An investigation into the effectiveness of relative and absolute atmospheric correction for retrieval the TSM concentration in inland waters

Abstract

The absolute atmospheric correction inputs are not always available, and then such parameters are assumed based on geographical location, acquisition time and sensor type. These assumptions can imply in errors in retrieving the remote-sensing reflectance (R rs ), and affects the optically active compounds estimates. As an alternative, relative atmospheric correction, i.e. radiometric normalization, can be used in cases where there is no information about atmospheric conditions. The main goal of this work was to perform a comparative analysis between absolute and relative atmospheric correction to estimate total suspended matter (TSM) concentrations in the Barra Bonita Hydroelectric Reservoir (São Paulo State, Brazil). The corrections were applied to the operational land imager, on-board Lansat-8 satellite. The R rs errors from each correction were computed considering in situ data, and the lowest error was obtained for green spectral band (RMSE absolute = 11.5 % and RMSE relative = 12.3 %). Using a regional algorithm that was developed using the in situ measurements (the model was TSM = 1742.7*B3 -5.42, with R 2 = 0.60, p-value = 0.05), the estimated TSM concentrations from absolute and relative corrections retrieved a RSME of 11 and 6 %, respectively. The errors from absolute correction can be originated from the input parameters that were adopted, such as CO 2 concentration, initial visibility, and water vapor information. The relative correction can be more appropriate in such cases because, besides atmospheric effects, the method try to minimize the illumination variability using normalization between temporal images, which improves the reflectances, and consequently, decreases TSM retrieval errors.

Mapping Agricultural Land in Afghanistan’s Opium Provinces Using a Generalised Deep Learning Model and Medium Resolution Satellite Imagery

“…Relative normalisation was used instead of an absolute radiometric calibration with an atmospheric model as no in-situ measurements, or approximations for historical images, are required and potential differences in sensor-calibration can be avoided [38]. IR-MAD has also been shown to reduce variability in retrieval of surface parameters compared to absolute correction [39]. Absolute atmospheric correction is worth investigation going forward as it will remove the need for reference images and any potential distortions in temporal patterns [40].…”

Mid-Term Monitoring of Suspended Sediment Plumes of Greek Rivers Using Moderate Resolution Imaging Spectroradiometer (MODIS) Imagery

“…This approach serves to establish a uniform comparison basis for the study [87]. Some researchers have claimed its superiority over absolute atmospheric correction methods, particularly for retrieval of total suspended matter concentrations in inland/coastal waters [88]. Several methods for the automatic selection of PIFs have been developed.…”

WaterNet: A Convolutional Neural Network for Chlorophyll-a Concentration Retrieval

“…The methods to remove atmospheric effects, called atmospheric correction, can be classified as absolute and relative correction. The absolute correction predicts and removes the scattering due to gases and aerosols in the atmosphere [15,16], and several models have been proposed for water bodies [17][18][19]. By contrast, the relative correction minimizes numerical differences among images using image processing techniques [20][21][22][23].…”