In situ radiometric mapping of soil erosion and field-moist bulk density on cultivated fields

Recent developments in in situ gamma ray spectrometry offer a new approach to measuring the activity of radionuclides such as Cs-137 and K-40 in soils , and thus estimating erosion or deposition rates and field moist bulk dens ity (rho (m)). Such estimates would be rapid and involve minimal site distu rbance, especially important where archaeological remains are present. This paper presents the results of a pilot investigation of an eroded field in Scotland in which a portable hyper pure germanium (HPGe) detector was used to measure gamma ray spectra in situ. The gamma (gamma) photon flux observe d at the soil surface is a function of the Cs-137 inventory, its depth dist ribution characteristics and rho (m). A coefficient, Qc(s), derived from th e forward scattering of Cs-137 gamma ray photons within the soil profile re lative to the Cs-137 full energy peak (662 keV), was used to correct the in situ calibration for changes in the Cs-137 vertical distribution in the pl oughed field, a function of tillage, soil accumulation and rho (m). Based o n only 8 measurements, the agreement between in situ gamma ray spectrometry and soil sample measurements of Cs-137 inventories improved from a non sig nificant r(2)=0.05 to a significant r(2)=0.62 (P<0.05). Erosion and deposit ion rates calculated from the corrected in situ Cs-137 measurements had a s imilarly good agreement with those calculated from soil cores. Mean soil bu lk density was also calculated using a separate coefficient, Q(K), derived from the forward scattering <gamma> photons from K-40 within the soil relat ive to the K-40 full energy peak (1460 keV). Again there was good agreement with soil core measurements (r(2)=0.64; P<0.05). The precision of the in s itu Cs-137 measurement was limited by the precision with which Qc(s) can be estimated, a function of the low Cs-137 deposition levels associated with the weapons testing fallout and relatively low detector efficiency (35%). I n contrast, the precision of the in situ <rho>(m) determination was only li mited by the spatial variability associated with soil sampling.