Fusion of Sentinel-1A and Sentinel-2A data for land cover mapping: a case study in the lower Magdalena region, Colombia

Nicola Clerici,Cesar Augusto Valbuena Calderón,Juan Manuel Posada

doi:10.1080/17445647.2017.1372316

Nicola Clerici, Cesar Augusto Valbuena Calderón + Show 1 more

Open Access

https://doi.org/10.1080/17445647.2017.1372316

Copy DOI

Abstract

ABSTRACTLand cover–land use (LCLU) classification tasks can take advantage of the fusion of radar and optical remote sensing data, leading generally to increase mapping accuracy. Here we propose a methodological approach to fuse information from the new European Space Agency Sentinel-1 and Sentinel-2 imagery for accurate land cover mapping of a portion of the Lower Magdalena region, Colombia. Data pre-processing was carried out using the European Space Agency’s Sentinel Application Platform and the SEN2COR toolboxes. LCLU classification was performed following an object-based and spectral classification approach, exploiting also vegetation indices. A comparison of classification performance using three commonly used classification algorithms was performed. The radar and visible-near infrared integrated dataset classified with a Support Vector Machine algorithm produce the most accurate LCLU map, showing an overall classification accuracy of 88.75%, and a Kappa coefficient of 0.86. The proposed mapping approach has the main advantages of combining the all-weather capability of the radar sensor, spectrally rich information in the visible-near infrared spectrum, with the short revisit period of both satellites. The mapping results represent an important step toward future tasks of aboveground biomass and carbon estimation in the region.

Highlights

Land cover mapping has been recognized as a fundamental task for deriving information for scientific, environmental management and policy purposes at global, regional and local scales (Feddema et al, 2005; Turner, Lambin, & Reenberg, 2007)
ALOS PALSAR and phenological information from the MODIS sensor were used by Qin et al (2016) to map forests in Monsoon Asia using decision tree algorithms
In this research we explored the integrated use of the recently launched Sentinel-1A and Sentinel-2A satellites for synergistic land cover mapping exploiting radar and optical data, for a case study in Colombia

Summary

Introduction

Land cover mapping has been recognized as a fundamental task for deriving information for scientific, environmental management and policy purposes at global, regional and local scales (Feddema et al, 2005; Turner, Lambin, & Reenberg, 2007). In this research we explored the integrated use of the recently launched Sentinel-1A and Sentinel-2A satellites for synergistic land cover mapping exploiting radar and optical data, for a case study in Colombia. Frequent revisit time is a major advantage over previous radar missions, especially for the mapping and analysis of phenological dynamics in vegetation and agricultural land covers, together with the dual polarization capability and rapid product delivery (Torres et al, 2012). The Sentinel-2 Multi-spectral Instrument (MSI) sensors provide radiometrically and geometrically superior multi-spectral high spatial resolution images over the global surface, at high revisit time (5 days at the Equator with two satellites in orbit) and a wide field of view covering 290 km with 13 bands in the optical NIR, SWIR parts of the electromagnetic spectrum (Drusch et al, 2012). The study was motivated by the need for both cost-effective and accurate land cover map production in a cloudy region (Colombian Andes), as part of an environmental characterization of the nationally funded project ‘Strategies for natural resources valuation and ownership as a climate change adaptation mechanism in the Lower Magdalena region, Colombia’

Study area

Satellite imagery pre-processing

Land cover mapping

Findings

Conclusion