Istanbul 2004 
ler when the 
ints’ center of 
evation in that 
ion (B-1). The 
e for. DIM 
to re-write the 
e coordinates 
ind y: 
(B-3) 
neans, that by 
1ding cofactor 
be computed. 
ly: 
(B-4) 
the plane is 
the arithmetic 
> computed of 
vided by the 
trian Science 
Digital 
Processing, 
Geometrische 
Laserscanner- 
ain models in 
data. ISPRS- 
ichter Laser- 
312-317. 
oer G., 2004. 
(DGMe) am 
Laserscanner- 
senschaftlich- 
Druck. 
IDENTIFICATION AND RISK MODELING OF AIRFIELD OBSTRUCTIONS 
FOR AVIATION SAFETY MANAGEMENT 
Chun Wang, Yong Hu, Vincent Tao 
GeoICT Lab, York University, 4700 Keele Street, Toronto M3J 1P3 - (chunwang, vhu, tao)@yorku.ca 
KEY WORDS: Lidar, Obstruction Identification Surface, Identification, Risk Model. Impact Analysis 
ABSTRACT: 
The air safety is critical to the national security and economy. The aeronautical community has recognized the need for accurate 3-D 
geospatial information in and around the airfield to identify obstructions, specifically for accurate runway positions, obstruction 
locations and heights, and topography around airfields. In this paper, we present an approach to identify the airfield obstructions and 
model the risks by using advanced airborne lidar processing techniques. Airfield objects required for the analysis of the obstruction 
identification surface (OIS) are digitized from the aerial imagery and the topographic map, and their heights are derived using lidar 
data and the lidar-derived DTM. The 3-D OISs are created for runways to identify airfield obstructions based on the latest safety 
specification. À risk modeling approach is developed to classify the obstructions into three risk levels by combining four risk factors 
in a multi-criteria evaluation to assist decision-making in managing the airfield obstructions. The result is a risk-rating map that 
illustrates the high-, median-, and low-risk obstructions in 3-D. The presented approach is of important value to the examine whether 
their airfields meet the new safety specification. 
1. INTRODUCTION 
On April 3", 1996. Secretary of Commerce Ron Brown, 32 
business executives, and US military personnel were killed in a 
plane crash in Dubrovnik, Croatia. This crash led US congress 
passed the Ron Brown Initiative in 2000 to survey glide slope 
obstructions on the approaches to some 7200 airports in the 
United States and an undetermined number of airports, 
worldwide. For a distance of 14km on each end of each runway, 
and for a specified distance around the entire airport complex, 
the geolocation and surveyed heights of all objects (building, 
trees, light poles, antennas and towers, power lines, and other 
physical features) must be recorded (Morain, 2001). 
The aeronautical community has recognized the acute need for 
accurate 3-D geospatial information in and around the airfield 
critical to flight safety, specifically for accurate runway 
positions, obstruction locations and heights, and topography 
around airfields. Knowing the location of an aircraft is only 
half of the solution. To bring an aircraft safely onto the runway 
with little else than satellite navigation, the pilot will need very 
accurate and reliable geodetic “coordinates for the landing 
runway. Currently, pilots are still expected to use conventional 
navigation aids or visual contact to direct the aircraft to the 
runway. 
Airborne lidar is suitable for collecting accurate terrain data 
and providing feature information for airport safety 
management. In addition to identifying obstructions and 
designing approach procedures, pilots will be able to use the 
generated 3-D airfield models for flight training, pre-flight 
flythrough familiarizations, as well as increasing overall 
aircrew situational awareness relating to mission planning. The 
airfield initiative document (AID) is a newly published 
specification for airfield obstruction identification (NIMA, 
2001). It describes the use of 3-D OISs to survey glide slope 
obstructions, and puts new requirements for a safer flying 
environment. “The OIS consists of several surfaces with certain 
dimensions related to a specific runway approach, including 
primary surface (PS), approach surface (AS), primary/approach 
transitional surface (P/ATS), inner horizontal surface (IHA), 
conical surface (CS), outer horizontal surface (OHS), and 
conical/outer horizontal approach transition surface 
(C/OHATS)" (NIMA, 2001). When approaches share the exact 
surface, only one OIS is required. An obstruction is any object 
that penetrates an OIS, except where no obstruction penetrates 
the OIS; it shall be the highest object within the area. The 
obstructions are often extracted from photogrammetric and 
survey data. In addition, to avoid airport incursion, the surface 
of vehicular traverse ways (SVTW) is also needed (NIMA, 
2001). 
"The geospatial data required for obstruction identification 
around an airfield include the airfield elevation model (AEM), 
airfield features, and different combinations of the highest, the 
most penetrating, the highest approach and the highest non 
man-made obstructions/objects tor analyzing each type of OIS 
surface” (NIMA, 2001). The AEM is in one arc second spacing 
each post having an absolute vertical accuracy of «30.0 meters 
with respect to reflective surface. This vertical accuracy is 
required throughout the entire project area except for those 
posts that fall within the primary, primary approach and 
primary/approach transitional surfaces (NIMA, 2001). Airfield 
features, such as runway ends, must be surveved to achieve 
high accuracy, for example, 0.3 m CE90 (circular error) and 
0.07 m LE90 (linear error) for runway points. The required 
accuracy within OIS is lower. For example, within the PS, it is 
6 m CE90 horizontally and ! m LE90 vertically for the highest 
obstruction and the highest non man-made obstruction/object in 
each 912-m section of the primary area on each side of the 
runway. The required accuracies for other surfaces are normally 
lower. 
2. STUDY AREA AND DATA DESCRIPTION 
The study airport, Santa Barbara Airport, was selected by US 
Department of Transportation and Airfield Initiative Remote 
Sensing Technologies Evaluation Project as initiated case study 
followed by several site selection criteria (TRB, 2002): 
119