LiDAR-derived surface roughness texture mapping: Application to mount St. Helens pumice plain deposit analysis

Abstract

Statistical measures of patterns (textures) in surface roughness are used to quantitatively differentiate volcanic deposit facies on the Pumice Plain, on the northern flank of Mount St. Helens (MSH). Surface roughness values are derived from a Light Detection and Ranging (LiDAR) point cloud collected in 2004 from a fixed-wing airborne platform. Patterns in surface roughness are characterized using co-occurrence texture statistics. Pristine-pyroclastic, reworked-pyroclastic, mudflow, boulder beds, eroded lava flows, braided streams, and other units within the Pumice Plain are all found to have significantly distinct roughness textures. The MSH deposits are reasonably accessible, and the textural variations have been verified in the field. Results of this work indicate that by affecting the distribution of large clasts and tens-of-meter scale landforms, modification of pyroclastic deposits by lahars alters the morphology of the surface in detectable quantifiable ways. When a lahar erodes a pyroclastic deposit, surface roughness increases, as does the randomness in the deposit surface. Conversely, when a lahar deposits material, the resulting landforms are less rough but more random than pristine pumice-rich pyroclastic deposits. By mapping these relationships and others, volcanic deposit facies can be differentiated. This new method of mapping, based on roughness texture, has the potential to aid mapping efforts in more remote regions, both on this planet and elsewhere in the solar system. © 1980-2012 IEEE.