Spatiotemporal pattern validation of chlorophyll-a concentrations in Lake Okeechobee, Florida, using a comparative MODIS image mining approach

Abstract

A comparative analysis was conducted using three types of data-mining models produced from Moderate Resolution Imaging Spectroradiometer (MODIS) Terra Surface Reflectance 1-day or 8-day composite images to estimate chlorophyll-a (chl-a) concentrations in Lake Okeechobee, Florida. To understand the pros and cons of these three models, a genetic programming (GP) model was compared to an artificial neural network (ANN) model and multiple linear regression (MLR) model with respect to two different data sets related to model formulation. The first data set included the MODIS Terra bands from 1 to 7; the second data set extended the first data set by adding environmental parameters such as Secchi disc depth (SDD), total suspended solids (TSS), wind speed, water level, rainfall and air temperature collected around the lake in 2003 and 2004. The GP algorithm, which has an advantage in machine learning allowing us to select the appropriate input parameters that significantly impact the prediction accuracy, outperformed the other two models based on four statistical indices. Specifically, the GP modelling outputs revealed interesting determinations of chl-a concentrations for MODIS bands 3, 5, 6 and 7, corresponding to wavelengths 459-479, 1230-1250, 1628-1652 and 2105-2155 nm, respectively. The number of training data points is limited; therefore, the inclusion of additional environmental variables cannot improve the prediction accuracy of the GP-derived chl-a concentrations.

A long term calibration and verification of a submerged aquatic vegetation model for Lake Okeechobee

“…SAV is an important habitat, providing refuge to juvenile fish and shellfish and providing a food source for fish and waterfowl. Consequently, the assessment of SAV provides a direct link between water quality (nutrients, chlorophyll a, and suspended sediments) (Chang et al 2012) and ecologically and economically important species. The extent of SAV in a water system varies directly with water clarity and inversely with water depth.…”

Can chlorophyll-a in meso-oligotrophic shallow waters be estimated using statistical approaches and empirical models from MODIS imagery?

“…However, our approach showed an improvement over previously published models (e.g., CHANG et al, 2012;OGASHAWARA et al, 2014;WU et al, 2009) due to the implementation of search criteria based on a sequential automatic search of exclusion-inclusion algorithms for selecting parameters. In addition, the estimation of Chl-a concentrations based on MLR presented advantages (e.g., few input variables and explicit calculation of the coefficients of selected independent variables) in comparison with the ANN and NPMR models, which is difficult to interpret the relationship between input and output variables ("blackbox" models) (WU et al, 2014).…”

“…Recent studies have applied empirical algorithms for retrieving Chl-a concentrations in lakes and reservoirs combining MODIS spectral bands (DUAN et al, 2017;CHANG et al, 2012;HUANG et al, 2014;SHI et al, 2017;TANG, 2011;WU et al, 2009;XIANG et al, 2015;ZHANG et al, 2016). However, these algorithms were developed for eutrophics waters (e.g., Lakes Taihu and Chaohu, China and Lake Okeechobee, USA) and for particular seasons, limiting their uses elsewhere and throughout the year (MATTHEWS, 2011;PALMER et al, 2015a).…”

“…However, these algorithms were developed for eutrophics waters (e.g., Lakes Taihu and Chaohu, China and Lake Okeechobee, USA) and for particular seasons, limiting their uses elsewhere and throughout the year (MATTHEWS, 2011;PALMER et al, 2015a). These approaches can be improved using the appropriate spectral remote-sensing reflectance values and combinations of spectral bands with either regression analysis (CHANG et al, 2012;OGASHAWARA et al, 2014) or an ANN (WU et al, 2009;XIANG et al, 2015). However, only the approach proposed by Ogashawara et al (2014) was developed in a meso-oligotrophic environment (Itumbiara Reservoir, Brazil).…”

Assessing water quality in lake Naivasha

“…Although the OC2 algorithm was first developed for SeaWiFSdata (Darecki and Stramski, 2004;O'Reilly et al, 1998;O'Reilly et al, 2000), it has proven to work in MODIS high resolution bands (500 m) in turbid waters (Hudson et al, 2000;Wang and Shi, 2008;Wang et al, 2007). OC2 algorithm has also been tested in turbid lake environments (Chang et al, 2012;Weghorst, 2008). Other studies have found the MODIS medium-resolution bands (250 and 500 m) to be 4 to5 times more sensitive than the Landsat 7 bands (Hu et al, 2004).…”