Earth Observation satellite systems are considered the main source of information used for delivering up-to-date land cover/use maps. Medium to high spatial resolution images, such as the ones provided by Sentinel-2 sensors, can improve significantly mapping and monitoring of vegetation communities and are utilized in a wide range of applications such as the management of natural resources and forest inventories. The aim of this work was to employ Sentinel-2 images for accurately classifying vegetation cover in selected areas of Greece that present diverse vegetation characteristics. Cloudfree Sentinel 2 (L2A) images were acquired for each area during 2021 for the months of February, June, and September, in order to capture the reflectance changes due to seasonal variations. Two machine-learning techniques, namely Random Forest (RF) and Support Vector Machines (SVM), were applied and assessed for their performance in mapping vegetation cover and species in the study areas. The training patterns, used as input in both classifiers, were acquired through photo-interpretation of stratified random points, distributed across forested areas. Consequently, validation of the classification results was performed, in order to estimate accuracy metrics for each model per site. More specifically, the kappa coefficient, overall (OA), user’s and producers’ accuracy were calculated. The accuracy results demonstrated higher scores for RF (OA over 90% for all areas) than SVM (OA ranging from 81 to 89%, respectively). Overall, our study demonstrates the capability of seasonal Sentinel-2 data to accurately discriminate vegetation communities over diverse biomes, when combined with advanced classification methods.
Burned area mapping is essential for quantifying the environmental impact of wildfires, for compiling statistics, and for designing effective short- to mid-term impact mitigation measures. The Sentinel-2 satellites are providing an unparalleled wealth of high-resolution remotely sensed information with a short revisit cycle, which is ideal for mapping burned areas both accurately and timely. However, the high detail and volume of the information provided actually encumbers the automation of the mapping process, at least for the level of automation required to map systematically wildfires on a national level. This paper presents a preliminary methodology for mapping burned areas using Sentinel-2 data, which aims to eliminate user interaction and achieve mapping accuracy that is acceptable for operational use. It follows an objectbased image analysis (OBIA) approach, whereby the initial image is segmented into a set of adjacent and non-overlapping small regions (objects). The most unambiguous of them are labeled automatically through a set of empirical rules that combine information extracted from both a pre-fire Sentinel-2 image and a post-fire one. The burned area is finally delineated following a supervised learning approach, whereby a Support Vector Machine (SVM) is trained using the labeled objects and subsequently applied to the whole image considering a set of optimally selected object-level features. Preliminary results on a set of recent large wildfires in Greece indicate that the proposed methodology constitutes a solid basis for fully automating the burned area mapping process.
The significance of forest ecosystems in terms of ecosystem processes and services and impacts on humanity is fully acknowledged. The constant exploitation of natural resources and the increasing anthropogenic pressure on ecosystems continue to put a strain on and irretrievably threaten global forest ecosystems. Global forest health is declining due to climate change, air pollution and increased human activities. Protecting and monitoring the health of forest ecosystems is a vital resource management function. The technological development in the field of remote sensing provides new tools and automated solutions for forest health monitoring. An effective web-based forest health monitoring platform can contribute to ecological, social, and economic aspects. This study aims to design rapid and automated workflows (Spatial Models-SMs) for time-series forest health monitoring with flexible parameterization and user-friendly interfaces ready for feeding WPS web-GIS platforms. Those include: i) SMs that ingest available time-series data and perform preprocessing activities, ii) SMs that calculate time-series of vegetation, soil and water indices from multispectral optical imagery, iii) SMs that create colored composite images from image algebra and SAR polarizations and vi) SMs that extract change detection maps from time-series SAR data. The study area is located in the wider region of the Mouzaki, Greece, where various types of forest species can be found. Sentinel-1 & 2 data were used while the ERDAS IMAGINE software was utilized for the design of the SMs. The results indicate the potential of the designed SMs to feed WPS web- GIS platforms promptly and efficiently.
Fire danger forecast constitutes one of the most important components of integrated fire management since it provides
crucial information for efficient pre-fire planning, alertness and timely response to a possible fire event. The aim of this
work is to develop an index that has the capability of predicting accurately fire danger on a mid-term basis. The
methodology that is currently under development is based on an innovative approach that employs dry fuel spatial
connectivity as well as biophysical and topological variables for the reliable prediction of fire danger. More specifically,
the estimation of the dry fuel connectivity is based on a previously proposed automated procedure implemented in R
software that uses Moderate Resolution Imaging Spectrometer (MODIS) time series data. Dry fuel connectivity estimates
are then combined with other ancillary data such as fuel type and proximity to roads in order to result in the generation of
the proposed mid-term fire danger index. The innovation of the proposed index—which will be evaluated by comparison
to historical fire data—lies in the fact that its calculation is almost solely affected by the availability of satellite data.
Finally, it should be noted that the index is developed within the framework of the National Observatory of Forest Fires
(NOFFi) project.
During the past decades, forest fires have increased both in frequency and severity thus, increasing the life threats for people and environment and leading countries to spend vast amounts of resources in fighting forest fires. Besides anthropogenic activities, climatic and environmental changes are considered as driving factors affecting fire occurrence and vegetation succession. Especially in the Mediterranean region, the development and existence of effective tools and services is crucial for assisting pre-fire planning and preparedness. The collaborative project NFOFRAS aims at introducing an innovative and effective system for rating forest fire risk, and is based on existing technology and standards that have been developed by countries with a long and a very successful involvement in this field. During the first phase of the project a detailed documentation of the proposed methodology was composed. In addition, Earth Observation (EO) and meteorological datasets were utilized for producing accurate pre-fire measurements over a selected study area in Greece.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.