Summary

It was the aim of this study to analyse the impact of land use and flooding on the diver-sity of arthropods living in temporarily flooded grassland habitats.
The "Lower Oder Valley National Park" was chosen as study area. The Lower Oder is a typical lowland river in Central Europe, where flooding events are frequent in winter but exceptional in summer. The floodplain mainly consists of grassland habitats. Parts of the grassland are extensively used for farming, whereas other parts were taken out of use in 1995. Currently, dykes bordering the river Oder prevent natural inundation of the floodplain during the summer months. During winter, floodgates integrated into the dykes stay open. Thus, the typical winter flooding can reach the floodplain. The phytophagous plant- and leafhoppers (Hemiptera: Auchenorrhyncha) and the pre-dacious spiders (Araneida) and ground beetles (Coleoptera: Carabidae) were chosen as representative groups.

In a descriptive field study, the impact of mowing and flooding on the diversity of plant- and leafhoppers was analysed (1).
Based on an experimental approach, it was investigated whether the typical wetland fauna repopulates the floodplain after each flooding event or overwinters in the habitats (2).
It was discussed, which impact the national park's plans of conservation management and conservation of processes might have on the corncrake as a chosen target species, on the anuran fauna as a vertebrate group typically found in wetlands and on plant- and leafhoppers as a key insect group in wet grasslands (3).
The results of the study on the impact of mowing and flooding on the diversity of plant- and leafhoppers were compared to those of studies on spiders and ground beetles, that were also carried out in the study area. Patterns common for plant- and leafhoppers and the other arthropod groups were explored. Based on these findings conclusions for the conservation of floodplain arthropods are drawn (4).

(1) The impacts of mowing and of flooding on the diversity of plant- and leafhoppers were investigated. Samples were taken by a motor-driven suction apparatus and pitfall traps four times between May and September 2001 and 2002, respectively. In 2001 plant- and leafhoppers were collected in a variety of sites differing in land use (fallows, mown sites) and in flooding impact (high, medium, low, none). In 2002 samples were only taken in fallows subject to different flooding regimes.
In fallows more species (43) were collected than in mown sites (33). The impact of flooding was only found in fallows, because mowing had a stronger impact on the plant- and leafhopper communities than flooding. In fallows the most species rich (29) com-munities occurred in sites not subject to flooding, whereas fewest species (21) were found in sites subject to regularly occurring long lasting winter floods. Flooding condi-tions in the latter sites resemble the natural flooding conditions most.
In both, mown sites and fallows, quite specific species assemblages were found. 15 spe-cies occurred predominantly in fallows, while eight species favoured mown grassland. In fallows two species were found to occur predominantly in sites subject to summer and winter flooding, 11 species in sites subject to long and regularly occurring winter floods, two in sites subject to short winter flooding and eight in sites not subject to in-undation. For a number of species, the distribution was either limited by food plant availability or by abiotic conditions.
In mown sites species were characterised by high mobility and a low degree of speciali-sation, whereas in fallows the communities differed in respect to flooding impact. In fallows that were subject to summer and winter floods, species communities were - like in mown meadows - characterised by mobile and less specialised species. Contrarily, in fallows that were flooded a long time during winter, very specialised species communi-ties were found which were, however, not very species rich.

(2) Arthropods living in river floodplains have to cope with a more or less regular cycle of wet and dry conditions. It was investigated whether the typical wetland fauna recolo-nises the floodplain after each flooding event or survives winter submersion in the habitat. Based on these findings, it was analysed whether flooding duration and fre-quency affect the distribution of arthropod species in the floodplain.
With receding water levels in May 2002, nine gauze-covered tents were placed into the floodplain to prevent colonisation of the space underneath. In July 2002 samples were taken from underneath the tents as well as from adjacent uncovered control plots with pitfall traps and a motor-driven suction apparatus. Additionally, samples were taken from nearby sites during the whole vegetation period. Planthoppers, leafhoppers, spiders and ground beetles were studied in detail.
Most species of plant- and leafhoppers (68%) tolerated submersion and overwintered in the floodplain, whereas in spiders (60%) and carabids (70%) most species immigrated into the floodplain with a receding water level. Some carabids followed both strategies. Furthermore, the distribution of immigrating and submersion tolerant species was de-pendent on the frequency and duration of flooding. Submersion tolerant plant- and leaf-hoppers as well as spiders dominated in sites that were regularly flooded for a long time during winter. Among these many are specialised and endangered species. Immigrating species of all three groups dominated in sites subject to summer and winter floods as well as in sites subject to short winter flooding. To create and maintain a maximum di-versity of habitats and specialists a semi-natural flooding regime needs to be achieved.

(3) Studies on insect and amphibian communities as well as on selected meadow bird species were carried out in the floodplain of the Lower Oder Valley National Park. The aim of these three studies was to analyse the effects of the present land use and man-agement of the flooding regime on the selected species.
The national park is still in the developmental stage. In plans for the future development of the national park, two main conservation strategies are discussed. On the one hand, conservation management is proposed with the aim to preserve and generate suitable habitats for the selected target species, corncrake (Crex crex). On the other hand, proc-ess conservation is suggested intending to create habitats suitable for the typical wetland fauna and flora by leaving some areas to natural processes and succession. On the basis of the three case studies the effects of the two conservation strategies on the selected taxa were discussed. On the one hand, Auchenorrhyncha and Anura will not necessarily benefit from conservation effort carried out for the benefit of the target spe-cies corncrake. These results cast doubt on the selection of the corncrake as umbrella species into question. On the other hand, conservation of processes negatively affects corncrake populations while the anuran community is neither positively nor negatively affected and insect communities benefit from the conservation strategy.

(4) Mowing and flooding were identified as main factors determining the distribution of plant- and leafhoppers in the floodplain. Patterns common for plant- and leafhoppers were compared to those found for spiders and ground beetles.
Short term effects of mowing on numbers of individuals were different in all three taxa. Numbers of individuals of plant- and leafhoppers and ground beetles decreased directly after the cutting event as a result of emigration activity, increased mortality and in Auchenorrhyncha additionally due to reduced food plant availability. Until September, individual numbers of plant- and leafhoppers increased again as a result of the higher nutritive value of the freshly grown plants. Individual numbers of ground beetles de-creased further, which correlated with the decreasing activity of the species during times of non-reproduction. Numbers of spiders were higher in mown sites than in fallows six weeks after the cutting due to reduced spatial resistance in the meadow. The latter leads to a higher activity of the spiders resulting in higher capture rates in the pitfall traps.
Mowing had a stronger impact on Auchenorrhyncha and Araneida communities, whereas flooding had a stronger impact on the carabid communities. Nevertheless, mowing negatively affected numbers of species in all three taxa. Furthermore, species of all three taxa occurring in mown sites were characterised by high mobility and a low degree of specialisation. In fallows that were subject to summer and winter floods, species communities of Auchenorrhyncha and Araneida were found to be characterised by mobile and less spe-cialised species. In ground beetles, also a mobile but very specialised community was found. The mobile specialists can successfully colonise a variety of small habitat patches on the river bank. These sites were dominated by immigrating species of all three taxa. Thus, mobility is the most important characteristic of species occurring in these sites.
In fallows that were flooded for a long time during winter, very specialised communi-ties of plant- and leafhoppers and spiders were found, which were not very rich in spe-cies. In these sites submersion tolerant species of both taxa dominated, which overwin-ter in the egg stage. In contrast, the typical wetland communities of ground beetles were dominated by immigrating and very mobile species. As ground beetles rarely overwin-ter in the egg stage, submersion tolerance is unlikely, thus flight is crucial to success-fully populate floodplain habitats. Concerning the conservation of typical floodplain arthropod communities, the results of this study reveal that it is crucial to stop land use and start restoring a more natural flooding regime. Additionally, the surrounding landscape of the conservation sites in question should be taken in account, as all immigrating species are dependent on non-flooded overwintering sites within as well as in the surroundings of the floodplain. Furthermore, this study reveals that the Lower Oder Valley National Park conservation management is beneficial for the target species corncrake, but will negatively affect arthropod communities. In contrast, those areas left for succession contribute to the con-servation of floodplain arthropods.

Location of the "Lower Oder Valley National Park"