Calibration interval scenario approach in spatial modeling of land cover change in East Kalimantan from 2016 to 2036

Abstract Spatial modeling can be used to predict future land cover changes based on past and present conditions. However, it is not yet known what extent this model the with reliable accuracy. Therefore, by using multi-temporal data, study aims build an optimal calibration interval scenario. The then analyze in East Kalimantan 2016–2036. 11 classified maps from Landsat Time Series Random Forest Google Earth Engine are 14 scenarios. A Change Modeler change driving variables. results of classification show a good level accuracy, Overall Accuracy value 71.43–85.14% Kappa 0.667–0.827. Then 2016–2021 one best scenarios 5-year intervals where accuracy predictions still relied upon for up three prediction iterations. scenario approach spatial decrease forest 2016 2021, deforestation rate 651 km 2 /year. 2036 estimates that remaining area 69.203 . It believed topography most influential variable Kalimantan.