University of Twente Student Theses

Login

Monitoring and Predicting Eutrophication of Inland Waters Using Remote Sensing

Mssanzya, Shabani Marijani (2010) Monitoring and Predicting Eutrophication of Inland Waters Using Remote Sensing.

[img] PDF
1MB
Abstract:Inland waters are human kind assets as they serve for both economical and ecological well being; however, their existence is compromised by raising eutrophication in these water bodies’ especially toxic cyanobacteria species. Existing in-situ water quality measurements have failed to offer required temporal and spatial coverage which cope with the dynamics of water quality. Therefore, the purpose of this thesis was to use remote sensing techniques to monitor and predict eutrophication with the focus on separating cyanobacteria from other Phytoplankton species. Eutrophication in water bodies is associated with growth of Phytoplankton biomass which is easily to be detected by satellite sensors. Moreover; as eutrophication process is taking place, Inherent Optical Properties (IOP) of water changes accordingly. It is these IOPs which can be related to their respective concentration using bio optical models. From observation of the water leaving reflectance, we can determine the IOPs of different water constituents, namely: chlorophyll a, dissolved organic matter, particulate matters and cyanobacteria. The differences in spectral characteristics of water constituents make it possible to quantify them separately using remote sensing techniques. In this thesis, GSM01 (Garver- Siegel- Maritorena) model was modified by removing band six of MERIS (centered at 620 nm) to exclude Phycocyanin pigments from derived Chlorophyll a concentration. This separation was possible because a Cyanobacterium which is determined by Phycocyanin pigments has maximum absorption at 620 nm. A simple analytical model was derived to retrieve absorption coefficients due to Phycocyanin pigments separately at 620. The method was validated using in-situ water quality measurements from Poyang Lake in China. The model gave good results on the relationships between derived absorption coefficients due to Phycocyanin pigments (apc) and Chlorophyll a concentration (R2= 0.9) which clearly prove the fact that at a certain concentration of chlorophyll a concentration (6.0 mg/m3to around 15 mg/m3according to the data set used) the relationship is strongly linear. Using derived products it was possible to develop Remote Sensed Eutrophication Indices (R.S.E.I.) by applying Principal Component analysis (PCA) which gave insight of having R.S.E.I which can replace in-situ derived indices. Proposed model gave a promising approach to separate Chl-a and apc as well as developing R.S.E.I. which can surrogate in-situ derived E.I. and be used as a tool for monitoring and predicting eutrophication in inland waters. However, model overestimated chlorophyll a concentration which needs adjustments using more in-situ measurements, and derived R.S.E.I. and apc need validation using more in-situ data. Finally it was found that is not feasible to use remote sensing techniques for water quality management in small water bodies.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/92292
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page