Land use/land cover classification using machine learning models

<p>An ensemble model has been proposed in this work by combining the extreme gradient boosting classification (XGBoost) model with support vector machine (SVM) for land use and land cover classification (LULCC). We have used the multispectral Landsat-8 operational land imager sensor (OLI) data with six spectral bands in the electromagnetic spectrum (EM). The area of study is the administrative boundary of the twin cities of Odisha. Data collected in 2020 is classified into seven land use classes/labels: river, canal, pond, forest, urban, agricultural land, and sand. Comparative assessments of the results of ten machine learning models are accomplished by computing the overall accuracy, kappa coefficient, producer accuracy and user accuracy. An ensemble classifier model makes the classification more precise than the other state-of-the-art machine learning classifiers.</p>

Assessment of land use cover changes, carbon sequestration and carbon stock in dry temperate forests of Chilas watershed, Gilgit-Baltistan

Land use and land cover change are affecting the global environment and ecosystems of the different biospheres. Monitoring, reporting and verification (MRV) of these changes is of utmost importance as they often results in several global environmental consequences such as land degradation, mass erosion, habitat deterioration as well as micro and macro climate of the regions. The advance technologies like remote sensing (RS) and geographical information system (GIS) are helpful in determining/ identifying these changes. In the current study area, changes in carbon stocks, notably in forest areas, are resulting in considerable dynamics of carbon stocks as a result of climate change and carbon sequestration. This study was carried out in the Diamer district of the Gilgit Baltistan (GB) Pakistan to investigate the change in cover change/land use change (particularly Forest Land use) as well as carbon sequestration potential of the forests in the district during almost last 25years. The land cover, temporal Landsat data (level 1, LIT) were downloaded from the USGS EROS (2016), for 1979-1989, 1990-2000 and 2001-2012. Change in land uses, particularly forest cover was investigated using GIS techniques. Forest inventory was carried out using random sampling techniques. A standard plot of size 0.1 ha (n=80) was laid out to determine the tree density, volume, biomass and C stocks. Simulation of C stocks was accomplished by application of the CO2FIX model with the data input from inventory. Results showed a decrease in both forest and snow cover in the region from 1979-2012. Similarly decrease was seen in tree volume, tree Biomass, dynamics of C Stocks and decrease was in occur tree density respectively. It is recommended we need further more like project such as BTAP (Billion Tree Afforestation Project) and green Pakistan project to increase the forest cover, to control on land use change, protect forest ecosystem and to protect snow cover.