ch05

As one of only a number of fallout radionuclides, ¹³⁷Cs has attracted particular attention for application in soil erosion investigations by virtue of its high affinity for sediment particles, its relatively long half-life, its ease of measurement and the well defined temporal pattern of fallout input. As such, it has effectively 'monopolized' most, if not nearly all, recent work and there have been very few attempts to use other fallout radionuclides that might offer similar or complementary opportunities. Thus, for example, the possibility of combining measurements of a natural (as distinct from man made) fallout radionuclide characterized by an essentially constant fallout input with those of ¹³⁷Cs whose input was primarily restricted to the period extending from the late 1950s into the early 1970s, could offer considerable scope for interpreting the erosional history of the study site. Furthermore, such a tracer could prove invaluable in areas where Chernobyl fallout has seriously complicated the interpretation of radiocaesium inventories. Equally, the availability of a fallout radionuclide with a short half-life could offer potential for investigating the short term behaviour and dynamics of erosional processes. Any fallout radionuclide used in this manner as a sediment tracer would clearly need to be rapidly and strongly fixed on reaching the soil surface; two which would appear to offer potential are ²¹⁰Pb and⁷Be. Figure 5, based on the work of Walling and Woodward [45], illustrates a typical example of the distribution of ¹³⁷Cs ²¹⁰Pb and⁷Be in soils from both cultivated and pasture areas near Exeter, Devon, UK, which emphasizes the similar behaviour of these three fallout radionuclides in terms of their fixation by the surface horizons of the soils involved. Some variations in their depth distributions are, however, evident and these reflect the different half-lives and the different fallout histories of the individual radionuclides. Further discussion of the potential for using ²¹⁰Pb and⁷Be in soil erosion investigations can usefully consider each of these radionuclides in turn.

5.1. Use of unsupported ²¹⁰Pb to estimate soil redistribution

Lead-²¹⁰ is a natural product of the 238U decay series, with a half-life of 22.26 years. It is derived from the decay of gaseous 222Rn, the daughter of ²²⁶Ra. Radium-226 occurs naturally in soils and rocks and will generate ²¹⁰Pb which will be in equilibrium with its parent. Diffusion of a small quantity of the 222Rn from the soil introduces ²¹⁰Pb into the atmosphere, and its subsequent fallout provides an input of this radionuclide to surface soils and sediments which is not in equilibrium with its parent ²²⁶Ra. This fallout component is termed 'unsupported' or 'excess' ²¹⁰Pb since it cannot be accounted for (or supported) by decay of the in situ parent. The amount of unsupported or atmospherically derived ²¹⁰Pb in a sediment sample can be calculated by measuring both ²¹⁰Pb and ²²⁶Ra and subtracting the supported or in situ component. Recent developments in the use of low background, low energy gamma spectrometry [49, 50] make such measurements relatively easy to undertake.

Although ²¹⁰Pb has been widely used for dating lake sediment cores [51, 52], its potential for estimating soil erosion rates has been essentially ignored to date. As a fallout radionuclide that is rapidly and strongly adsorbed by the surface soil, unsupported ²¹⁰Pb will behave in a similar manner to ¹³⁷Cs except that its fallout input is essentially constant through time and the supply to the soil surface is being continuously replenished. This explains the slight differences in the depth distributions of the two radionuclides evident in Fig. 5. Because of the continuous surface replenishment, unsupported ²¹⁰Pb concentrations are greatest at the surface, whereas in the case of ¹³⁷Cs the lack of significant replenishment over the past 20 years, coupled with biological activity within the soil and some downward diffusion, have caused the level of maximum concentration to be displaced a few centimetres below the surface. Soil inventories associated with unsupported ²¹⁰Pb are generally somewhat greater than those of ¹³⁷Cs

FIG. 6. Spatial pattern of (a) ¹³⁷Cs and (b) unsupported ²¹⁰Pb inventories in a field at Butsford Barton, near Colebrooke, Devon, UK

FIG. 7 Long term soil redistribution rates within a field at Butsford Barton, near Colebrooke, Devon, UK, estimated from the unsupported ²¹⁰Pb inventory data presented in Fig. 6.

Figure 6 emphasizes the similar response of ¹³⁷Cs and unsupported ²¹⁰Pb to erosional redistribution within a small field at Butsford Barton near Colebrooke, Devon, UK. The maps presented are based on data assembled for a grid of > 200 soil cores and in this location the reference inventories for ¹³⁷Cs and unsupported ²¹⁰Pb were estimated to be about 275 mBq/cm² and about 600 mBq/cm², respectively. Considering the field as a whole, the loss of ¹³⁷Cs and unsupported ²¹⁰Pb relative to their respective reference inventories amounts to 12.7 and 16.7%, respectively. The increased loss and the slightly greater range of inventory values associated with the unsupported ²¹⁰Pb measurements reflect the continuous input of this radionuclide and thus the annual potential for erosion of newly accumulated ²¹⁰Pb from the soil surface prior to cultivation, and the longer period available for removal of this radionuclide. The unsupported ²¹⁰Pb inventory values depicted in Fig. 6(b) have been converted to estimates of the long term soil redistribution rate by using a theoretically derived calibration relationship, similar to that employed by Walling and Quine [11] to convert measured ¹³⁷Cs inventories to estimates of erosion and deposition rates. In this case, the model employed incorporated values for the annual ²¹⁰Pb deposition flux derived from the reference inventory, the plough depth, and the initial depth distribution of fresh fallout prior to incorporation into the profile by tillage, and assumed that cultivation had been continuous for > 100 years. The result is presented in Fig. 7, which shows clear evidence of sediment accumulation in the central depression that traverses the field and against the field boundaries. In addition, the influence of tillage in displacing soil from the slope convexities is also clearly evident. Further scope undoubtedly exists for interpreting the relative magnitude of the unsupported ²¹⁰Pb and ¹³⁷Cs inventories for individual sites to provide information on the erosional history of the site. Whereas the former will have been influenced by soil redistribution over the past 100 years or more, the latter will reflect only soil redistribution that has occurred since the late 1950s.

5.2. Use of⁷Be to investigate soil erosion dynamics

Beryllium-7 is a cosmogenic radionuclide produced in the upper atmosphere by cosmic ray spallation of nitrogen and oxygen. In this case, the radionuclide is extremely short lived (half-life of 53.3 days) relative to ¹³⁷Cs and ²¹⁰Pb and it offers potential for investigating soil erosion dynamics over much shorter timescales. Figure 5 indicates that⁷Be is typically concentrated in the upper 5 mm of the soil profile and it is therefore capable of providing good discrimination between sediment derived from the immediate soil surface and that derived from depths > 10 mm, where concentrations will be effectively zero. Burch et al. 153] and Wallbrink and Murray [54] working in Australia were, for example, able to use this radionuclide to distinguish sediment mobilized by sheet erosion and rill erosion in erosion plot experiments and to demonstrate the progressive development of rills during the course of a single simulated runoff event. Scope clearly also exists for coupling measurements of⁷Be activity with equivalent measurements of ¹³⁷Cs and unsupported ²¹⁰Pb to provide increased capacity for sediment source discrimination [45]. Monitoring of the spatial distribution of⁷Be activity across small plots, immediately after storm events, could also provide a basis for investigating local erosion patterns in response to microtopographical control.