LiDAR Surveying: A Tool for Flood Risk Mapping and Engineering

Jim Hartman, PE

Advancements in multi-func­tionality mapping that uti­lize emerging tech­nol­ogies, such as remote sens­ing systems, are now providing other “opportunities” to better manage watershed resources. In fact, these new technologies can now be used as important tools for addressing policies such as balancing the natural environment and growth expectations. In addition to providing current land use planning and natural hazards management information, such as the delineation of floodplains, unstable slope and erosion-prone areas, these technologies are being used to “monitor” the natural environment and potential hazards that could impact social and economic well-being.
The technology of airborne LiDAR (sometimes called laser scanning) is being used increasingly for a variety of engineering studies where detailed topography of the ground is required. While many studies have used LiDAR data (including a recent flood-risk mapping pilot project by the Ganaraska Region Conservation Authority in Ontario involving Greenland International Consulting), very few studies have actually shown a comparison of LiDAR results with “real world” ground elevations. Indeed, several articles that have been published in the USA and abroad have shown a propensity to obfuscate the results with complicated error budget formulae and qualifications.
This article reports on a definitive comparison between a LiDAR survey and a quality control ground survey conducted “independently.” The study was a result of long-term mapping deficiencies for the Community of Angus (Township of Essa), about 15 kilometers southwest of Barrie, Ontario, Canada.
In the Community of Angus, about three-fifths (1,695 sq. km) of surface water drainage converges from the Nottawasaga Valley Watershed. A decrease in gradient results in low river flow velocities and a thick, stable ice cover in winter months. A stable ice cover, in conjunction with some form of obstruction (either natural—such as a river bend or confluence, or man-made—such as a bridge), will frequently result in the formation of ice jams during spring break-up. Such is the case in Angus where the Nottawasaga River meanders through the community and converges with the Pine and Mad Rivers from the west and Bear Creek from the east.
Since the 1980’s, floodplain mapping and flood protection studies of the Nottawasaga River and Pine River systems have identified that a number of buildings and structures in the community would be affected not only by rainfall and snowmelt induced flooding, but also by ice jam factors during spring freshet periods.
Unfortunately, floodplain management for the community was hampered by what appeared to be potentially inaccurate topographic mapping that had been prepared in 1989 as part of a Floodline Mapping Study about Angus.
Indeed, a few years after the production of the Flood Damage Reduction Program (FDRP) 1:2,000 topographic mapping, several indications of mapping discrepancies were noted. Two developers re-surveyed areas within the Nottawasaga Valley Conservation Authority (NVCA) floodplain and demonstrated that the FDRP mapping contours were uniformly 0.5 meters high.
The NVCA adopted an interim policy to account for these discrepancies and to regulate development. The interim policy utilized the floodlines as des­cribed on the FDRP mapping; that is, plotted as a distance from the riverbank, not as an elevation. Furthermore, lot lines were regulated to remain outside of the FDRP plotted floodline, and all residences within any new development were to be floodproofed to the Regional Storm elevation as indicated on the FDRP mapping.
A survey was ordered to verify if there were discrepancies within the existing 1:2,000 topographic floodplain mapping through the Community of Angus. NVCA and the Township of Essa retained Greenland International Consulting, a professional consulting engineering firm, to undertake a comprehensive topographic field survey, as part of an on-going Flood Remediation Study. Appropriate locations on the existing floodplain mapping were identified in order to survey river cross-sections and spot elevations. Geodetically referenced sewer plan and profile information was obtained from the Township of Essa and used as a reference for cross-section and spot elevation information. The results of the leveling survey completed in November 2001 indicated that discrepancies did exist throughout the mapping.
As a result, the December 2001 Angus Flood Remediation Study, Phase 1 Background Review Final Report by Greenland recommended that new topographic mapping should be prepared for the study area. In March 2002, the NVCA and Township of Essa authorized the preparation of new topographic mapping by Lasermap Image Plus/GPR using LiDAR technology and new imagery. This mapping approach was proposed by Greenland.
The mapping process was completed in the summer of 2002 using the latest generation of LiDAR equipment and color photography. The color photography was rectified and delivered as a digital orthophoto map.

LiDAR Mapping Verification Analysis Methodology
Verification of the LiDAR topographic mapping received from Lasermap Image Plus was completed according to the 1984 Ontario Ministry of Natural Resources (OMNR) Map Interpretation Requirements. To fulfill these requirements, the engineering consultant is responsible for field checking map sheets. At a minimum the consultant must:
-- Field check at least 10% of the maps for a particular project; however, at least one map sheet must be selected for detailed inspection.
-- For each map, select ten spot elevations and ten identifiable elevation contour crossings (with roads, railways, etc.) to be inspected. Using existing bench marks as datum, compare the map elevation with field elevations for the points selected. The map meets the required accuracy standards if 90% of the spot elevations checked are within 33% of the contour interval, and if 90% of the contour crossings are within 50% of the contour interval of the map.
-- Select three well-defined, identifiable and accessible features from the map. These three points should be at least 20 cm apart at the scale of the map (1:2,000). Using monu­ment­­ed survey stations, establish the true position of the selected points by field survey method. The map sheets meet the re­quired accuracy standards if the orthophoto map points are within a 1.0 mm radius of their true position.
The field verification survey was completed by Greenland in the fall of 2002. The Angus Flood Remediation Study area is covered by a total of 13 maps. A LiDAR mapping tile (Map 11) with the largest number of easily accessible identifiable spot elevations and contours was chosen for the complete set of OMNR field verification re­quirements (Figure 1). This included contours and spot elevations located on the bank of the Nottawasaga River, as well as in wooded areas. Both of these land features represent areas of potential discrepancy in the existing FDRP mapping for Angus (and in floodline mapping projects generally), making this map a very appropriate choice for verification mapping. However, in order to be conservative, spot and contour elevations were also checked on two other mapping tiles (Maps 7 and 8). Lasermap, incidentally, had not been informed that any field checking was to take place.

LiDAR Mapping Verification Analysis Results
The LiDAR mapping for Community of Angus provided elevation contours at an interval of a half-meter. Therefore, as dictated in the OMNR Requirements for mapping verification, spot elevations on the LiDAR mapping must be within 33% of a half-meter, or 0.17m of the surveyed elevation, and all contours must be within 50% of a half-meter, or 0.25 m of the surveyed elevation. Table 1 presents the results from verification analysis.
As shown in the table, 90% of all the spot and contour elevations are within the required OMNR tolerance for floodline mapping for Map 11 including the spot elevations on the banks of the Nottawasaga River (Spot 102) and in wooded areas (Spot 121). It should be noted that the average difference between all verified field surveyed spot elevations and the LiDAR mapping elevations is 0.07 m and the average difference between all verified surveyed contour locations and the contours produced from the LiDAR is 0.15 m.
It should also be noted that the map sheet contours in this instance were generated from the LiDAR data by using a software package. The positions of contours are, therefore, interpolated by the algorithms in the software and it is to be expected discrepancies may be slightly higher on contours than on specific LiDAR points.

Conclusions
Based on the mapping verification analysis presented in this study, the LiDAR mapping provided by Lasermap Image Plus/GPR meets the OMNR requirements for mapping used in Flood Remediation Projects. In fact, the mapping results exceeded expectations of the study’s consulting engineers, Greenland International Consulting.
In addition to flood risk mapping projects and a wonderful “tool” for presenting accurate data to the Public, Greenland International Consulting also recognizes these other potential benefits from new multi-functional mapping technologies, such as LiDAR systems:
-- Municipal planning
-- Hydrologic modeling
-- Stream geometry and geomorphology studies
-- Natural heritage (forests, wetlands, etc.) assessments
-- Verification of aggregate extractions
-- Environmentalandwatershedhealth monitoring
-- Coastal and shoreline management
-- Infrastructure servicing—including engineering design tasks (for example, water main design)

About the Author
Jim Hartman is a Senior Project Manager with Greenland International Consulting. He would like to acknowledge the Nottawasaga Valley Conservation Authority and Township of Essa for their support in the preparation of this article.

Back