In each of the eight research catchments, belt transects, consisting of contiguous 10 x 20 m quadrats, have been established at the discharge site, usually a saltlake. Typically, each transect is deployed at right angles from the shoreline, across the fringing samphire zone, into adjacent woodland/shrubland. Lengths vary from 50 m to 130 m. At each discharge site, there are at least two transects. An additional two transects act as controls at a nearby wetland, not exposed to groundwater discharge, i.e. in control-impact terms, there is one control transect for every impact transect in the study.
Surveys are carried out at annual intervals, ideally with a baseline survey immediately before construction begins. Along each transect, the pattern of vegetation is described – community composition (with identifications to species or subspecies level), percentage cover, height, layering. Topographic features (e.g. depressions, crests, slope changes) and soil profiles are also described.
At the start of the monitoring program, we expected that impacts on the wetland vegetation (probably associated with severe changes in hydroperiod and with chemical pollutants) would be gross and conspicuous, such as massive death, or clear shifts in community zonation. On the contrary, and somewhat unexpectedly, we have observed no vegetation changes that can be ascribed unequivocally to groundwater discharge. If discharge is indeed affecting the vegetation, then its influence is more subtle than anticipated, and will be revealed only by rigorous statistical analysis.