Dissertation > Environmental science, safety science > Environmental Quality Assessment and Environmental Monitoring > Analysis and Evaluation of Environmental Quality > Biological assessment, ecological assessment

A Study on the Predictive Model of Health Assessment in Li River, Guiling, Guangxi

Author ZhangJie
Tutor WangBeiXin
School Nanjing Agricultural College
Course Agricultural Entomology and Pest Control
Keywords Benthic fauna River health assessment Prediction model Biological assessment O / E values Lijiang
CLC X826
Type Master's thesis
Year 2010
Downloads 22
Quotes 0
Download Dissertation

RIVPACS models were widely used for river health assessment. We used the data of benthic assemblages and environmental variables collected at Li River, Guiling, Guangxi autonomous region in March and October,2008, to develop, validate and test a RIVPACS type model. Furthermore, we try to give a criteria of assessment using the output of our model and to evaluate the health condition of Li River.1. Eleven environmental variable that related to the distribution of benthic macroinvertebrate:stream wet width, depth, velocity, conductivity, DO, pH, water temperature, altitude, longitude, latitude and substratum were measured at each sampling site. The substrates were divided into five types by particle size:substrate sizeⅠ(<2mm),Ⅱ(2-8mm),Ⅲ(8~64mm),Ⅳ(64~256mm) andⅤ(>256mm).2. A total of 98 benthic macroinvertebrate taxa were indentified from 27 sampling sites collected in Li River, in March and October,2008. It contains 88 taxa in Arthropoda(89.8%),6 taxa in Mollusca(6.1%),3 taxa in Annelida (3.1%) and 1 taxa in Platyhelminthes. Among insecta,24 taxa in Ephemeroptera,11 in Ttrichoptera,15 in Plecoptera,16 in Odonata,4 in Megaloptera,8 in Diptera,6 in Coleoptera,1 in Hemiptera and 3 in Lepidoptera.3. The establishment of our predictive model was based on the methodology and processes of RIVPACS development. The genus level taxa data and environmental data of 48 samples were used to develop the predictive model, of which 32 samples were identified as reference, and 27 of which were used to calibrate the model,5 remain reference samples.10 slightly to moderately disturbed and 6 heavily disturbed samples were used to validate and test the model. Firstly,27 reference samples were clustered into 3 groups using Bray-Curtis coefficient. Then, 5 environmental variables (mean stream velocity, width/depth, water temperature, substrate sizeⅠ(<2mm) andⅡ(2-8mm)) were determined as the best discriminated variables among 3 sample groups through a stepwise discriminant function analysis. Then, the observed (O) and expected (E) biota, and O/E ratio of all 48 samples were obtained by the model, The O/E values of 27 reference samples used in model development had no significant difference with 5 remain reference samples for validation, but had significant difference with 10 slightly to moderately and 6 heavily disturbed samples, and 10 slightly to moderately disturbed samples were also significantly different from 6 heavily disturbed samples.4. Model II was established using the data of family taxa. Fourty reference samples were clustered into 3 groups using Bray-Curtis coefficient. Then,4 environmental variables (width/depth, water temperature, substrate size I (<2mm) and pH) were determined as the best discriminated variables among 3 sample groups through a stepwise discriminant function analysis.5. From outputs of O/E values, we calculated the mean value and distribution of the O/E values in 27 reference samples, and established the O/E value criteria of the output of RIVPACS models for Li River health assessment. There are five health levels:best, healthy, sub-healthy, normal condition and worst. The results of reference sites assessed by model I were disagreement with model II at 35.42%. We suggested that further study was urgently needed to establish a robust and scientifical predictive model for water quality biomonitoring and assessment. Moreover, the health condition of 7 reference sites assessed by models and B-IBI were different. Collectively, the health condition upstream branches of Li River were in healthy in 2008, while the health condition downstream were degraded in some degree, mainly casued by anthropogenic disturbance. The health condition of 9 sites located in the upstream of Li River collected in April,2009 were unhealthy using our models, indicated that it might had close relation to a reduction of discharge caused by a six-month-drought from October 2009 to March,2010.

Related Dissertations
More Dissertations