Assessing the Utility of Sentinel-1 Coherence Time Series for Temperate and Tropical Forest Mapping

Ignacio Borlaf-Mena,Ovidiu Badea,Mihai Andrei Tanase

doi:10.3390/rs13234814

Abstract

This study tested the ability of Sentinel-1 C-band to separate forest from other common land use classes (i.e., urban, low vegetation and water) at two different sites. The first site is characterized by temperate forests and rough terrain while the second by tropical forest and near-flat terrain. We trained a support vector machine classifier using increasing feature sets starting from annual backscatter statistics (average, standard deviation) and adding long-term coherence (i.e., coherence estimate for two acquisitions with a large time difference), as well as short-term (six to twelve days) coherence statistics from annual time series. Classification accuracies using all feature sets was high (>92% overall accuracy). For temperate forests the overall accuracy improved by up to 5% when coherence features were added: long-term coherence reduced misclassification of forest as urban, whereas short-term coherence statistics reduced the misclassification of low vegetation as forest. Classification accuracy for tropical forests showed little differences across feature sets, as the annual backscatter statistics sufficed to separate forest from low vegetation, the other dominant land cover. Our results show the importance of coherence for forest classification over rough terrain, where forest omission error was reduced up to 11%.

Highlights

Forest ecosystems host a large portion of terrestrial biodiversity, and provide many ecosystem services, such as timber and food production, risk mitigation, and climate regulation, as forests hold a large portion of terrestrial biomass, and its growth and degradation play an essential role on climate and atmospheric CO2 dynamics
We used preexisting land cover maps as data sources to generate a consistent layer for training andpreexisting validation purposes including: We land cover maps as data sources to generate a consistent layer for Weused used preexisting land cover maps as data sources to generate a consistent layer and validation purposes including:
Richer feature sets were tested starting with annual backscatter statistics and adding long-term coherence as well as short-term coherence statistics

Summary

Introduction

Forest ecosystems host a large portion of terrestrial biodiversity, and provide many ecosystem services, such as timber and food production, risk mitigation (i.e., flood, erosion), and climate regulation, as forests hold a large portion of terrestrial biomass, and its growth and degradation play an essential role on climate and atmospheric CO2 dynamics. This has prompted several international agreements to preserve forest services and biodiversity, along with specific procedures to track forest cover and status. Extent ofofthe study areas ((A), temperate, Romania; (B),(B), tropical, Brazil). The the white outline represents the extent of the sites. Background imagery is courtesy of Google Satellite. We used preexisting land cover maps as data sources to generate a consistent layer for training andpreexisting validation purposes including: We land cover maps as data sources to generate a consistent layer for Weused used preexisting land cover maps as data sources to generate a consistent layer and validation purposes including: training

Objectives

Methods

Results

Discussion

Conclusion