Construction of TP quantitative models based on diatoms and cladocera for the Irish ecoregion using palaeolimnological techniques
MetadataShow full item record
Title of thesis: Construction of TP Quantitative Models Based on Diatoms and Cladocera for the Irish Ecoregion Using Palaeolimnological Techniques Author of thesis: Guangjie Chen This study uses palaeolimnological techniques and numerical methods to construct TP transfer functions for the Irish Ecoregion based on both phytoplankton (diatoms) and zooplankton (Cladocera) remains and applies them to the sediments of seven impacted lakes to identify the pre-impacted reference conditions and help inform lake restoration targets. A 72-lake diatom training-set and a 33-lake Cladocera training-set were developed along a TP gradient (max. 142.3 μg l-1 TP). Seventeen related environmental variables were available for data exploration. A variety of exploratory and multivariate data analyses were used to investigate environmental and biological data structure and their relationships using the R program. Detrended Correspondence Analysis (DCA) revealed a high degree of species heterogeneity in the diatom data compared to the Cladocera data. Both datasets were used to assess the viability of the Irish Lake Typology physicochemical classification scheme using hierarchical clustering method and both organisms provided good biological verification. Ordination analyses showed that nutrient gradient was among the most significant variables in determining both the diatom assemblages of 72 lakes and the Cladocera assemblages of 33 lakes. TP transfer functions were constructed using Weighted Averaging (WA)-related and linear modelling methods. A diatom TP transfer function produced a jack-knifed coefficient of determination (r2 jack) of 0.743 with a root mean squared error (RMSEP) of 0.213 based on untransformed diatom data from 70 lakes. A TP transfer function built on square root transformed Cladocera data from 31 lakes yielded an r2 jack of 0.729 with a RMSEP of 0.206. A sub-set of lakes where both diatom and Cladocera data were available were examined to compare the predicted TP from both models and they displayed a strong correlation (r = 0.685) for log-transformed TP. TP models based on both indicators were applied in top-bottom analyses of seven impacted lakes. Diatom and Cladocera results indicated a similar trend of nutrient enrichment between the current (top samples) and reference status (bottom samples) for most of the lakes. Therefore the use of combined TP transfer functions based on diatom and Cladocera provided a comprehensive insight into the reference conditions and ecological status for lake restoration due to their differential positions in the community dynamics and distinctive sensitivity to water quality of lakes.
KeywordsTP quantitative models
Language (ISO 639-3)eng
PublisherMary Immaculate College, University of Limerick
- Geography (Theses) 
This item is protected by original copyright: