Analysis and Predicting the Trend of Land Use/Cover Changes Using Neural Network and Systematic Points Statistical Analysis (SPSA)

Kalkhajeh, Reza Ghorbani; Jamali, Ali Akbar

doi:10.1007/s12524-019-00995-7

Cited by 13 publications

(1 citation statement)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite its simplicity, it can exhibit remarkably rich behavior and is efficient in extracting realistic simulations of land use classes and other spatial structures (Marceau et al, 2013). This algorithm has been widely used in previous studies to identify LC classes (see Etemadi et al, 2018; Feng et al, 2012; Ghorbani Kalkhajeh & Jamali, 2019; Islam et al, 2018; Ozturk, 2015). These studies have shown that CA models are efficient in producing practical simulations of land cover patterns and other spatial structures.…”

Section: Introductionmentioning

confidence: 99%

Analysis and prediction of land cover changes using the land change modeler (LCM) in a semiarid river basin, Iran

et al. 2021

View full text Add to dashboard Cite

Predicting future land cover (LC) changes is an important step in the proper planning and management of watersheds. As a susceptible area to salinity and desertification, receiving only about 195 mm rainfall annually, the Hable‐Rud River basin is especially sensitive to land use/cover changes. Based on corrected LANDSAT satellite images for the years 1986, 2000, and 2017, the LC were extracted using the maximum likelihood (ML) method. LC changes were predicted by applying the land change modeler (LCM) for the basin. The kappa index for classification in 1986, 2000, and 2017 was 75, 78, and 81%, respectively. Using LCM, the prediction was accomplished for the year 2017 with a kappa index of above 74%. The LC map was predicted for year 2040. The analysis indicates that over the past 32 years, bare land, saline land, agricultural, industrial, and residential areas have increased by about 8, 6.2, 2.7, 0.63, and 0.48%, respectively; while rangeland area was decreased by 18%. Results also indicate that given the predicted LC for year 2040 in comparison with the reference year (2017), saline land, agricultural, industrial, and residential areas will be likely to continue to increase by about 3, 1.5, 0.7, and 0.8%, respectively, whereas bare land and rangeland will most likely decrease by about 4.55 and 1.45%, respectively. The findings of this study assist in analyzing the future trends of LC changes in the basin. This information can be used as a guide for land planners and managers in future land use planning of the area.

show abstract