Estimation of chlorophyll-a concentration with semi-analytical algorithms using airborne hyperspectral imagery in Nakdong river of South Korea

• Published in: Spatial information research (Online) Vol. 27; no. 1; pp. 97 - 107
• Main Authors: Jeon, Eui-Ik, Kang, Seong-Joo, Lee, Keum-Young
• Format: Journal Article
• Language: English
• Published: Singapore Springer Singapore 01.02.2019; 대한공간정보학회
• Subjects: Academia and Industry collaboration on the Spatial Information; Database Management; Earth and Environmental Science; Geographical Information Systems/Cartography; Geography; Remote Sensing/Photogrammetry; 공학일반; Chlorophyll; Hyperspectral imagery; Remote sensing; Airborne; Semi-analytical algorithms
• ISSN: 2366-3286; 2366-3294
• DOI: 10.1007/s41324-018-0204-0

In this study, semi-analytical algorithms such as two-band and three-band models were used to estimate the chlorophyll- a (Chl- a ) concentration in the turbid river using an airborne hyperspectral imagery. In order to select the optimal wavelength band to be used in the empirical equation, surface water was collected at the same time of acquisition of the aerial hyperspectral imagery. The spectral characteristic of the Chl- a , PC, CDOM, NAP, and phytoplankton were analyzed from by collected samples. The concentrations of PC and CDOM which affect the spectral characteristics to Chl- a were low and there was no change over time. So the range of wavelengths was able to broaden than the existing cases. As the result of widening the wavelength band, the two-band and three-band models were found to be higher R 2 than the results obtained by using the existing formula. Because the three-band model is more statistical significance than the two-band model, it is more appropriate to estimate the chlorophyll- a concentration in the turbid river. However, the Chl- a concentration of this study was relatively low at 45 mg/m 3 , and the effect of PC and CDOM also was small. To estimate the correct Chl- a concentration, data such as airborne hyperspectral imagery and water sample need to be accumulated in different years and the correlation between optical properties and concentration should be thoroughly analyzed.