Use of Sentinel-2 Time-Series Images for Classification and Uncertainty Analysis of Inherent Biophysical Property: Case of Soil Texture Mapping

Abstract

The Sentinel-2 mission of the European Space Agency (ESA) Copernicus program provides multispectral remote sensing data at decametric spatial resolution and high temporal resolution. The objective of this work is to evaluate the ability of Sentinel-2 time-series data to enable classification of an inherent biophysical property, in terms of accuracy and uncertainty estimation. The tested inherent biophysical property was the soil texture. Soil texture classification was performed on each individual Sentinel-2 image with a linear support vector machine. Two sources of uncertainty were studied: uncertainties due to the Sentinel-2 acquisition date and uncertainties due to the soil sample selection in the training dataset. The first uncertainty analysis was achieved by analyzing the diversity of classification results obtained from the time series of soil texture classifications, considering that the temporal resolution is akin to a repetition of spectral measurements. The second uncertainty analysis was achieved from each individual Sentinel-2 image, based on a bootstrapping procedure corresponding to 100 independent classifications obtained with different training data. The Simpson index was used to compute this diversity in the classification results. This work was carried out in an Indian cultivated region (84 km2, part of Berambadi catchment, in the Karnataka state). It used a time-series of six Sentinel-2 images acquired from February to April 2017 and 130 soil surface samples, collected over the study area and characterized in terms of texture. The classification analysis showed the following: (i) each single-date image analysis resulted in moderate performances for soil texture classification, and (ii) high confusion was obtained between neighboring textural classes, and low confusion was obtained between remote textural classes. The uncertainty analysis showed that (i) the classification of remote textural classes (clay and sandy loam) was more certain than classifications of intermediate classes (sandy clay and sandy clay loam), (ii) a final soil textural map can be produced depending on the allowed uncertainty, and iii) a higher level of allowed uncertainty leads to increased bare soil coverage. These results illustrate the potential of Sentinel-2 for providing input for modeling environmental processes and crop management.

Bio-Inspired Hybridization of Artificial Neural Networks: An Application for Mapping the Spatial Distribution of Soil Texture Fractions

“…With respect to the clay index of Sentinel-2, there was a negative relationship with soil clay content [77]. The soil texture mapping using Sentinel-2 data performance had a good discrimination for extremely different soil texture classes (e.g., sandy loam versus clay), whereas neighboring soil texture classes (e.g., sandy clay loam and clay class) had high uncertainty [75].…”

“…RS-derived covariates were not strongly relevant for predicting PSFs in this study. The main reason was attributed to the effect of surficial soil moisture and roughness, the presence of vegetation canopy, and, in some cases, due to the low spatial resolution of the RS data [75,76]. All covariates derived from Landsat and MODIS imagery were classified as being weakly to slightly relevant covariates, whereas two covariates derived from the Sentinel-2 were moderately important.…”

The Impact of Acquisition Date on the Prediction Performance of Topsoil Organic Carbon from Sentinel-2 for Croplands

“…None of the recent works have investigated the impact that acquisition date may have on prediction performance for SOC content. Gomez et al [31] showed that several single-date S2 data acquired over a same study area may provide different performances of soil texture prediction, but they did not investigate the reasons for these differences. This is the purpose of the present study, considering an area where encouraging performance was previously obtained from a single spring date of March, for the Versailles Plain [17]: are there optimal dates for predicting topsoil SOC and what are the main factors disturbing the SOC prediction?…”

“…Remote Sens. 2018, 10, x FOR PEER REVIEW 10 of Histograms of Sentinel-1 (S1)-derived root mean square surface height (Hrms) values obtained fromBaghdadi et al (2018) [31] at the soil sample locations for five dates (red line: median value).…”

Evaluation of Vegetation Index-Based Curve Fitting Models for Accurate Classification of Salt Marsh Vegetation Using Sentinel-2 Time-Series

“…The above five VIs were chosen based on full consideration of the characteristics of salt marsh vegetation. Substantial variations in plant density among different salt marsh vegetation species were observed during our field surveys, ranging from >100 plants/m 2 (e.g., S. alterniflora , P. australis ) to <20 plants/m 2 (e.g., T. chinensis , S. salsa ), thereby necessitating that the impacts of vegetation saturation and soil background be considered [ 36 ]. Besides, given that salt marsh vegetation is frequently submerged by tides and that pioneer vegetation has a low plant height and density (e.g., S. mariqueter in particular), the water-related VI was employed.…”