Lidar Elevation Data Research Articles

The Challenge of Multispectral Remote Sensing for Mapping Kentucky Wetlands

Accurate and up-to-date data on the location and characterization of Kentucky's wetlands are lacking. The goal of this research was to assess the potential of low-cost, multispectral imagery for the delineation and classification of wetlands in three focal river basins: the Kentucky River Basin, Licking River Basin, and the Salt River Basin. Methods included the classification of mid-resolution multispectral Landsat 8 and ASTER remotely sensed imagery. These data were evaluated in conjunction with ancillary watershed, soil, and hydrologic data layers in a GIS. The field-verified results demonstrate the challenges of classifying Kentucky's wetlands using mid-resolution multispectral imagery due to the spectral similarity of wetlands and other land cover classes. The majority of commission errors occurred in inundated or partially inundated areas with abundant vegetation, notably agricultural fields and forests. Future research may improve classification by incorporating LiDAR slope and elevation data.

Wet channel network extraction by integrating LiDAR intensity and elevation data

AbstractThe temporal dynamics of stream networks are vitally important for understanding hydrologic processes including surface water and groundwater interaction and hydrograph recession. However, observations of wet channel networks are limited, especially in headwater catchments. Near‐infrared LiDAR data provide an opportunity to map wet channel networks owing to the fine spatial resolution and strong absorption of light energy by water surfaces. A systematic method is developed to map wet channel networks by integrating elevation and signal intensity of ground returns. The signal intensity thresholds for identifying wet pixels are extracted from frequency distributions of intensity return within the convergent topography extent using a Gaussian mixture model. Moreover, the concept of edge in digital image processing, defined based on the intensity gradient, is utilized to enhance detection of small wet channels. The developed method is applied to the Lake Tahoe area based on eight LiDAR snapshots during recession periods in five watersheds. A power law relationship between streamflow and wetted channel length during recession periods is derived, and the scaling exponent ( ) is within the range of reported values from fieldwork in other regions.

A Flood Risk Assessment of the LaHave River Watershed, Canada Using GIS Techniques and an Unstructured Grid Combined River-Coastal Hydrodynamic Model

A flexible mesh hydrodynamic model was developed to simulate flooding of the LaHave River watershed in Nova Scotia, Canada, from the combined effects of fluvial discharge and ocean tide and surge conditions. The analysis incorporated high-resolution lidar elevation data, bathymetric river and coastal chart data, and river cross-section information. These data were merged to generate a seamless digital elevation model which was used, along with river discharge and tidal elevation data, to run a two-dimensional hydrodynamic model to produce flood risk predictions for the watershed. Fine resolution topography data were integrated seamlessly with coarse resolution bathymetry using a series of GIS tools. Model simulations were carried out using DHI Mike 21 Flexible Mesh under a variety of combinations of discharge events and storm surge levels. Discharge events were simulated for events that represent a typical annual maximum runoff and extreme events, while tide and storm surge events were simulated by using the predicted tidal time series and adding 2 and 3 m storm surge events to the ocean level seaward of the mouth of the river. Model output was examined and the maximum water level for the duration of each simulation was extracted and merged into one file that was used in a GIS to map the maximum flood extent and water depth. Upstream areas were most vulnerable to fluvial discharge events, the lower estuary was most vulnerable to the effect of storm surge and sea-level rise, and the Town of Bridgewater was influenced by the combined effects of discharge and storm surge. To facilitate the use of the results for planning officials, GIS flood risk layers were intersected with critical infrastructure, identifying the roads, buildings, and municipal sewage infrastructure at risk under each flood scenario. Roads were converted to points at 10 m spacing for inundated areas and appended with the flood depth calculated from the maximum water level subtracted from the lidar digital elevation model.

Effective LiDAR-based modelling and visualisation of managed retreat scenarios for coastal planning: An example from the southern UK

This study presents a visualisation of proposed managed retreat options for a scientifically important estuarine area within the South Downs National Park, United Kingdom. A range of GIS and remote sensing techniques are used to facilitate the manipulation of LiDAR data for landscape visualisation. Landscape and flood modelling methods are incorporated and developed. The extent of inundation and tidal prism are assessed following proposed coastal planning and landscape changes.Ground-truthed LiDAR elevation data are used to model a series of proposed managed retreat scenarios (scenarios A, B, C and D) for the river Cuckmere estuary. Following detailed landscape visualisation and manipulation, a 'pour-point' flood modelling method is used to simulate tidal inundation and calculate the tidal prism for proposed managed retreat scenarios using current tidal data. Further analysis was undertaken to assess the potential impact of sea level rise (for the years 2025, 2050, 2075 and 2100) in this estuarine area if the managed retreat proposals are actioned.The use of GIS-based techniques to alter and visualise landscape features under proposed managed retreat alternatives proved successful. Clear differences in tidal inundation and the tidal prism were observed following the modelling of large scale engineering features. These differences in inundation were apparent both using current tidal data and predicted sea level rise data. The tidal prism was strongly affected by the choice of the managed retreat scenario. Subtle changes in landscape features were shown to strongly affect the water accommodation potential of the estuary. The study demonstrates that high resolution data facilitated the detailed visualisation and modelling of fine-scale landscape change. Furthermore, the alteration of landscape features suggested that the area of inundation could potentially vary by over 20% (>100 ha); and that these landscape changes would noticeably alter tidal impact in this estuary.

Quality assessment of TanDEM-X DEMs using airborne LiDAR, photogrammetry and ICESat elevation data

Abstract. TanDEM-X mission has been acquiring InSAR data to produce high resolution global DEM with greater vertical accuracy since 2010. In this study, TanDEM-X CoSSC data were processed to produce DEMs at 6 m spatial resolution for two test areas of India. The generated DEMs were compared with DEMs available from airborne LiDAR, photogrammetry, SRTM and ICESat elevation point data. The first test site is in Bihar state of India with almost flat terrain and sparse vegetation cover and the second test site is around Godavari river in Andhra Pradesh (A.P.) state of India with flat to moderate hilly terrain. The quality of the DEMs in these two test sites has been specified in terms of most widely used accuracy measures viz. mean, standard deviation, skew and RMSE. The TanDEM-X DEM over Bihar test area gives 5.0 m RMSE by taking airborne LiDAR data as reference. With ICESat elevation data available at 9000 point locations, RMSE of 5.9 m is obtained. Similarly, TanDEM-X DEM for Godavari area was compared with high resolution aerial photogrammetric DEM and SRTM DEM and found RMSE of 5.3 m and 7.5 m respectively. When compared with ICESat elevation data at several point location and also the same point locations of photogrammetric DEM and SRTM, the RMS errors are 4.1 m, 3.5 m and 4.3 m respectively. DEMs were also compared for open-pit coal mining area where elevation changes from -147 m to 189 m. X- and Y-profiles of all DEMs were also compared to see their trend and differences.

Vineyards in Terraced Landscapes: New Opportunities from Lidar Data

AbstractVineyard landscapes are a relevant part of the European culture, and several authors concluded that they are the agricultural practice that causes the highest soil loss. Grape quality depends on the availability of water, and soil erosion is an important parameter dictating the vineyard sustainability; therefore, soil and water conservation measures are often implemented. Among them, the construction of terraces is the most widely used system. However, while favouring agricultural activities, terraces if not properly maintained can lead to local instabilities creating hazards for settlements and cultivations, and for the related economy. Terraced fields are also served by agricultural roads that can have deep effects on water flows triggering surface erosion. The goal of this research is to use lidar elevation data for a hydro‐geomorphological analysis of terraced vineyards. The work is divided in two parts. At first, the Relative Path Impact Index is tested in two vineyards to identify terrace‐induced and road‐induced erosions. Statistical thresholds of the Relative Path Impact Index are then defined to label the most critical areas. On the second step, using the index and the defined thresholds, we simulate different scenarios of soil conservation measures, establishing the optimal solution to reduce erosion. The results highlight the effectiveness of high‐resolution topography in the analysis of surface erosion in terraced vineyards, when the surface water flow is the main factor triggering the instabilities. The proposed analysis can help in scheduling a suitable planning to mitigate the consequences of the anthropogenic alterations induced by the terraces and agricultural roads. Copyright © 2014 John Wiley & Sons, Ltd.

Soil dispersion tests combined with topographical information can describe field‐scale sediment and phosphorus losses

AbstractMethods for estimating sediment and phosphorus (P) transfer from agricultural land to surface waters are needed to mitigate their adverse impact on water quality. This study quantified sediment and P losses from five agricultural fields and evaluated the patterns observed based on the intrinsic risk of sediment and P mobilization from the soil, together with field topographical and hydrological data. The mobilization risk was estimated using the environmental soil dispersion test DESPRAL. High‐resolution LiDAR elevation data and crop management data were used to describe transport and delivery of the material mobilized. Annual flow‐weighted suspended solids (SS) concentration in drainage water ranged from 48 to 374 mg/L, total P (TP) from 0.12 to 0.39 mg/L and unreactive P (UP) from 0.08 to 0.33 mg/L. The mobilization risk, measured as turbidity in the aliquot recovered from the dispersion test, varied from 781 to 2310 nephelometric turbidity units (NTU). The method proved to be efficient in describing and differentiating sediment and P mobilization potential between fields. The topographical data also showed large differences between fields, with the length–slope (LS) parameter varying from 0.037 to 0.999. Based on assessments of source (mobilization risk) and transport factors (LS, crop management data), it was possible to characterize fields as favoured or limited in terms of source and transport and to explain the long‐term SS and P losses observed at field scale.

Analysis of the accuracy of horizontals estimated by different mapping methods

The objective of this research is to analyse the accuracy of horizontals which are formed by different mapping methods and to create an accuracy analysis model of surface elevation formed by different methods. The research was carried out in three stages: 1) Choosing the territory and collecting spatial data; 2) Data processing; 3) Analysis of data. The research object is the relief of some part of the Pypliai Village which is in the Kaunas District. In respect of the relief it is a varied place of different expressiveness situated by the Nemunas River. After the comparison of elevation data got by geodetic and analogical methods it was estimated that the total intersect area was 0.9581 ha (32% of all investigative territory). When comparing the elevation model made by the geodetic method and the LiDAR method, it was estimated that elevation intersected in 1.2762 ha of the territory (43% of all investigative territory). In comparison of the elevation data of all three methods it was measured that the area of intersection was 0.5926 ha (only 20% of all investigative territory). It can be confirmed that the data is accurate in this territory – it intersects in all three mapping methods. The LiDAR elevation data is more accurate than the analogical model data. Nevertheless, the utmost 10 meters error is considerable while comparing to the geodetic model data, it occurs 10 times of 74 cases of comparison.

Quantitative Methods for Comparing Different Polyline Stream Network Models

Two techniques for exploring relative horizontal accuracy of complex linear spatial features are described and sample source code (pseudo code) is presented for this purpose. The first technique, relative sinuosity, is presented as a measure of the complexity or detail of a polyline network in comparison to a reference network. We term the second technique longitudinal root mean squared error (LRMSE) and present it as a means for quantitatively assessing the horizontal variance between two polyline data sets representing digitized (reference) and derived stream and river networks. Both relative sinuosity and LRMSE are shown to be suitable measures of horizontal stream network accuracy for assessing quality and variation in linear features. Both techniques have been used in two recent investigations involving extraction of hydrographic features from LiDAR elevation data. One confirmed that, with the greatly increased resolution of LiDAR data, smaller cell sizes yielded better stream network delineations, based on sinuosity and LRMSE, when using LiDAR-derived DEMs. The other demonstrated a new method of delineating stream channels directly from LiDAR point clouds, without the intermediate step of deriving a DEM, showing that the direct delineation from LiDAR point clouds yielded an excellent and much better match, as indicated by the LRMSE.

Forest Mapping Through Object-based Image Analysis of Multispectral and LiDAR Aerial Data

The objective of this research was to establish a set of rules for using Object-based Image Analysis (OBIA) to semi-automatically map a given densely forested area. Aerial images (RGB and NIR bands) and LiDAR elevation data provided the raw data and were analysed using eCognition Developer 8 software. All the non-forested areas, including built-up areas, water surfaces or agricultural land, were identified first. The forested areas were then classified as stands composed principally of broadleaf trees, coniferous trees, mixed forest or clear-cuts, which was achieved with an accuracy of almost 90%. Subsequently the stands were classified on the basis of height, using 5 metres intervals, and this was achieved with an accuracy of just over 70%.

A NOVEL METHOD FOR AUTOMATION OF 3D HYDRO BREAK LINE GENERATION FROM LIDAR DATA USING MATLAB

Abstract. Water body detection is necessary to generate hydro break lines, which are in turn useful in creating deliverables such as TINs, contours, DEMs from LiDAR data. Hydro flattening follows the detection and delineation of water bodies (lakes, rivers, ponds, reservoirs, streams etc.) with hydro break lines. Manual hydro break line generation is time consuming and expensive. Accuracy and processing time depend on the number of vertices marked for delineation of break lines. Automation with minimal human intervention is desired for this operation. This paper proposes using a novel histogram analysis of LiDAR elevation data and LiDAR intensity data to automatically detect water bodies. Detection of water bodies using elevation information was verified by checking against LiDAR intensity data since the spectral reflectance of water bodies is very small compared with that of land and vegetation in near infra-red wavelength range. Detection of water bodies using LiDAR intensity data was also verified by checking against LiDAR elevation data. False detections were removed using morphological operations and 3D break lines were generated. Finally, a comparison of automatically generated break lines with their semi-automated/manual counterparts was performed to assess the accuracy of the proposed method and the results were discussed.

Ecohydrological and socioeconomic integration for the operational management of environmental flows

Investment in and operation of flow control infrastructure such as dams, weirs, and regulators can help increase both the health of regulated river ecosystems and the social values derived from them. This requires high-quality and high-resolution spatiotemporal ecohydrological and socioeconomic information. We developed such an information base for integrated environmental flow management in the River Murray in South Australia (SA). A hydrological model was used to identify spatiotemporal inundation dynamics. River ecosystems were classified and mapped as ecohydrological units. Ecological response models were developed to link three aspects of environmental flows (flood duration, timing, and inter-flood period) to the health responses of 16 ecological components at various life stages. Potential infrastructure investments (flow control regulators and irrigation pump relocation) were located by interpreting LiDAR elevation data, digital orthophotography, and wetland mapping information; and infrastructure costs were quantified using engineering cost models. Social values were quantified at a coarse scale as total economic value based on a national survey of willingness-to-pay for four key ecological assets; and at a local scale using mapped ecosystem service values. This information was integrated using a constrained, nonlinear, mixed-integer, compromise programming optimization model and solved using a stochastic Tabu search algorithm. We tested the model uncertainty and sensitivity using 390 Monte Carlo model runs at varying weights of ecological health vs. social values. Integrating ecohydrological and socioeconomic information identified environmental flow management regimes that efficiently achieved both ecological and social objectives. Using an ecologically weighted efficient and socially weighted efficient scenario, we illustrated model outputs including a suite of cost-effective infrastructure investments and an operational plan for new and existing flow control structures including dam releases, weir height manipulation, and regulator operation on a monthly time step. Both the investments and management regimes differed substantially between the two scenarios, suggesting that the choice of weightings on ecological and social objectives is important. This demonstrates the benefit of integrating high-quality and high-resolution spatiotemporal ecohydrological and socioeconomic information for guiding the investment in and operational management of environmental flows.

Combining airborne lidar and Landsat ETM+ data with photoclinometry to produce a digital elevation model for Langjökull, Iceland

An incomplete airborne lidar survey of Langjökull, Iceland's second largest ice cap (˜900 km2) and the surrounding area was undertaken in August 2007. Elevation data were interpolated between the lidar swaths using the technique of photoclinometry (PC), which relates Sun-parallel slope angles to image brightness. A Landsat Enhanced Thematic Mapper Plus (ETM+) image from March 2002 was used for this purpose. Different bands and band combinations were assessed and Band 4 (760–900 nm) was found to be the most appropriate. Parameters in the slope–brightness equation were derived empirically by comparing the image brightness with lidar elevation data in a 4 km × 4 km region in the centre of the ice cap. This relationship was then used to calculate the slopes, and, by integration between tie points of known lidar elevation, the elevations of the 30 m pixels that were not surveyed by lidar. The root-mean-square (RMS) precision (repeatability) of lidar elevations was 0.18 m and the accuracy was estimated to be 0.25 m. The 68.3% quantile of absolute difference relative to lidar (analogous to root-mean-square error (RMSE)) of all interpolated areas where PC assumptions are met was 5.44 m (4.66 m and 8.73 m for on- and off-ice areas, respectively). Where one or more PC assumptions were not met (e.g. self-shading, sensor saturation), the 68.3% quantile of absolute difference relative to lidar was 27.89 m (18.52 m on the ice cap and 32.91 m off-ice). These accuracies were applicable to 63%, 31%, and 6% of the ice cap and 59%, 28%, and 13% of the final digital elevation model (DEM), respectively. The area-weighted average 68.3% quantiles were 2.89 m for the ice cap and 6.75 m for the entire DEM. The PC technique applied to satellite imagery is a useful and appropriate method for interpolating a lidar survey of an ice cap.

DTM generation with Radarsat-2 Data without GCP

Abstract. Digital terain models (DTMs) were extracted from high-resolution Radarsat-2 stereo high-resolution images. The metadata supplied by MacDonald, Dettwiler and Associates Ltd. were used to replace the ground control data used to compute for the deterministic geometric model. The DTMs are then evaluated with 0.2-m accurate lidar elevation data. Because DTMs included the height of land covers, elevation linear errors with 68 and 90 percent confidence level (LE68 and LE90) were both computed over the full study site and the bare surfaces. Comparisons were also performed with the same rigorous model using accurate differential GPS as ground control points (GCPs). Using the radargrammetric model using eight dGPS GCPs achieved the best results (3.3 m LE68) than using only the metadata (3.9 m LE68). However, this accuracy is compensated by the fact that the user does not have to collect any ground data, which offers a strong advantage in remote and harsh environments.

Remote detection of small wetlands in the Atlantic coastal plain of North America: Local relief models, ground validation, and high-throughput computing

Isolated wetlands are ecologically important freshwater ecosystems that occur frequently throughout the Atlantic coastal plain ecoregions of North America. Known to support 86 species recognized by the US Fish and Wildlife Service as threatened or endangered, isolated wetlands are a conservation priority in the United States and elsewhere. They are often obscure and methods to detect them at the spatial scales necessary for systematic conservation planning and forest management have been time consuming, cost ineffective, or too coarse-filter. To fill existing information gaps and develop a repeatable, high-resolution methodology, we subjected LiDAR elevation data to custom relief models designed to elucidate fine-scale geomorphology, specifically small, localized changes in concavity, as a location predictor. Because fine grain size and large spatial extent can impose processing limits in landscape-level analysis, we executed our workflow in a high-throughput computing (HTC) environment, which achieved a 91× time-savings over our 55,000ha study area. We conducted field validation at 114 randomly selected sites to measure model commission (14.9%), approximate omission (5.3%) error rates and estimate wetland boundaries. Depressional wetlands predicted in this study (n=4610) were mostly small (x¯=0.37±0.69ha) and previously unmapped sites. The mapping accuracy of this effort (85.1%) suggests that local relief models captured slight geomorphologic changes that successfully predict wetland boundaries in low-relief ecosystems. Many small wetlands are centers of biodiversity in forested landscapes and such analyses will provide information and improved methods for landscape-scale management and conservation.

3D PHOTOGRAMMETRIC PROCESSING OF WORLDVIEW-2 DATA WITHOUT GCP

Abstract. A hybrid model based on 3D deterministic Toutin's model developed at the Canada Centre for Remote Sensing and taking full advantages of the image metadata was used for the 3D photogrammetric processing of WorldView-1 stereo images without in situ ground control points (GCP) collection. The accuracy of the modelling results in monoscopy and stereoscopy was around 0.7-1.8 m in planimetry and 2.7 m in altimetry computed over 40 dGPs checked points. Elevations were thus extracted and compared to 0.2 m accurate lidar elevation data. Elevations linear error with 68 percent confidence level (LE68) computed over bare surfaces was 2.6 m, with small biases. These results were compared with the solution using 8 dGPS GCPs and demonstrated there is no significant difference in auccuracy. Consequently, the new hybrid model will offer a strong advantage of no-control collection in operational conditions, mainly in remote and harsh environments or when cartographic or ground control data do not exist.

Impact of no GCP on elevation extraction from WorldView stereo data

A hybrid model based on 3D deterministic Toutin’s model developed at the Canada Centre for Remote Sensing and taking full advantages of the image metadata was used to geometrically process WorldView-1 and -2 stereo images without in situ ground control points (GCPs) collection. Elevations were thus extracted and compared to 0.2-m accurate lidar elevation data to see the impact of no-GCP. Elevations linear errors with 68% confidence level (LE68) computed over bare surfaces were 2.6m and 2.1m for WorldView-1 and -2, respectively with small biases. The not-significant difference in LE68 (around 10–20cm), when compared to the solution using accurate GCPs, will offer strong advantages of no-control collection in operational conditions, mainly in remote and harsh environments or when cartographic or ground control data do not exist.

ON GENERATING DIGITAL ELEVATION MODELS FROM LIDAR DATA – RESOLUTION VERSUS ACCURACY AND TOPOGRAPHIC WETNESS INDEX INDICES IN NORTHERN PEATLANDS

Global change and GHG emission modelling are dependent on accurate wetness estimations for predictions of e.g. methane emissions. This study aims to quantify how the slope, drainage area and the TWI vary with the resolution of DEMs for a flat peatland area. Six DEMs with spatial resolutions from 0.5 to 90 m were interpolated with four different search radiuses. The relationship between accuracy of the DEM and the slope was tested. The LiDAR elevation data was divided into two data sets. The number of data points facilitated an evaluation dataset with data points not more than 10 mm away from the cell centre points in the interpolation dataset. The DEM was evaluated using a quantile-quantile test and the normalized median absolute deviation. It showed independence of the resolution when using the same search radius. The accuracy of the estimated elevation for different slopes was tested using the 0.5 meter DEM and it showed a higher deviation from evaluation data for steep areas. The slope estimations between resolutions showed differences with values that exceeded 50%. Drainage areas were tested for three resolutions, with coinciding evaluation points. The model ability to generate drainage area at each resolution was tested by pair wise comparison of three data subsets and showed differences of more than 50% in 25% of the evaluated points. The results show that consideration of DEM resolution is a necessity for the use of slope, drainage area and TWI data in large scale modelling.

Radarsat-2 DSM Generation With New Hybrid, Deterministic, and Empirical Geometric Modeling Without GCP

Digital surface models (DSMs) extracted from high-resolution Radarsat-2 stereo-images using different geometric modeling (deterministic, new hybrid, and empirical) are evaluated. The 3-D deterministic models are Toutin's and hybrid Toutin's models (TM and HTM) developed at the Canada Centre for Remote Sensing, and the empirical model is the rational function model (RFM). TM is computed with one and eight ground control points (GCPs), HTM without GCP and RFM supplied by MacDonald, Dettwiler and Associates Ltd. is postprocessed with 3-9 GCPs depending of degrees of 2-D polynomial functions. The DSMs are then generated and compared to 0.2-m accurate lidar elevation data. Because DSMs included the height of land covers, elevation linear errors with 68% and 90% confidence level (LE68 and LE90) are computed and compared over bare surfaces only. LE90 results are: TM with eight GCPs achieves the best results (6.3 m), then HTM with no GCP (7 m), TM with one GCP (8.6 m), and finally RFM the worst (9.7 m) whatever the polynomial degree and GCP number. HTM is the only modeling not using any GCP, which offers a strong advantage in operational environments.

Comparing Habitat Models Using Ground-Based and Remote Sensing Data: Saltmarsh Sparrow Presence Versus Nesting

Remote sensing data can represent various habitat characteristics, and thus can substitute for detailed ground sampling when constructing habitat models. To predict saltmarsh sparrow (Ammodramus caudacutus) distribution and nesting activity, we compared Bayesian hierarchical models in which variables were generated from field or remote sensing data, at a scale of 1-ha plots and at the landscape scale. Field data consisted of plant structure and plant composition variables. Data derived from remote sensing included high and low marsh classifications, LiDAR elevation data, and a classification derived from spectral characteristics specifically associated with saltmarsh sparrow habitat use. The best sparrow presence model used a variable derived from spectral reflectance values associated with plots where sparrows did not occur, indicating that the remote sensing data included additional information about conditions associated with saltmarsh sparrow occurrence than was provided by plant composition, structure, or community classes. In contrast, nest presence was modeled best using vegetation structure variables that required data collection on the ground, although the best remote-sensing model was almost as good. These results reinforce the value of remote-sensing data in habitat modeling, and highlight the need to distinguish between sites that contribute to reproduction and sites where a species is merely present.