An analysis of solute transport on a harvested hillslope in the southern Appalachian Mountains

Interest in transport of dissolved nitrogen (N) and carbon (C) in forested ecosystems is growing because of potential effects of these solutes on streamwater quality and implications for C sequestration. Additional research will further the understanding about the dynamics of these soil solutes, particularly in response to harvesting of forests. Also, the purported role of riparian buffers, where logging is restricted along stream channels, in retaining soil solutes is not well studied in the steeply sloping terrain of the southern Appalachian Mountains. I examined solute transport in a first-order watershed in the Nantahala National Forest in North Carolina that was harvested in February 2006 with retention of a 10-m riparian buffer.

To quantify the movement of dissolved inorganic nitrogen (DIN), dissolved organic nitrogen (DON), and dissolved organic carbon (DOC), four transects of lysimeters, approximately 30 m apart, were installed perpendicular to the stream on one hillslope. Porous ceramic cup (2-bar) lysimeters were installed in each transect 1, 4, 10, 16, 30, and 50 m from the stream in the A horizon and B horizon, and 4, 16, and 50 m from the stream in the saprolite layer. Samples were removed from the lysimeters 24 hr after 50 centibars of tension were placed on them, and riparian groundwater well and stream samples were collected at the same time as lysimeter samples. Collection of samples from the lysimeters, wells, and stream occurred every four to six weeks for one calendar year beginning March 2007. A 16-wk laboratory N mineralization study was conducted on A horizon soils.

Mean nitrate values in the soil solution of the A horizon in the spring were 1.53mg-N/L and decreased through the growing season to 0.030mg-N/L. Mean soil solution nitrate values in the B horizon and saprolite layer were 0.40mg-N/L in the spring and summer and decreased to 0.031mg-N/L in the winter. Mean soil solution ammonium concentrations were higher in the A horizon (0.090mg-N/L) than the B horizon and saprolite layer (0.034mg-N/L) and were lowest during the summer and fall. Dissolved organic C was significantly higher in the A horizon, with values ranging from 2.3mg/L to 599mg/L, than in the relatively stable B horizon and saprolite (1.9mg/L to 36.6mg/L). Dissolved organic C was logarithmically correlated to DON (r2 = 0.64), and DON values were highest in the A horizon (0.70mg/L). Cumulative N mineralization potential ranged from 48.1mg-N/kg to 75.6mg-N/kg and was not a useful predictor for nitrate soil solution values.

Nitrate leached vertically, and a large percentage of nitrate was stored in the B horizon and saprolite. Ammonium, DON, and DOC did not appear to leach vertically because they did not increase in the B horizon or saprolite layer. Ammonium, DON, and DOC are less mobile in soil solution than nitrate. The 10-m riparian zone had little impact on nitrate, ammonium, DON, and DOC removal. Nitrate remaining in the A horizon was likely removed through plant uptake in the harvested area before reaching the riparian zone. There was no detectable difference between ammonium concentrations in the harvested area and riparian zone likely because of limited mobility. The riparian zone did not remove excess DON or DOC, and in some transects was a source of DON and DOC. Nitrate and DOC concentrations were highly variable among transects and locations within transects. This may be caused by sensitivity of these solutes to site heterogeneity. This suggests that a large number of lysimeters should be used to account for this variability in future studies to ensure accuracy.

This study observed limited vertical leaching of ammonium, DON, and DOC through the profile. However, excess nitrate was observed moving from the A horizon into the B horizon and saprolite layer, suggesting the potential for delivery to the stream via subsurface transport and the need for attenuation of nitrate by the riparian zone. Because of low concentrations of nitrate entering the riparian zone during this study, the capacity for riparian attenuation of nitrate was not demonstrated.