Changes in vegetative communities and water table dynamics following timber harvesting in small headwater streams

In order to better understand the relationship between vegetation communities and water table in the uppermost portions (ephemeral–intermittent streams) of headwater systems, seasonal plot-based field characterizations of vegetation were used in conjunction with monthly water table measurements. Vegetation, soils, and water table data were examined to determine potential indicator species of near-surface water that could be used in rapid delineation of Streamside Management Zones (SMZs) by forest managers. Twelve watersheds were instrumented with three hundred screened wells, installed in grids of 25 per sub-watershed. Well locations were used to monitor water table and vegetation communities. Species were classified according to their wetland indicator status for Region 2; communities were evaluated using a prevalence index (PI). As part of a larger study, the uppermost reaches of the headwater systems were treated one of four treatments: (1) removal of all merchantable stems leaving understory intact with minimum surface soil disturbance (BMP1), (2) the same as treatment BMP1with the addition of logging debris in the drainage channel (BMP2), (3) total harvest with no BMPs applied (clearcut) and (4) no harvest (reference). Post-harvest increases in water table ranged from 1.6 cm in BMP1 to 28.2 cm in clearcut treatments during 2008, from 10.5 cm in BMP1 to 54.2 cm in BMP2 during 2009. PI differed significantly between channel and hillslope positions and represent distinctive vegetation communities. Forest clearcutting affected vegetation communities through combined direct and indirect disturbances. PI in the clearcut did not respond directly to changes in water table. In the two treatments where BMPs were employed, changes in vegetative communities corresponded to both changes in water table and changes in the microclimate as a result of harvesting intensity and changing stand heterogeneity. A vegetative indicator analysis, based on the presence of saturated soil conditions and water table elevation, was used to determine potential indicators of the true hydrologic boundaries of the headwater streams. Three potential indicator species (Viola blanda, Ludwigia glandulosa, and Arundinaria gigantean Ssp. Tecta) were more prevalent within ‘‘wet’’ channel positions but exhibited less total frequency of occurrence across the study areas than the fern species Polystichum acrostichoides, which is locally used for rapid estimation of intermittent stream extent). The combined use of the strong indicator species identified in this study and the ‘‘fern line’’ used by local industry foresters provides a means for rapid assessment of hydrologically functional SMZs in these headwater streams. 2012 Elsevier B.V. All rights reserved.