Object-based gully feature extraction using high spatial resolution imagery Rajesh B.V. Shruthi ⁎, Norman Kerle, Victor Jetten Department of Earth Systems Analysis, Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente, Netherlands abstract article info Article history: Received 22 September 2010 Received in revised form 17 June 2011 Accepted 4 July 2011 Available online 21 July 2011 Keywords: Gully erosion Object-based image analysis Land degradation High spatial resolution imagery Gully erosion is responsible for a substantial amount of soil loss and is generally considered an indicator of desertification. Hence, mapping these gully features provides essential information needed on sediment production, identification of vulnerable areas for gully formation, land degradation, and environmental and socio- economical effects. This paper investigates the use of object-oriented image analysis (OOA) to extract gully erosion features from satellite imagery, using a combination of topographic, spectral, shape (geometric) and contextual information obtained from IKONOS and GEOEYE-1 data. A rule-set was developed and tested for a semi-arid to sub-humid region in Morocco. The percentage of gully system area indicated negligible overestimations between the reference area and the OOA area in two sub-watersheds (0.03% and 1.77%). We also observed that finer gully- related edges within the complex gully systems were better identified semi-automatically than was possible by manual digitization, suggesting higher detection accuracy. OOA-based gully mapping is quicker and more objective than traditional methods, and is thus better suited to provide essential information for land managers to support their decision making processes, and for the erosion research community. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Erosion by surface runoff has been receiving substantial attention from researchers, conservationists and policy makers. It comprises sheet or inter-rill, rill and gully erosion. Amongst these forms of soil removal, gully erosion in both ephemeral and permanent gullies is responsible for a substantial amount of soil loss, and is generally considered an indicator of desertification (UNEP, 1994). Fig. 1 illustrates a typical gully formation situation, with incisions frequently cutting through different soil horizons, and their form and shape being guided by the hydrological and mechanical properties of these soil layers. A second commonly occurring gully formation process is the backward extension of a gully in the hillslope, which occurs as a combination of water incision and small mass movement on the sides and head of a gully. Extensive reviews on the initiation, controlling factors and impacts of gullying have been provided by Poesen et al. (2003) and Valentin et al. (2005). Poesen et al. (2006) and Vrieling et al. (2007) also identified that most research has focused on sheet (inter-rill) and rill erosion, and that little is known about gully erosion and its importance at large spatial scales. One of the reasons is that gullies, once formed, can remain unaltered for extended periods of time, especially in semi-arid climates. Although they are evidence of severe land degradation, their dimensions may not be easily related to current rainfall (Seeger et al., 2009) and surface runoff (Marzolff and Ries, 2007). Moreover, the timeframe at which they formed and changed is often unclear. Sustainable land management fundamentally requires knowledge of the landscape and its processes, for which an efficient way of understanding, surveying and monitoring is needed. Given that gullies are one of the main drivers for soil loss in the landscape system, there is an imperative need for detailed monitoring and better prediction of gully locations. This study focuses on rill/ephemeral and permanent gully erosion. The gully features investigated are discontinuous, and much narrower (b 10 m) than gullies on a river bank (alluvial gullies) with widths of 20 to 140 m (Brooks et al., 2009; Perroy et al., 2010). This constitutes a real challenge for the semi-automatic detection of gullies, because of not only the size, shape and distribution of gullies but also the presence of various land cover, land use, shadow and illumination. This study attempts to address the two existing problems: 1) mapping gully systems through field work and manual image digitization are difficult and time consuming, and 2) there is a lack of a generic algorithm to identify gullies from images. Mapping gullies and erosional activity is crucial for monitoring erosion and studying its impacts including sediment production, land degradation, and other socio-economical influences. Field-based methods were used in the past until aerial photos and later satellite imagery became more readily available. Remote sensing-based mapping is the only practical approach for mapping gully features over large areas, given the variability in gully size, shape and occurrence (Knight et al., 2007), as well as the dynamic nature of gully-affected landscapes. A review of different methods used to map and monitor gully erosion features is given below. It has been recognized that accurate identification of gullies is not possible without additional data or expert knowledge (Bocco and Valenzuela, 1993). In addition auxiliary information, such as geometric properties (shape, dimension, orientation and texture) and the Geomorphology 134 (2011) 260–268 ⁎ Corresponding author. Tel.: + 31 53 4874504; fax: + 31 53 4874336. E-mail address: shruthi@itc.nl (R.B.V. Shruthi). 0169-555X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.geomorph.2011.07.003 Contents lists available at ScienceDirect Geomorphology journal homepage: www.elsevier.com/locate/geomorph