Spatio-temporal evolutional characteristics of landscape patterns in the Loess Plateau in China — A landscape metrics-based assessment

Knowledge on the local landscape pattern changes can provide the necessary information for developing strategies for sustainable land management. Landscape metrics have been widely used to investigate landscape patterns quantitatively in the last decades. Based on the TM images for three temporal stages (2000, 2005 and 2010) in the Baota district, Yanan city of Shannxi province, this study applied a hill-slope analytical model and landscape ecological method to examine three main types of land use/cover changes (LUCC) and their effects on landscape structure since the implementation of the Grain for Green (GFG) project (2000–2010). The LUCC in the Loess Plateau have witnessed significant changes since 2000. Results showed that between 2000 and 2010, farmland, grassland and woodland were the main types of land covers in the Baota district, and the area of farmland has decreased whereas that of grassland and woodland has increased. Compared to 2000, the overall landscape structure in 2010 has become more tight, stable and aggregated, mainly due to the effect of the GFG project. Land-cover changes mainly occurred on regions with a slope less than 15 degrees and ranging between 15 and 25 degrees. Over the last decade, the farmland in the Baota district showed a more aggregated spatial pattern, whereas woodland and grassland exhibited a more fragmented pattern. The effect of the GFG project on the LUCC in the period 2005–2010 was more remarkable than that in the period 2000–2005. Furthermore, the scattered farmlands were gradually transferred into grassland or woodland. Further estimations showed that about 10,375 hectares of farmland on the slopes greater than 25 degrees still need to implement the GFG project, which indicates that the project has a great potential at steep slope zones for the improvement of ecological reconstruction in the Loess Plateau in the future.