Novel shape indices for vector landscape pattern analysis

The formation of an anisotropic landscape is influenced by natural and/or human 2 processes, which can then be inferred on the basis of geometric indices. In this 3 study, two minimal bounding rectangles in consideration of the principles of 4 mechanics (i.e. minimal width bounding (MWB) box and moment bounding 5 (MB) box) were introduced. Based on these boxes, four novel shape indices, 6 namely MBLW (the length-to-width ratio of MB box), PAMBA (area ratio 7 between patch and MB box), PPMBP (perimeter ratio between patch and MB 8 box) and ODI (orientation difference index between MB and MWB boxes), were 9 introduced to capture multiple aspects of landscape features including patch 10 elongation, patch compactness, patch roughness and patch symmetry. Landscape 11 pattern was, thus, quantified by considering both patch directionality and patch 12 shape simultaneously, which is especially suitable for anisotropic landscape 13 analysis. The effectiveness of the new indices were tested with real landscape 14 data consisting of three kinds of saline soil patches (i.e. the elongated shaped 15 slightly saline soil class, the circular or half-moon shaped moderately saline soil, 16 and the large and complex severely saline soil patches). The resulting 17 classification was found to be more accurate and robust than that based on 18 traditional shape complexity indices. 19