Pleistocene and Holocene [82,83].

All the ancient geological studies show that the Burullus region in the north of the governorate was less arid than our present era with the presence of many plants. Likewise, the shore of the delta region was mainly composed of silt, with swamps and depressions increasing in it. In the flood season, these depressions were filled with fresh water, forming

a series of small lakes and wetlands. These bogs were filled with organic matter and sediments resulting from the analysis of plant remains, so most of this water was devoid of oxygen. It was also filled with the shells of some bivalve mollusks, especially the cardium type. The coastal area consists of a sandy beach as a result of the sediments that were carried by the waves of the Mediterranean [82,83].

Climate: In general, the climate in Kafr el-Sheikh Governorate is an arid climate (classified as BWh) by the Köppen–Geiger system. The warmest month is August (31 ◦C), and the coldest month is January (9.4 ◦C). The total number of rainy days in a year is 31 days, where January is the wettest month while July is the driest month (0.0 mm/0 inches).

A climate diagram (Figure 2) is based on 30 years of available data from the study area. From this indication of typical climate patterns and conditions of temperature and precipitation, it is clear from the figure that the rain is very little, at less than 25 mm, and most of it occurs in January. Generally, the governorate has a Mediterranean climate, and the temperature varies between 13.2 ◦C in January (winter) and 26.6 ◦C in July (summer). The amount of precipitation ranges from 140 mm to 250 mm per year. The winds are generally western and northwest.

**Figure 2.** Climate parameters of the studied area.

Agricultural land and irrigation water: Soil texture in the study area is classified from heavy clay to sandy soil. Surface irrigation systems use the Nile water or drainage reuse, and it has electrical conductivity values between 0.31 and 1.86 dS m<sup>−</sup>1.

The soil texture classes of the researched area differ between sandy and heavy clay, according to field surveys and laboratory investigations. Cation-exchangeable capacity (CEC) was strongly associated with clay content and ranged from 7.36 to 44.87 cmolc kg<sup>−</sup>1. These soils ranged from being non-saline to being extremely salty according to the salinity levels, which ranged from 0.81 to 10.80 dS m−1. ESP and pH values ranged from 1.02 to 36.20 and 7.83 to 8.81, respectively. The study area's bulk density and soil depth were between (1.11 and 1.63 Mg m<sup>−</sup>3) and (120 to 150 cm), respectively. Organic matter generally is on average 16.2 g kg−1. The high temperature in dry and semi-arid locations, which causes the decomposition of fresh residuals, is to blame for the low value of OM. CaCO3 content is 7.30 g kg−<sup>1</sup> on average.

The area of agricultural land in the governorate represents about 7.5% of the total agricultural area in Egypt [84–86]. Winter rains that fall on the northern coast of the area are unreliable. Groundwater cannot be used due to its excessive salinity as a result of seawater intrusion as well as to limit its effects. The efficiency of the applied surface irrigation system does not exceed 60% [84]. Recently, the governorate's share of fresh water available for agriculture was about 3.15 billion cubic meters, and agricultural drainage water is mixed with freshwater canals to meet agricultural water requirements [85]. Drainage water with low-quality water such as wastewater is used for irrigation [86], especially in the areas located at the end of irrigation networks that receive inadequate fresh water. There are some main drains in the area such as West El-Burullus, Gharbia, El-Khashaah, Tirrah, and El-Hoks [87]. The principal pollutants are biological oxygen demand (BOD), chemical oxygen demand (COD), and NO3-N. While NH4-N and NO3-N values fall within the normal range for irrigation, they are in the abnormal range according to Egyptian standards. BOD and COD values are rated as bad to moderate and moderate. According to Egyptian standards, the values are within the usual range for irrigation. Furthermore, except for Ni, whose readings are within the normal range, the levels of the heavy metals Cu, Mn, Pb, and Cd are higher than what is permitted for irrigation. B values in water samples range from poor to excellent. In the meantime, irrigation-appropriate pH values ranged from 7.33 to 8.15, EC values ranged from 1.87 to 4.71 ds m<sup>−</sup>1, and SAR values ranged from 5.86 to 9.32.

#### *2.2. Soil Analysis*

Soil sampling was conducted in the study area, where 66 soil samples were collected at 0–0.3 m depths. The samples were dried, grounded, and passed through a 2.0 mm sieve in the laboratory (Table 1). The soil reaction (pH) and soil electrical conductivity (ECe, dS m<sup>−</sup>1) were identified according to the Page method [87]. The soil organic matter (SOM, g kg−1) was determined according to Nelson and Sommers method [88]. The Ca2+ as carbonate (CaCO3, g kg<sup>−</sup>1) was measured volumetrically using a Collins calcimeter method [89]. The exchangeable sodium percentage (ESP) was calculated using the [33] equation:

$$\text{ESP} = \frac{100 \times (-0.0126 + 0.01475 \,\text{SAR})}{1 + (-0.0126 + 0.01475 \,\text{SAR})} \tag{1}$$

where SAR (sodium adsorption ratio) is a measure of the amount of sodium (Na+) relative to calcium (Ca2+) and magnesium (Mg2+) in the water extracted from a saturated soil paste. It is the ratio of the Na concentration divided by the square root of one-half of the Ca + Mg concentration. SAR is calculated from the equation:

$$\text{SAR} = \text{Na}^+ / [(\text{Ca}^{2+} + \text{Mg}^{2+}) / 2]^{0.5}$$

**Table 1.** Basic variables for the sixty-six soil samples' studied soil properties.


Soils that have values for sodium adsorption ratio of 13 or more may have an increased dispersion of organic matter and clay particles, reduced saturated hydraulic conductivity and aeration, and a general degradation of soil structure.

#### *2.3. Soil Amelioration*

Leaching requirements (LR, depth cm) were calculated using the [90] equation:

$$\text{LRR} = \frac{\text{EC}\_{\text{iw}}}{5 \times (\text{EC}\_{\text{e}} - \text{EC}\_{\text{iw}})} \tag{2}$$

where ECiw is the electrical conductivity of the irrigation water and ECe is the soil electrical conductivity.

Gypsum requirements (GR, Mg ha<sup>−</sup>1) were calculated using the [33] equation.

$$\text{GR} = \frac{\text{ESP}\_{\text{i}} - \text{ESP}\_{\text{f}}}{100} \times \text{CEC} \times 4.1 \tag{3}$$

where ESPi: actual ESP of the soil; ESPf: ESP required to be reached by reclamation; and CEC: cation exchange capacity (cmolc kg<sup>−</sup>1).

#### *2.4. Image Preprocessing and Analysis*

#### 2.4.1. Remote-Sensing Data

Remote sensing provides spatial coverage by measuring reflected and emitted electromagnetic radiation from the earth's surface and surrounding atmosphere over a wide range of wavelengths. Remote sensing implies collecting data without making physical contact with the studied object. This study used Landsat 8 (OLI) images (path 177, row 38) in May 2021.

#### 2.4.2. Image Preprocessing

Image distortions and degradations occur during the acquisition process of remotely sensed images. Preprocessing satellite data are required to remove sensor errors during data acquisition and display the correction, band selection, data dimensionality reduction, and computing complexity reduction. The team conducted radiometric to eliminate radiometric problems in images such as nonuniformity, stripe noises, and defective lines, for proper conversion of digital numbers to reflectance values, geometric, and atmospheric corrections on the studied Landsat OLI images to increase the visual distinction between features.

#### 2.4.3. Atmospheric Correction Using FLAASH Tool

The team conducted atmospheric correction using The FLAASH (fast line/of/sight atmospheric analysis of spectral hypercubes) tool in ENVI 5.1 software to have better reflectance. The team used the native file in a BSQ format for the correction and converted the images to BIL and BIP format to be compatible with the FLAASH tool.

#### *2.5. Surface Interpolation Using the Ordinary Kriging Technique*

The team used the interpolation method to determine the spatial variability and pattern of the soil characteristics in two-dimensional soil data sets in the topsoil. The geostatistical analyst extension (Arc GIS 10.4.1) [91] was used to develop the semi-variogram between each pair of points and interpolate between the sampling locations using the kriging method to predict the soil salinity in the study area. Ordinary kriging was used to estimate the value of continuous soil salinity (*z*) at an unsampled location (u) using only data on this characteristic [*z*(uα), α = 1, n] as a linear combination of neighboring observations:

$$\mathcal{Z}\_{\text{ok}}^{\*}\left(\mathbf{U}\right) = \sum\_{\alpha=1}^{n(\mathbf{u})} \lambda\_{\alpha}(\mathbf{u}) \mathcal{Z}\left(\mathbf{u}\_{\alpha}\right) \tag{4}$$

The ordinary kriging weights were chosen to minimize the estimation or error variance,

$$
\sigma \frac{2}{E} (\mathbf{u}) = \text{Var}[Z(\mathbf{u}) - Z(\mathbf{u})] \tag{5}
$$

The weights were obtained by solving a system of linear equations:

$$\begin{cases} \sum\_{\beta=1}^{n(\mathbf{u})} \lambda\_{\beta}(\mathbf{u}) \gamma(\mathbf{u}\_{\infty} - \mathbf{u}\_{\beta}) - \mu(\mathbf{u}) = \gamma(\mathbf{u}\_{\infty} - \mathbf{u})\\ \sum\_{\beta=1}^{n(\mathbf{u})} \lambda\_{\beta}(\mathbf{u}) = 1 \text{ a} = 1, \dots, \text{ n (u)} \end{cases} \tag{6}$$

To ensure the estimator was unbiased, constraining the weights to sum to one requires the definition of the Lagrange parameter m (u).

#### *2.6. Soil Salinity Indices*

The team examined fourteen different spectral salinity indices related to salt detection and soil salinity mapping developed in numerous studies. The most commonly used salinity indices taken into account in this study (NDSI, SIA, SI 1, SI 2, SI 3, SI 4, SI 5, SI 6, SI 7, SI 8, SI 9, NDVI, SAVI, and VSSI) are presented in Table 2.


**Table 2.** Soil salinity indices based on different band ratios of Landsat.

Where: B, G, R, NIR, SWIR1, and SWIR2 refer to the reflectance in visible blue, green, red, near-infrared, Shortwave infrared 1, and 2, respectively.

The processing steps of mapping soil salinity using Landsat 8 image by the superior index among the used indices through assessing soil salinity from soil samples are shown in Figure 3. Firstly, FLAASH model was applied to remove the atmospheric effects [98–100]. Then, image processes were applied, i.e., image morphology, conversion from digital number to reflectance value, cloud filtering [101], and image enhancement. Subsequently, indices were computed and analyzed. Then, indices were normalized in Excel software. Secondly, the soil samples were analyzed with the spectral reflectance of the image. The soil salinity is estimated by the measured laboratory EC. Based on the results, the relationship was determined between reflectance values and indices of soil salinity to estimate the soil salinity from the image. It was noted that various soil types reflect solar radiation differently. The variation in reflectance makes it possible to identify the type of soil at the surface layer. Validation samples were taken from different land use/land cover types, including Sabkha, water bodies, waterlogged, bare soil, and cultivated land. The sample locations are selected at different salinity intrusion degrees. Finally, leaching water requirements and Gypsum requirements were calculated as shown in Figure 3.

**Figure 3.** The workflow of saline soil amelioration using Landsat 8 OLI image.

#### **3. Results**

#### *3.1. The Relationship between ECe and Remote-Sensing Spectral Indices*

Various spectral indices derived from the initial Landsat bands in the study area validated the developed spatial model of soil salinity; Figure 4. The statistical correlation between soil electrical conductivity (ECe) and remote-sensing spectral index revealed that the salinity index (SIA), salinity index 3 (SI3), vegetation soil salinity index (VSSI), and salinity index 9 (SI9) had a significant correlation with ECe (R<sup>2</sup> = 0.90, 0.89, 0.87, and 0.83, respectively). However, salinity index 5 (SI5) had a low correlation (R2 = 0.21). The models with the highest R<sup>2</sup> values, indicating a high correlation between field measurement data and satellite data, were chosen as the best regression model to produce the soil salinity map of the study area. Overall, the brightness index with bands (R and NIR) of the image dated May 2021 had the highest correlation of 90%. Therefore, this obtained regression equation was used for soil mapping, while the density-slicing method was used to classify the different salinity levels, according to the different salinity classes. These salinity classes were defined using the international salinity thresholds.

The SIA, SI2, SI3, SI4, SI7, SI8, and SI9 indices are positively correlated between the actual ECe and the modeled ECe, and negatively correlated with NDS1, SI6, and SI5. All of the studied indicators followed a normal distribution for the low-salinity class (2 and 4 dS m<sup>−</sup>1), while the higher-salinity class resulted in negative skewness. It was noted that there was a high level of uncertainty and variability for all indicators with higher salinity levels, but a lower level of uncertainty with lower salinity content (Figure 2).

**Figure 4.** *Cont*.

**Figure 4.** (**a**) The correlation between actual ECe and soil salinity indices. (**b**) The pair plot (**up**) and correlation (**down**) between the actual ECe and that extracted from satellite images.

#### *3.2. Land Use*

The land use in the study area was 373,191 ha and can be classified into four dominant classes: cultivated land, water bodies and lake Burrulus, fish ponds, and urban area. The cultivated land was the main class, the second class was water bodies and lake Burrulus, the third class was fish bonds, and the fourth class was the urban area, which covered about 72, 12, 10, and 6% of the total area, respectively (Table 3).

**Table 3.** The main land use categories in the study area.


#### *3.3. Soil Quality Index (SQI)*

The results related to classifying the different salinity levels by the salinity index SIA with bands (R and NIR) are given in Table 4 and Figure 5. The results of the proposed model revealed that the assessment of salinity levels was classified into four classes: the low salinity (4 < dS m−1) class occupies 118,580 ha (44% of the cultivated area); the moderatesalinity class (4–8 dS m<sup>−</sup>1) occupies 73,177 ha (27% of the cultivated area); the high-salinity class (8–16 dS m<sup>−</sup>1) represents 77 ha (28% of the cultivated area); and the extreme-salinity class (>16 dS m<sup>−</sup>1) represents 145 ha (1% of the cultivated area). The results indicated that the salt-affected soils in the study area represent 56% of the cultivated land.

**Table 4.** The main salinity class categories in the study area.


**Figure 5.** Results of soil salinity index for arid and semi-arid conditions (SIA) in the study area.

#### *3.4. Soil Amelioration*

#### 3.4.1. Leaching Water Requirements

Figure 6 shows the different leaching requirements that should be added to the soil to reduce soil salinity. The different leaching water requirement depths were classified into six classes: (1) 0.01 to 0.1 m ha−<sup>1</sup> for an area of about 27,607 ha (10% of the cultivated area); (2) 0.1 to 0.2 m ha−<sup>1</sup> for an area of about 62,335 ha (23% of the cultivated area); (3) 0.2 to 0.3 m ha−<sup>1</sup> for an area of about 66,775 ha (25% of the cultivated area); (4) 0.3 to 0.4 m ha−<sup>1</sup> for an area of about 83,453 ha (31% of the cultivated area); (5) 0.4 to 0.6 m ha−<sup>1</sup> for an area of about 15,447 ha (5% of the cultivated area); (6) 0.6 to 0.9 m water depth ha−<sup>1</sup> for an area of about 14,012 ha (5% of the cultivated area).

**Figure 6.** Different leaching requirements by satellite images for the study area.

#### 3.4.2. Gypsum Requirements

Figure 7 illustrates the different gypsum requirements that should be added to the soil to reduce soil sodicity. As is clear from the figure, the gypsum requirement was classified into six classes: (1) 0.10 to 1 ton ha−<sup>1</sup> for an area of 41,346 ha (15% of the cultivated area); (2) 1 to 2 ton ha−<sup>1</sup> for an area of about 69,754 ha (26% of the cultivated area); (3) 2 to 3 ton ha−<sup>1</sup> for an area of about 65,86 ha (24% of the cultivated area); (4) 3 to 4 ton ha−<sup>1</sup> for an area of about 49,171 ha (18% of the cultivated area); (5) 4 to 6 ton ha−<sup>1</sup> for an area of about 34,234 ha (13% of the cultivated area); (6) 6 to 9 ton ha−<sup>1</sup> for an area of about 9265 ha (3% of the cultivated area).

**Figure 7.** Different gypsum requirement assessments by GIS for the study area.

#### **4. Discussion**

Assessment of salt-affected soil using remote sensing and GIS is beneficial due to its low cost and efficiency. This will improve the management of the salt-affected soil [1]. It was clear from the results obtained that using Landsat images can capture soil salinity with a significant correlation between the ECe values and bands of the Landsat images [40–46].

Our results indicate that the correlation between the remote-sensing spectral index and ECe (salinity index (SIA), salinity index 3, vegetation soil salinity index, and salinity index 9) was highly significant. This agrees with the views of [102,103] who emphasized that the salinity index (SI) has the highest correlation with soil salinity based on the image enhancement method. Elhag [104] indicated that the SI-3 and SI-9 have a high correlation with soil salinity indices.

The results showed that the salt-affected soils in the study area represent 56% of cultivated land. These results agree with one study [80,93], which stated that more than 50% of the soil in Kafr El-Sheikh Governorate suffers from land degradation. Additionally, good or non-saline soils in the study area decreased by 33% during the period 1961 to 2016 [81]. Another study conducted by Enar [105] indicated that the low-salinity soil increased by 0.46%, while moderate salinity, high salinity, and extreme soil salinity increased by 16%, 52%, and 20% during the period 2000 to 2020, respectively.

The added value of the paper is in mapping the soil water leaching and gypsum requirements using remote-sensing and GIS techniques. Limited studies have covered this topic so far. The method used to calculate the leaching water requirements (cm, depth) and gypsum requirements (GR, Mg ha<sup>−</sup>1) according to the concentration of salts is reliable and accurate [33,90]. The leaching water depths required to reduce the ECe ranged between 0.01 and 0.90 m ha<sup>−</sup>1, while the gypsum required to reduce the initial ESP in different study zones ranged from 0.1 to 9 tons ha<sup>−</sup>1.

The investigated salt-affected soils are formed as a result of climate and inappropriate soil management. This is in addition to the effect of irrigation water, water logging, and saline water intrusion of the Mediterranean. Saline, saline–sodic, and sodic soils have a strong presence in the area with an average of 55% of the total cultivated soils (Table 2). The south part of the area is threatened by sodicity according to the low-salinity soils and highly carbonated irrigation water, while the north of the area contains the highest area of saline and saline–sodic soils, reaching 33%. Poor drainage in addition to reuse of saline drainage water supports the buildup of salinity and sodicity [106]

The most popular method for reclaiming salt-affected soils in the region is a gypsum amendment (CaSO4 2 H2O) combined with intermittent leaching. Another two ways for adapting and mitigating salinity and sodicity accumulation in the region are furrow irrigation and rice production under ponding.

The salts are spread in the soil profile, especially in the northern areas of the study area, adjacent to the water-logged areas, Sabkha, and along the coastline. The dominant salts in the delta are saline, and sodic soils are sodium sulfate (Na2SO4) and sodium carbonate and bicarbonate (Na2CO3 and NaHCO3). The solubility of these salts decreases sharply with temperature decreases; accordingly, the reclamation and leaching processes should be applied during the summer warm season only. Improving drainage and preventing industrial and sanitary wastes in the agricultural drain is a must [107]. Land degradation in coastal areas, increased distribution of soil salinity, and reduced crop productivity in the region are the manifestations of climate change that have already appeared in the region from rising sea levels, coastal erosion, reduced Nile flow, increased summer temperature, changing rainfall patterns [108–112], and increased evapotranspiration [113].

In the study case, saline soils should be treated, and subsequently sodic soils, in order to reduce the concentration of salts to the appropriate degree for the growth of plant roots and even the appropriate depth for the roots [114–120].

Proper agricultural practices should be followed such as: adding organic fertilizers, developing and maintaining drainage, following an appropriate agricultural cycle, using a digger plow, and choosing salt-tolerant crops and an appropriate irrigation system.

Nitrate, phosphate, and potassium fertilization increase the resistance of plants to salinity. Salt-tolerant crops such as barley and sorghum are the most salt-resistant cereal crops, followed by rice and wheat, while maize is the least resistant. Cotton and sugar beet are the most important salt-tolerant crops, while sugar cane, fava beans, and peas are the least tolerant of salinity. Most vegetable crops are moderately resistant to salinity, while most fruit crops, especially deciduous, are sensitive to salinity [114–124].

#### **5. Conclusions**

The physico-chemical characteristics of 60 soil profiles were investigated. The results showed remarkable differences among various sites. The differences were very clear between the southern regions, where they are often affected by a slight or medium degree of deterioration and are often concentrated in soil compaction, while the northern regions were greatly degraded due to the presence of most types of soil deterioration such as salinity, alkaline, and waterlogging. The GIS, Landsat OLI satellite images, and multi-temporal satellite image analysis were used to estimate the rate and extent of salinized areas. These proven tools are handy for providing accurate and timely geospatial information depicting soil conditions. The results reflected that 56% of cultivated land of the Kafr El-Sheikh Governorate suffers from salinization. Zoning or classifying the area into zones can lead to better management and amelioration of the different salinity zones. Therefore, using this technology improves the management of salt-affected soil on a large scale and can be regarded as one of the best management strategies for increasing crop production. The causes of salinity in investigated soils are thought to be a result of seawater intrusion—especially

in the coastal zone of the area—high water table level, accumulation of salt in the upper soil layers due to unsuitable irrigation management, and inadequate drainage conditions. Salinity problems require sustainable management strategies, including identifying and further developing crop cultivation practices adapted to saline conditions, enhanced drainage systems, using salt-tolerant varieties/crops, and exchanging knowledge and transferring practical and adaptive solutions. Water is fundamental in agriculture; different sources of pollution such as sewage and industrial wastewater are discharged onto the drains. So, the water in this drain has very low quality, which in turn may cause hazards to soil and grown crops. It could be concluded that drains may be used for irrigation purposes under controlled precautions with good soil management, e.g., good tillage, deep plowing, land leveling, applying soil and water amendments, and finally a suitable cropping system.

**Author Contributions:** Conceptualization, H.M.A. and M.A.E.A.; methodology, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; software, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A. and T.H.K.; validation, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; formal analysis, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A. and T.H.K.; investigation, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; resources, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; data curation, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; writing—original draft preparation, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; writing—review and editing, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S.; visualization, H.M.A.; supervision, H.M.A. and M.A.E.A.; project administration, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A. and T.H.K.; funding acquisition, H.M.A., M.A.E.A., A.M.S.K., M.S.M.E., K.A.A., T.H.K. and A.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The manuscript presents a participation between the scientific institutions in two countries (Egypt and Italy), and in particular, the authors are grateful for their support in carrying out the work to: (1) National Authority for Remote Sensing and Space Sciences (NARSS), Cairo 11769, Egypt. (2) Soil, Water and Environment Research Institute (SWERI), Agricultural Research Centre (ARC), Giza 12112, Egypt. (3) SAFE-Università degli Studi della Basilicata.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

