The Role of Correction Factors in Sediment Source Fingerprinting of the Lake Urmia Sand Dunes

  Over the last decades, sediment fingerprinting technique against the experimental models for erosion and deposition processes is now routinely used for its higher reliability and lower uncertainties. Its reliable information give the best indication of sediment yield produced by spatial sources of a catchment and let authorities know how take conservative operations and proper actions across the catchment to stop the soil erosion. Therefore, identification of the dominant processes and sources generating the sediment within its catchment are vital. The western shore of the Lake Urmia, NW Iran, the world’s second largest hyper-saline lake has now retreated more than 7 Km from the shore and across its western margin, sand dunes and sand ridges have been appearing. Finding out the geomorphological/lithological units as the sediment feeders out of its western catchment using geochemical data was done. As the main aim of the present research, need to correction factors including particle size, organic matter and tracer discriminatory weighting was applied in recognition of potential changes in fingerprint properties during sediment delivery.
. . . . . . . . . . . . . . .
Material and The mixing model algorithm was used to estimate the relative contributions from the potential sediment sources by minimising the sum of squares of the weighted relative errors. OF=∑_(i=1)^n▒0((S_Sink-(∑_(j=1)^m▒〖S_Source .P_s.Z_s.O_s)〗)/S_Sink )^2 W_i
0≤P_S≤1 ∑_(j=1)^m▒〖P_S=1〗
SSink: concentration of fingerprint property (i) in the sediment collected from the outlet;
PS: the percentage contribution from source category (s);
SSource: mean concentration of fingerprint property (i) in source category (s);
ZS: particle size correction factor for source category (s);
OS: organic matter content correction factor for source category (s);
Wi: tracer discriminatory weighting;
n: number of fingerprint properties comprising the composite fingerprint;
m: number of sediment source categories;
. . . . . . . . . . . . . .
The above algorithm has incorporated three correction factors to reflect the impact of element concentration in given sediment load size. The effect of the correction factors into the fluvial and alluvial sediment loads has been approved, what is has not been well understood for aeolian sediments and desert environments. Therefore, the role of the correction factors in estimation of proportion of each potentially sediment source is inferred. Paired t-student statistical method was applied to be seen whether differences between being of correction factors and not being correction factors is seen.
. . . . . . . . . . . . . . As the paired t-student method results show, there is not significant differences between the source contribution before using the correction factors and after using the correction factors. But it is an statistically result and mixing model or objective function results has another story. According to Table 2, before using the correction factors, Qmf and Qt geomorphological/lithological units with 47.76% and 52.24% have the highest proportion in generating the sediment load of the catchment respectively. After implementation of the correction factor, Qf and Klshi geomorphological/lithological units with 67.5% and 32.5% have the highest contribution respectively. So, different source proportion was seen with no significant statistic results.
. . . . . . . . . . . . . . . The present research successfully interpreted the impact of correction factors on sediment source contribution of the Lake Urmia sand dunes. These correction factors are now widely used into the mixing model or objective function to improve the comparability of source and sediment samples. It is inferred the organic matter correction factor could have used while mineral-magnetism properties of samples are known as the tracers. The particle size correction factor due to its strong influence on many of the tracers used for fingerprinting is applied, when the relation of grain size to each tracer's concentration is tested. With generating a scatter plot of particle size or organic matter content against tracer concentration for each source group, necessity of correction factor is evaluated. Generally, it is interpreted applying the correction factors is vital when some other parameters including sediment environments, tracer properties, chronology of sediments, particle size of sediment loads etc. is preliminary evaluated