Unsupervised classification of lava flows in Harrat Lunayyir using remote sensing and GIS

Abstract

Mapping of lava flows based on remote sensing data of high accuracy has become a common tool for exploring volcanic eruptions in greater detail. Mapping data based on remote sensing data provides information on the location and flow direction of lava flows as well as their areas and enables the localisation of volcanic vents-valuable knowledge for pre-and post-eruption volcanic activity and behaviour estimations. The current research seeks to expand the understanding of the volcanic and tectonic processes in Harrat Lunayyir, Saudi Arabia. In view of this, remote sensing data was classified using iterative self-organising data analysis technique (ISODATA) algorithms. The aim of the classification is to identify spectral top-of-atmosphere (TOA) reflectance values to distinguish layers of old and recent lava flows based on differences. As a result, three distinct basaltic units were identified to have the following ages: 15.1 ± 6.1 ka (4%), 15.0 ± 8.4 ka (6%), and 14.6 ± 23.1 ka (10%). The differences in the calculated areas of the lava flows as mapped based on the remote sensing data and earlier modelling are captured and identified in relation to the local geomorphologic and geologic structures. The differences can be partly explained as being related to a considerable weathering of the observed geological formations at 800-1600 nm, with reflectance of 12%. The limitation of the methodology relates to the lack of accurate geochronological timeline, that is, an inability to identify accurate age of the samples.

Geothermal Exploration Using Remote Sensing, Surface Temperature, and Geophysical Data in Lunayyir Volcanic Field, Saudi Arabia

“…The proximity to metropolitan areas is advantageous for power production, as it reduces construction and power line costs. The Harrats (e.g., Rahat, Lunayyir, and Khaybar) are promising, although the distribution of permeability remains uncertain [32,44,45].…”

“…The seismic activity and its location indicate that Harrat Lunayyir could potentially have areas with high permeability that could be exploited for geothermal energy [44]. The study area has a potential area for geothermal resources due to its location in a tectonically active region with associated volcanic activity.…”

“…Temperature Range ( The temperature anomaly patterns in the LVF most likely reflect the emplacement of a very shallow dyke intrusion [44]. Geothermal exploration requires the identification of areas with high subsurface temperatures.…”

“…Statistical analysis of the subsurface temperature data from the surrounding area outside the LVF at the well location (Source of data:[44]). …”

Unsupervised Classification of Landsat-8 Satellite Imagery-Based on ISO Clustering

“…When novel data is presented, the model employs the traits it has already learned to identify the data's class. Its primary applications are feature reduction and clustering [1,20,[31][32][33][34]. When provided with an unidentified pixel class training set, an unsupervised classification or clustering algorithm aggregates or groups image pixels into a predetermined number of natural clusters in the feature space.…”