Controls on Near‐Surface Hydraulic Conductivity in a Raised Bog

• Published in: Water resources research 2019-02, Vol.55 (2), p.1531-1543
• Main Authors: Morris, Paul J., Baird, Andy J., Eades, Phil A., Surridge, Ben W. J.
• Format: Article
• Language: English
• Subjects: Bogs; Brackishwater environment; Bulk density; Decomposition; Drainage ditches; Drainage measurement; Estuaries; Estuarine environments; Groundwater table; Hydraulic conductivity; Hydraulics; loglinear; multiple regression; Multiple regression models; Peat; peatland; Peatlands; permeability; Prediction models; Regression analysis; Regression models; Restoration; Shallow water; Soil; Stocks; von post; Water resources; Water table
• ISSN: 0043-1397; 1944-7973
• DOI: 10.1029/2018WR024566

Shallow water tables protect northern peatlands and their important carbon stocks from aerobic decomposition. Hydraulic conductivity, K, is a key control on water tables. The controls on K, particularly in degraded and restored peatlands, remain a subject of ongoing research. We took 29 shallow (~50 cm) peat cores from an estuarine raised bog in Wales, UK. Parts of the bog are in close‐to‐natural condition, while other areas have undergone shallow peat cutting for fuel and drainage, followed by restoration through ditch blocking. In the laboratory we measured horizontal (Kh) and vertical (Kv) hydraulic conductivity. We fitted linear multiple regression models to describe log10‐transformed Kh and Kv on the basis of simple, easy‐to‐measure predictors. Dry bulk density and degree of decomposition were the strongest predictors of Kh and Kv. Perhaps surprisingly, the independent effect of hummocks was to produce higher‐Kv peat than in lawns; while the independent effect of restored diggings was to produce higher‐K peat than in uncut locations. Our models offer high explanatory power for Kh (adjusted r2 = 0.740) and Kv (adjusted r2 = 0.787). Our findings indicate that generalizable predictive models of peat K, similar to pedotransfer functions for mineral soils, may be attainable. Kh and Kv possess subtly different controls that are consistent with the contrasting roles of these two properties in peatland water budgets. Our near‐surface samples show no evidence for the low‐K marginal peat previously observed in deeper layers at the same site, indicating that such structures may be less important than previously believed. Key Points Peat hydraulic conductivity is well explained by a small number of simple, easy‐to‐measure descriptors Horizontal and vertical hydraulic conductivity possess subtly different controls that reflect contrasting roles in peatland water budgets Generalizable models of peat hydraulic conductivity using simple descriptors may be attainable through meta‐analysis of published data