科研之友
微信
新浪微博
Facebook
分享链接
摘要
This study first sought to isolate a select group of landscape metrics particularly well-suited for describing dryland Mediterranean landscapes in Jordan. We examined the response of 50 landscape metrics to a large range of imagery grain sizes. Most of the metrics exhibited an expected behavior, similar to what has been previously reported in literature such as (a) a predictable (linear or power law) response to changing grain size, and (b) an unpredictable (staircase-like or erratic) response to changing grain size. Some metrics, however, exhibited a domain of scale effect, in particular the core area metrics. Using correlation analysis, the original 50 metrics were placed into 19 groups such that all metrics within a group were strongly correlated with each other, and were represented by a single representative metric. Using these representative metrics in the context of principal components analysis, we then found that six factors explained 95.35% of the total variation found in the landscape pattern. The highest loadings for these six factors, in order, were the number of patches (NP), mean proximity index (PROX_MN), largest patch index (LPI), patch cohesion index (COHESION), total core area (TCA), and the proximity index coefficient of variation (PROX_CV). It was concluded that east Mediterranean landscapes with a long history of anthropogenic-driven change showed a domain of scale for core area metrics. We also recommend that the majority of the pattern in dry Mediterranean landscapes, particularly those in Jordan, can be described with six metrics. We suggest that our procedure for landscape metric selection can be utilized in other regions of study as well.
- 出版日期2011-3
