Monitoring Soil Contamination by Hydrocarbons and Heavy Metals in Parts of Basra Southern Iraq Using Remote Sensing and GIS Techniques

Abstract


Introduction
Oil pollution is one of the most severe problems in the natural and human environment because of its direct impact on living organisms.Oil and its derivatives are considered a high toxicity hazard due to the emission of gases upon evaporation or decomposition of spilled oil particles and to contain crude oil (Nasr and Suha, 2016).In addition, it has toxic gases such as hydrogen sulfide H 2 S and others.Pollution is measured by its size, concentration, or damage.Environmental damage arises from various human activities, particularly industrial activities.This study was conducted in the Al-Faw District in Basra, southern Iraq, which it is considered one of the most important places from an economic point of view.It is located to the southeast of Basra (Fig. 1), about 100 km between latitudes 29⁰ 97′ N north and longitudes 48⁰ 47′ E east.One of the large oil spills is located next to the main Al-Faw street close to the Al-Faw oil depot studied, the oil spills appears annually due to rain.The study area is located within regions of the Mesopotamian Zone of the unstable shelf of the Arabian plate according to the tectonic division of Buday and Jassim (1987).Stratigraphically setting, it is represented by the sediments of the Quaternary period extending between the Pleistocene and Holocene Epochs, which are characterized as deltaic and alluvial sedimentary plain area (Al-Sayyab et al.,1982).In this period, the stages of sedimentation and erosion were repeated.
Accumulation of toxins, salts, chemicals, radioactive materials, and other disease-causing agents in the soil has a detrimental effect on plant growth as well as the health of animals and people.Pollutants are present because heavy metals have accumulated in the soil in higher concentrations than allowed.The environment faces a hidden threat from soil pollution, which is difficult to observe by the human eye.Soil contamination may not have been remarkable in the past, and the interest was little, but it has become attractive to us.It has received significant interest from many companies in detecting pollution and determining its causes, then determining the depths of pollution and trying to get rid of pollution by treating the soil in different and varied ways to reduce the risk of this pollution, which has a long-term effect.When oil is mishandled, spills can occur on land and in water, and the poisonous compounds that left behind can contaminate the environment for years (Burger, 1997).Because of the lack of local studies that focused on studying, monitoring, and creating maps of oil spills in time and place, we turned to this study.Globally, there are few studies on this subject.The only practical method for obtaining fast and accurate data for mapping geographic areas is through the use of Remote Sensing (RS) and Geographic Information System (GIS).

Office Work
The fieldwork was preceded by the investigation of the study area using Google Earth and some geological maps, in addition to some geological, hydrological, and geographical studies of the studied area location to help get more details about the studied area.In the beginning, the programs that will be F1 F1 F1 F1 used in the study was prepared (Google Earth Pro, GIS Desktop 10.8, ENVI V.5.3), then the base map was reviewed, and the environmental changes that have occurred in the area, such as oil spills, are monitored if the same situation continues or is remediating.The in which the contamination occurred and how widespread it is in the place.

Fieldwork
The fieldwork included a reconnaissance trip on 30/04/2021 to locate the sampling sites.and as a result, Soil samples were collected for contamination tests.They were using sterile polyethylene containers capacity of 250 ml.The jars were filled and emptied air and tightly closed.from each site.Soil samples for Particle Size Distribution were collected in 1 kg bags, and all sampling sites were identified using GPS device type (GPS 128 Flush Garmin Ks) (Fig. 2).Table 1 shows the location area.

Grain size analysis
Specified hydrometer methods were utilized to determine the size distribution of fine particles (mud and clay size) by sedimentation.It takes measurements of the soil suspension's density at various time intervals.According to the procedures (BS 1377-2:1990 Part 2: Classification Soil).Used BS hydrometer, sieve 36 mm, 2 mm, balance readable to 0.01 g, drying oven with temperatures between 60 and 65 degrees Celsius and 105 and 110 degrees Celsius, and sodium hexametaphosphate solution, 33g of sodium hexametaphosphate, 7g of sodium carbonate, and distilled water should be dissolved.When analyzing inorganic soils, pretreatment with hydrogen peroxide should typically be skipped when organic matter is in significant quantity.The test was done in the laboratory of the RSK company branch in Iraq.

pH
Soil (pH) expresses the acidity or alkalinity of the soil and gives a clear idea of the properties and composition of the soil.The acidity of soil samples was measured using saturated paste extract using a pH device Meter after titration with standard solutions.The equipment used for the test: Balance, capable of weighing to 2 decimal places, Shaker, Extraction tubes, and pH meter, a Eutech 2700 is suitable.

Electrical conductivity (EC)
The term "soil electrical conductivity" describes the capacity of soil water to carry electrical current (EC).Electrical conductivity, an electrolytic process, mainly occurs through water pores.Salts dissolved in soil water produce cations (Ca 2+ , Mg 2+ , K + , Na + , and NH4 + ) and anions (SO4² -, Cl -, NO3 -, and HCO3 -) that transmit electricity and carry electrical charges.Accordingly, the soil's EC is determined by the ion concentration.EC device Meter after titration with standard solutions.The equipment used for the test Balance can weigh three decimal places, a Conductivity meter, and Volumetric flasks of various volumes.

Heavy metals
Transmission of heavy metals to the surrounding ecosystem is one of the detrimental repercussions of oil spills.The toxicity and concentration of each metal in the crude oil determine if it is contaminated by heavy metals (Al-Khafaji, 2020).Heavy metals are considered dangerous contamination because they cause problems to the environment in the long run, and their stay in the soil takes long periods.If it is sandy soil, it descends to the bottom until it reaches the groundwater and causes pollution, but in clay soils, it takes a long time in the soil (Grim, 1968).To determine the amounts of heavy metals, the soil of the study region was analyzed.The two samples were sent to the approved Element Laboratory in Dubai, where the results were the elements that were chosen to conduct the examination (As, Co, Cr, Cu, Fe, Mn, Ni, Pb, Zn), and they were compared with the permissible limits.

Total petroleum hydrocarbons (TPH)
TPH is a standard metric for estimating the extent of environmental hazards brought on by various petroleum hydrocarbon products, including fuels, oils, lubricants, waxes, and others.It is common knowledge that petroleum hydrocarbons are neurotoxic to humans and animals.Hexane, benzene, toluene, xylenes, naphthalene, and other chemicals found in gasoline, jet fuel, mineral oil, and other petroleum products are among the substances found in TPH.The TPH are carbon chains with a C6 to C35 range.Table 2. Aliphatic (compounds with a straight carbon chain) and aromatic (compounds with a carbon ring) products include TPH.The equipment used for the test Balances, Rapid Trace sample tubes, 2ml autosampler vials, pipettes, Shaker, Various calibrated pipettes, Various calibrated dispensers, 500ml glass bottles, 30ml vials, 2ml autosampler vials, Centrifuge, Desiccator, 5.10Aglient 6890 GC-FID.

Download imagery
In this study, Sentinel 2B for monitoring oil spills and contaminated soils will be used.Images were downloaded from the USGS (United States Geological Survey) website.

Pre-processing
Pre-processing is a significant and diverse collection of satellite images preparation technologies that work to correct issues with the band data and recalculate Digital Numbers (DN) values that minimize these issues (Al-Ajmi and Uddin, 2009).The two most frequent preprocessing procedures to correct external defects in remotely sensed imagery are atmospheric and geometrical corrections.

• Radiometric preprocessing
Geometric correction of satellite images is critical for any digital analysis process, according to (Jensen, 1996).When employing geometrically uncorrected images, the classification process of satellite images can be misleading, according to (Dai and Khorram,1998).It has been discovered that the lack of geometric correction in studies of satellite imagers of varied terrain affects product correctness (Stow, 1998).regarding the present study, using Sentinel-2 image, the Sentinel-2 satellite imagery been geometrically rectified.

• Atmospheric correction
To correct the impact of the atmospheric effect on remote sensing data, numerous techniques have been developed.Relative scattering correction approaches are helpful and relatively easier to control scattering issues.Depending on the work being done, atmospheric correction of remote sensing data may not always be necessary, according to studies by (Song et al. 2001).a few researchers vehemently advocate minimizing the impacts caused by light scattering of atmospheric scattering (Weng, 2012).This is especially true for the visible few of the electromagnetic spectrum.The Dark Object Subtraction (DOS)DOS and Sun elevation correction methods are used for Atmospheric correction.

• Dark Object Subtraction (DOS)
The additive scattering-induced haze component can be removed from remote sensing data using the Dark Object Subtraction (DOS) approach, which is an image-based technique (Chavez, 1988).The geospatial community has determined that this technique for reducing light scattering in remote sensing data is data-dependent and widely accepted (Song et al., 2001).One of the most common and wellknown straightforward atmospheric correction techniques is based only on the image itself.A specific empirical atmospheric correction technique called "dark object subtraction" is available in ENVI and is based on the assumption that air scattering accounts for a significant portion of the reflectance from dark objects.Dark object subtraction looks for the darkest pixel value in each band.This subtraction helps reduce hill shadows, cloud coverage, and floating particle accumulation.

• Sun Elevation Correction
It is a kind of radiometric correction; generally, converting the DN values to spectral units is far more beneficial.The radiometric correction pattern applied to satellite imagery varies greatly depending on the sensors because the radiation measured on any object is affected by several factors, including changes in the scene's illumination, the state of the atmosphere, and the characteristics of the sensor's response.The sun elevation correction should be performed when using both the near-infrared and visible portions of the electromagnetic spectrum (Lillisand and kiefer, 1987).Part of the downloaded satellite image file can be used to gain information on the sun's elevation angle (Kumar, 2004).Radiometric aberrations can be caused by alterations in scene lighting, atmospheric conditions, visibility mode, and sensor response traits (Al-Ali, 2019).

Minimum noise fraction
MNF is advanced hyperspectral analysis; the performance of this method is valuable for decisionmakers to choose better bands with more information for more accurate information extraction.(One technique to reduce noise on hyperspectral imagery is known as MNF) (Syarif and Kumara, 2018).Furthermore, little research has attempted to demonstrate the impact of the MNF transform on multispectral data.The MNF transformation is used to determine the inherent dimensionality of image data, separate noise in the data, and reduce the computational requirements for further processing, and the MNF transformation is used (Boardman and Kruse, 1995).The (MNF) transform is also used to: remove residual noise from the spectral data and provide a convenient mechanism for choosing prototype spectra.Envi software is used to carry out MNF.

Image Classification
The process of extracting information known as classification analyzes spectral signatures and classifies pixels into thematic clusters based on similarities in their Digital Numbers (DN) values.This paper used Support Vector Machines SVM classification to monitor the oil spill and contaminated soil.Support vector machines (SVM) are a potential machine learning research advancement that is not frequently used in remote sensing.

Grain Size Analysis
Results of the study's samples' grain size analysis are provided.Sandy silty CLAY with gypsum, which displays the percentages of clay (50%) and silt (34%), and sand (16%).Partical Size Distribion PSD analysis is important in studying environmental pollution because it is related to the concentrations of heavy elements in the soil; in clay soils, the absorption of heavy elements increases (Grim, 1968).The volumetric analysis is important in determining the concentrations and quantities of heavy metals and hydrocarbons.

pH
The pH results are shown in (Table 2) the class soil was Alkaline with higher pH values from 8.3 to 8.85 The presence of sodium carbonate or high exchanged sodium.Still, when the pH ranges from 8.0 to 8.5, this usually increases the concentration of calcium carbonate salts in the soil (Al-Sabah, 2007).

Electrical Conductivity (EC)
The salinity category in the study area is between moderately saline -very saline) so that the presence of sodium becomes dangerous to plants in soils with high EC because of the high sodium content, and these soils typically have poor structure and drainage (Table 3).

Heavy Metals
They are increasing the concentration of heavy metals in the study area is due to oil spills and clay soil.Therefore, the attention of the elements in the study area is increasing at rates higher than the internationally recommended concentrations, as shown in Table 4.The Table 4 data indicates the high concentrations of As exceeded the permissible environmental limit of 1.5 ppm (Al-Saraifi, 2021).This poses a danger that its pollution reaches the groundwater if there is no remediation; Arsenic is toxic to the liver, as it causes cirrhosis of the liver and affects the bone marrow and cellular elements in the blood.Its effects on the fetus appear as the activity of enzymes decreases (Patlolla, 2005).It can also cause cancerous diseases (skin, intestine, bladder, and kidney cancers).The laboratory analysis results indicated that high Co concentrations (Al-Saraifi, 2021) mentioned the permissible limit of 2.12.The toxicity resulting from the excess of cobalt metal in the human body causes myocardial infarction due to the lack of oxygen gas access, which disrupts the body's chemical and vital functions (Rahma and Suha, 2013).The concentrations of high of Cr in Al-Faw soil.It was observed that the concentration of this element in the surface soil increased as a result of adsorption of this element by clay minerals, especially the mineral montmorillonite, which is prevalent in the soil of seasonal contrasted climate (Chamly, 1989;Mohammed, 2019) dry and semi-arid areas (Al-Obaidi, 2000).As it is known that the sediments of the Iraqi Mesopotamian plain are characterized by a high degree of interaction with an essential medium, (Al-Rawi et al., 1983), which leads to an increase in the adsorption of heavy elements.Thus, the results showed an increased chromium concentration in the soils by a substantial amount of the study area, which indicates the contamination of this area with chromium which indicates the contamination of this area with chromium exceeded the permissible environmental limit between (1-5 ppm) (WHO, 2003).Chronic exposure to chromium causes kidney and liver failure, and chromium is one of the carcinogens; and causes cases of diarrhea, bloody bleeding, and a decrease in the body's immunity and causes severe poisoning.And increased exposure to causes rhinitis, damage to cell walls, ulcers in the lung, damage to nerve tissue, and reduces the hemoglobin level in the blood, thus leading to death (Kleefstra, 2004).Copper is one of the quickreacting elements with air, especially with abundant water vapor, which forms the toxic, green-colored copper oxide.High concentrations in the study area were higher than the permissible limits15ppm (Al-Saraifi, 2021).Copper is one of the metals that cause cancer.It causes rapid physiological changes and harmful effects that affect many human body organs, such as cirrhosis, dermatitis, and neurological disorders (Olaifa, 2005).The results showed that the iron element values reached the highest value in Al-Faw.
The microorganisms in the biosphere that consume organic residues and decomposing plant residues in the soil in oxidation and reduction Iron are sources of energy that cause the release of iron ions.These ions may remain in the soil due to their adsorption from clay minerals and organic materials, as indicated by this (Al-Hadithi, 2001).Continuous exposure to iron leads to liver damage, heart failure, diabetes, genetic effects, and in some cases, thalassemia.The highest Nickel concentration in the study area's soil has adverse effects, as it affects humans with lung and nose cancer, causes inflammation in the airways, and causes inflammation in the outer layer of the skin (Sapongi et al., 2007).The results, shown in Table 4, indicated that the concentrations and values of Pb in the soil of the study area are less than the global determinants set by the World Health Organization WHO 2003, about 50 ppm.Exposure to lead leads to anemia and a decrease in the total number of red blood cells, as well as high blood pressure in the kidneys, and causes heart attacks that may lead to death and also causes memory loss and many diseases.One of the rare elements similar to copper and nickel like its presence in the rocks of the earth's crust.Zinc is involved in the metabolism of both animals and plants, and zinc is necessary for the growth of humans and animals.However, a deficiency, especially in the early stages of growth and development, is tiny despite the need for it; this amount causes weakness in the bones and joints.As shown in Table 4, the values and concentrations of zinc in the surface soil models showed in the Al-Faw District.It was less than the permissible limits (150-300) WHO 2003.Zinc affects human health and growth and confuses the performance of functions.It also leads to dermatitis of the extremities due to enteropathy.Excess exposure to zinc causes acute poisoning, nausea, vomiting, diarrhea, lethargy, and fever (Baker et al., 2013).

Total Petroleum Hydrocarbons (TPH)
Pollution from oil hydrocarbons One of the most critical problems facing the environment is one because of large quantities.It's just leaky but also because of toxicity.According to the results of the laboratory analysis referred to in Table 5.There are apparent spatial variations in the concentrations of hydrocarbons; in the soil studied in Al-FAO, the TPH concentration of deviation in the locations from which the models were taken was highest at the first site F1depth 0.00 (4793.58ppm)and slightly lower with depth (4515.59ppm).while it was lowest at the fourth location (17.05ppm)This indicates that soils far from oil spills do not contain hydrocarbon contaminated.

Minimum Noise Fraction (MNF)
Few studies have attempted to demonstrate the impact of the MNF transform on multispectral data.The MNF transformation is used to separate noise from signal in the data collection, reducing the computational needs for further processing and estimating the underlying dimensionality of imagery data.(Boardman and Kruse, 1995).This process was applied in the study area to determine the amount of noise in the bands.It was noted that bands (1,2 and 3) contain information and can be used to obtain imagery with the lowest percentage of noise compared to other bands (4-12).Table 6, Figs. 3 and 4. show the amount of noise with each band.The MNF eigenvalues plots (Fig. 4) corresponding to the aforementioned typical items were extracted after the first three bands on the MNF spectra were chosen.Plots demonstrated that as the number of bands increased, MNF Eigenvalues steadily decreased, indicating that the lower-ranked bands held useless information (Fig. 3).Through the analysis of MNF eigenvalue plots and images, it can be seen (Fig. 5) that the oil spill is red and the contaminated soil is yellowish green.The present study shows that Fig. 7 is a successful machine learning SVM classifier mapping three classes.The results showed that the oil spill in red color is due to oil leakage from oil pipelines or near the industrial area, which caused soil contamination and appears clear in blue, In Fig. 6 the blue polygon repersnts the residential area clear in green.This classification helped by detailedly studying the contaminated area and the oil spills in the places that were not visited.The presence of the red color is noticeable even in the river areas.The pollution of this river is due to the oil leakage from the oilcarrying ships, which caused the river to be contaminated with hydrocarbons and heavy metals.In addition.Regarding the study of (Al-Tamimi, 2021) conducted on the Shatt Abu Al-Khasib, an extension of this river (Shatt Abu Al-Khasib), the Shatt Abu Al-Khasib suffers from being polluted with hydrocarbons and heavy metals due to Ship carriers.This classification succeeded in using remote sensing technology to monitor the environment and determine the area occupied by contamination, whether on the soil or water the Fig. 7 showing the result of SVM classification.

Discussion
By searching for the study of pollution in the Al-Faw district, we did not find any investigation related to the area regarding hydrocarbon contamination or heavy metals.Detailed research should be carried out on the extent of the contamination of the site and the depth to which the contamination has reached.No study used remote sensing techniques and GIS to monitor oil spills.This study was considered the first of its kind for a region that studied the area in terms of pollution using remote sensing systems and GIS.Companies and concerned parties rely on this technology because of its importance in observing the land surface with a short time and at a lower cost to study the area, especially the hardto-reach areas.Oil companies should care and force them to use RS and (GIS) technologies to monitor their work, especially in the case of oil spills, because of their devastating impact on the environment if they are not addressed and neglected.

Conclusions
Most of the soil is silty clay with gypsum and halite mineralization.Alkaline soils suffer from high salinity and contain high concentrations of heavy metals (As, Co, Cr, Cu, Fe, Mn, Ni, Pb, Zn) higher than the recommended concentration.The soil of the study area suffers TPH severe pollution needs to remediation.Utilizing remote sensing technology is important in determining the areas of oil spills, the cause of the spill, and the extent of soil contamination with this spill.Using MNF and SVM classification, we could identify the areas of oil spills and contaminated soils that were not visited.

Fig 1 .
Fig 1. Study area in AL-Faw district

Fig. 2 .
Fig. 2. Oil slicks in the study area in Al-Faw near the main street

Fig. 3 .
Fig. 3. MNF Eigenvalues in 12 bands of the image.The image is represented by bands 1, 2, and 3, which show less noise

Fig. 4 .
Fig. 4. Eigen values of MNF transformed bands and bands number

Fig . 5 .
Fig .5.MNF results show the oil spill red color and the contaminated soil yellowish green

Fig. 6 .
Fig.6.Training area.The green with red boundary polygon is for the oil spill, the red polygon is for soil contamination, and the blue is for a residential area.

Fig. 7 .
Fig.7.SVM classification for the study area

Table 1 .
Locations of collected soil samples in the study area Fig 2: Oil slicks in the study area in Al-Faw near the main street.

Table 2 .
The pH result and class of soil in the study area.

Table 3 .
The EC result and class of soil in the study area

Table 4 .
The concentrations of Heavy metals (ppm) in the study area

Table 5 .
The TPH result in the study area

Table 6 .
MNF bands, the information decreased drastically, to almost nearly 98% less than the first MNF band