Sources Identification of Heavy Metals in Surface Soil with the Approach of Statistical Multivariate Analysis (Case Study of Hamadan Province)

Introduction
The expansion of industries, the development of cities, the increase in population, and the excessive human intervention in nature have led to the destruction and pollution of the environment. Soil pollution, especially soil pollution with heavy metals, is one of the environmental issues of concern to many researchers. The geographical distribution of heavy elements in the soil, either naturally or through human activities, will bring issues and problems. Heavy metals can enter the soil environment through the weathering of parent materials or external pollutants. For some metals, such as lead, pollution from external sources is much more than from natural sources. Among the natural factors, parent materials are the main factors that determine the concentration of heavy metals in the soil, and among the unnatural factors, industrial and agricultural activities are among the main sources of heavy metals entering the soil and are known as human sources. Human sources of heavy metal entry are divided into primary and secondary categories. Heavy metals are added to the soil as impurities along with primary sources including sewage sludge, chemical fertilizers and pesticides. Secondary sources include the entry of heavy metals through mines, industrial activities or through deposition from the air. The purpose of this study is to identify the sources of heavy metals (chromium, cobalt, zinc, vanadium, copper, lead, nickel, antimony, arsenic and cadmium) in the surface soil of Hamedan province (0-20 cm) using multivariate statistical methods. (Analysis of main and cluster components).


Methodology
In this study, to investigate sources of ten heavy metals (Cr, Cd, Zn, V, Cu, Pb, Ni, Sb, As, Cd,) in soil surface layer in Hammadan province, a systematic random method was utilized for soil sampling and 286 soil samples were collected from 0-25cm depth. For this purpose, first, the region was divided into 5 x 5 km grids. In the next step, based on the knowledge of the study area and the variety of uses and activities, the grid spacing was 2.5 x 2.5 km in the areas where the intensity of land use was high, and the grid spacing was 10 x 10 km in the areas where the intensity of land use was less. And the intersection of networks was considered as sampling points. A total of 286 soil samples were collected from the study area by a combined method in sampling after spotting, at the sampling site a plot of 20x20 meters (macroplot) and inside it three plots of 3x3 meters (microplot) was applied, then 5 soil samples were taken crosswise from each microplate from a depth of 0-20 cm and after mixing the soil, a 2-3 kg composite sample was taken from it. Also, by completing the sampling form, information about the use of the sampling location, the type of cultivation (irrigated or rainfed), the appearance of the land, the type of product, geographical coordinates and the nearest village was completed. Soil sampling was done from pristine and untouched areas, mountainous areas, salt marshes, irrigated or rainfed agricultural lands, vineyards, gardens and lands around villages.
Soil samples were dried and prepared for analysis after passing through a 2 mm sieve. Extraction was done to determine the total concentration of heavy metals in the soil using HCl and HNO3. The concentration of heavy elements lead, zinc, arsenic, antimony, chromium, cobalt, vanadium, nickel and copper was measured using ICP-MS device and cadmium was measured by graphite furnace method and using atomic absorption device.
In this study, the mean, median, skewness, skewness and Kolmogorov-Smirnov statistics were used to examine the distribution and test the normality of the data at the 95% confidence level. Principal component analysis is used to classify relationships between measured variables. In cases where the number of evaluated variables is large, it is difficult to analyze them through the correlation coefficient.
In this study, the accumulation method was used to group the variables. This method starts with one variable, so at the beginning of processing, the number of groups is equal to the number of studied variables (target variables). Then, the similar and homogeneous groups are combined with each other and with decreasing homogeneity, all the subgroups are piled up and condensed and finally form a single group.
Conclusion
Results of multivariate statistical analysis showed that the main origin of the chromium, cobalt, nickel, zinc, lead and vanadium is parent material and copper cadmium, arsenic and antimony have common origins of natural material and anthropogenic sources (agricultural activities). Finding appropriate information about possible resources of heavy metals can be used for the monitoring and evaluation processes of agricultural soils in the study area. The results obtained from measuring the concentration of total heavy metals in the surface soil of Hamadan province showed that by using factor analysis, the source of heavy metal accumulation can be identified. The results of factor analysis showed that chromium, cobalt, nickel, iron, zinc, lead, and vanadium were the first factor with the highest connection, according to the results of factor analysis. The close correlation between these elements could explain their shared origin. The high concentration of these components in the surface soil of the mother matter appears to be the key reason, according to the geological structure and agricultural maps of the study area. However, because of the high consumption of chemical fertilizers (average urea fertilizer consumption is 500-700 kg/ha per year, potash fertilizer consumption is 200-330 kg/ha per year, and phosphorus fertilizer consumption is 300-558 kg/ha per year), and the possibility of these elements in the chemical structure of urea, phosphate, and potash fertilizers, increasing the element concentration in the soil by them is not far from the mind. Copper, antimony, cadmium, and arsenic are among the components present in the second factor. According to the geological structure and land use map, the high concentrations of these elements are due to both natural and human activity (agricultural activities). The results of a cluster analysis confirm these findings. Overall, the findings of this study revealed that studying the relationship between heavy metals, investigating their origins, and analyzing them using multivariate statistical methods (factor analysis and cluster analysis) would be very accurate and simple.