Channel widths, landslides, faults, and beyond: The new world order of high-spatial resolution Google Earth imagery in the study of earth surface processes

Abstract

International audienceThe past decade has seen a rapid increase in the application of high-resolution imagery and geographic-based information systems across every segment of society from security intelligence to product marketing to scientific research. Google Earth has positioned itself at the forefront of this spatial information wave by providing free access to high-resolution imagery through a simple, user-friendly interface. Whereas Google Earth imagery has been widely exploited across the earth sciences for spatial visualization, education, and place-based searches, few studies have utilized the high-resolution imagery to yield quantitative insights about the processes and mechanisms acting at the earth's surface. In this paper, we detail the benefits of the underlying high-resolution imagery available within Google Earth, review the limited published research to date, and utilize this imagery to quantitatively illuminate previously difficult and unresolved questions within the discipline of geomorphology involving: (1) channel-width variability and scaling relations in the tectonically active Himalaya; (2) landslide characteristics related to large magnitude climatic and tectonic events in Haiti; and (3) identification and quantification of laterally offset geomorphic features within eastern California. In each example, we compare analyses using freely available Google Earth imagery with standard imagery and techniques (e.g., Landsat, ASTER, lidar) to demonstrate the potential benefits of using high-spatial resolution Google Earth imagery over established methodologies. In addition, we discuss the potential limitations and problems with using the imagery currently available in Google Earth and propose favorable future applications, namely studies in remote terrains and those requiring high-resolution imagery across a large spatial extent, where purchasing such imagery in an academic environment would be cost-prohibitive. Whether as a supplement, for reconnaissance, or as the primary data set, high-resolution Google Earth imagery, when properly applied, holds great promise for quantitatively tackling previously unresolved problems in the study of earth surface processes

Understanding erosion rates in the Himalayan orogen: A case study from the Arun Valley

“…However, small, poorly mixed catchments or samples collected shortly after recent landslide events may result in calculated denudation rates that do not reflect the time‐averaged denudation rate [ Yanites et al ., ]. To quantify the potential impact of landsliding on our samples, we mapped visible landslide scars in all sampled tributary catchments using high‐resolution imagery from Google Earth (see data repository Figure S1 in the supporting information for further information) [ Fisher et al ., ]. While small landslides are ubiquitous in the study area, we observed five catchments with a pronounced amount of landsliding as a percentage of total catchment area: three small catchments (<10 km 2 ) in the topographic transition zone (ARU‐11‐10, ARU‐11‐11, and BBRS01) and one small catchment (~10 km 2 ) and one medium‐sized catchment (~100 km 2 ) in the Higher Himalaya (ARU‐12‐09 and ARU‐12‐19A, respectively).…”

“…This approach results in channel width increasing nonlinearly as upstream area increases. However, the low concavity values we derived from χ analysis of the fluvial network indicate that many channels in the Arun continue to have steep channels gradients in their lower reaches and thus narrow channel widths [ Yanites et al ., ; Fisher et al ., ]. Discharge scaling is therefore likely an inaccurate estimate of channel width in this environment and may explain why SSP poorly describes denudation rates in the Arun compared to other orogens [e.g., Bookhagen and Strecker , ].…”

Characterizing the transient geomorphic response to base‐level fall in the northeastern Tibetan Plateau

“…In contrast, several recent field studies document systematic variations in channel width in channels adjusted to varying rates of base‐level fall or rock uplift [ Duvall et al ., ; Finnegan et al ., ; Amos and Burbank , ; Yanites et al ., 2010; Kirby and Ouimet , ]. Moreover, channel narrowing appears to be a primary means by which rapidly incising fluvial systems initially respond to base‐level fall [ Amos and Burbank , ; Whittaker et al ., , , ], a response that may be amplified by variations in rock erodibility [ Attal et al ., ; Fisher et al ., , ; Allen et al ., ]. Although the process of width adjustment remains incompletely understood, recent models that explicitly consider channel cross‐sectional geometry [e.g., Stark , ; Wobus et al ., ; Turowski et al ., ; Yanites and Tucker , ] suggest that spatially variable bedrock wear associated with the degree of sediment cover on the bed [e.g., Sklar and Dietrich , ] is a primary determining factor.…”

Quantifying River Channel Stability at the Basin Scale

“…Channel width has been shown to be estimated expediently via remote sensing methods (e.g., [126,[148][149][150][151][152][153][154]), air photo interpretation (e.g., [155]) and application of Google Earth imagery (e.g., [156,157]) with considerable benefits over established methodologies (see [158] for issues with map-based width measurement). However, where data are available two problematic areas remain: accuracy where banklines are obscured in images or where scans are unreliably filtered for vegetation [141]; wetted width at the time of survey is merely a snapshot of the geometry and might not represent either the bankfull or a temporally-averaged condition [143].…”