Αξιολόγηση Της Οικολογικής Ποιότητας Των Ποταμών Αχέροντα Και Λούρου Της Δ. Ελλάδας Και Της Λεκάνης Απορροής Τους Με Χρήση Υδρόβιων Μακροφύτων Ως Βιολογικών Δεικτών

Abstract

The main objective of the present thesis was the ecological assessment of two river basins, Acheron and Louros; based on aquatic plant communities, water quality and hydromorphological characteristics, as well as, the development of a methodology for collecting, analyzing and assessing field data, specific adapted to Greek conditions. The specific objectives of the doctoral thesis include: a) the investigation of abiotic parameters influencing the distribution of aquatic macrophytes along the two river basins (Chapter A); b) the investigation of riparian zone structure and the assessment of ―key‖ environmental gradient and human intervention (Chapter B); c) abiotic typology for reducing the spatial variability, monitoring the temporal and seasonal variation in of aquatic macrophytes distribution, and determination of reference communities in each geomorphological unit of the riverine ecosystem (Chapter C), and d) the development of the Macrophyte Multimetric Index (MMI) which is described step by step in Chapter D.Data were collected from 32 sampling sites, which were chosen according to the following criteria: i) covering a wide gradient of land used i.e. from natural to artificial; ii) selection of more than one site with similar geomorphological characteristics; iii) the constant distance between sampling sites; iv) the homogeneity of the sampling sites, and v) the accessibility to the sampling points throughout the duration of monitoring. Based on the above criteria, 15 sites were selected along Acheron river basin [11, Acheron river (A1-8 & A11-13); 1, Cocytos (A9); 1, Vouvo Rema (A15); 2, irrigation canals (A10 & A14)]. Additionally, 17 sampling sites were selected along Louros river (S1-S15). The survey area was subdivided into 3 zones, a) the wetted part of the channel, b) the marginal-active channel and c) the riparian woodland plot. The survey area length was standardized at 100 m according to the widely accepted methods [e.g. Mean Trophic Rank, (MTR); Riparian Forest Quality, (QBR)]. On the other hand, the width of each plot includes the area between, the end of the active channel and the riparian woodland.The field surveys were conducted during the vegetation periods (April to September) of the years 2005 to 2007. The coverage of each macrophyte species was visually estimated using DAFOR 5-point scale (1: Rare, 2: Occasional, 3: Frequent, 4: Abundant and 5: Dominant). Also, water physicochemical parameters such as temperature, DO, pH, conductivity, were measured in situ using portable equipment (WTW340i/SET). Additionally, surface water samples were collected for determination of nutrients such as nitrate and nitrite nitrogen, total nitrogen, ammonium, phosphorus nutrients (soluble reactive phosphorus and total phosphorus), Chl-a, and alkalinity (HCO-3, CO3=). Nutrient samples were collected and analyzed according to APHA standard analytical method (APHA, 1989).For the statistical processing of the data were used: 1) Hierarchical cluster analysis (Bray-Curtis), for the biotic typology of macrophyte data along river basins and to determine the structure of the riparian vegetation; 2) Indicator Species Analysis, was used to describe the hierarchical structure of each group to distinguish the indicator species (Ind_Sps) for each vegetation group; 3) Ordination, which were used both Indirect and Direct gradient analyses. Indirect gradient analyzes were used: a) Principal Component Analysis (PCA), to identify Geomorphological Units and to determining the pressure gradients; and b) Detrended Correspondence Analysis (DCA), for researching the spatial distribution of species abundance, and for seasonal and inter-annual variability of the reference communities. For direct gradient analyzes were used: a) Redundancy Analysis (RDA), for correlation between, water physicochemical parameters and Indicator species; and b) Canonical Correspondence Analysis (CCA), for correlation between geomorphological characteristics and indicator species, as well as, to determine the key variables best explain the degradation of the riparian zone.The results showed that along the two river basins, six vegetation groups can be distinguished, with similar environmental gradient pattern (springs-estuaries). The dominant species for the upper part of Acheron and Louros river basins were: Pteridium aquilinum, bryophyte taxa and Carex species (C. acuta and C. pendula respectively). Moreover, four different vegetation groups were characterized the middle part of both river basins. The different degree of anthropogenic alteration of river estuaries (Acheron and Louros), reflects in the composition of the plant communities. The estuaries of Acheron were characterized by the dominant of green algae and the tolerant species Potamogeton pectinatus, while the estuaries of Louros river was characterized by the absence of green algae and the dominance of Potamogeton nodosus. The most important physicochemical variables that significantly affect the distribution of aquatic macrophytes along the river basins were: pH, conductivity, water temperature, and mean water velocity, concentrations of nitrogen nutrients (ΝΟ3-Ν, ΝΖ4-Ν, ΝΟ2-Ν) and orthophosphate (PO4-P), as well as, Chl-a. Whereas, the most important geomorphological factors were: altitude, bed substrate, channel width, channel shading, habitat type, water depth.Hierarchical cluster analysis distinguished five vegetation groups of the riparian zone of the study area. The following groups were indentified and characterized by the dominant riparian species. The upper reaches were characterized by the Vegetation Group Ia: Platanus orientalis and Ib: Quercus coccifera-P. orientalis; the middle reaches by II: Salix alba and Populus alba; and lowland areas until estuaries by III: Phragmites australis. Vegetation Group IV was characterized by the dominant of non native species Eucalyptus camaldulensis and the giant reed Arundo donax. According to the results, species abundance was low in all Vegetation Groups. Also, the applied indices Riparian quality assessment (QBR), Habitat Quality Assessment (HQA) and Habitat Modification Score (HMS) revealed that the ecological integrity of the riparian zone increases the overall ecological quality of the riparian ecosystem.According to the hydromorphological characteristics, the sampling sites were grouped into three Geomorphological Units: Upper –Middle reaches-Estuaries. In each Geomorphological Unit (GU) a pressure gradient analysis was performed, and the reference sites were identified. For GU I the number of reference sites was 7, for GU II 3 reference sites were identified, while, for the estuaries the number of the reference sites were only 2. Also in each GU the seasonal and inter annual changes in reference communities, were investigated, and the structure of the reference community was determined. The results showed that the composition of the reference communities in lowland rivers appears to remain constant in the physical fluctuations. On the other hand, the slight differences which were recorded, in the GU I, during spring and summer sampling, referred to the reduction of bryophytes. This reduction might be cause due to the late spring (May), at the upper part of Louros river. Also, differences were recorded between autumn and spring samplings, in the GU III (estuaries). Those differences referred to the reduction of pteridophytes, maybe due to the high temperatures that prevail during the summer months. Finally, the sampling period could be set between late April and late September, since, during this period, any field visit will give the same composition of the reference communities.We investigated, to the greatest possible degree, all the potential parameters and pressures, which could influence the composition and distribution of aquatic macrophytes along the river basins. The most significant effect, from the current pressures, in the studied area, is the alteration of the hydromorphological characteristics of the ecosystem. The Multimetric approach was chosen, as the most holistic and reliable evaluation method of the ecological status of the study area. A preliminary list of potential plant metrics was compiled from a review of the literature, and in situ observations of plant community patterns. The metrics were grouped into five categories, so that each one representing a different ecological aspect of aquatic and riparian plant communities: 1) Richness/Diversity, 2) Trophic status, 3) Composition, 4) Riparian Integrity, and 5) Sensitivity/Tolerance. A total of 86 metrics were tested, of which 7 belonged to the category of "Abundance/Diversity", 4 to the category of "Trophic‖ status, 46 metric referred to ―Composition‖ of plant communities, 11 referred to "Riparian Integrity‖, and finally, 18 belonged to ―Sensitivity/Tolerance‖ category. The estimation of the pressure gradient was performed using Principal Component Analyses (PCAs), with 36 hydromorphological degradation variables in 2 spatial scales: macroscale and microscale.Candidate metrics, which can be identified as robust and most informative, are scrutinized further, in the process of selecting core metrics. To be selected as a core metric the following aspects have to be considered: 1) core metrics should cover different metric types (see above); 2) metrics should not give redundant information. Inter-correlation tests between candidate metrics were carried out to detect redundant metrics (threshold value Spearman‘s r<0.800). In case of redundancy we further investigate: the correlation of each metric with stress gradient (threshold value Spearman‘s r<0.05); the correlation of each of the pair of metrics with the other metrics in order to finally omit the one that showed the higher overall mean correlation; and how well they separate stressed from unstressed sampling sites (graphical analysis of box-whisker plots). The final index includes 6 metrics out of 86 candidated metrics. The core metrics are: 1) IBMR, 2) QBR, 3) Number of Bryophyte, 4) % Reference Species, 5) % Nitrophilous taxa and 6) % Helophyte taxa (Phe_herbids).The different numerical scales of each core metric (e.g. abundance class, number of individuals) were normalized to unitless scores between 0 and 1. The upper and lower anchors mark the indicative range of a metric. The upper anchor corresponds to the upper limit of the metric‘s value under reference conditions, and it was set as the 75th of the unstressed sites. The lower anchor corresponds to the lower limit of the metric‘s value under the worst attainable conditions and it was set as 5th percentile of stressed sites.The multimetric index was calculated as the arithmetic mean of the normalized metrics (Böhmer et al., 2004). The final Multimetric Index provides a score that represents the overall relationship between the combined values of the biological parameters observed for a given site and the expected value under reference conditions. This score is – as for single metrics – expressed as a numerical value between zero and one. This range can be subdivided into any number of categories corresponding to various levels of impairment. To determine the boundaries of Multimetric Index, the 25th percentile of the unstressed sites (0,623) was set as the boundary for the high/good ecological class. We propose quality classes with equal ranges, to provide five ordinal rating categories for assessment of impairment in accordance with the demands of the WFD. The application of Macrophyte Multimetric Index (MMI) in the medium lowland rivers of Greece led to the ecological classification in five quality classes. The results indicated that the 29.6% of the sampling sites (8 sites) were classified in High ecological status, 4% (4 sites) in Good ecological status; the 22.3% (6 sites) in Moderate, 29.6% (8 sites) in the Poor and only 3.7% (1 site) was characterized as Bad ecological quality.The current doctoral thesis was the first attempt in the Greece, for establishing an integrated monitoring system of ecological quality of running water ecosystems, using aquatic macrophytes as biological quality element. The methodology developed, was pilot implemented in two river basins, Acheron and Louros, in Epirus, following the requirements and guidelines of the European Water Framework Directive, 2000/60/EE. Also, the evaluation system was developed to be proposed for implementation at all similar river types of Greece, since both the list of candidate metrics and the Multimetric Index itself, can be tested in other types of Greek rivers (RM-1, RM-3, RM-4 & RM-5). Finally, the results, will contribute to the sustainable management and conservation of riverine ecosystems, as well as, to the implementation of the water environmental policy in Greece.