Land cover changes have far-reaching effects on climate change and the evolution of ecological environments, making them important components of global change research. Based on MODIS land cover data (MCD12Q1), statistical parameters such as land cover dynamic degree, transition probability, and transition matrix were employed to analyse the spatiotemporal characteristics of global land cover changes from 2001 to 2020. The results indicated that land cover exhibits a latitudinal distribution in mid-latitude regions and a longitudinal distribution in high and low-latitude regions. During the study period, the areas of evergreen needleleaf forests, deciduous needleleaf forests, deciduous broadleaf forests, closed shrublands, and permanent wetlands exhibited fluctuating changes. Evergreen broadleaf forests and barren lands have decreased annually, whereas mixed forests, woody savannas, grasslands, urban and built-up areas, and cropland/natural vegetation mosaics have increased annually. In terms of the dynamic degree, the most significant changes occurred in deciduous broadleaf forests, closed shrublands, permanent wetlands, and cropland/natural vegetation mosaics. Additionally, there were significant variations in the types and regions of land cover changes and transitions. Evergreen needleleaf forests transitioned into grassland types, and the conversion between grassland types was the main transition from mid- to high- latitude regions. In tropical areas, evergreen broadleaf forests, open shrublands, and croplands have shifted towards grassland types as primary transitions. These findings provide significant implications for the in-depth exploration of land-atmosphere interactions, climate change, and numerical models.
Keywords: Land cover changes; MODIS; Regional variations; Spatial-temporal distributions.
Copyright © 2023 Elsevier B.V. All rights reserved.