Hydraulic conductivity in upland blanket peat: measurement and variability

A key parameter used in wetland hydrological and landform development models is hydraulic conductivity. Head recovery tests are often used to measure hydraulic conductivity but the calculation techniques are usually confined to rigid soil theory. This is despite reports demonstrating the misapplication of rigid soil theory to non-rigid soils such as peats. While values of hydraulic conductivity calculated using compressible techniques have been presented for fenland peats these data have never, to the authors’ knowledge, been compared to such calculations in other peat types. Head recovery tests (slug withdrawal) were performed on piezometers at depths ranging from 10 cm to 80 cm from the surface on north Pennine blanket peats. Results were obtained using both rigid and compressible soil theories allowing comparison of the two techniques. Compressible soil theory gives values for hydraulic conductivity that are typically a factor of five times less than rigid soil calculations. Hydraulic conductivity is often assumed to decrease with depth in upland peats but at the study site in the northern Pennines it was not found to vary significantly with depth within the range of peat depths sampled. The variance within depth categories was not significantly different to the variance between depth categories showing that individual peat layers did not have characteristic hydraulic conductivity values. Thus large lateral and vertical differences in hydraulic conductivity over short distances creates problems for modelling but may help account for the high frequency of preferential flow pathways within what is otherwise a low matrix hydraulic conductivity peat. Hydraulic conductivity was found to vary significantly between sampling sites demonstrating that hillslope or catchment-scale variability may be more important than plot-scale variability. Values for compressibility of the peats are also reported. These generally decline with depth and also vary significantly between sampling sites. There are implications for the way in which measurements of hydraulic conductivity and other properties of blanket peat are interpreted as the effects of environmental change in one part of a peat catchment may be very different to those in another.