The potential of 3D gully monitoring with GIS using high-resolution aerial photography and a digital photogrammetry system

• Published in: Geomorphology (Amsterdam, Netherlands) Vol. 111; no. 1; pp. 48 - 60
• Main Authors: Marzolff, I., Poesen, J.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.10.2009
• Subjects: Close range aerial photogrammetry; DEM (Digital Elevation Model); Freshwater; Gully erosion monitoring; High resolution images; Kite and blimp platforms; Spain; ANW, Atlantic, The Gully; ANE, Spain; Spain; High resolution images; Close range aerial photogrammetry; Kite and blimp platforms; DEM (Digital Elevation Model); Gully erosion monitoring; Spain
• ISSN: 0169-555X; 1872-695X
• DOI: 10.1016/j.geomorph.2008.05.047

Although gully erosion is generally considered a major process of land degradation, its contribution to total soil loss by erosion has recently been a subject of much discussion. The lack of adequate methods for the documentation and monitoring of gullies resulted in the shortage of quantitative data. Therefore, a high-resolution remote sensing system for aerial surveys by blimp or kite has been developed which meets spatial and temporal image resolutions required for short-term gully monitoring. The objective of this paper is to examine the potential of a method using non-metric digital photogrammetry and GIS for gully surface modelling and monitoring. Two bank gullies representing different morphological types of gullies (V-shaped and U-shaped) were chosen from a dataset of gully systems in semi-arid Spain. The considerable relief energy and complex topography of the gullies in a natural, vegetated landscape were found to be a challenge to digital photogrammetric techniques, introducing errors which inhibit fully automated DEM generation. Using a hybrid method combining stereomatching for mass-point extraction with manual 3D editing and digitizing, high-resolution DEMs (5 and 7.5 cm pixel size) were created for the study sites. GIS analysis of the DEMs for different monitoring periods (2 to 4 years) allowed the computation of gully area and volume, as well as their changes with an accuracy and detail sufficient to represent the geomorphological forms and processes involved. Furthermore, the spatially continuous survey of the entire form offered the possibility of distinguishing different zones of activity both at the gully rim and within the gully interior, identifying patterns of erosion and deposition which indicate the limited use of headcut retreat rates for the assessment of sediment production on a short-term basis.