University of Twente Student Theses


Water Quality Monitoring of Roxo reservoir using Landsat Images and In-situ Measurements

Olet, Emmanuel (2010) Water Quality Monitoring of Roxo reservoir using Landsat Images and In-situ Measurements.

[img] PDF
Abstract:Water quality monitoring in fresh water bodies incorporating the use of earth observation products has become a major component in many a water quality monitoring program. This is majorly due to the inadequacies of traditional methods imposed by the lack of comprehensive and reliable in-situ datasets. In this study we explored the possibility to use Landsat imagery data to quantify pollutant indicators through estimation of water quality parameters of the Roxo reservoir waters. Chlorophyll-a (Chla) and suspended particulate matter (SPM) were the parameters of interest. Landsat images for the Roxo reservoir acquired on 05-09-2000, 04-06-2001 and 01-07-2002 were obtained from the Landsat data archive. Owing to a lack of in-situ radiometric, Chla and SPM measurements on the Roxo reservoir, a dataset acquired on the Poyang Lake during a field excursion in October 2008 was incorporated in the study. The atmospheric correction algorithms and water quality models developed were trained on the Poyang Lake dataset and implemented on Roxo reservoir as a case study. An image based atmospheric correction was implemented on the Landsat imagery data and validated using in-situ measured water leaving reflectance (WLR) on the Poyang Lake. The R2 and RMSE values between the atmospherically corrected satellite derived WLR and in-situ measured values were 0.797 and 0.117 respectively. Following a matrix inversion technique, a semi-analytical bio-optical model was used to estimate Chla, CDOM and SPM from atmospherically corrected WLR. However, this application yielded physically unrealistic estimates of Chla and SPM concentration. No analysis was done on CDOM due to a lack of CDOM in-situ measurements. The Area Chlorophyll-a Concentration Retrieved Model was then used to empirically estimate Chla, from which SPM was inferred. Chla concentrations derived were in the range 0.31 μgl-1 to 0.35 μgl-1 with R2 = 0.295 and RMSE = 0.603μgl-1. SPM concentrations on the other hand were in the range of 9 mgl-1 to 34 mgl-1 with an R2=0.5 and RMSE =0.145 mgl-1 showing relatively high correlation between the satellite derived and in-situ measured concentrations of SPM. This study has shown the possibility to implement image-based atmospheric correction on Landsat imagery and limitations of Landsat imagery for Chla and SPM retrieval using a semi-analytical bio-optical model. However, following an empirical approach Chla and SPM can still be derived. Landsat can, therefore, still play a critical role in water quality estimation for monitoring purposes on small inland lakes and other water bodies. Keywords: water quality monitoring; remote sensing; Landsat; atmospheric correction; suspended particulate matters; Chlorophyll-a; Roxo reservoir
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:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page