Premium
Using geostatistics to elucidate temporal change in the spatial variation of aeolian sediment transport
Author(s) -
Chappell Adrian,
Mctainsh Grant,
Leys John,
Strong Craig
Publication year - 2003
Publication title -
earth surface processes and landforms
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.294
H-Index - 127
eISSN - 1096-9837
pISSN - 0197-9337
DOI - 10.1002/esp.463
Subject(s) - aeolian processes , spatial variability , sediment transport , geostatistics , kriging , sediment , environmental science , sampling (signal processing) , scale (ratio) , geology , erosion , hydrology (agriculture) , soil science , geomorphology , geotechnical engineering , computer science , mathematics , statistics , geography , cartography , filter (signal processing) , computer vision
Little is known about the spatial and temporal scales of variation in aeolian processes. Studies that aim to investigate surface erodibility often sample aeolian sediment transport at the nodes of a regular grid of arbitrary size. Few aeolian transport investigations have the resources to obtain sufficient samples to produce reliable models for mapping the spatial variation of transport. This study reports the use of an innovative nested strategy for sampling multiple spatial scales simultaneously using 40 sediment samplers. Reliable models of the spatial variation in aeolian sediment transport were produced and used for ordinary punctual kriging and stochastic simulated annealing to produce maps for several wind erosion events over a 25 km 2 playa in western Queensland, Australia. The results support the existence of a highly dynamic wind erosion system that was responding to possibly cyclic variation in the availability of material and fluctuations in wind energy. The spatial scale of transport was considerably larger than the small scale expected of the factors controlling surface erodibility. Thus, it appears that transport cannot be used as a surrogate of erodibility at the scale of this investigation. Simulation maps of transport provided considerably more information than those from kriging about the variability in aeolian sediment transport and its possible controlling factors. The proposed optimal sampling strategy involves a nested approach using ca 50 samplers. Copyright © 2003 John Wiley & Sons, Ltd.