Digital Elevation Model Error Research Articles

SRTM DEM Correction in Vegetated Mountain Areas through the Integration of Spaceborne LiDAR, Airborne LiDAR, and Optical Imagery

The Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) is one of the most complete and frequently used global-scale DEM products in various applications. However, previous studies have shown that the SRTM DEM is systematically higher than the actual land surface in vegetated mountain areas. The objective of this study is to propose a procedure to calibrate the SRTM DEM over large vegetated mountain areas. Firstly, we developed methods to estimate canopy cover from aerial imagery and tree height from multi-source datasets (i.e., field observations, airborne light detection and ranging (LiDAR) data, Geoscience Laser Altimeter System (GLAS) data, Landsat TM imagery, climate surfaces, and topographic data). Then, the airborne LiDAR derived DEM, covering ~5% of the study area, was used to evaluate the accuracy of the SRTM DEM. Finally, a regression model of the SRTM DEM error depending on tree height, canopy cover, and terrain slope was developed to calibrate the SRTM DEM. Our results show that the proposed procedure can significantly improve the accuracy of the SRTM DEM over vegetated mountain areas. The mean difference between the SRTM DEM and the LiDAR DEM decreased from 12.15 m to −0.82 m, and the standard deviation dropped by 2 m.

Satellite-derived Digital Elevation Model (DEM) selection, preparation and correction for hydrodynamic modelling in large, low-gradient and data-sparse catchments

• SRTM DEM had higher accuracy than ASTER GDEM. • Vegetation smoothed DEM improved the hydrodynamic model results. • Hydrologically-corrected DEM improved the hydrodynamic model results. • The hydrodynamic model reached optimum performance at grid size of 120 m. Digital Elevation Models (DEMs) that accurately replicate both landscape form and processes are critical to support modelling of environmental processes. Topographic accuracy, methods of preparation and grid size are all important for hydrodynamic models to efficiently replicate flow processes. In remote and data-scarce regions, high resolution DEMs are often not available and therefore it is necessary to evaluate lower resolution data such as the Shuttle Radar Topography Mission (SRTM) and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) for use within hydrodynamic models. This paper does this in three ways: (i) assessing point accuracy and geometric co-registration error of the original DEMs; (ii) quantifying the effects of DEM preparation methods (vegetation smoothed and hydrologically-corrected) on hydrodynamic modelling relative accuracy; and (iii) quantifying the effect of the hydrodynamic model grid size (30–2000 m) and the associated relative computational costs (run time) on relative accuracy in model outputs. We initially evaluated the accuracy of the original SRTM (∼30 m) seamless C-band DEM (SRTM DEM) and second generation products from the ASTER (ASTER GDEM) against registered survey marks and altimetry data points from the Ice, Cloud, and land Elevation Satellite (ICESat). SRTM DEM (RMSE = 3.25 m,) had higher accuracy than ASTER GDEM (RMSE = 7.43 m). Based on these results, the original version of SRTM DEM, the ASTER GDEM along with vegetation smoothed and hydrologically corrected versions were prepared and used to simulate three flood events along a 200 km stretch of the low-gradient Thompson River, in arid Australia (using five metrics: peak discharge, peak height, travel time, terminal water storage and flood extent). The hydrologically corrected DEMs performed best across these metrics in simulating floods compared with vegetation smoothed DEMs and original DEMs. The response of model performance to grid size was non-linear and while the smaller grid sizes (⩽120 m) improved the hydrodynamic model results, these offered only slight improvements at very significant computational costs compared to grid size of 120 m, with grid sizes 250 m and greater decreasing in model accuracy. This study highlights the important impact that the quality of the underlying DEM has, and in particular how sensitive hydrodynamic models are to preparation methods and how important vegetation smoothing and hydrological correction of the base topographic data for modelling floods in low-gradient and multi-channel environments.

InSAR Image Regularization and DEM Error Correction With Fractal Surface Scattering Model

This paper presents a method for removing spikes in digital elevation models (DEMs) caused by residues in interferometric synthetic aperture radar (InSAR) phase image. We consider that the scattering mechanism is properly modeled by the small perturbation method for fractal surfaces and present a model that relates the phase and magnitude in InSAR image. This data model provides the regularization term of the method, without directly enforcing smooth phase or magnitude. Noise models are given by additive Gaussian for the phase and multiplicative non-unit-mean gamma for the magnitude. Experiments with simulated and real L-band data show that the proposed method considerably improves DEM accuracy and simultaneously suppresses speckle and phase noise.

Digital stereo photogrammetry for grain-scale monitoring of fluvial surfaces: Error evaluation and workflow optimisation

Grain-scale monitoring of fluvial morphology is important for the evaluation of river system dynamics. Significant progress in remote sensing and computer performance allows rapid high-resolution data acquisition, however, applications in fluvial environments remain challenging. Even in a controlled environment, such as a laboratory, the extensive acquisition workflow is prone to the propagation of errors in digital elevation models (DEMs). This is valid for both of the common surface recording techniques: digital stereo photogrammetry and terrestrial laser scanning (TLS). The optimisation of the acquisition process, an effective way to reduce the occurrence of errors, is generally limited by the use of commercial software. Therefore, the removal of evident blunders during post processing is regarded as standard practice, although this may introduce new errors. This paper presents a detailed evaluation of a digital stereo-photogrammetric workflow developed for fluvial hydraulic applications. The introduced workflow is user-friendly and can be adapted to various close-range measurements: imagery is acquired with two Nikon D5100 cameras and processed using non-proprietary “on-the-job” calibration and dense scanline-based stereo matching algorithms. Novel ground truth evaluation studies were designed to identify the DEM errors, which resulted from a combination of calibration errors, inaccurate image rectifications and stereo-matching errors. To ensure optimum DEM quality, we show that systematic DEM errors must be minimised by ensuring a good distribution of control points throughout the image format during calibration. DEM quality is then largely dependent on the imagery utilised. We evaluated the open access multi-scale Retinex algorithm to facilitate the stereo matching, and quantified its influence on DEM quality. Occlusions, inherent to any roughness element, are still a major limiting factor to DEM accuracy. We show that a careful selection of the camera-to-object and baseline distance reduces errors in occluded areas and that realistic ground truths help to quantify those errors.

Incorporating DEM uncertainty in coastal inundation mapping.

Coastal managers require reliable spatial data on the extent and timing of potential coastal inundation, particularly in a changing climate. Most sea level rise (SLR) vulnerability assessments are undertaken using the easily implemented bathtub approach, where areas adjacent to the sea and below a given elevation are mapped using a deterministic line dividing potentially inundated from dry areas. This method only requires elevation data usually in the form of a digital elevation model (DEM). However, inherent errors in the DEM and spatial analysis of the bathtub model propagate into the inundation mapping. The aim of this study was to assess the impacts of spatially variable and spatially correlated elevation errors in high-spatial resolution DEMs for mapping coastal inundation. Elevation errors were best modelled using regression-kriging. This geostatistical model takes the spatial correlation in elevation errors into account, which has a significant impact on analyses that include spatial interactions, such as inundation modelling. The spatial variability of elevation errors was partially explained by land cover and terrain variables. Elevation errors were simulated using sequential Gaussian simulation, a Monte Carlo probabilistic approach. 1,000 error simulations were added to the original DEM and reclassified using a hydrologically correct bathtub method. The probability of inundation to a scenario combining a 1 in 100 year storm event over a 1 m SLR was calculated by counting the proportion of times from the 1,000 simulations that a location was inundated. This probabilistic approach can be used in a risk-aversive decision making process by planning for scenarios with different probabilities of occurrence. For example, results showed that when considering a 1% probability exceedance, the inundated area was approximately 11% larger than mapped using the deterministic bathtub approach. The probabilistic approach provides visually intuitive maps that convey uncertainties inherent to spatial data and analysis.

A Baseline-Combination Method for Precise Estimation of Ice Motion in Antarctica

Differential synthetic aperture radar interferometry (D-InSAR) is a powerful method for measuring surface deformation, such as in studies of the earthquake cycle, volcano deformation monitoring, land subsidence monitoring, and glaciological studies. However, its application to glaciological studies is limited by the lack of accurate digital elevation models (DEMs), particularly over the Antarctic ice sheet. Previous studies on ice motion using D-InSAR are mostly based on short-baseline interferograms because these data sets are insensitive to DEM errors. Unfortunately, short-baseline interferograms are often unavailable. In this paper, we refine the InSAR technique by using a combination of two interferograms to make accurate ice-flow velocity measurements. The refined technique is tested in the Grove Mountains area, East Antarctica. Ice-flow velocities from the baseline-combination method are in good agreement with those measured by short-baseline interferograms. This method is also capable of reducing phase errors by combining the appropriate data sets. The reliability of the data sets is assessed by defining a baseline-combination parameter and ensuring that it is less than or equal to 1.0. With this method, we are able to extend the usefulness of D-InSAR for glaciological studies.

A very fast phase inversion approach for small baseline style interferogram stacks

The recently developed interferometric time-series analysis techniques have shown great potential in ground surface deformation monitoring applications. Such techniques overcome the drawbacks of the traditional differential radar interferometry (DInSAR) and can achieve millimeter-level measurement accuracy. One of the most important operations in interferometric time-series analysis techniques — referred to as phase inversion — is to estimate relative deformation velocity and digital elevation model error from a double-differenced interferometric phase time-series. Unfortunately, current phase inversion methods generally exhibit a low computational efficiency due to their high non-linearity, especially in the case when the dimension of an interferogram stack is large. In this paper, a new approach is proposed to efficiently resolve phase inversion problems defined on stacks constructed by interferograms with small baselines. The approach separates an estimation procedure into two parts. First, preliminary estimates are obtained by weighted least squares. Then, the estimates are refined by optimizing the corresponding ensemble phase coherence function. The proposed approach was applied to simulated and real data. Experimental results demonstrate that it can accurately address the phase inversion problem with a very high computational performance.

Mitigating systematic error in topographic models derived from UAV and ground‐based image networks

ABSTRACTHigh resolution digital elevation models (DEMs) are increasingly produced from photographs acquired with consumer cameras, both from the ground and from unmanned aerial vehicles (UAVs). However, although such DEMs may achieve centimetric detail, they can also display systematic broad‐scale error that restricts their wider use. Such errors which, in typical UAV data are expressed as a vertical ‘doming’ of the surface, result from a combination of near‐parallel imaging directions and inaccurate correction of radial lens distortion. Using simulations of multi‐image networks with near‐parallel viewing directions, we show that enabling camera self‐calibration as part of the bundle adjustment process inherently leads to erroneous radial distortion estimates and associated DEM error. This effect is relevant whether a traditional photogrammetric or newer structure‐from‐motion (SfM) approach is used, but errors are expected to be more pronounced in SfM‐based DEMs, for which use of control and check point measurements are typically more limited. Systematic DEM error can be significantly reduced by the additional capture and inclusion of oblique images in the image network; we provide practical flight plan solutions for fixed wing or rotor‐based UAVs that, in the absence of control points, can reduce DEM error by up to two orders of magnitude. The magnitude of doming error shows a linear relationship with radial distortion and we show how characterization of this relationship allows an improved distortion estimate and, hence, existing datasets to be optimally reprocessed. Although focussed on UAV surveying, our results are also relevant to ground‐based image capture. © 2014 The Authors. Earth Surface Processes and Landforms published by John Wiley & Sons Ltd.

COMPARISON ANALYSIS OF INTERPOLATION TECHNIQUES FOR DEM GENERATION USING CARTOSAT-1 STEREO DATA

Digital Elevation Model (DEM) can be generated using several techniques such as photogrammetric technique, interferometry, Lidar, etc. In photogrammetric technique, a DEM generation using stereo images, accuracy of generated DEM is also dependent on interpolation techniques. The process of interpolation is conducted to generate DEM as a continuous data from the point map that contained height information as a discrete data. In this research, point map was extracted from Cartosat-1 stereo image and from geodetic single frequency GPS in differential mode. Different interpolation techniques were applied on these data sets with different combination within these data sets. In this study, analysis of DEM interpolation was conducted with deterministic interpolators such as inverse distance weighted (IDW), global polynomial, local polynomial, and radial basis functions (RBF); and probabilistic interpolators such as simple kriging, ordinary kriging, universal kriging, indicator kriging, probabilistic kriging, disjunctive kriging, and cokriging. The accuracy of generated DEMs through different interpolation techniques were evaluated with ground point data collected from geodetic single frequency GPS in differential mode. Based on the analysis, the range error of DEMs generated was between 1.29 m to 2.96 m. Interpolation method with the least error was ordinary kriging using point map data and GPS points, while the highest error was obtained from global polynomial method.

Impact of uncertainty in remote sensing DEMs on topographic amplification of seismic response and Vs 30

Impact of topography on spatial variation of seismic response is well-observed synthetically, experimentally, and visually during seismic events. Numerical and experimental investigations for predicting topographic impact on seismic response are often limited to isolated and/or synthetic hills and ridges. Furthermore, most of these studies only focus on one of the many terrain parameters necessary for evaluating the impact of topographic features on amplification or de-amplification of seismic response. Seismic events located in rough terrain, like the 2005 Kashmir earthquake in northern Pakistan, exhibit intensified seismic response and associated devastation at hill ridges and on inclined slopes. Satellite remote sensing-derived digital elevation models (DEMs) are frequently and effectively used to compute topographic attributes and seismic parameters to evaluate topographic seismic response. However, the influence of DEM random errors on computed topographic attributes and seismic response is often overlooked. This study uses the Shuttle Radar Topography Mission (SRTM) DEM (90 m) and the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) GDEM (30 m) to quantify uncertainties in the computed topographic attributes and seismic parameters and ultimately the impact of these uncertainties on topographic seismic response. Accuracy assessment of the DEMs shows root mean square error (RMSE) of 13.78 m in the ASTER DEM and 23.71 m in the SRTM DEM. The influence of DEM errors on derived topographic attributes quantified through Monte Carlo simulations shows higher uncertainty in slope and aspect computed from ASTER DEM than from SRTM DEM. The influence of uncertainty in the SRTM and ASTER DEMs shows significant impact on the computed seismic parameters of slope, relative height, and V s 30 and ultimately derived topographic seismic response.

DEM Corrections Before Unwrapping in a Small Baseline Strategy for InSAR Time Series Analysis

Synthetic aperture radar interferometry (InSAR) is limited by temporal decorrelation and topographic errors, which can result in unwrapping errors in partially incoherent and mountainous areas. In this paper, we present an algorithm to estimate and remove local digital elevation model (DEM) errors from a series of wrapped interferograms. The method is designed to be included in a small baseline subset (SBAS) approach for InSAR time series analysis of ground deformation in natural environment. It is easy to implement and can be applied to all pixels of a radar scene. The algorithm is applied to a series of wrapped interferograms computed from ENVISAT radar images acquired across the Himalayan mountain range. The DEM error correction performance is quantified by the reduction of the local phase dispersion and of the number of residues computed during the unwrapping procedure. It thus improves the automation of the spatial unwrapping step.

A Comparitive Study On The Perfermance Of The InSAR Phase Filtering Approches In The Spatial And The Wavelet Domains

Abstract. In this paper, we studied and tested different filtering approaches of the SAR interferograms in the spatial and wavelet domains. In the spatial domain, we applied the classic Lee filter and the Weighted Median Filter WMF. In the wavelet domain, we tested a noise reduction algorithm WInP proposed by López and Fàbregàs and its enhanced version FAMM developed by Abdelfattah and Bouzid. Those filters are validated with different SAR interferograms provided by Radarsat-2, Envisat, ERS-2 and COSMO-SkyMed SLC data acquired over regions of Mahdia and Ben Guerden in Tunisia. The aim of this study is to select the optimal filtering approach with respect to the fringe pattern in the interferogram. This selection is based on the Digital Elevation Model error computed between the filtered unwrapping phase image and the Global ASTER DEM of the same regions and verified with simulated interferograms.

A methodological intercomparison of topographic survey techniques for characterizing wadeable streams and rivers

Fine-scale (submeter) resolution digital elevation models (DEMs) created from high precision (subcentimeter) instruments (e.g., total station, rtkGPS, and laser scanning) have become ubiquitous in the field of fluvial geomorphology. They permit a diverse range of spatially explicit analyses including hydraulic modeling, habitat modeling, and geomorphic change detection. While previous studies have assessed the quality of specific topographic survey methods at individual sites or across a limited number of sites, an intercomparison of survey technologies across a diverse range of wadeable streams could help clarify which techniques are feasible, as well as which work best under what circumstances and for what purposes. Although a wealth of existing studies and protocols explain how to undertake each individual technique, in this study we seek to provide guidance on what techniques to use in which circumstances. We quantified the relative quality and the amount of effort spent collecting data to derive bare earth topography from an array of ground-based and airborne survey techniques. We used topographic survey data collected over the summer of 2010 from six sample reaches of varying complexity in the Lemhi River basin, Idaho, USA. We attempted to conduct complete, replicate surveys at each site using total station (TS), real-time kinematic (rtk) GPS, discrete return terrestrial laser scanner (TLS), and airborne LiDaR surveys (ALS). We evaluated the precision and accuracy of derived bare earth DEMs relative to the higher precision total station point data. Discrepancies between pairwise techniques were calculated using propagated DEM errors thresholded at a 95% confidence interval. Mean discrepancies between total station and rtkGPS DEMs were relatively low (≤0.05m), yet TS data collection time was up to 2.4 times longer than rtkGPS. The ALS DEMs had lower accuracy than TS or rtkGPS DEMs, but the aerial coverage and floodplain context of the ALS data set was superior to all other techniques. The TLS bare earth DEM accuracy and precision were lower than any other technique because of vegetation returns misinterpreted as ground returns. Our results are helpful for understanding the strengths and weaknesses of different approaches.

New Progress in the Investigation of Spatial Distribution for SRTM DEM Errors

The Shuttle Radar Topography Mission (SRTM) is an international research effort that obtains digital elevation models (DEM) over 80% of the Earths land surface. SRTM DEM plays a key role in geosciences and GIS. In order to investigate the vertical accuracy of SRTM DEM, we evaluated the root mean square error (RMSE) of height between SRTM DEM and 1:50,000 scale topographic map within north China, and extracted the related topographic factors including height, slope and aspect. Then, we analyzed the relationship between the topographic factors and SRTM DEM errors. The results show that the SRTM DEM errors not only have a tendency to get larger in areas of large slope and complex topography, but also have a strong correlation with aspect. Furthermore, this correlation appears increasingly strong with great slope.

Uncertainty due to DEM error in landslide susceptibility mapping

Terrain attributes such as slope gradient and slope shape, computed from a gridded digital elevation model (DEM), are important input data for landslide susceptibility mapping. Errors in DEM can cause uncertainty in terrain attributes and thus influence landslide susceptibility mapping. Monte Carlo simulations have been used in this article to compare uncertainties due to DEM error in two representative landslide susceptibility mapping approaches: a recently developed expert knowledge and fuzzy logic-based approach to landslide susceptibility mapping (efLandslides), and a logistic regression approach that is representative of multivariate statistical approaches to landslide susceptibility mapping. The study area is located in the middle and upper reaches of the Yangtze River, China, and includes two adjacent areas with similar environmental conditions – one for efLandslides model development (approximately 250 km2) and the other for model extrapolation (approximately 4600 km2). Sequential Gaussian simulation was used to simulate DEM error fields at 25-m resolution with different magnitudes and spatial autocorrelation levels. Nine sets of simulations were generated. Each set included 100 realizations derived from a DEM error field specified by possible combinations of three standard deviation values (1, 7.5, and 15 m) for error magnitude and three range values (0, 60, and 120 m) for spatial autocorrelation. The overall uncertainties of both efLandslides and the logistic regression approach attributable to each model-simulated DEM error were evaluated based on a map of standard deviations of landslide susceptibility realizations. The uncertainty assessment showed that the overall uncertainty in efLandslides was less sensitive to DEM error than that in the logistic regression approach and that the overall uncertainties in both efLandslides and the logistic regression approach for the model-extrapolation area were generally lower than in the model-development area used in this study. Boxplots were produced by associating an independent validation set of 205 observed landslides in the model-extrapolation area with the resulting landslide susceptibility realizations. These boxplots showed that for all simulations, efLandslides produced more reasonable results than logistic regression.

Quantifying local flow direction uncertainty

Absolute elevation error in digital elevation models (DEMs) can be within acceptable National Map Accuracy standards, but still have dramatic impacts on field-level estimates of surface water flow direction, particularly in level regions. We introduce and evaluate a new method for quantifying uncertainty in flow direction rasters derived from DEMs. The method utilizes flow direction values derived from finer resolution digital elevation data to estimate uncertainty, on a cell-by-cell basis, in flow directions derived from coarser digital elevation data. The result is a quantification and spatial distribution of flow direction uncertainty at both local and regional scales. We present an implementation of the method using a 10-m DEM and a reference 1-m lidar DEM. The method contributes to scientific understanding of DEM uncertainty propagation and modeling and can inform hydrological analyses in engineering, agriculture, and other disciplines that rely on simulations of surface water flow.

DEM Error Correction in InSAR Time Series

We present a mathematical formulation for the phase due to the errors in digital elevation models (DEMs) in synthetic aperture radar (SAR) interferometry (InSAR) time series obtained by the small baseline (SB) or the small baseline subset method. We show that the effect of the DEM error in the estimated displacement is proportional to the perpendicular baseline history of the set of SAR acquisitions. This effect at a given epoch is proportional to the perpendicular baseline between the SAR acquisition at that epoch and the reference acquisition. Therefore, the DEM error can significantly affect the time-series results even if SB interferograms are used. We propose a new method for DEM error correction of InSAR time series, which operates in the time domain after inversion of the network of interferograms for the displacement time series. This is in contrast to the method of Berardino (2002) in which the DEM error is estimated in the interferogram domain. We show the effectiveness of this method using simulated InSAR data. We apply the new method to Fernandina volcano in the Galapagos Islands and show that the proposed DEM error correction improves the estimated displacement significantly.

A robust regression-based linear model adjustment method for SAR interferogram stacking techniques

Differential interferometric synthetic aperture radar (DInSAR) is recognized as an effective remote-sensing technique for a variety of ground deformation mapping applications. Centimetre-level measurement accuracy can be achieved with the DInSAR technique. However, two key limitations – temporal decorrelation and phase delay due to atmospheric inhomogeneities – might decrease the accuracy of deformation measurements. To overcome such problems, interferogram stacking techniques, which extend the DInSAR technique, have been developed in recent years. In most implementations of such techniques, a so-called ‘linear model adjustment’ step is required to obtain the relative linear deformation rate and digital elevation model error from the double-differenced phase observations along the stack. In this step, since a non-linear system has to be resolved, the traditional least squares method cannot be directly applied. In order to overcome this problem, several methods have been developed in recent years. In this article, a new method has been developed to deal with the problem of linear model adjustment. This method repeatedly uses robust regression to resolve the non-linear system and is much easier to implement compared with other methods. This method is applied to both simulated and real data, and the results demonstrate that it can be efficiently used for linear model adjustment.

Improving maps of ice-sheet surface elevation change using combined laser altimeter and stereoscopic elevation model data

AbstractWe combine the complementary characteristics of laser altimeter data and stereoscopic digital elevation models (DEMs) to construct high-resolution (∼100 m) maps of surface elevations and elevation changes over rapidly changing outlet glaciers in Greenland. Measurements from spaceborne and airborne laser altimeters have relatively low errors but are spatially limited to the ground tracks, while DEMs have larger errors but provide spatially continuous surfaces. The principle of our method is to fit the DEM surface to the altimeter point clouds in time and space to minimize the DEM errors and use that surface to extrapolate elevations away from altimeter flight lines. This reduces the DEM registration errors and fills the gap between the altimeter paths. We use data from ICESat and ATM as well as SPOT 5 DEMs from 2007 and 2008 and apply them to the outlet glaciers Jakobshavn Isbræ (JI) and Kangerdlugssuaq (KL). We find that the main trunks of JI and KL lowered at rates of 30–35 and 7–20 m a−1,respectively. The rates decreased inland. The corresponding errors were 0.3–5.2 m a−1for JI and 0.3–5.1 m a−1for KL, with errors increasing proportionally with distance from the altimeter paths.

GIScience Research at the Thirty‐second Annual Esri International User Conference

GIScience Research at the Thirty‐second Annual Esri International User Conference