# Also in the Article

2.3. ZSD modeling and statistical analysis
This protocol is extracted from research article:
Remote sensing of water transparency variability in the Ibitinga reservoir during COVID-19 lockdown
Remote Sens Appl, Apr 17, 2021;

Procedure

For the Secchi Disk Depth, Z SD (m), estimation of the semi-analytical model from Lee et al. (2013, 2015) was performed based on the diffuse attenuation coefficient of downwelling irradiance (K d) considering its applicability in a wide range of environments (Eq. (1)).

where $Min(Kdtr)$ is the minimum value within the visible spectral domain (443–665 nm) of the attenuation coefficient of downwelling irradiance of the transparent window, and $Rrstr$ is the remote sensing reflectance at the same wavelength chosen for $Kdtr$ (Lee et al., 2015). $Kd(λ)$ can be estimated by (Eq. (2)):

where $θs$ is the solar zenith angle, here considered to be 30° and $bbw$ is the backscattering coefficient for water molecules based on Smith and Baker (1981); $m0−3$ and $γ$ are equal to 0.005, 4.26, 0.52, 10.8 and 0.265, respectively; $a$ is the total absorption coefficient and $bb$ is the backscattering coefficient, obtained by the quasi-analytical algorithm (QAA) and here based on the version 5 (QAA_v5, Lee et al., 2009). The reference wavelength used in QAA_v5 was 561.41 nm.

The QAA_v5 was selected based on Rodrigues et al. (2017) results; the authors used data from Barra Bonita reservoir, which is near our study area. They tested the QAA_v5 and adjusted their own QAA (QAAM14) and find a better result using the QAA_v5.

A boxplot of ZSD was used to compare the images through Two Sample t-Tests and confidence level of 0.05. The boxplot was based on 40 pixels selected manually in the reservoir taking care of avoiding the shallows areas; this is an important step to avoid the adjacency effects.

Note: The content above has been extracted from a research article, so it may not display correctly.

Q&A