Monday, December 9, 2013

LIDAR AND IFSAR for Flood Plain Mapping

Building accurate maps of flood planes has real impact on individuals and insurance costs. Is FEMA doing a good job. How frequently do these maps need to be updated and how accurate is good enough? Could use of UAV's lower the cost enough to make it more feasible? Is multi-sensor fusion of LIDAR and foliage penetrating IFSAR needed? The Scientific American article says FEMA knew the NYC and NJ maps were inaccurate, develops the impact, but does not quantify accuracy required.

Federal Flood Maps Left New York City Unprepared for Sandy, and FEMA Knew It: Scientific American

According to documents and interviews, state, local and federal officials had been aware for years that the crucial maps of flood risks were inaccurate; some feared they understated the dangers in New York City's low-lying areas.

The flaws in the maps had significant impact. Developers relied on FEMA's assessment of risks when they built new homes near the water. And homeowners and businesses made crucial decisions about where to buy or lease property on the assurance that they were outside of the high-risk zones.
Thousands of the buildings incorrectly identified as outside the flood zone were damaged when seawater surged ashore as Hurricane Sandy made landfall on Oct. 29, 2012.

State and city officials had been asking FEMA for years to revise the maps with technology and modeling methods that didn't exist when they were first drawn in the 1980s. William S. Nechamen, New York State's floodplain chief, warned FEMA in a 2005 letter that the failure to do so "will lead to higher than necessary flood damages and more expenses placed on individuals and on FEMA."
The Intermap IFSAR survey appears to show that the California coast and the eastern half of the country, particularly the Atlantic coast and gulf coast are poorly covered (excepting South Carolina and Georgia). 

INTERMAP_FEMA_Risk_MAP_Suitability_English.pdf

Intermap Technologies®’ interferometric synthetic aperture radar (IFSAR)-generated NEXTMap® digital elevation dataset satisfies FEMA’s Risk MAP requirements for the majority of the United States. In accord with FEMA’s standards for LiDAR and other high-quality digital topography datasets, the IFSAR-generated NEXTMap data meets the decile requirements for accuracy and density for the low and medium decile watersheds.

Our IFSAR-generated NEXTMap digital terrain model (DTM) is hydro-enforced to ensure structures over water bodies (such as bridges) are removed, water surfaces are flat, and watercourses flow downstream. It is an ideal dataset for water resource and floodplain management applications for:
  • Entire watersheds, counties, or states
  • Areas surrounding critical floodplains
  • Isolated areas that largely have low flood risk, but may have population clusters with significant risks
  • Other areas in critical need of map updates where funding is limited
DroneMetrex conducts UAV mapping operation - News - Shephard

According to the company, the client had a requirement for an aerial mapping solution to monitor existing waste mounds under a licence from the Environment Protection Authority (EPA). The client also required base maps to be created to inform future planned expansion of the waste site. DroneMetrex supplied its photogrammetric mapping TopoDrone-100 UAV system to capture 2.5cm GSD aerial imagery with 25mm absolute accuracy of the 4sq.km site.
DroneMetrex has provided aerial mapping services with its TopoDrone-100 UAV system for a new recycling and waste management client in Australia.

Elevation Data for Floodplain Mapping


Floodplain mapping involves determining amount of flooding (hydrology), the height of flooding (hydraulics), and the land areas impacted by flooding. A floodplain or flood hazard map has two key inputs: (1) imagery and/or cartographic line work to provide land surface reference information (base map) and (2) a digital elevation model representing the earth’s surface or “terrain.” Responding to concerns expressed by the Congress, the National Academy of Sciences established this committee to respond to the following statement of task:
1.      Identify the current mapping technologies being used by the Federal Emergency Management Agency (FEMA) to develop flood hazard maps;
2.      Identify mapping technologies that are currently available; and
3.      Determine if newer technologies are appropriate and would be of additional benefit to floodplain mapping.


Land surface elevation data for flood management studies of individual streams and rivers has traditionally been derived by land surveying, but the very large areal extent of FEMA floodplain mapping, which covers nearly 1 million miles of the nation’s streams and shorelines, means that land surface elevation data for Flood Map Modernization are mostly derived from mapped sources, not from land surveying.
FEMA floodplain mapping standards for detailed study areas call for elevation data of 2-foot equivalent contour accuracy in flat areas and 4-foot equivalent contour accuracy in rolling or hilly areas, but FEMA does not have a defined standard for approximate study areas. The corresponding root mean square errors of these elevation data are 0.61 feet (18.5 centimeters) for flat areas, and 1.22 feet (37.0 centimeters) for rolling or hilly areas.
These standards apply to “bare-earth” elevation, that is, the land surface with buildings and vegetation removed. Accurate elevation data are needed for precise depiction of the shape of the land surface in the floodplain to support hydraulic engineering computation of floodwater elevation. Except for some special cases, FEMA does not generally support the cost of new elevation data collection. Some communities and a few states, most notably North Carolina, have undertaken elevation mapping programs that provide data of the required accuracy to meet floodplain mapping standards.
 

Evaluation of the Utility and Accuracy of LIDAR and IFSAR Derived Digital Elevation Models for Flood Plain Mapping 

This South Carolina study seems to show that neither LIDAR or IFSAR are sufficiently accurate
NCGIA_Final_Report.doc - NCGIA_Final_Report.pdf 

The NCCGIA ARC project tested the efficacy of LIDAR and IFSAR for accurate floodplain mapping. The state of North Carolina desired an unbiased assessment and comparison of LIDAR and IFSAR accuracy. This research illustrated the significant advantage in accuracy provided by the LIDAR system. However, cost differences between the two systems must be considered, and the advantage of LIDAR is reduced. 
The table below summarizes the accuracy and costs that were determined from this project:


System
Accuracy (RMSE)
Cost /km2
LIDAR Aeroscan
0.93 m
$500
IFSAR Star-3i
10.63 m
$35
USGS Level 2 DEM
1.64 m
Not available

The LIDAR accuracy is impressive, but at this time does not meet the FEMA standards for LIDAR-derived DEMs used for floodplain mapping. As this data was acquired during leaf-on conditions, it is
expected that a leaf-off product will produce acceptable results.


P1041 - FUSING LIDAR AND IFSAR DEMS: A SEVEN-STEP METHODOLOGY

FUSING LIDAR AND IFSAR DEMS: A SEVEN-STEP METHODOLOGY

by
James J. Damron
Geographer
U.S. Army Engineer Research and Development Center
Topographic Engineering Center
Topography, Imagery and Geospatial Research Division
Geospatial Applications Branch
Alexandria, VA 22315-3864
 703-428-8168  (work), 703-428-8176 (fax)
jdamron@tec.army.mil
Fusing, the merging of traditional cartographic and photogrammetric-derived Digital Elevation Models (DEM), has numerous problems. Problems encountered with merging traditionally derived DEMs are not present in the emerging technologies used in single pass Interferometric Synthetic Aperture Radar (IFSAR) and LIght Detection And Ranging (LIDAR). IFSAR and LIDAR technologies utilize common INU and GPS devices. This makes the merging of DEM data straightforward using any raster-based GIS software package. Problems are encountered when datum, ellipsoid, and geoid models are not documented and supported within a GIS.

        Esri software ArcInfo, with extension ArcGRID, demonstrates how this seven-step methodology can be used by any raster-based GIS software to fuse or merge IFSAR and LIDAR DEMs. This methodology has been tested to provide assistance to the Federal Emergency Management Agency (FEMA) for floodplain mapping. The main goal is to provide a cost-effective mix of wide area coverage with an IFSAR DEM and urban/floodplain coverage with a LIDAR DEM and fuse the two DEMs into one data set. This ongoing work is presented as an applied approach to GIS users involved in floodplain mapping and can be used for other earth science applications.

No comments: