{"title":"Mapping of LC/LU changes inside the Aghdam district of the Karabakh omics region applying object-based satellite image analysis","authors":"A. Rasouli, M. Asgarova, S. Safarov","doi":"10.29228/jlsb.22","DOIUrl":null,"url":null,"abstract":"Identification of the environmental consequences of the 30-year occupation of Karabakh and its adjacent territories by the Armenian armed formations is an important and urgent research task. Object-Based Image Analysis (OBIA) procedures were accordingly applied to examine the condition and changes in landcover and landuse (LC/LU) in the territories of Karabakh liberated from Armenian occupation within the Aghdam District. Firstly, Dynamic Thresholds Indexing (DTI) algorithms were operated to display the main LC by developing several spectral NDWI, NDVI, NBRI, and AVBI indices. At the next step, to recognize precise LU changes inside the study area, a rule-based Nearest Neighbour Classify (NNC) was considered by accompanying an advanced supervised classification technique within the Trimble eCognition setting (eCognition Developer, 2019). DTI results indicated that from 2016 to 2021 inside the Aghdam District, LC changes are quite meaningful. A significant decrease in vegetated cover (10.2 %), increases in the non-vegetated area (11.8 %), and the most noticeable changes are observed in vulnerable lands of about 45.1 km (26.8 %). Subsequently, the rule-based NNC method approved various negative LU changes inside the study area that had occurred predominantly to the mixed forest-pasture classes (9.8 %). Besides, the areas of degraded lands have increased by 35 % and barren lands by 4.4 % according to the study. It should be noted that water and agricultural LU demonstrate the least changes overall of 3.4 % and 0.3 %, respectively. The overall accuracy of 0.95 and Kappa statistics of 0.93 confirmed the significant changes in the final LC/LU productions. Consequently, accurate image processing and mapping of the current situation of the liberated regions of Azerbaijan have to be the most urgent tasks of the geographers, ecosystem scientists, and remote sensing specialists prior to the start of reconstruction and rehabilitation projects by government officials and decision-makers.","PeriodicalId":344223,"journal":{"name":"Journal of Life Sciences and Biomedicine of ANAS","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Life Sciences and Biomedicine of ANAS","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.29228/jlsb.22","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Identification of the environmental consequences of the 30-year occupation of Karabakh and its adjacent territories by the Armenian armed formations is an important and urgent research task. Object-Based Image Analysis (OBIA) procedures were accordingly applied to examine the condition and changes in landcover and landuse (LC/LU) in the territories of Karabakh liberated from Armenian occupation within the Aghdam District. Firstly, Dynamic Thresholds Indexing (DTI) algorithms were operated to display the main LC by developing several spectral NDWI, NDVI, NBRI, and AVBI indices. At the next step, to recognize precise LU changes inside the study area, a rule-based Nearest Neighbour Classify (NNC) was considered by accompanying an advanced supervised classification technique within the Trimble eCognition setting (eCognition Developer, 2019). DTI results indicated that from 2016 to 2021 inside the Aghdam District, LC changes are quite meaningful. A significant decrease in vegetated cover (10.2 %), increases in the non-vegetated area (11.8 %), and the most noticeable changes are observed in vulnerable lands of about 45.1 km (26.8 %). Subsequently, the rule-based NNC method approved various negative LU changes inside the study area that had occurred predominantly to the mixed forest-pasture classes (9.8 %). Besides, the areas of degraded lands have increased by 35 % and barren lands by 4.4 % according to the study. It should be noted that water and agricultural LU demonstrate the least changes overall of 3.4 % and 0.3 %, respectively. The overall accuracy of 0.95 and Kappa statistics of 0.93 confirmed the significant changes in the final LC/LU productions. Consequently, accurate image processing and mapping of the current situation of the liberated regions of Azerbaijan have to be the most urgent tasks of the geographers, ecosystem scientists, and remote sensing specialists prior to the start of reconstruction and rehabilitation projects by government officials and decision-makers.