Conference PaperPDF Available

ASSESSMENT OF SPATIAL INTERPOLATION TECHNIQUES FOR RIVER BATHYMETRY GENERATION AND INTEGRATION IN LIDAR DTM: IMPACTS IN RIVER FLOW SIMULATIONS

Authors:
  • Geo-SAFER Agusan

Abstract and Figures

LiDAR DTMs produced by conventional LiDAR sensors and techniques cannot accurately represent terrain covered by water due to the incapability of the lasers emitted by the LiDAR sensor to penetrate water especially at high flow conditions. Intermediate steps are usually taken to integrate or merge river bed elevation/bathymetry data gathered from field surveys into the DTM, prior to using it as input into flood simulation models. Nowadays, Acoustic Doppler Current Profiler (ADCP), echosounder and total station instruments are basically used to measure and determine river bathymetry but generated datasets is mainly discrete format. Interpolation techniques are generally utilized to developed statistical surface of river bathymetry. In this paper, assessment of different spatial interpolation techniques has been carried out to determine the appropriate method for bathymetry generation. IDW, Kriging, Kernel Interpolation with Bariers, Local Polynomial Interpolation and Topo to Raster have been used to generate the bathymetry of Surigao river in Mindanao, Philippines as a case study area. The analysis reveals that the IDW generated the best interpolation with the lowest RMSE among the different techniques. On the other hand, the impacts of the interpolation techniques in river flow simulations were also assessed by integrating each of the generated river bathymetry into a LIDAR DTM. This bathymetry-integrated LIDAR DTMs were then used as inputs into a 2D numerical model for river flow simulations. Results show that the different spatial interpolation techniques produced different simulation results in terms of depth, velocity, and extent of inundation.
Content may be subject to copyright.
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
ASSESSMENT OF SPATIAL INTERPOLATION TECHNIQUES FOR RIVER
BATHYMETRY GENERATION AND INTEGRATION IN LIDAR DTM: IMPACTS IN
RIVER FLOW SIMULATIONS
Linbert C. Cutamora, Jesiree L. Serviano, Sherwin P. Pulido, Jojene R. Santillan, Meriam M. Makinano-Santillan,
Ronald M. Makinano, Melfred Anthony Berdera, Almer Cris N. Estorque, Jared P. Culdora
CSU Phil-LiDAR 1 Project, Caraga Center for Geo-Informatics,
College of Engineering and Information Technology, Caraga State University,
Ampayon, Butuan City, 8600, Philippines
E-mail: cutamoralinbert@gmail.com
KEY WORDS: River bathymetry, interpolation techniques, LIDAR DTM, bathymetry integration, river flow
simulation
ABSTRACT: LiDAR DTMs produced by conventional LiDAR sensors and techniques cannot accurately represent
terrain covered by water due to the incapability of the lasers emitted by the LiDAR sensor to penetrate water
especially at high flow conditions. Intermediate steps are usually taken to integrate or merge river bed
elevation/bathymetry data gathered from field surveys into the DTM, prior to using it as input into flood simulation
models. Nowadays, Acoustic Doppler Current Profiler (ADCP), echosounder and total station instruments are
basically used to measure and determine river bathymetry but generated datasets is mainly discrete format.
Interpolation techniques are generally utilized to developed statistical surface of river bathymetry. In this paper,
assessment of different spatial interpolation techniques has been carried out to determine the appropriate method
for bathymetry generation. IDW, Kriging, Kernel Interpolation with Bariers, Local Polynomial Interpolation and
Topo to Raster have been used to generate the bathymetry of Surigao river in Mindanao, Philippines as a case study
area. The analysis reveals that the IDW generated the best interpolation with the lowest RMSE among the different
techniques. On the other hand, the impacts of the interpolation techniques in river flow simulations were also
assessed by integrating each of the generated river bathymetry into a LIDAR DTM. This bathymetry-integrated
LIDAR DTMs were then used as inputs into a 2D numerical model for river flow simulations. Results show that
the different spatial interpolation techniques produced different simulation results in terms of depth, velocity, and
extent of inundation.
1. INTRODUCTION
Light Detection and Ranging (LiDAR) technology has made possible the availability of very high spatial resolution
topographic datasets, particularly Digital Terrain Models (DTM), allowing flood modellers to generate highly
detailed flood hazard maps (Turner et al., 2013). LiDAR DTMs, with spatial resolution of 1x1 m or better, are
usually used as inputs into one-dimensional (1D), two-dimensional (2D) or even three-dimensional (3D) flood
simulation models as source of topographic information necessary to simulate such processes like river flow
hydraulics and flood routing. However, it is an accepted fact that LiDAR DTMs, particularly those produced by
conventional LiDAR sensors and techniques (i.e., those without bathymetric mapping capabilities), cannot
accurately represent terrain covered by water due to the incapability of the lasers emitted by the LiDAR sensor to
penetrate water especially at high flow conditions (Caviedes-Voullième, 2014).
Prior to using it as input into flood simulation models, river bed elevation data gathered from field surveys are
usually interpolated and integrated into the DTM, (Mandlburger et al., 2009). Field surveys using equipment like
single or multi-beam SONAR (Sound Navigation And Ranging) combined real time kinematic Global Positioning
System (GPS) are commonly employed to measure discrete points or cross-sections representing river bed
elevation (Hilldale and Raff, 2007; Merwade, 2009). Since these techniques have limitations to produce continuous
data, interpolation techniques are commonly applied such as Inverse Distance Weighted (IDW), Kriging, Kernel
Interpolation with Bariers, Local Polynomial Interpolation, and Topo to Raster. These techniques estimate
elevation values at unmeasured/unsampled points from measurements made at surrounding sites (known values of
sampled points) (Weng, 2006).
The accuracy of interpolated river bed surfaces must be ensured before it is integrated into the LiDAR DTM and
utilized as input into flood models, especially that river bed topography plays a critical role in numerical modelling
of flow hydrodynamics (Merwade, 2009). Commonly available interpolation methods such as triangulation, IDW,
splines or kriging are reported to yield inaccurate river bed topography (Goff and Nordfjord , 2004; Merwade et al.
2006). Other study showed that kriging yielded better altitude estimations than IDW irrespective of the landform
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
type and sampling pattern (Zimmerman et al.,1999). In other studies neighborhood approaches such as IDW or
radial basis functions were found to be as accurate as krigging or even better (Guarneri et al., 2012). Although there
have been many studies on the accuracy of interpolation techniques for the generation of river bathymetry but there
are still no consistent findings about the performances of the spatial interpolators (Panhalkar et al., 2015)
In this work, we assessed the Inverse Distance Weighted (IDW), Kriging, Kernel Interpolation with Bariers, Local
Polynomial Interpolation and Topo to Raster interpolation techniques to determine the appropriate method for
bathymetry generation particularly in this area using the surveyed bathymetry data. In addition, the study also
assessed the impacts of the interpolated river bed integrated into the LiDAR DTM used as inputs into a 2D
numerical model for river flow simulations.
2. DATASETS AND METHODS
2.1 River Bathymetric Survey Data
To illustrate how spatial interpolation techniques affect the river flow simulations, we focused on generation of
river bed topography of a portion of the Surigao River with LiDAR DTM located in Surigao River Basin, Surigao
del Norte, Mindanao in Philippines (Figure 1). This portion is approximately 800 meters in length, with an average
width of 90 m. The bathymetric data of this portion was collected through a combination of on-boat and manual
bathymetry surveys. On-boat surveys were conducted using an integrated RTK GNSS (South S86T ) and Single-
beam Echosounder (South SDE-28S) equipment (Figure 2). Riverbed elevations were referred to the Mean Sea
Level (MSL) datum. Manual surveys were conducted using RTK GNSS equipment in shallow areas where boat
navigation is difficult. The surveys followed pre-determined bathymetry route in a manner shown in Figure 3.
Basically, bathymetry route covered the leftmost and rightmost portions of the river near the banks, its centerline,
and zigzag direction at 100-m interval including sea portion near at the mouth of the river.
Figure 1. Map of the study area.
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
Figure 2. Set-up of the bathymetric surveys.
.
Figure 3. Route during the conduct of echosounder and manual bathymetry surveys.
2.2 River Bed Topography Generation and Accuracy Assessment
The surveyed bathymetric data (Figure 4) was partitioned into interpolation and validation points (Figure 5). A total
of 4,345 (or 95% of the total collected points) were used for interpolation while the remaining 5% or 229 points
were used as input for validating the interpolated surface.
We used the IDW (n=2), Kriging, Kernel Interpolation with Bariers, Local Polynomial Interpolation and Topo to
Raster interpolation methods. The parameters that were set for each interpolation method are shown in Table 1.
The accuracies of the interpolations were assessed using 5% validation points. We used Root Mean Square Error
which is one of commonly used quantitative methods to assess accuracy (Merwade, 2009). Since the validation
points are unique, these points can be used as basis in determining which interpolation methods are the most
superior in terms of accuracy. In addition, we also compared the actual cross-section data conducted along the river
with the interpolation results. The interpolated river bed surfaces were then be clipped and integrated into the 1 m x
1 m resolution LiDAR DTM using available tools in ArcGIS 10. Overall, a total of 5 bed-integrated DTMs were
generated.
Table 1. Parameters that were used for each interpolation methods.
Interpolation Methods
Parameters
Inverse Distance Weighted
(IDW)
Power = 2; Major axis = 125; Minor axis = 75; Angle = 65
Max and Min Neighbors = 3;Sector Type = 4 sector with 450 offset
Kriging
Kriging Method = Ordinary; Search Radius = Variable
Number Points = 6
Kernel Interpolation With
Barriers
Kernel Function = EPANECHNIKOV; Bandwidth = 60
Order of polynomial = 1; Ridge Parameter = none
Output surface type = Prediction
Local Polynomial
Interpolation
Order of polynomial = 1; Search Neighborhood = Smooth Circular
Smoothing Factor = 0.2; Kernel Function = EPANECHNIKOV
Bandwidth = 60
Topo to Raster
Maximum Number of Iterations = 20
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
Figure 5. Map showing the interpolation and validation points of bathymetric data.
Figure 4. Map showing the LiDAR Digital Terrain Model (DTM) and river bathymetric data.
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
2.3 River Flow Simulation using HEC RAS 2D Hydraulic Model
The 5 bed-integrated DTM were each used as input into a 2D hydraulic model based on the latest version of HEC
RAS (Version 5.0.1; USACE HEC, 2016). The model was constructed for it to use the topographic data provided
by the bed-integrated DTM in simulating river flow. The model domain was focused only on the portion of the
river where the bed elevations were interpolated (Figure 6). We used a single Manning’s roughness value of 0.04
for all portions of the model domain. We used a computational mesh size of 2 m x 2 m containing 27,608 cells. The
decision of not using the full resolution of 1 m x 1 m as computational mesh size of the model is order to have
faster computation times.
Each of the configured 2D models is then used to simulate the flow of water entering the upstream portion of the
river. The river was set as initially dry before the simulation. A 24-hour hydrograph with a peak flow of 986.62
m3/s was utilized as the inflow boundary condition (Figure 7). This hydrograph was computed by a calibrated
hydrologic model for this portion. For the downstream portion, we assigned a “Normal Depth” boundary condition
(friction slope = 0.004 based on the average longitudinal slope of the river bed). For a stable model simulation, we
set the computational time interval to 10 seconds.
Figure 6. The HEC RAS 2D computational domain overlaid in the LiDAR DTM.
Figure 7. The inflow boundary condition utilized in the 2D river flow simulation
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
2.4 Assessment of the 5 Bed-Integrated DTM used as Inputs to River Flow Simulation
The effect of different interpolation techniques was assessed by examining the result in terms of the simulated
depth, velocity, and flood extent by the 5 2D hydraulic models with different bed-integrated DTMs as inputs. By
comparing each model output, we determined the similarities or differences in maximum depth, flood extent and
velocity. The differences in depth and velocity between the 5 bed-integrated DTM were assessed using the 3 cross-
sectional and 3 longitudinal (one near left bank, one at a center and one at right bank) transects. Each cross-
sectional and longitudinal transects containing approximately 100 points at 1-m interval was used to extract the
simulated depth and velocity. The locations of the transected cross-sectional and longitudinal are shown in Figure
8.
Figure 8. Location of cross-sectional and longitudinal transects.
3. RESULTS AND DISCUSSION
3.1 Interpolated River Bed Surfaces
Figure 9 shows the result of the IDW, Kriging, Kernel Interpolation with Bariers, Local Polynomial Interpolation
and Topo to Raster interpolation methods. For the interpolation using IDW, differences in elevation ranges were
least obvious than those interpolation using Kriging and Topo to Raster except Kernel Interpolation with Bariers
and Local Polynomial Interpolation. The Kernel Interpolation with Bariers and Bariers and Local Polynomial
Interpolation interpolations displayed almost flat interpolated surfaces.
The RMSEs of these interpolated surfaces based on independent set of validation data points range from 0.18 m to
0.35 m (Figure 10). These computed RMSEs show that IDW is much appropriate interpolation techniques as per
the various validation methods. The graph showing the comparison of the cross-section data and the interpolated
river bed surfaces is shown in Figure 11. Surveyed cross-section data is plotted against IDW, Kriging, Kriging,
Kernel Interpolation with Bariers, Local Polynomial Interpolation and Topo to Raster generated cross-sections, it
also shows that IDW generated cross-section are much more than accurate than others.
3.2 River Flow Simulation Results
Figure 12 and 13 show the maximum flood depth and velocity simulated by 2D hydraulic model using different
bed-integrated DTM. As observed, the maximum flood depth simulated using Kernel Interpolation with barriers
was underestimated compared to IDW, Kriging and Topo to Raster in which they produced maximum flood depths
that were similar to each other. Among the 5 bed-integrated DTM, IDW, Kriging and Topo to Raster interpolation
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
techniques produced almost same areas of the flood extent. Local Polynomial Interpolation got bigger area of the
flood extent. It can be observed that each bed-integrated DTM produced different model results (Figure 14 to
Figure 25).
Figure 9. Results of the IDW, Kriging, Kernel Interpolation with Bariers, Local Polynomial Interpolation and Topo
to Raster interpolation methods.
Figure 10. Root Mean Square Error (RMSE) of the interpolated river bed surfaces using the validation points.
0.188 0.192
0.353 0.353
0.240
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
IDW KRIGING KERNEL INTERPOLATION
WITH BARRIERS LOCAL POLYNOMIAL
INTERPOLATION TOPO TO RASTER
RMSE Value
Interpolation Methods
RMSE of the Interpolated River Bed Surface
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
Figure 11. Comparison of actual surveyed cross-section and cross-sections generated from interpolated river bed
surfaces.
Figure12. Maximum flood depths simulated by the 2D hydraulic model using the 5 bed-integrated DTM.
-4
-2
0
2
4
6
8
121 41 61 81 101 121 141 161
Rvier Bed Elevation (m)
Distance (m)
Comparison of Cross-sections
ACTUAL SURVEYED CROSS-SECTION
IDW
KRIGING
KERNEL INTERPOLATION WITH BARRIERS
LOCAL POLYNOMIAL INTERPOLATION
TOPO TO RASTER
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
Figure13. Maximum flood velocities simulated by the 2D hydraulic model using the 5 bed-integrated DTM.
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91 101
Flood Deoth (m)
Distance (m)
Comparison of Flood Depth at Cross-sectional
Transect 2
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91
Flood Depth (m)
Distance (m)
Comparison of Flood Depth at Cross-sectional
Transect 3
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91 101 111 121
Flood Depth (m)
Distance (m)
Comparison of Flood Depth at Longitudinal Transect
1
IDW
KRIGING
KERNEL
INTERPOLATION
WITH BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91 101 111
Flood Depth (m)
Distance (m)
Comparison of Flood Depth at LongitudinalTransect
3
IDW
KRIGING
KERNEL INTERPOLATION
WITH BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91
Flood Deoth (m)
Distance (m)
Comparison of Flood Depth at Cross-sectional
Transect 1
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91 101 111 121
Flood Depth (m)
Distance (m)
Comparison of Flood Depth at Longitudinal Transect
2
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
Figure 16. Comparison of model generated flood depth
at cross-sectional transect 3.
Figure 17. Comparison of model generated flood depth
at longitudinal transect 1.
Figure 15. Comparison of model generated flood depth
at cross-sectional transect 2
Figure 14. Comparison of model generated flood depth
at cross-sectional transect 1.
Figure 18. Comparison of model generated flood depth
at longitudinal transect 2.
Figure 19. Comparison of model generated flood depth
at longitudinal transect 3.
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
0
0.5
1
1.5
2
2.5
3
3.5
4
1 11 21 31 41 51 61 71 81 91
Velocity (m/s)
Distance (m)
Comparison of Velocity at Cross-sectional Transect 1
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
1
2
3
4
5
6
1 11 21 31 41 51 61 71 81 91 101
Velocity (m/s)
Distance (m)
Comparison of Velocity at Cross-sectional Transect 2
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL INTERPOLATION
TOPO TO RASTER
0
0.5
1
1.5
2
2.5
3
3.5
1 11 21 31 41 51 61 71 81 91 101 111 121
Velocity (m/s)
Distance (m)
Comparison of Velocity at Longitudinal Transect 1
IDW
KRIGING
KERNEL INTERPOLATION
WITH BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
0.5
1
1.5
2
2.5
3
3.5
4
1 11 21 31 41 51 61 71 81 91 101 111 121
Velocity (m/s)
Distance (m)
Comparison of Velocity at Longitudinal Transect 2
IDW
KRIGING
KERNEL INTERPOLATION
WITH BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
0.5
1
1.5
2
2.5
3
3.5
1 11 21 31 41 51 61 71 81 91 101 111
Velocity (m/s)
Distance (m)
Comparison of Velocity at Longitudinal Transect 3
IDW
KRIGING
KERNEL INTERPOLATION
WITH BARRIERS
LOCAL POLYNOMIAL
INTERPOLATION
TOPO TO RASTER
0
0.5
1
1.5
2
2.5
3
1 11 21 31 41 51 61 71 81 91
Velocity (m/s)
Distance (m)
Comparison of Velocity at Cross-sectional Transect 3
IDW
KRIGING
KERNEL INTERPOLATION WITH
BARRIERS
LOCAL POLYNOMIAL INTERPOLATION
TOPO TO RASTER
Figure 21. Comparison of model generated flood
velocity at cross-sectional transect 2.
Figure 20. Comparison of model generated flood
velocity at cross-sectional transect 1.
Figure 23. Comparison of model generated flood
velocity at longitudinal transect 1.
Figure 22. Comparison of model generated flood
velocity at cross-sectional transect 3.
Figure 25. Comparison of model generated flood
velocity at longitudinal transect 3.
Figure 24. Comparison of model generated flood
velocity at longitudinal transect 2.
37th Asian Conference on Remote Sensing 2016, 17-21 October 2016, Colombo, Sri Lanka
CONCLUSION
In this paper, we determined the accuracy of different spatial interpolation techniques and the result of the 2D
hydraulic model utilized the different bed-integrated DTMs. The results show that IDW is the suitable and accurate
interpolation technique for generating the river bed topography of our study area. It also concluded that generated
river bathymetry from different interpolation techniques and integrated with LiDAR DTM also produced different
2D model simulation results in terms of depth, velocity, and extent of inundation.
ACKNOWLEDGEMENT
This work is an output of the Caraga State University (CSU) Phil-LiDAR 1 project under the “Phil-LiDAR 1.
Hazard Mapping of the Philippines using LiDAR” program funded by the Department of Science and Technology
(DOST). The LiDAR DTM used in this work was provided by the University of the Philippines Disaster Risk and
Exposure for Mitigation (UP DREAM)/Phil-LIDAR 1 Program.
REFERENCES
Caviedes-Voullième, D., Morales-Hernández, M., López-Marijuan, I., García-Navarro, P., 2014. Reconstruction of
2D river beds by appropriate interpolation of 1D cross-sectional information for flood simulation. Environmental
Modelling & Software, 61, pp. 206-228.
Goff, J.A., Nordfjord, S., 2004. Interpolation of fluvial morphology using channel oriented coordinate
transformation: a case study from the New Jersey shelf. Mathematical Geology, 36 (6), pp. 643658.
Guarneri, J. C., Weih Jr., R. C., 2012. Comparing methods for interpolation to improve raster digital elevation
models. Journal of the Arkansas Academy of Science, 66, pp. 7781.
Hilldale, R. C., Raff, D., 2008. Assessing the ability of airborne LiDAR to map river bathymetry. Earth Surface
Processes and Landforms, 33(5), pp. 773-783.
Mandlburger, G., Hauer, C., Höfle, B., Habersack, H., Pfeifer, N., 2009. Optimization of LiDAR derived terrain
models for river flow modelling. Hydrology and Earth System Sciences, 13(8), pp. 1453-1466.
Merwade, V., 2009. Effect of spatial trends on interpolation of river bathymetry. Journal of Hydrology, 371(1), pp.
169-181.
Panhalkar, S.S., Jarag., A.P., 2015. Assessment of spatial Interpolation techniques for river bathymetry generation
of Panchganga River Basin using geoinformatic techniques. Asian Journal of Geoinformatics, 15(3), pp. 9-15.
Turner, A. B., Colby, J. D., Csontos, R. M., Batten, M., 2013. Flood modeling using a synthesis of multi-platform
LiDAR data. Water, 5(4), pp. 1533-1560.
USACE HEC, 2016. HEC RAS River Analysis System 2D Modeling User’s Manual Version 5.0, Hydrologic
Engineering Center, United States Corps of Engineers.
Weng, Q., 2006. An evaluation of spatial interpolation accuracy of elevation data, In: Progress in Spatial Data
Handling, Springer-Verlag, Berlin, pp. 805-824.
Zimmerman, D., Pavlik, C., Ruggles, A., Armstrong, M., 1999. An experimental comparison of ordinary and
universal krigging and inverse distance weighting, Mathematical Geology, 31, 1999, pp. 375-390.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Abstract River geometry is very crucial for various hydrological applications. However, generation of river geometry data is cumbersome process. DGPS, Total station and echosounder instrument are basically used to generate river bathymetry but generated datasets is mainly in discrete format. Interpolation techniques are generally used to develop statistical surface of river bathymetry. In the present paper, assessment of various interpolation techniques has been carried out to suggest appropriate method for bathymetry generation. Panchganga river of Kolhapur district for a stretch of 50 km (from Balinga to Ichalkaranji) has been selected for the present work. Forty-eight cross sections have been generated by using DGPS and Total station instrument. To generate bathymetry of Panchganga river, some of the cross section have been used. IDW, Krigging and Topo to Raster techniques have been used to generate river bathymetry. RMS and Standard RMS techniques are used for cross validation. The analysis reveals that IDW is much appropriate techniques for river bathymetry generation. Key words: River Bathymetry, Cross-section data, interpolation techniques, RMS.
Article
Full-text available
This study examined the utility of a high resolution ground-based (mobile and terrestrial) Light Detection and Ranging (LiDAR) dataset (0.2 m point-spacing) supplemented with a coarser resolution airborne LiDAR dataset (5 m point-spacing) for use in a flood inundation analysis. The techniques for combining multi-platform LiDAR data into a composite dataset in the form of a triangulated irregular network (TIN) are described, and quantitative comparisons were made to a TIN generated solely from the airborne LiDAR dataset. For example, a maximum land surface elevation difference of 1.677 m and a mean difference of 0.178 m were calculated between the datasets based on sample points. Utilizing the composite and airborne LiDAR-derived TINs, a flood inundation comparison was completed using a one-dimensional steady flow hydraulic modeling analysis. Quantitative comparisons of the water surface profiles and depth grids indicated an underestimation of flooding extent, volume, and maximum flood height using the airborne LiDAR data alone. A 35% increase in maximum flood height was observed using the composite LiDAR dataset. In addition, the extents of the water surface profiles generated from the two datasets were found to be statistically significantly different. The urban and mountainous characteristics of the study area as well as the density (file size) of the high resolution ground based LiDAR data presented both opportunities and challenges for flood modeling analyses.
Article
Full-text available
A factorial, computational experiment was conducted to compare the spatial interpolation accuracy of ordinary and universal kriging and two types of inverse squared-distance weighting. The experiment considered, in addition to these four interpolation methods, the effects of four data and sampling characteristics: surface type, sampling pattern, noise level, and strength of small-scale spatial correlation. Interpolation accuracy was measured by the natural logarithm of the mean squared interpolation error. Main effects of all five factors, all two-factor interactions, and several three-factor interactions were highly statistically significant. Among numerous findings, the most striking was that the two kriging methods were substantially superior to the inverse distance weighting methods over all levels of surface type, sampling pattern, noise, and correlation.
Article
Full-text available
Airborne LiDAR (Light Detection And Ranging) combines cost efficiency, high degree of automation, high point density of typically 1–10 points per m2 and height accuracy of better than ±15 cm. For all these reasons LiDAR is particularly suitable for deriving precise Digital Terrain Models (DTM) as geometric basis for hydrodynamic-numerical (HN) simulations. The application of LiDAR for river flow modelling requires a series of preprocessing steps. Terrain points have to be filtered and merged with river bed data, e.g. from echo sounding. Then, a smooth Digital Terrain Model of the Watercourse (DTM-W) needs to be derived, preferably considering the random measurement error during surface interpolation. In a subsequent step, a hydraulic computation mesh has to be constructed. Hydraulic simulation software is often restricted to a limited number of nodes and elements, thus, data reduction and data conditioning of the high resolution LiDAR DTM-W becomes necessary. We will present a DTM thinning approach based on adaptive TIN refinement which allows a very effective compression of the point data (more than 95% in flood plains and up to 90% in steep areas) while preserving the most relevant topographic features (height tolerance ±20 cm). Traditional hydraulic mesh generators focus primarily on physical aspects of the computation grid like aspect ratio, expansion ratio and angle criterion. They often neglect the detailed shape of the topography as provided by LiDAR data. In contrast, our approach considers both the high geometric resolution of the LiDAR data and additional mesh quality parameters. It will be shown that the modelling results (flood extents, flow velocities, etc.) can vary remarkably by the availability of surface details. Thus, the inclusion of such geometric details in the hydraulic computation meshes is gaining importance in river flow modelling.
Article
Airborne bathymetric LiDAR was collected for 220 river kilometres in the Yakima and Trinity River Basins in the USA. Concomitant with the aerial data collection, ground surveys of the river bed were performed in both basins. We assess the quality of the bathymetric LiDAR survey from the perspective of its application toward creating accurate, precise and complete streambed topography for numerical modelling and geomorphological assessment. Measurement error is evaluated with respect to ground surveys for magnitude and spatial variation. Analysis of variance statistics indicate that residuals from two independent ground surveys in similar locations do not come from the same population and that mean errors at different study locations also come from different populations. Systematic error indicates a consistent bias in the data and random error falls within values of expected precision. Published in 2007 by John Wiley & Sons, Ltd.
Article
We present a new methodology for interpolating channel morphology that incorporates a transformation from geographic to channel-based coordinate systems. Interpolation in the transformed space enables enforcement of downstream continuity of morphology and edge delineation through any changes in channel direction. The transformation is guided by a channel center line, which approximately tracks the path of the channel through geographic space; coordinates are given in distance along and across the center line. Accurate interpolation requires a track line density sufficient to unambiguously trace channels from one track line to the next. Channel continuity is ensured by first interpolating along paths defined by the channel thalweg and edges, which must be chosen by the user, and along several interim paths between the edges and thalweg. The completed interpolations for each channel are transformed back into geographic coordinates, and channel confluence is handled through a maximum depth criterion. The method is applied here to shallowly buried channels mapped with high-resolution chirp seismic data on the New Jersey shelf, but should be applicable to a wide range of subaerial and buried fluvial systems.
Chapter
This paper makes a general evaluation of the spatial interpolation accuracy of elevation data. Six common interpolators were examined, including Kriging, inverse distance to a power, minimum curvature, modified Shepard’s method, radial basis functions, and triangulation with linear interpolation. The main properties and mathematical procedures of the interpolation algorithms were reviewed. In order to obtain full evaluation of the interpolations, both statistical (including root-mean-square-error, standard deviation, and mean) and spatial accuracy measures (including accuracy surface, and spatial autocorrelation) were employed. It is found that the accuracy of spatial interpolation of elevations was primarily subject to input data point density and distribution, grid size (resolution), terrain complexity, and interpolation algorithm used. The variations in interpolation parameters may significantly improve or worsen the accuracy. Further researches are needed to examine the impacts of terrain complexity in details and various data sampling strategies. The combined use of variogram models, accuracy surfaces, and spatial autocorrelation represents a promising direction in mapping spatial data accuracy.
Article
Continuous surface of river bathymetry (bed topography) is typically produced by spatial interpolation of discrete point or cross-section data. Several interpolation methods that do not account for spatial trend in river bathymetry produce inaccurate surfaces, thus requiring complex interpolation methods such as anisotropic kriging. Although isotropic methods are unsuitable for interpolating river bathymetry, issues that limit their use remain unaddressed. This paper addresses the issue of effect of spatial trend in river bathymetry on isotropic interpolation methods. It is hypothesized that if the trend is removed from the data before interpolation, the results from isotropic methods should be comparable with anisotropic methods. Data from six river reaches in the United States are used to: (i) interpolate bathymetry data using seven spatial interpolation methods; (ii) separate trend from bathymetry; (iii) interpolate residuals (bathymetry minus trend) by using all seven interpolation methods to get residual surfaces, (iv) add the trend back to residual surfaces; and (v) compare resulting surfaces from (iv) with surfaces created in (i). Quantitative and qualitative comparison of results through root mean square error (RMSE), semi-variograms, and cross-section profiles show that significant improvement (as much as 60% in RMSE) can be accomplished in spatial interpolation of river bathymetry by separating trend from the data. Although this paper provides a new simple way for interpolating river bathymetry by using (otherwise deemed inappropriate) isotropic methods, the choice of trend function and spatial arrangement of discrete bathymetry data play a vital role in successful implementation of the proposed approach.
Comparing methods for interpolation to improve raster digital elevation models
  • J C Guarneri
  • Weih Jr
Guarneri, J. C., Weih Jr., R. C., 2012. Comparing methods for interpolation to improve raster digital elevation models. Journal of the Arkansas Academy of Science, 66, pp. 77–81.