International Journal of Advanced Research and Publications ISSN: 2456-9992 Volume 3 Issue 7, July 2019 www.ijarp.org 83 Drone-Based DSM For Multiscale Geometrical Characteristics Of Ephemeral Gullies. Jean A. Doumit, Souhail F. Awad Lebanese university, Faculty of Letters and Human Sciences, Department of Geography Second branch, Fanar, Lebanon jeandoumit@ul.edu.lb Lebanese university, Faculty of Letters and Human Sciences, Department of Geography Second branch, Fanar, Lebanon souheil.awad @ul.edu.lb Abstract: The advance uses of drones in geosciences by producing very high spatial resolution Digital Surface Models (DSMs) and Digital Ortho Models (DOM), at various flight heights, led to different digital models scales. Relief plays an important role in the formation of Ephemeral Gullies (EG), this study focuses on the prediction of multiscale EG location using the compound topographic index (CTI) and analyzed their geometrical characteristics as length, depth and volume of the three different spatial resolutions DSM processed from different drone flights height. Ephemeral Gully extracted from the three flight heights of 120, 240 and 360 meters were compared with each other to understand the effect of generalization at different scales. The results highlight the presence of two scales, small scale ephemeral gully expressed by the flight height 240 and 360 m and a much smaller scale in the level of micro relief of the flight height 120 m. Keywords: drone, ephemeral gully, CTI, DSM, Multi-scale. 1. Introduction The appearance of drones and the very fast evolution in close range photogrammetry and their applications for digital surface extraction leads to multiscale high-resolution terrain analysis. The scale is predominantly considered as a function of the resolution of Digital Surface Models (DSM) [12, 15]. Pike et al. (2009) remarked that no digital elevation models derived map is definitive, as the generated parameters differ with algorithms and can vary with resolution and scale [7]. Soil erosion from water runoff occurs predominantly by three processes: sheet erosion, rill erosion, and ephemeral gullying [8]. Doumit and Awad (2019) analyzed the effect of multiscale DEM on the delineation of sheet and rill erosion whereas in this paper we focus only on ephemeral gullies (EG) erosion based on a multiscale drone based DSM. Ephemeral gullies are small channels eroded by concentrated flow in the same location due to subsequent runoff events [3]. Time series aerial photography and DEM were used to map gully erosion [19,24,22,26]. More recently, drone-based close-range photogrammetry and structure from motion (SfM) including different spatial scales have been used in geomorphic studies, [14, 29, 30,9]. Close range photogrammetry is based on computer visualization tools for three-dimensional surface reconstruction algorithms. Structure from Motion (SfM) creates massive point clouds based on pixel matching from which highly accurate digital surface models (DSM), and orthophotos can be derived [9,14]. Several studies have documented the relationship between ephemeral gully formation and runoff erosivity using terrain derivatives, especially drainage area and local slope [ 23,25,17]. From the terrain derivatives, stream power is widely used to identify the location of ephemeral gullies which depends on ―generation of concentrated surface runoff of sufficient magnitude and duration to initiate and maintain erosion, leading to channelization‖ [27,6]. Plan curvature provides the measure of the degree of flow convergence along a flow path leading to an ephemeral gullies formation models [32]. Zevenbergen (1989) describes five factors as influencing ephemeral gully formation including: 1. Overland flow discharge and duration; 2. Slope and flow depth determining the magnitude of the flow‘s downslope; 3. Planform curvature determining the convergence of the flow; 4. Soil characteristics determining the erodibility of the soil; 5. Vegetation characteristics, reducing the soil susceptibility to erosion [31]. This study is based on drone DSM multiscale and it focused only on topographic controls of overland flow influencing on ephemeral gully development taking into account only the first three Zevenbergen‘s terrain factors: discharge, slope and plan curvature, which is the basic key of topographic controls in ephemeral gully formation process [27]. Plan curvature as second terrain derivative contributes to ephemeral gully formation in multiple ways, a) Convergence runoff and discharge are related to slope length to a power greater than unity [31]. b) plan curvature degree determines local flow geometry and the degree of flow concentration. c) plan curvature represents the degree of concentration of stream power. For the identification of ephemeral gullies, Thorne et al. (1986) used these parameters to calculate a Compound Topographic Index (CTI). Parker et al. (2007) tested the CTI model to predict ephemeral gullies locations in a GIS environment for different sites. Moore et al. (1988) used two methods to estimate EG location: the slope area index and the wetness topographic index, Casalí,.et al. (2006) tested several methods for estimating ephemeral gullies and concluded that CTI is probably the most widely used approach for predicting ephemeral gullies location. After we presented the fact of the studies concerning the identification of ephemeral gully erosion, we submit the specific objectives of this study were:  to apply the work carried out by Zevenbergen and Thorne, utilizing Geographical Information Science (GIS) for multiscale ephemeral gullies delineation from CTI parameters.