The catastrophic dieback of Typha domingensis in a drained and restored East Mediterranean wetland: re-examining proposed models.

We experimentally tested two geochemical models which have been proposed as potential mechanisms leading to the catastrophic dieback of Typha domingensis in constructed Lake Agmon within the drained Hula peatland, northern Israel. An elaborate lysimeter station simulating the conditions imposed by sulphide toxicity and P deficiency models was used to test them experimentally. Rhizosphere pH and redox potential were monitored in situ in real time throughout the summers of 2007 and 2008. A comparative study of redox-reduction simulation was coupled with periodic sampling and analysis of sulphide and other reducing species. N and P were determined in the plant tissues to compute the N:P nutrient limitation index and evaluate nutrient deficiencies. The sulphide toxicity model was not accepted as a viable mechanism because Typha domingensis stands did not show any signs of stress, even when growing in rhizosphere with a sulphide concentration three times that during the actual dieback in 1996. The P limitation model was not supported by the N:P index, which indicated N (<14) rather than P (>16) limitation. Since Typha has now returned and is thriving in Lake Agmon and adjacent drainage canals, we suggest a self-thinning mechanism followed by normal succession of macrophytes in this relatively young constructed wetland as the most logical mechanism to explain the observed dieback.