Comparison of Hydrology of Two Atlantic Coastal Plain Forests

Abstract

Abstract. This article compares the short-term and long-term hydrology of two typical forests in the humid Atlantic Coastal Plain, including a relatively undisturbed forest with natural drainage in South Carolina (SC) and a drained pine plantation in North Carolina (NC), using monitoring and modeling approaches. Highly dynamic outflow (O) from both of these systems is driven by the water table (WT) position, as influenced by rainfall (R) and evapotranspiration (ET). The annual runoff coefficient (ROC) varied from 5% in dry years to 56% in wet years, depending on the soil water storage (SWS), with a significantly higher average value for the NC site despite its deeper WT, on average, than the SC site. Although both sites behaved similarly in extreme climate conditions, the change in SWS above the WT influenced the annual RO, ROC, and ET. The 17-year average annual ET of 1114 mm (R – O, assuming annual balanced SWS) for the SC site was significantly higher (p = 0.014) than the ET of the drained NC site (997 mm) despite the SC site’s lower mean annual R of 1370 mm, compared to 1520 mm for the NC site. This may be due to both the higher potential ET (PET) and soil water-holding capacity of the SC site. The SC site had higher frequency and duration of WT near the surface during winter, deeper summer WT, and higher correlation of annual ET to annual R (r2 = 0.90 vs. 0.15), suggesting that the SC site was often moisture-limited, particularly during the growing season. Most of the streamflow in these systems occurred during winter, with low ET demands. However, summer periods with tropical storms also resulted in large RO events, generally with higher frequency and longer durations at the drained NC site. These results are similar to an earlier short-term comparison with an unstable behavior period at the SC site after Hurricane Hugo (1989). This study highlighted (1) the differences in hydrology between coastal forests drained for silvicultural production and undrained natural forests managed only for restoration, (2) the importance of long-term monitoring and the effects of regeneration as well as vegetation management on flow regime, and (3) the application and limitations of two widely used models (MIKESHE and DRAINMOD) in describing the hydrology of these forests. Long-term studies can be a basis for testing new hypotheses on water yield, stormwater management, wetland hydrology, vegetation restoration, bioenergy production, and climate change, in addition to applications of proper models for assessing the eco-hydrologic impacts of land use and climate change on freshwater coastal forests linked with downstream riparian rivers and estuaries affected by tidal fluxes and sea level rise.HighlightsOutflow, driven by water table position on these forest systems, is highly variable, depending on its soil water storage.The hydrologic responses of both forest sites were similar during extreme climatic events or disturbances.Effect of forestry drainage on runoff was obscured by its large interannual differences.Long-term monitoring provides better insights on climate and vegetation management effects on flow regime and model validation Keywords: Drainage, Evapotranspiration, Hydrologic models, Pine forest, Poorly drained soils, Runoff coefficient, Water table.