Low phosphorus release but high nitrogen removal in two restored riparian wetlands inundated with agricultural drainage water

TY - JOUR

T1 - Low phosphorus release but high nitrogen removal in two restored riparian wetlands inundated with agricultural drainage water

AU - Hoffmann, Carl Christian

AU - Heiberg, Lisa

AU - Audet, Joachim

AU - Schønfeldt, Boris

AU - Fuglsang, Ann

AU - Kronvang, Brian

AU - Ovesen, Niels Bering

AU - Kjærgaard, Charlotte

AU - Hansen, Hans Chr. Bruun

AU - Jensen, Henning S.

PY - 2012/9

Y1 - 2012/9

N2 - Re-established riparian wetlands used to mitigate nitrogen (N) loss from agricultural soils to surface water may lose phosphorus (P) from the top soils that often have received fertilizers. This could lead to eutrophication of lakes and estuaries. For a 2-year period we established mass balances of N and P in two restored riparian wetlands of ~0.6ha situated on mineral soil. Monitoring began 5 years after restoration. Both wetlands received drainage water from upland agricultural fields rich in nitrate (1.5-12.3mgNL -1) and low in total P (TP) (0.016-0.04mgPL -1). Water balances were reasonably accounted for (15% imbalance at most). Water passed the wetlands as sheet flow without exchange with groundwater because of clay horizons in sub-soils, and sheet pilings along the stream banks allowed continuous measurements of inflow and outflow. The Egeskov riparian wetland (wetland:upland ratio 0.13) removed 121 and 28kgNha -1yr -1 (43 and 75% of the load) and retained 0.08kgPha -1 (6% of the load) in year one and had a net release of 0.15kgPha -1 (25% of the load) in year two. The Stor å riparian wetland (wetland:upland ratio 0.02) removed 229 and 158kgNha -1yr -1 (32 and 26%). Net releases of P were 0.33 and 0.90kgPha -1yr -1 (22 and 127%). Nitrogen removal rates are on par with published rates for similar wetlands, while the P release rates appear surprisingly low. Phosphate outlet concentrations resembled the equilibrium concentrations (EPC 0) where no phosphate exchange occurred between top soils and drainage water, suggesting that P release or retention was controlled by phosphate adsorption. This value was 0.015mgPL -1 for Egeskov and 0.047mgPL -1 for Stor å. The high phosphate affinity was probably governed by high ratios between oxidized iron and iron-bound P. The top soils (10cm) contained 87 and 201kgPha -1 as iron-bound P and herbaceous vegetation accumulated 10.7 and 16.5kgPha -1yr -1. These figures are 55-136 and 8-11 times higher than the annual P-load to the wetlands, and we suggest that annual harvest of vegetation could maintain or even improve the P retention capacity of these wetlands.

AB - Re-established riparian wetlands used to mitigate nitrogen (N) loss from agricultural soils to surface water may lose phosphorus (P) from the top soils that often have received fertilizers. This could lead to eutrophication of lakes and estuaries. For a 2-year period we established mass balances of N and P in two restored riparian wetlands of ~0.6ha situated on mineral soil. Monitoring began 5 years after restoration. Both wetlands received drainage water from upland agricultural fields rich in nitrate (1.5-12.3mgNL -1) and low in total P (TP) (0.016-0.04mgPL -1). Water balances were reasonably accounted for (15% imbalance at most). Water passed the wetlands as sheet flow without exchange with groundwater because of clay horizons in sub-soils, and sheet pilings along the stream banks allowed continuous measurements of inflow and outflow. The Egeskov riparian wetland (wetland:upland ratio 0.13) removed 121 and 28kgNha -1yr -1 (43 and 75% of the load) and retained 0.08kgPha -1 (6% of the load) in year one and had a net release of 0.15kgPha -1 (25% of the load) in year two. The Stor å riparian wetland (wetland:upland ratio 0.02) removed 229 and 158kgNha -1yr -1 (32 and 26%). Net releases of P were 0.33 and 0.90kgPha -1yr -1 (22 and 127%). Nitrogen removal rates are on par with published rates for similar wetlands, while the P release rates appear surprisingly low. Phosphate outlet concentrations resembled the equilibrium concentrations (EPC 0) where no phosphate exchange occurred between top soils and drainage water, suggesting that P release or retention was controlled by phosphate adsorption. This value was 0.015mgPL -1 for Egeskov and 0.047mgPL -1 for Stor å. The high phosphate affinity was probably governed by high ratios between oxidized iron and iron-bound P. The top soils (10cm) contained 87 and 201kgPha -1 as iron-bound P and herbaceous vegetation accumulated 10.7 and 16.5kgPha -1yr -1. These figures are 55-136 and 8-11 times higher than the annual P-load to the wetlands, and we suggest that annual harvest of vegetation could maintain or even improve the P retention capacity of these wetlands.

U2 - 10.1016/j.ecoleng.2012.04.039

DO - 10.1016/j.ecoleng.2012.04.039

M3 - Journal article

SN - 0925-8574

VL - 46

SP - 75

EP - 87

JO - Ecological Engineering

JF - Ecological Engineering

ER -