Dec 16, 2016 • Today at #AGU16, Enner Alcantara presented a poster showing that Landsat 8 can map the colored dissolved organic matter of inland tropical waters.

Here is what he shared with us:

Presentation Title
Landsat-8/OLI images has the potential to estimate the CDOM absorption coefficient in tropical inland water

What are the major findings of this research?
An empirical model was adjusted between the remote sensing reflectance and the colored dissolved organic matter (CDOM) absorption coefficient at 440 nm. The validation using in situ data of the aCDOM440 algorithm indicated a goodness of fit, R2 = 0.70, with a root mean square error (RMSE) of 10.65%. This regional-scale model was applied to a series of OLI images in order to obtain the aCDOM440 spatial distribution.

What are the implications of your findings?
The aCDOM440 is often used to establish a relationship with dissolved organic carbon concentration in water, therefore the OLI/Landat-8 can be used to estimate the carbon content in inland waters and consequently contribute to carbon cycle understanding.

What insight did you gain from Landsat that would have been impossible to glean otherwise?
Inland waters bodies are usually small, but the 30m of spatial resolution of Landsat is considered adequate. Also, the Landsat series allowed us to study the carbon content in a historical way. What is the difference between the carbon content in 1980 and in 2016?

Does the addition of Landsat 8’s new band 1 (the coastal/aerosol band) help you map colored dissolved organic matter? Why is this spectral band more useful than the longer wavelength bands?
Atmospheric correction is considered a challenge for water quality studies.

In order to reduce the influence of path radiance, we used Fast Line-of-sight Atmospheric Analysis of Hypercubes (FLAASH). Including the Landsat 8 OLI band 1 (the coastal/aerosol band) helped the FLAASH method give better results making the aCDOM map more accurate. The OLI band 1 is useful because atmospheric scattering increases when the wavelengths of the radiation shortens. So, it is expected that the atmospheric correction methods take advantage of this information.

map of aCDOM440

Spatial distribution of aCDOM440 in Barra Bonita reservoir (Tietê River, São Paulo State, Brazil), using OLI/Landsat-8 images from: (a) 30 January 2014, (b) 11 September 2014, (c) 13 October 2014, (d) 29 October 2014 and (e) 16 December 2014.


Nariane Bernardo

UNESP Sao Paulo State University

Anyone can freely download Landsat data from the USGS EarthExplorer or LandsatLook.

Further Reading:
+ Landsat Abounds at #AGU16