Fallout Radionuclides - Application of Geochemical Tracers to Fluvial Sediment - page 67

3.3.2.2 Uncertainties

The assessment of erosional and depositional magnitudes using of the FRN inventory
approach has been applied in a wide range of geologic, physiographic, and climatic
settings, and is nowwidely accepted bymany, if not most, in the scientific community.
Acceptance of the method is based in part on the validation of its results using data
provided by (1) traditional approaches such as erosion plots, erosion pins, and the
monitoring of catchment sediment yields (e.g., Elliott et al. 1990 ; Porto et al. 2001 ,
2003a , b ;Mabitetal. 2002 ; Wallbrink and Croke 2002 ; Schuller et al. 2006 ; Ceaglio
et al. 2012 ), empirical erosion models (e.g., the USLE and RUSLE) (Di Stefano et al.
1999 ; Saç et al. 2008 ; Mabit et al. 2008 ), and other FRNs (e.g., 210 Pb ex ,

240 Pu)
(Hoo et al. 2011 ; Ketterer et al. 2011 ; Matisoff and Whiting 2011 ). Moreover, trends
in the generated results often fit what is intuitively expected on the basis of land-
use and physiographic characteristics of the site. Nonetheless, some investigators
have questioned the validity of the generated results. Parsons and Foster ( 2011 ),
for example, concluded that no current rates of soil erosion that are based upon the
use of this technique are reliable, and that 137 Cs cannot be used to provide reliable
information about rates of soil erosion (see Mabit et al. 2013 for a rebuttal). Many of
concerns raised by Parsons and Foster ( 2011 ) are linked to three of the fundamental
assumptions upon which the FRN method (including 137 Cs) is based, including that:
(1) 137 Cs was uniformly distributed over the landscape, and its distribution was not
modified before it was bound to sedimentary particles, (2) 137 Cs is rapidly, strongly
and irreversibly bound to sediment and/or organic materials, such that the observed
differences in its inventories reflect the movement of soil particles, and (3) estimates
of soil erosion or deposition can be accurately estimated frommeasurements of 137 Cs
inventories using the available conversion models. These assumptions are explored
in more detail below.

239 ,

1. Cesium-137 (and presumably other FRNs) are Initially Distributed Uniformly
Across the Study Area
An important assumption inherent in the approach is that the FRNs are initially
distributed uniformly across the study area. This allows for any spatial deviation in
inventories from the reference site to be assigned to the redistribution of the FRNs
after their deposition on and sorption to soil particles. Parsons and Foster ( 2011 )
argued that the assumption of an initially uniform distribution of 137 Cs (and,
presumably other FRNs) is invalid. To support their contention, they argued that if
the assumption was strictly observed the variability in FRNs between cores of the
reference sites would be similar. However, in the case of 137 Cs, spatial variations
in inventories of as much as 40% have been recorded (Wallbrink et al. 1994 ;
Sutherland 1994 ; Wallbrink and Murray 1996 ). Parsons and Foster ( 2011 ) note
that part of the variability is related to the non-uniform atmospheric deposition of
137 Cs (and presumably 210 Pb ex ,

7 Be, etc.) by means of wet deposition as a result
of topographic and geographical patterns in precipitation amounts and intensities.
In other words, variations in precipitation across the landscape are likely to result
in the spatially non-uniform fallout of 137 Cs and other FRNs to the soil surface.