 |  |
| Bildiri Özetleri | |
| Ana Sayfaya Dönüş |
| Back To The Main List |
|
Şenay Özden 1 , Günay Erpul 2 1 Soil and Fertilizer Research Institute, Ankara, Turkey 2 Ankara University, Faculty of Agriculture, Department of Soil Science, Ankara, Turkey Abstract The Revised Universal Soil Loss Equation (RUSLE) is an attempt to improve the capability of USLE in portioning dynamic hydrological and erosional processes and the flexibility of USLE in adjusting process parameters to account for spatial and temporal changes. It is therefore expected from RUSLE that valuable information on complex interrelation of parameters or processes give rise to more accurate soil loss predictions by the model. The vulnerability of soil to the soil erosion is a multifaceted property that greatly influenced by the variations in antecedent soil-water and soil-surface conditions and consequently in soil properties. In this paper, we closely looked into the data obtained from direct measurement on natural runoff plots of 13 locations of 9 Research Institutes of General Directorate of Rural Service, Turkey to approximately calculate soil erodibility factor (Ki) at any time ti. The calculated Ki values were statistically compared with the observed K values of natural runoff plots. Finally, the average annual values of soil erodibility for each institute were estimated. Introduction USLE is an empirically based erosion prediction technology in which factors affecting soil erosion are function of climate, soil properties, topography, soil surface conditions, and human activities (Wischmeier and Smith, 1978). These factors are expressed in an equation as the lumped parameters, mostly without delineating distinct processes. For example, the soil erodibility factor is a lumped value that stands for an average annual value of soil reaction to a large number of erosion and hydrologic processes (USDA-ARS, 1997), primarily to raindrop and runoff detachment and to the infiltration. Because of having the empirically derived and the lumped parameters, the USLE can only predict time-averaged, or long-term soil losses. In other words, some of the model parameters are insensitive to the spatial and temporal changes possible to occur during the projected prediction span (Nearing et al., 1990). However, precisely defining the processes and parameters considered influential on the soil erosion and their interactions plays very significant role in the process-based models. In this connection, RUSLE aims to make transition to a process-based model as new technologies have been used, and research has supplied a greater number of data about parameters for portioning hydrological and erosional processes. The improved soil erodibility factor of RUSLE provides a great opportunity for making itself to be compatible with the physically based models (Römkens et al., 1986), and accommodates the spatial and temporal variability in topography, surface roughness, soil properties, and rainwater infiltration etc. Indeed, seasonal K values consider the changes in antecedent soil-water and soil surface conditions and the seasonal variations in soil properties and by this way, offers a prediction of soil vulnerability for specific events as well as for long-term averages. Mutchler and Carter (1983) suggested a trigonometric cosine function to account for seasonal variations in K values: Kr = 1 + a cos ( bt-c ) [1] where, Kr is the ratio of the average monthly K value to the average annual K value; t is the mean monthly temperature; a is the peak value of soil erodibility for a given soil and for a given cycle; b is the cycle that the graph of function has through one period, and c is the phase shift of the periodic function. Clearly, magnitudes of a, b, and c vary with the location. The methodology of calculating K values for a given soil depends upon temporal variation in soil erodibility, and it is expressed as an exponential decay function: Ki = Kmax e f (t) [2] where, Ki is the soil erodibility factor at any time ti ; Kmax is the maximum value of soil erodibility factor for a given soil; f (t) is the time function whose parameters are the times at which K max and K min occur and the length of growing period (Dt). Time function varies with location and soil. In fact, Eq. [2] assumes an exponential decrease in soil erodibility as the growing season proceeds (USDA-ARS, 1997). On the other hand, application of a model depends not only on versatility and suitability of the model to various conditions but also on availability of data. Our analysis for the seasonal K values is based on the direct measurements on natural runoff plots in different locations of Turkey, using case formulae developed for the eastern United States. Although the data is seemingly adequate for the required calculations, they were not designed for the seasonal K values, but average long-term K values, and obtained with no replications. That's why, this analysis is a preliminary and, to some degree, incomplete, but its future objective is to develop formulae for the seasonal K values adaptable to Turkey conditions. Methodology The case formulae given in Table 1 (USDA-ARS, 1997) were used to estimate the soil erodibility factor at any time using long-term data from the direct measurements on natural runoff plots collected by 13 locations of 9 Research Institutes of General Directorate of Rural Service, Turkey (Oğuz et al., 2002). It is important here to note that the constant 0.009 of equations of Table 1 was obtained for the eastern United States, and it is much likely to be incompatible to Turkey climates. An analysis is thus required to determine the constant used in the equations, which is not included in this paper. Moreover, in our calculations, growing period was assumed to be  180 days, therefore, in locations where the growing periods are considerably different from 6 months, the calculated values need further adjusting, which is also not in the scope of the paper.Using the estimated Ki , the average annual value of soil erodibility (Kav) was calculated by:  where, Eli is the storm erosivity for a event at time ti. Results and Discussion The calculated average K values from seasonal soil erodibility factors (Ki) are shown in Table 2, and Figure 1 also illustrates the plots of the observed Ki values versus the Ki values calculated by Eq. [3]. In the cases of Tokat, Konya-Beyşehir, Konya-Seydişehir, Kırklareli, and Şanlıurfa, the coefficients of determination, R2, were reasonably well and more than 0.78 (Table 2). This result suggested that the data were in a good agreement with the time span between Kmax and Kmin, which was taken as 6 months. Obviously, Ki attained its minimum at the end of growing period. For the cases of Eskişehir, Menemen, Samsun and Yozgat, the calculated values of Ki explained more than 50% of the variations in observed Ki values. Nevertheless, those of Erzurum, Kütahya, and Bilecik, the R2 were 0.11, 0.33, and 0.39, respectively, meaning that the time span between Kmax and Kmin was significantly different from 6 months, and distinctly, the periodic changes in the calculated Ki did not agreed with the variability of the observed values. In fact, we had very limited data on statistical variability of the observed Ki values since they are obtained with no replicate.  However, Rüttimann et al. (1995) recommended as many replications as possible for erosion plots to have representative data of erosion rates for a given location, and the data from replicated natural rainfall-erosion plots showed several important properties of erosion plot variance (Nearing et al., 1999, Risse et al., 1993). Therefore, it is very difficult to give a reason for these smaller values without having information on the data variability. For example, for these institutes the growing period would be expected to not greatly differ from 6 months, and it is extremely possible that Kmax and Kmin would be unrepresentative for the seasonal changes in the soil erodibility. Overall, the analysis was promising but imposed a further study with regard to the effects of regional climate pattern on the soil susceptibility to erosion.  Conclusion Our analysis with the long-term data showed that the ability to more accurately predict the seasonal soil erodibility factors depends closely upon precisely delineating of cycles, in which soil erodibility greatly changes. This additionally requires integrating some climatic parameters in to the model.  References . Mutchler, C. K., and C. E. Carter. 1983. Soil erodibility variation during the year. Trans. ASAE 26: 1102-1104, 1108. . Nearing, M. A., G. Govers, L. D. Norton. 1999. Variability in soil erosion data from replicated plots. Soil Science Society of America Journal 63: 1829-1835. . Nearing, M. A., L. J. Lane, E. E. Alberts and J. M. Laflen. 1990. Prediction technology for soil erosion by water: Status and research needs. Soil Sci. Soc. Am. J. 54: 1702-1711. . Oğuz, İ, Ş. Özden, H., Cebel, M. Demiryürek, E. Ayday. 2002. Türkiye Üniversal Denklem Toprak Kaybı Eşitliği Rehberi. Köy Hizmetleri Genel Müdürlüğü , Ankara. (In press). . Risse, L. M., M. A. Nearing, A. D. Nicks and J. M. Laflen. 1993. Assessment of error in the Universal Soil Loss Equation. Soil Science Society of America Journal 57: 825-833. . Römkens, M. J. M., S. N. Prasad, and J. W. A. Poesen. 1986. Soil erodibility and properties. In Proc. 13th Congr. Int. Soil Sci. Soc., Vol. 5, pp. 492-504. Hamburg, Germany. . Rüttimann, M., D. Schaub, V. Prasuhn, W. Ruegg. 1995. Measurement of runoff and soil erosion on regularly cultivated fields in Switzerland-some critical considerations. Catena 25: 127-139. . U.S. Department of Agriculture, Agricultural Research Service. 1997. Predicting soil erosion by water: A guide to conservation planning with the Revised Universal Soil Loss Equation (RUSLE). Agric. Handbook No. 703. . Wischmeier, W. H. and D. D. Smith. 1978. Predicting rainfall erosion losses: A guide to conservation planning. U. S. Dep. Agric., Agric. Handbook No. 537. |