Semi-Automatic Methodology for Fire Break Maintenance Operations Detection with Sentinel-2 Imagery and Artificial Neural Network

Uncovering Vegetation Changes in the Urban–Rural Interface through Semi-Automatic Methods

Forest fires are considered by Portuguese civil protection as one of the most serious natural disasters due to their frequency and extent. To address the problem, the Fire Forest Defense System establishes the implementation of fuel management bands to aid firefighting. The aim of this study was to develop a model capable of identifying vegetation removal in the urban–rural interface defined by law for fuel management actions. The model uses normalised difference vegetation index (NDVI) of Sentinel-2 images time series and is based on the Welch t-test to find statistically significant differences between (i) the value of the NDVI in the pixel; (ii) the mean of the NDVI in the pixels of the same land cover type in a radius of 500 m; and (iii) their difference. The model identifies a change when the t-test points for a significant difference of the NDVI value in the ‘pixel’ as comparted to the ‘difference’ but not the ‘mean’. We use a moving window limited to 60 days before and after the analysed date to reduce the phenological variations of vegetation. The model was applied in five municipalities of Portugal and the results are promising to identify the places where the management of fuel bands was not carried out. This indicates which model could be used to assist in the verification of the annual management of the fuel bands defined in the law.

Single-Image Super-Resolution of Sentinel-2 Low Resolution Bands with Residual Dense Convolutional Neural Networks

Sentinel-2 satellites have become one of the main resources for Earth observation images because they are free of charge, have a great spatial coverage and high temporal revisit. Sentinel-2 senses the same location providing different spatial resolutions as well as generating a multi-spectral image with 13 bands of 10, 20, and 60 m/pixel. In this work, we propose a single-image super-resolution model based on convolutional neural networks that enhances the low-resolution bands (20 m and 60 m) to reach the maximal resolution sensed (10 m) at the same time, whereas other approaches provide two independent models for each group of LR bands. Our proposed model, named Sen2-RDSR, is made up of Residual in Residual blocks that produce two final outputs at maximal resolution, one for 20 m/pixel bands and the other for 60 m/pixel bands. The training is done in two stages, first focusing on 20 m bands and then on the 60 m bands. Experimental results using six quality metrics (RMSE, SRE, SAM, PSNR, SSIM, ERGAS) show that our model has superior performance compared to other state-of-the-art approaches, and it is very effective and suitable as a preliminary step for land and coastal applications, as studies involving pixel-based classification for Land-Use-Land-Cover or the generation of vegetation indices.

An Integrated Decision Support System for Improving Wildfire Suppression Management

Wildfires are expected to increase in number, extent, and severity due to climate change. Hence, it is ever more important to integrate technological developments and scientific knowledge into fire management aiming at protecting lives, infrastructure, and the environment. In this paper, a decision support system (DSS) adapted to the Portuguese context and based on multi-sensor technologies and geographic information system (GIS) functionalities is proposed to leverage operational data, enabling faster and more informed decisions to reduce the impact of wildfires. Here we present a flexible and reconfigurable DSS composed of three components: an ArcGIS online feature service that provides operational data and enables a collaborative environment of users that share operational data in near real-time; a mobile client application to interact with the system, enabling the use of GIS technology and visualization dashboards; and a multi-sensor device that collects field data providing value to external services. The design and validation of this system benefitted from the feedback of wildfire management specialists and a partnership with an end-user in the municipality of Mação that also helped establish the system requirements. The validation results demonstrated that a robust system was achieved with fully interoperable components that fulfill the defined system requirements.

Assessing the Natural Recovery of Mangroves after Human Disturbance Using Neural Network Classification and Sentinel-2 Imagery in Wunbaik Mangrove Forest, Myanmar

In this study, we examined the natural recovery of mangroves in abandoned shrimp ponds located in the Wunbaik Mangrove Forest (WMF) in Myanmar using artificial neural network (ANN) classification and a change detection approach with Sentinel-2 satellite images. In 2020, we conducted various experiments related to mangrove classification by tuning input features and hyper-parameters. The selected ANN model was used with a transfer learning approach to predict the mangrove distribution in 2015. Changes were detected using classification results from 2015 and 2020. Naturally recovering mangroves were identified by extracting the change detection results of three abandoned shrimp ponds selected during field investigation. The proposed method yielded an overall accuracy of 95.98%, a kappa coefficient of 0.92, mangrove and non-mangrove precisions of 0.95 and 0.98, respectively, recalls of 0.96, and F1 scores of 0.96 for the 2020 classification. For the 2015 prediction, transfer learning improved model performance, resulting in an overall accuracy of 97.20%, a kappa coefficient of 0.94, mangrove and non-mangrove precisions of 0.98 and 0.96, respectively, recalls of 0.98 and 0.97, and F1 scores of 0.96. The change detection results showed that mangrove forests in the WMF slightly decreased between 2015 and 2020. Naturally recovering mangroves were detected at approximately 50% of each abandoned site within a short abandonment period. This study demonstrates that the ANN method using Sentinel-2 imagery and topographic and canopy height data can produce reliable results for mangrove classification. The natural recovery of mangroves presents a valuable opportunity for mangrove rehabilitation at human-disturbed sites in the WMF.

National Scale Land Cover Classification for Ecosystem Services Mapping and Assessment, Using Multitemporal Copernicus EO Data and Google Earth Engine

Land-Use/Land-Cover (LULC) products are a common source of information and a key input for spatially explicit models of ecosystem service (ES) supply and demand. Global, continental, and regional, readily available, and free land-cover products generated through Earth Observation (EO) data, can be potentially used as relevant to ES mapping and assessment processes from regional to national scales. However, several limitations exist in these products, highlighting the need for timely land-cover extraction on demand, that could replace or complement existing products. This study focuses on the development of a classification workflow for fine-scale, object-based land cover mapping, employed on terrestrial ES mapping, within the Greek terrestrial territory. The processing was implemented in the Google Earth Engine cloud computing environment using 10 m spatial resolution Sentinel-1 and Sentinel-2 data. Furthermore, the relevance of different training data extraction strategies and temporal EO information for increasing the classification accuracy was also evaluated. The different classification schemes demonstrated differences in overall accuracy ranging from 0.88% to 4.94% with the most accurate classification scheme being the manual sampling/monthly feature classification achieving a 79.55% overall accuracy. The classification results suggest that existing LULC data must be cautiously considered for automated extraction of training samples, in the case of new supervised land cover classifications aiming also to discern complex vegetation classes. The code used in this study is available on GitHub and runs on the Google Earth Engine web platform.

Fully Automated Countrywide Monitoring of Fuel Break Maintenance Operations

Fuel break (FB) networks are strategic locations for fire control and suppression. In order to be effective for wildfire control, they need to be maintained through regular interventions to reduce fuel loads. In this paper, we describe a monitoring system relying on Earth observations to detect fuel reduction inside the FB network being implemented in Portugal. Two fast automated pixel-based methodologies for monthly monitoring of fuel removals in FB are developed and compared. The first method (M1) is a classical supervised classification using the difference and postdisturbance image of monthly image composites. To take into account the impact of different land cover and phenology in the detection of fuel treatments, a second method (M2) based on an innovative statistical change detection approach was developed. M2 explores time series of vegetation indices and does not require training data or user-defined thresholds. The two algorithms were applied to Sentinel-2 10 m bands and fully processed in the cloud-based platform Google Earth Engine. Overall, the unsupervised M2, which is based on a Welch t-test of two moving window averages, gives better results than the supervised M1 and is suitable for an automated countrywide fuel treatment detection. For both methods, two vegetation indices, the Modified Excess of Green and the Normalized Difference Vegetation Index, were compared and exhibited similar performances.