1116
() RccolonwllIon or Tidal Flat;, Alter Disturbance
those of adult benthic organisms by 3-5 times. Due to this recruitment. individual
densities rose until the end of the study period in September.
Multivariate analysis showed distinct benthic assemblages before and after the
major settlement event in early June. Until the end of May. ostracods and
P. elegans accounted for 88 % of the individuals in the benthic assemblage. which
was composed of 15 species that spring. S. armiger and E. longa were the next
most abundant species. A higher diversity characterized the assemblage again in
summer and autumn 1996. when overall 25 species were recorded. P. elegans
remained the dominant organism (57 % of the individuals). Besides S. armiger and
E. long a the molluscs Mytilus edulis and Macoma halthica were frequent members
of the benthic assemblage.
6.5
Discussion
The sandflats in the backbarrier system of Spiekeroog were recolonized by benthic
diatoms. meio- and small macrofauna after both experimentally simulated disturbances and an ice winter event. Neither the experimental nor the natural disturbance caused any long-lasting changes in the benthic assemblages. Given the size
of the experimentally disturbed plots in ELA W AT. the study period was not long
enough to follow all successional stages. The two prominent macrobenthic species
found on the tidal tlats. A. marina and L. conchilega, were recorded on the disturbed plots with relatively few individuals only. Also, few adults recolonized the
disturbed plots within the study period.
The defaunation experiments in ELA W AT are the largest of such experiments
(20 m' sized plots each) carried out in benthic ecology to date. Previous experiments using a similar methodological approach took place on plots of several cm 2
up to m 2 in size, whereby Thrush et al. (1996) could show a scale-dependent
recolonization pattern. Also Smith & Brumsickle (1989) proved that the smaller
the disturbed site. the faster the recolonization. Varying with the size of disturbed
plots is the relative contribution of larval, postlarval or adult stages to the recolonization process (Smith & Brumsickle 1989; Gunther 1992; Whitlatch et al. 1998).
To assess the recolonization in the intertidal zone after larger-scale defaunations than those experimentally simulated in ELA W AT would require natural or
human-related disturbance events. Thus the disturbance of the ice winter 1995/96
was fortunate for the project (Chap. 7). A further large-scale disturbance occurred
in June 1996. mainly in the backbarrier system of the island of Baltrum, and was
induced by the bloom of Coscinodiscu.l' cOllcinnus. It was later inferred that algaproduced lipids had covered the sediment like an oilfilm and interrupted the watersediment gas exchange, resulting in a mass mortality of benthic organisms on
several km
2
of tidal tlats (Hapner 1996). In retrospect, it was a missed opportunity
that this event could not be studied during the course of the ELA W A T project.
The design of the large-scale experiments in ELA W AT had artefacts. The
sediments covered by tarpaulins for several weeks in 1994. compacted and formed
depressions in the sandflat, which were visible for several months. At each low
tide, these depressions filled with residual water after the tide receded, which
() RccolonwllIon or Tidal Flat;, Alter Disturbance
those of adult benthic organisms by 3-5 times. Due to this recruitment. individual
densities rose until the end of the study period in September.
Multivariate analysis showed distinct benthic assemblages before and after the
major settlement event in early June. Until the end of May. ostracods and
P. elegans accounted for 88 % of the individuals in the benthic assemblage. which
was composed of 15 species that spring. S. armiger and E. longa were the next
most abundant species. A higher diversity characterized the assemblage again in
summer and autumn 1996. when overall 25 species were recorded. P. elegans
remained the dominant organism (57 % of the individuals). Besides S. armiger and
E. long a the molluscs Mytilus edulis and Macoma halthica were frequent members
of the benthic assemblage.
6.5
Discussion
The sandflats in the backbarrier system of Spiekeroog were recolonized by benthic
diatoms. meio- and small macrofauna after both experimentally simulated disturbances and an ice winter event. Neither the experimental nor the natural disturbance caused any long-lasting changes in the benthic assemblages. Given the size
of the experimentally disturbed plots in ELA W AT. the study period was not long
enough to follow all successional stages. The two prominent macrobenthic species
found on the tidal tlats. A. marina and L. conchilega, were recorded on the disturbed plots with relatively few individuals only. Also, few adults recolonized the
disturbed plots within the study period.
The defaunation experiments in ELA W AT are the largest of such experiments
(20 m' sized plots each) carried out in benthic ecology to date. Previous experiments using a similar methodological approach took place on plots of several cm 2
up to m 2 in size, whereby Thrush et al. (1996) could show a scale-dependent
recolonization pattern. Also Smith & Brumsickle (1989) proved that the smaller
the disturbed site. the faster the recolonization. Varying with the size of disturbed
plots is the relative contribution of larval, postlarval or adult stages to the recolonization process (Smith & Brumsickle 1989; Gunther 1992; Whitlatch et al. 1998).
To assess the recolonization in the intertidal zone after larger-scale defaunations than those experimentally simulated in ELA W AT would require natural or
human-related disturbance events. Thus the disturbance of the ice winter 1995/96
was fortunate for the project (Chap. 7). A further large-scale disturbance occurred
in June 1996. mainly in the backbarrier system of the island of Baltrum, and was
induced by the bloom of Coscinodiscu.l' cOllcinnus. It was later inferred that algaproduced lipids had covered the sediment like an oilfilm and interrupted the watersediment gas exchange, resulting in a mass mortality of benthic organisms on
several km
2
of tidal tlats (Hapner 1996). In retrospect, it was a missed opportunity
that this event could not be studied during the course of the ELA W A T project.
The design of the large-scale experiments in ELA W AT had artefacts. The
sediments covered by tarpaulins for several weeks in 1994. compacted and formed
depressions in the sandflat, which were visible for several months. At each low
tide, these depressions filled with residual water after the tide receded, which