Abstract
Introduction: Loess deposits are the silt dominated aeolian deposits. Loess-paleosol sequences have been frequently used to determine and reconstruct paleoclimate and paleoenvironment of quaternary. The first step for this purpose, is to identify the sequence of loess and paleosols and describe their pedological, physical and chemical characteristics. Particle size distribution is the most important characteristics of loess deposits. This characteristic is a function of distance from source and pedological processes. Carbonates are the main constituents of the loess deposits which can act as a cement and affect the particle size distribution. Main loess areas in Iran are located in the north of country, especially in Golestan province. However, there are loess deposits in Sari which there is little information about their pedological and morphological properties. The objectives of this study were to 1) identify the loess - palaeosol sequences and 2) determine the effect of carbonates on particle size distribution of two loess-paleosol sequence in southern Sari.
Materials and Methods: There is a loess containing area in southern Sari. During field surveying, two loess-paleosol sections were explored and the loess layers and paleosols (horizons and pedogenic features) were described. The sections were sampled from 10 cm intervals. Calcium carbonate equivalent (CCE), soil organic carbon, pH and particle size distribution of the samples were measured. Particle size distribution was determined using pipette method. The fractions of sand, coarse silt, medium silt, fine silt and clay fractions were measured. To determine the effect of carbonate on particle size distribution, sand, silt (coarse silt, medium silt and fine silt) and clay contents of the samples were measured before and after carbonate removing. Diluted HCl was used to remove the carbonates.
Results and discussion: The thickness of the sections were 408 and 627 cm and were morphologically similar. Both sections contained a modern soil (MS) and three palaeosol which separated by loess layers. This indicated periodic alternation of soil formation and loess accumulation. In the studied sections, the modern soils composed of cambic (Bw) and calcic (Bk) horizons. Paleosols 1 (PS1) and paleosols 2 contained argillic-calcic (Btk) horizons. Paleosols 3 (PS3) indicated vertic properties (Bsskg). In contrast to modern soil in loess in humid regions of Golestan province which contained argillic horizon, the modern soil in the studied sections were not highly developed. Slickenside and wedge shape were the two main vertic properties which observed in PS3. Seasonal saturation in PS3 due to low permeability of Bsskg horizon produced stagnic properties. Accumulation of Mn oxides and to some extent Fe oxides revealed the intermediate reduction states.
Carbonate calcium equivalent (CCE) in the sections ranged from 2 to 48.75 %. High amount of secondary carbonates were found in loess layers and was higher than overlying paleosol. This indicated that in humid periods of soil formation the carbonates translocated from the paleosols. Particle size distribution results indicated that sand fraction mostly composed of hard carbonate nodules and consequently the amount of sand fraction reached to 42 % which is unusual in loess deposits. The amount of sand, silt and clay in the sections were 2 to 48.75 %, 36.2 to 78.2 % and 16.6 to 43.8 % which after removing the carbonates changed to 0.3 to 3.8 %, 47.1 to 81.5 % and 17.6 to 50.3 % respectively. Particle size distribution firstly indicated the distance from the origin.  Dominance of medium silt showed that the sediments have been originated from relatively large distance. Weathering and translocation of fine particles during humid periods increase the clay content in the paleosols. The highest amount of clay content and the lowest amount of CCE occurred in PS2 palesols which indicate suitable weathering conditions during PS2 formation. Furthermore, the thick Btk horizons in PS2 paleosols of both sections is another evidence of suitable conditions for soil formation.
Conclusion: Results of the present study indicated that the studied sections are suitable for future paleoclimate and paleoenvironment studies. The sequences of loess-paleosols in southern Sarin was similar to those in humid areas of Gorgan which indicated they experienced similar sequences of climate variations. On the other hand, secondary carbonates accumulated as hard nodules increase d the amount of sand content which may cause misleading interpretation. Therefore, it is suggested removing the carbonates of loess sections in humid areas of northern Iran.