A set of equations linking the time-spread of a laser altimeter echo-profile, commonly known as the pulse-width, to the variance of topography within the pulse-footprint are tested by comparing pulse-width data to surface characteristics measured from high-resolution Digital Terrain Models. The research is motivated by the advent of high-resolution Digital Terrain Models over Mars, which enables the calibration of Mars Orbiter Laser Altimeter pulse-widths, and evolves to include lunar and terrestrial data in an attempt to validate the theory and develop new methods. Analysis of Mars Orbiter Laser Altimeter pulse-width data reveals mixed results. Over homo- geneously rough terrain, at kilometre-scales, these pulse-widths show some correlation to surface characteristics, once poor pulse data has been removed. However, where roughness is highly vari- able over short baselines, little correlation is observed, which is attributed to a mix of georeferencing errors and instrument methods. In a similar study, Lunar Orbiter Laser Altimeter pulse-widths are shown to produce only poor correlations with surface characteristics over local study sites. Instead, the observed correlations differ from orbit to orbit, with the majority of those used appearing to contain poor quality pulse- width data - attributed to the instrument methods - and only 14 % revealing correlations similar, or better, than observed over Mars. Finally, an examination of the relationship between footprint-scale surface characteristics and pulse-width estimates derived from smoothed Ice, Cloud, and land Elevation Satellite echo-profiles enables different pulse-width thresholds to be tested. Here, pulse-widths measured using a 10 % Peak Energy threshold are shown to produce greater correlations than those observed using the Mars Orbiter Laser Altimeter and the Lunar Orbiter Laser Altimeter data, which use a Full Width Half Maximum threshold. To conclude, pulse-widths can show strong correlations to surface roughness and slope within the pulse-footprint. However the assumption that detrended surface roughness can be derived by applying a slope contribution effect is shown to be unfounded. The principal recommendation is for future instruments to use a full echo-profile in estimating pulse-width values at a 10 % Peak Energy threshold, providing both efficient noise removal and a better correlated dataset.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:668478 |
| Date | January 2015 |
| Creators | Poole, W. D. B. |
| Publisher | University College London (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://discovery.ucl.ac.uk/1469361/ |
Page generated in 0.0418 seconds