Signature extension as a machine learning strategy for mapping plantation forest genera with Sentinel-2 imagery

Aug 5, 2026

Plantation forest inventory data, including tree genus, is essential for forestry planning, water use assessment, rotation and harvest monitoring, and policy decisions, yet it is often collected in the field, making it time-consuming and expensive. Machine learning with remote sensing can map genera at regional scales, but training labels are costly to acquire across large areas. This study evaluates signature extension to train random forest models on Sentinel-2 imagery across space, enabling classification among acacia, eucalyptus, and pine.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Forest plantation inventory and mapping needs

Limits of in-field data collection

Machine learning and remote sensing for forest species

Article focus and methods

Signature extension concept and experiments

Data tiling and model training with Sentinel-2

Accuracy evaluation using distance and environmental variation

Remote Sensing Applications: Society and Environment 33 (2024) 101136 Contents lists available at ScienceDirect Remote Sensing Applications: Society and Environment journal [homepage:](homepage: www.elsevier.com/locate/rsase)[ www.elsevier.com/locate/rsase](homepage: www.elsevier.com/locate/rsase) | Signature extension as a machine learning strategy for mapping plantation forest genera with Sentinel-2 imagery
*
Caley Higgs , Adriaan van Niekerk
Geography and Environmental Studies, Stellenbosch University, Stellenbosch, South Africa | | | | | --- | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords: Plantation forests Signature extension Remote sensing Machine learning | | Plantation forest inventory data such as tree genus is important for supporting forestry decisions and policymaking, but such data are traditionally collected in-field, which is time-consuming and costly. Although machine learning and remote sensing technologies have shown great potential to reduce the time and effort for mapping forest plantation genera at regional scales, they rely on training (labelled) data, which are also costly to collect over large areas. One approach to reducing the effort of training data collection is to make use of signature extension, whereby training data collected in one area is used to train and apply a machine learning model in a different area. This study aimed to evaluate the viability of training data signature extension for constructing random forest (RF) machine learning models to differentiate between acacia, eucalyptus and pine trees using Sentinel-2 imagery as input. The study was carried out over a large area (about 4920 km2) in South Africa. The study area was divided into 19 tiles of 100 × 100 km (each tile coincides with the footprint of a Sentinel-2 tile) from which three were chosen for sourcing (collecting) training data. Four experiments were conducted. In the first experiment, a fixed number (3000 per genera) of training samples were collected in the first source tile and used to build an RF model. The resulting model was then applied and assessed in all 19 tiles. This protocol was repeated in the second and third experiments using the training data collected in the other two source tiles respectively. In the final experiment, training data from all three source tiles were combined and applied to all 19 tiles. The mean overall classification accuracy of each classified tile was compared to the extension distance (i.e. the distance between the target and source tile), differences in rainfall seasonality, and variation in the mean annual temperature among tiles to gain an understanding of how signature extension efficiency is influenced by distance and environmental conditions. The results show that signature extension is viable (~70% overall accuracies) over distances of up to 500 km, but only if the source and target tiles represent areas with similar rainfall regimes. | | 1. Introduction Forest plantation inventory data includes the location, extent, planting date, tree species/genus/clone, water use, and yield at individual compartmental levels (Mati and Dawaki, 2015). Such data are fundamental for forest management, which involves planning, land management, analysing growth rates to maximise production, assessing water use, monitoring rotations and harvests. Forest inventories are useful to monitor the status of forests, identify trends in the forestry industry, model climate change, carry out * Corresponding author. E-mail [address:](address: higgscaley@gmail.com)[ higgscaley@gmail.com](address: higgscaley@gmail.com) (C. Higgs). [https://doi.org/10.1016/j.rsase.2023.101136](https://doi.org/10.1016/j.rsase.2023.101136) Received 22 May 2023; Received in revised form 27 November 2023; Accepted 29 December 2023 Available online 3 January 2024 2352-9385/© 2024 Elsevier B.V. All rights reserved. C. Higgs and A. van Niekerk Remote Sensing Applicatio

FAQs

What problem does the study address in plantation forest mapping?

It addresses the high cost of collecting labeled in-field training data for machine learning models used to map plantation forest tree genus at regional scales.

How does signature extension work in this research?

Training samples collected in one Sentinel-2 tile are used to build a random forest model, which is then applied to other tiles to classify acacia, eucalyptus, and pine.

Under what conditions does signature extension remain accurate?

The results indicate viability over distances up to about 500 km, but only when the source and target tiles share similar rainfall regimes.

Aug 5, 2026

Upload by Evangeline