Ahmad Nohegar; Mohamad Kazemi; S.Javad Ahmadi
Volume 4, Issue 12 , December 2017, , Pages 67-87
Abstract
Extent Abstract Introduction Considering the impact of accelerated rates of sediment yield and soil erosion on catchments, which results from land clearance and poor land management (Palazón et al., 2015) including soil degradation, environmental pollution, and sedimentation in dam reservoirs, ...
Read More
Extent Abstract Introduction Considering the impact of accelerated rates of sediment yield and soil erosion on catchments, which results from land clearance and poor land management (Palazón et al., 2015) including soil degradation, environmental pollution, and sedimentation in dam reservoirs, the reduction of the sedimentation is required to implement appropriate methods of sediment control and soil conservation in the critical areas of sediment resources in the catchment (Patrick et al, 2015). In addition, the recognition and identification of the relative importance of the sediment resources and their contribution (Chen et al., 2016) to sedimentation are necessary for identifying appropriate methods and proper implementation of soil conservation programs. The sediment finger printing is a direct approach to identify the relative contribution of each source and provides a direct approach for quantifying sources of sediment. A fingerprint of sediment sources is obtained using radionuclides, tracer metals, or other sediment properties, which enables the determination of the relative source contributions (Motha et al., 2004). Including the erosion and sediment, the important thing is to choose a model or method to estimate the actual loss or erosion of soil and the contribution of each source to its value. There are few studies estimating the level of GOF and ME on sediment fingerprinting approach to determine the relative contribution of each of the resources. GOF and ME allow making better informed decisions on sediment management (Minella et al., 2008) and can reliably determine mixing the contribution of each sediment source in mixing models (Collins et al., 2010). Methodology The study area, Tange Bostank catchment, covers an area of 81.73 km2 and is located at about 80 km far from the Northwest of Shiraz, at the geographical location of 52° 03' 43'' to 52° 13' 36'' in the East and 30° 16' 33'' to 30° 25' 18'' in the North. Geological formations maps were provided as Razak, Kashkan, Bakhtiari, Quaternary, PabdehGurpi, and Asmari formations using SFF method. Land use maps were also provided as rangelands, forests, gardens, and irrigations using ML method with Landsat satellite image 8 of OLI sensor. Discriminating sediment sources to confirm the discrimination of the potential sediment sources was done in two steps. The first step was based on the use of the Kruskal–Wallis H-test to discriminate the potential sources by the fingerprint properties. In the second step, stepwise multivariate discriminant function analysis (DFA) was used to identify the optimum combination of the tracers passing the Kruskal–Wallis H-test and to maximize discriminating between the potential sources. The multivariate mixing model (Walling, 2005) involves minimizing the sum of the squares of the residuals between predicted tracer values for each source in sediment samples and the observed values. The sediment source apportionment involved a comparison of the results obtained using several multivariate mixing models. Using an optimization source proportion minimizes the errors in mixing models. We minimized the sum of the squares of the relative errors (R) in the objective functions( Eq.s 1-5). Eq.1: Eq. 2: Eq. 3: Eq. 4: Eq. 5: where: ci = concentration of fingerprint property (i) in sediment samples; Sij = concentration of fingerprint property (i) in source category (j); X j = percentage contribution from source category (j); Z j = particle size correction factor for source category (j); Oj = organic matter content correction factor for source category (j); Wi = tracer discriminatory weighting or tracer specific weighting; SVji = weighting representing the within-source variability of fingerprint property (i) in source category (j); VARij = variance of the measured values of tracer in source area j; mj = the total number of samples for an individual source; n = number of fingerprint properties; m = number of sediment source categories. Genetic Algorithm optimization (GA) was employed to find the optimal source sediments contribution. In addition, goodness of fit (GOF) equation and Mean Error (ME) were used to determine the results of each of the mixing models (Eq.6 and Eq.7) Eq.6: Discussion Soil erosion and sediment yield are the most destructive phenomena that cause a lot of damages in different regions. However, in order to combat them, it is needed to be aware of the sediment sources location in the region. Sediment fingerprinting technique, based on geochemical tracers, organic and isotopic ratios, and various mixing models, is used in the recognition of the contribution of the different sediment sources in an area. In this study, the optimum combination of organic and rare tracers was used to separate the different sources. In addition, to determine the contribution of this erosion and sediment yield resource, Collins, Collins modified, Motha, Landwehr and Slattery models associated with genetic algorithm optimization were used. The results of the discriminant analysis showed Compounds of C, Cu Si, and Ti as tracers for land uses and four tracers (Nd143/144, Cu, Si,Ti) to discriminate between geology formation’s source categories. To determine the best model, GOF and ME indexes were used. Tables1-4 render the results of applying the ME and GOF indices to select the best models in formation and land use units. The M Collins and Collins mixing models with GOF and ME indices of 99.95%, 99.996% and 99.16%, 99.977% were respectively selected as the best models in land use and formation units. According to ME and GOF results, the calculated relative contributions of the range lands and the Asmari formation with 65% and 56.5% were the highest. Moreover, sedimentation rates of sub basins number 6 and 5 with 59.11% and 58.7% were very important in the management of the soil conservation (the highest proportion in sediment and erosion basins) and sub basins number 31 with 7.54% were not important in the management of the soil conservation (minimal role in Sediment yield of Tange Bostanak watershed). Conclusion Soil erosion and sediment yield are the most destructive phenomena that cause a lot of damages in different regions. However, in order to combat them, it is needed to be aware of the sediment sources location in the region. Sediment fingerprinting technique, based on geochemical tracers, organic and isotopic ratios, and various mixing models, is used in the recognition of the contribution of the different sediment sources in an area. In this study, the optimum combination of organic and rare tracers was used to separate the different sources. In addition, to determine the contribution of this erosion and sediment yield resource, Collins, Collins modified, Motha, Landwehr and Slattery models associated with genetic algorithm optimization were used. The results of the discriminant analysis showed Compounds of C, Cu Si, and Ti as tracers for land uses and four tracers (Nd143/144, Cu, Si,Ti) to discriminate between geology formation’s source categories. The M Collins and Collins mixing models with GOF and ME indices of 99.95%, 99.996% and 99.16%, 99.977% were respectively selected as the best models in land use and formation units. According to ME and GOF results, the calculated relative contributions of the range lands and the Asmari formation with 65% and 56.5% were the highest. Moreover, sedimentation rates of sub basins number 6 and 5 with 59.11% and 58.7% were very important in the management of the soil conservation (the highest proportion in sediment and erosion basins) and sub basins number 31 with 7.54% were not important in the management of the soil conservation (minimal role in Sediment yield of Tange Bostanak watershed).