Stream network modelling using lidar and photogrammetric digital elevation models: a comparison and field verification

Digital elevation models (DEM) are mathematical representations of the Earth's bare surface in computer-readable format. The underlying measurements are often obtained by remote sensing and photogrammetry methods and processed into continuous raster data. Each of these data sources, however, provides imperfect information, and further processing steps often increase the degree of imperfection. Consequently, the process of DEM generation cumulates in uncertainty, which affects subsequent hydro- and geomorphological analyses and modelling (e.g., stream network delineation, flowpath distribution, erosion modelling). In many DEM-based studies, however, the aspect of uncertainty related to the DEM data source has been neglected. Therefore, we propose a new approach for quantifying the effects of DEM uncertainty on hydro-geomorphological modelling based on Gaussian white noise, a concept widely used in signal processing to map noise in signals and extract the actual message context. The basic idea is to add noise to the original DEM values by means of a Gaussian distribution whose parameters are determined from the mean value of the elevation values in a moving window and the device-specific properties (precision and accuracy). We postulate that such an approach can be used to determine uncertainties and their effect on subsequent analysis steps of hydro-geomorphological modelling. It is conceivable to create DEM ensembles depending on known parameters such as the accuracy and precision of the measuring instrument, as is used operationally in weather forecasting. Using such ensembles, probability ranges for terrain and catchment hydro-geomorphological properties can be determined and uncertainty ranges can be specified. Thus, the currently mostly deterministic approach of digital terrain modelling will be replaced by a more probabilistic understanding. Overall, our approach will help decision-makers and scientists to better assess the results of digital terrain analysis. Furthermore, it will also facilitate determining whether a result of DEM-based hydro-geomorphological analysis is sufficiently certain to answer specific research questions.

Download Full-text

Drainage divide networks – Part 1: Identification and ordering in digital elevation models

Earth Surface Dynamics ◽

10.5194/esurf-8-245-2020 ◽

2020 ◽

Vol 8 (2) ◽

pp. 245-259 ◽

Cited By ~ 2

Author(s):

Dirk Scherler ◽

Wolfgang Schwanghart

Keyword(s):

Drainage Basin ◽

Average Distance ◽

Digital Elevation Models ◽

Stream Network ◽

San Gabriel Mountains ◽

Digital Elevation ◽

River Confluences ◽

Characteristic Values ◽

Downstream Flow ◽

Basin Boundaries

Abstract. We propose a novel way to measure and analyze networks of drainage divides from digital elevation models. We developed an algorithm that extracts drainage divides based on the drainage basin boundaries defined by a stream network. In contrast to streams, there is no straightforward approach to order and classify divides, although it is intuitive that some divides are more important than others. A meaningful way of ordering divides is the average distance one would have to travel down on either side of a divide to reach a common stream location. However, because measuring these distances is computationally expensive and prone to edge effects, we instead sort divide segments based on their tree-like network structure, starting from endpoints at river confluences. The sorted nature of the network allows for assigning distances to points along the divides, which can be shown to scale with the average distance downslope to the common stream location. Furthermore, because divide segments tend to have characteristic lengths, an ordering scheme in which divide orders increase by 1 at junctions mimics these distances. We applied our new algorithm to the Big Tujunga catchment in the San Gabriel Mountains of southern California and studied the morphology of the drainage divide network. Our results show that topographic metrics, like the downstream flow distance to a stream and hillslope relief, attain characteristic values that depend on the drainage area threshold used to derive the stream network. Portions along the divide network that have lower than average relief or are closer than average to streams are often distinctly asymmetric in shape, suggesting that these divides are unstable. Our new and automated approach thus helps to objectively extract and analyze divide networks from digital elevation models.

Download Full-text

Quantifying and modeling of stream network using digital elevation models

Ain Shams Engineering Journal ◽

10.1016/j.asej.2015.09.002 ◽

2017 ◽

Vol 8 (3) ◽

pp. 311-321 ◽

Cited By ~ 6

Author(s):

Daisy Paul ◽

V. Ravibabu Mandla ◽

Tejpal Singh

Keyword(s):

Digital Elevation Models ◽

Stream Network ◽

Digital Elevation

Download Full-text

Identification and ordering of drainage divides in digital elevation models

10.5194/esurf-2019-51 ◽

2019 ◽

Author(s):

Dirk Scherler ◽

Wolfgang Schwanghart

Keyword(s):

Drainage Basin ◽

Average Distance ◽

Digital Elevation Models ◽

Great Distance ◽

Stream Network ◽

Natural Landscape ◽

Common Location ◽

Time Period ◽

Digital Elevation ◽

Basin Boundaries

Abstract. We propose a novel way to measure and analyse networks of drainage divides from digital elevation models. We developed an algorithm that extracts drainage divides, based on the drainage basin boundaries defined by a stream network. In contrast to streams, there is no straightforward approach to order and classify divides, although it is intuitive that some divides are more important than others. We thus propose a divide-network metric that orders divides based on the average distance one would have to travel down on either side of a divide to reach a common stream location. Because measuring these distances is computationally very expensive, we instead sort divide segments in a tree-like network, starting from endpoints at river junctions. The sorted nature of the network allows assigning distances to points along the divides, which can be shown to scale with the average distance downstream to the common location. Furthermore, because divide segments tend to have characteristic lengths, an ordering scheme in which divide orders increase by one at junctions, mimics these distances. We applied our new algorithm to a natural landscape and to results from landscape evolution model experiments to assess which parameters of divides and divide networks are diagnostic of divide mobility. Results show that stable drainage divides strive to attain a constant hillslope relief as well as flow distance from the nearest stream, provided a distance of > ~ 5 km from endpoints. Disruptions of such pattern can be related to mobile divides that are lower than stable divides, closer to streams, and often asymmetric in shape. In general, we observe that drainage divides high up in the network, i.e., at great distance from endpoints, are more vulnerable than divides closer to endpoints of the network and are more likely to record disturbance for a longer time period. We compared different topographic metrics to assess drainage divide mobility and found that cross-divide differences in hillslope relief proved more useful for assessing divide migration than other tested metrics. Finally, we introduced a new metric to assess divide junction stability, based on the connectivity of junctions with adjacent drainage divide segments.

Download Full-text