**19. Conclusion**

146 Soil Erosion Studies

Table 8 shows Pearson Correlation Coefficient between different measured variables with runoff volume, sediment production and sediment concentration. With statistical investigation, it was found that slope with %80 and %88 correlation coefficients, has the highest direct relationship with the amount of produced sediment and runoff, respectively, these results are similar to a part of Ribolzi et al. (2010) results. After that is the amount of silt with %53 correlation coefficient with sediment production and %54 correlation coefficient with runoff production. Materials having higher amount of silt are easily dispread and transported and are more erodible (Meyer and Harmon, 1984). The amount of clay has negative relationship with sediment and runoff production. With increase in the amount of sand, permeability is increased and lower amount of runoff is produced, in addition, despite having low adhesion and easy to be separated due to their coarser sizes, sand grains resist to transportation by runoff and produce lower amount of sediment. This result is similar to Vanesland et al. (1987) result. Vegetation cover has a negative correlation coefficient of %36 and the highest adverse relationship with sediment production and meaningful negative relationship with runoff. The reason is decrease of rainfall drop energy and velocity of surface runoff by vegetation (Yu et al. 2006). Percentage of Kaolinite has the highest negative correlation coefficient with runoff volume (%53). Other variables have weaker relationship with sediment and runoff productions. Chemical characteristics also affect runoff and sediment production so that the amount of organic matter has reverse relationship with sediment and runoff productions. In the studied area, the effect of Mg++ ion on decreasing sediment production is lower than Ca++ ion, so that it does not have meaningful correlation. Two factors of SAR and ESP, although they are not in regression models, have positive relationships with runoff and sediment production. These two are affected by other characteristics such as cations which are important in soil aggregate stability, soil infiltration and formation of surface crusts. The investigation of correlation matrix shows that none of the parameters can solely describe all observed variations in the

amount of sediment.

<sup>3</sup>Total

**Dependent variable** 

Sediment concentration (gr/lit)

**Adjusted** 

1 Runoff(cm3) 0.812 .000 YRun.= 54.254Sl–

sediment(gr/m2) 0.809 .000 YSed.= 2.144Sl+ 3.792CEC–

Table 9. Regression analyses and obtained models for runoff and sediment

Co = Chlorite mineral (%), YTurb.= turbidity(gr/lit), CEC= cation exchage capacity(c mol/kg),

For anticipating variation of sediment production based on physical and chemical properties and determination of share of each variable on explanation of the amount of

YRun. = Total runoff volume (cm3), Sl = slope (%), Ka = kaolinite mineral (%),

Ca= Calcium cation(meq/lit), YSed.= Total sediment (gr/m2), Si= silt (%)

**R2 Sig. Regression Model Standardized** 

31.937Ka+18.704Co+ 739.80

1.084Ca-18.393

2.071Ca+ 0.968Si – 92.90

0.771 .000 YTurb.= 0.798Sl+ 2.153CEC–

**regression model** 

YRun.= 0.788Sl– 0.180Ka + 0.118Co

YTurb.= 0.707Sl+ 0.416CEC– 0.282Ca

YSed.= 0.740Sl+ 0.285CEC– 0.209Ca+ 0.161Si

**Model no.** 

2

Although it seems that a lot of factors are effective on runoff and sediment production of loesses, but investigation of the results show that a few number of key parameters are more important in the studied area and that other parameters have indirect effect on this matter. Slope is the most important factor in sediment and runoff production. Presence of Kaolinite decreases and the increase in the amount of silt increases sediment and runoff productions. Meyer and Harmon (1984) and Vanesland et al. (1987) also found similar results. Among chemical properties which were analyzed in this research, CEC and SAR have direct relationship and calcium cation, the amount organic matter have reverse relationships with sediment production. These are similar to a part of Hasanzade Nafuti et al. (2009), Vitharana et al. (2008) and Mahmoodabadi et al. (2009) results. Among clay mineral, smectite having week bounds between layers and being highly expandable, has direct relationship and Kaolinite being a stable clay mineral, has negative relationship with runoff and sediment production. Zhang et al. (2004) in studying loesses of China found similar relationships between erosion of loesses and the kind and amount of clay minerals. The results of regression analyses show that between the amount of runoff and sediment productions as dependent variables and independent variables with significant correlation, there is a significant relationship at %1 level. Among independent variables, only four factors: Percentage of slope, CEC, soluble Calcium cation and amount of silt, have important role in sediment production and slope and kaolinite and chlorite percentages have important role in runoff production so that these factors control %80 of sediment variations and %81 of runoff volume variations and other %20 relates to factors which are not studied in this research. Among the variables which were entered in the model, slope factor is more important so that one unit change in standard diversion of this factor produces 0.74 unit change in standard diversion of sediment production and 0.79 unit change in standard deviation of runoff volume which is because of its effect on increase of velocity, surface runoff and rain drop impact (Toy et al., 2002). Silt from the view point of size and cohesion is susceptible to erosion. CEC is effective on size and stability of soil aggregation. Calcium cation causes chemical bounds between aggregates and flocculates the grains, therefore has inverse relationship with sediment production of loesses.

In investigation of physical and chemical characteristics of loesses between surface and rill erosional features, it is found that there are not significant differences between these characteristics and the formation of special kind of erosion feature is mainly affected by slope amount and landuse type. The range lands are under overgrazing and without a plan throughout the year. This subject has caused many problems for natural reproduction of important and effected rangeland plants, the result of which being acceleration of erosion and sediment production. Construction of roads for access to cultivated lands across rangelands also has caused accelerated erosion. Also in recent years with promotion of technology, native people of the area use tractor and plough the lands in the direction of slope. These also cause intensification of erosion and sediment production.

Investigation of Effective Factors on Runoff Generation and

of china, 12th ISCO Conference Beijing 2002, pp. 1-10

Sediment Yield of Loess Deposits Using Rainfall Simulator 149

Kehl, M., Sarvati, R., Ahmadi, H., Frechen, M., Skowronek, A., 2005. On loess deposits of northern Iran, Quarterly Geographical Journal of Territory, 1(4), 1-11. Keli Z., L. Shuangcai, Wenying, P., 2002. Erodibility of agricultural soils in the loess plateau

Khaje, M., 2003, Study of sedimentology, sedimentary environment and sediment

Khormali, F., Kehl, M., 2010. Micro morphology and development of loess-derived surface

Mahmoodabadi, M., Charkhabi, A. H., Refahi, H., 2007. The effect of soil physical and

simulator, Agricultural Engineering Research Journal, 8, 1-16.(in Persian) Meyer, L. D., Harmon, W., 1984. Susceptibility of agricultural soils to interrill erosion, Soil

Moghaddam, M., 2007. Range and Range Management, University of Tehran Press. 350p.(in

Montgomery D. C., Pech, E., Geffry, G., 2001. Introduction to linear regression analysis,

Moore, D., Reynolds, R., 1989. X-ray diffraction and the identification and analysis of clay

Okhravi, R., Amini, A., 2001. Characteristics and provenance of the loess deposits of the Gharatikan watershed in Northeast Iran, Global and Planetary Change, 28,11–22. Pashaee, A., 1998. Investigation of physical and chemical characteristics and source of loess

Qiangguo, C., 2002. The relationship between soil erosion and human activity on the loess

Rhoton, F., Duiker, S., 2008. Erodibility of a soil drainage sequence in the loess uplands of

Rozoycki S. Z., 1991. Loess and loess-like deposits, The Publishing House of the polish

Saghafian, B., 2003. Investigation and determination of criteria for rain simulator

Schiettecatte, W., Jin, K., Yao, Y., 2005. Influence of simulated rainfall on physical properties

Sheklabadi, M., Charkhabi, A., 2003. Runoff generation and sediment yield in soil with

Refahi, H., 2001. Soil erosion and control, University of Tehran Press. 551p.(in Persian) Ribolzi, O., Patin, J., Bresson, L.M., Latsachack, K.O., Mouche, E., Sengtaheuanghoung, O.,

deposits in Gorgan and Rasht Areas. Journal of Geological Science, 23 &24, 67-

Silvera, N., 2010. Impact of slope gradient on soil surface features and infiltration on steep slopes in northern Laos, Geomorphology, Number of Pages 11.(Article in

construction based on climatic conditions of Iran, Soil conservation and Watershed

different parent materials( Galabad Basin, Ardestan), Journal of Olum va Fonune

Azad Islamic University, Science and Research Branch.(in Persian)

International Journal, Number of pages15, Article in press.

Liu, T., 1988. Loess in China, Springer Verlag Publication, 224p.

science society of America Journal, 48, 1152- 1157.

Third edition, Jon Wiley & Sons publication, 350p.

minerals, Oxford University Press, 332p.

plateau, 12th ISCO Conference Beijing 2002.

Mississippi, Catena, 75, 164–171.

academy of Sciences, Wroclaw, 187p.

Management Research Center Press.85p.(in Persian)

Keshavarzi va Manabe Tabiee, 2, 88-100.(in Persian)

of a conventionally tilled loess soil, Catena, 64, 209-221.

Persian)

78.(in Persian)

press)

production in Gorganrood Catchment (Il Chashmeh and Ghurchay), Ph.D. Thesis,

and buried soils along a precipitation gradient in Northern Iran. Quaternary

chemical properties on runoff generation and sediment yield using rainfall
