Land use and land cover change in Duzce region following the major earthquake: implications for ANN and Markov Chain Analysis

摘要

Earthquakes are natural events that affect the spatial and temporal dynamics of land use/land cover. Here, we have focused on the city of Duzce, NW of Turkey, which has experienced the M＞7 doublet earthquakes in 1999. In order to conduct land cover change analysis and future projection, we have utilized integrated the multilayer perceptron neural network-based Markov chain model combined with geographical information systems. Our results show that most of the land use in 1990 was 77,601 hectares (65.89% of the region) consist of forest and shrubs. By 2020, agricultural land was the most expanding land class in the region with 48.80%. The 1999 earthquake caused the city to grow horizontally in the long run, and this expansion has ended the occupation in grade-one-agricultural lands that prevailed in the region. This horizontal development also caused the residential areas that covered 2.76% of the study area in 1999 expand to the ratio of 20.37% in 2050. It has been determined that the most effective driver in these changes in the land cover is the distance from urban with a Cramer's value of 0.4299. For this reason, the area that is most exposed to land cover change is the forest and shrub areas occupied by the city, which has spread to the northeast due to the earthquake. It is concluded that the earthquakes can change the land cover of the city significantly in various aspects in the long term, contrary to the approved zoning plans.