The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
180 
Hence, besides range accuracy, the lidar system user should 
also account for possible data voids because of a limited 
dynamic range of intensities of a particular lidar system. The 
lidar user may have to check the data dropout rate during data 
collection to avoid unacceptable data voids. If the lidar 
system’s dynamic range is insufficient to handle strong signal 
variations resulting from the highly variable reflective 
properties of the targets, additional passes may be required to 
cover areas with problematic targets. 
3.2 Laser Footprint Size and Data Accuracy 
Laser footprint size is an additional factor that should be clearly 
understood regarding lidar accuracy. One cannot directly 
specify the accuracy of a lidar system without taking into 
account the finite size of the laser footprint. Under actual 
survey conditions, there is always uncertainty in where a laser 
spot of finite size hits the target. For example, in the case of a 
building edge, uncertainty in the horizontal position is 
determined by the laser footprint size, which is typically about 
25 cm for a 1-km flying height. This factor only brings the 
horizontal accuracy of the building edge down to the 1/4000 
level regardless of other lidar subsystem performance 
characteristics such as the scanner, rangefinder, and GPS/INS 
system. 
Mathematical modelling (Ussyshkin, 2007b) of the vertical and 
horizontal positional errors due to the finite size of the laser 
footprint on the ground shows that this consideration becomes 
critical for the accuracy of targets producing partial signal 
return, tilted targets, or sloped terrain. Figure 5 shows some 
results of this modelling where the elevation error Az and cross 
track component Ay of the horizontal error are calculated as a 
function of the scan angle and the slope of the terrain. These 
results show that to maintain reasonable data accuracy for data 
collected over sloped and highly non-uniform terrains, the user 
should reduce scan angles and flying height and plan a project 
accordingly. 
Modeled elevation error az and cross-track component Ay of the horizontal 
error due to the laser footprint size on the ground 
4.50 
4.00 
3.50 
3.00 
2.50 
2.00 
1.50 
1.00 
0.50 
0.00 
slope=0; 1-kmAGL 
sk>pe=30,1-kmAGL 
stope=40,1-kmAGL 
- - • slope =0; 3-km AGL 
- - slope=30; 3-km AGL 
slope =40; 3-km AGL 
o slope =0; 1-km AGL 
o stope=30,1-km AGL 
slope =40,1-km AGL 
• stope=0; 3-km AGL 
• sk>pe=30; 3-km AGL 
siope=40, 3-kmAGL 
10 15 20 
Scan angle (deg) 
Figure 5. Modelled vertical and horizontal positional errors 
caused by the finite size of the laser footprint on the ground, if 
the beam divergence is 0.3 mrad (full angle). Solid and dashed 
lines represent elevation error Az; solid and empty circles 
represent Ay (cross-track) component of the horizontal error. 
3.3 FOV and Data Accuracy 
The analysis of the impact of the laser footprint size on data 
accuracy presented above shows that both vertical and 
horizontal accuracy strongly depend on the scan FOV (or scan 
angle). In fact, as the scanner FOV (or maximum scan angle) 
widens, the expected deterioration of data accuracy around the 
scan edges generally becomes more significant. Additionally, 
errors in the position and orientation system (POS) data also 
contribute to the deterioration of data accuracy at large scan 
angles (Ussyshkin et al., 2008a). 
As the lidar user might be aware, some manufacturers explicitly 
specify the lidar’s scan FOV while others may not. In either 
case, a question about the accuracy numbers given on the 
specification sheet remains open: Does the specification sheet 
indicate scan nadir accuracy (the best), scan edge accuracy (the 
worst), or something in between? Over-emphasizing accuracy 
numbers without indicating the scan angle can lead to a 
misinterpretation of system capabilities and wrong expectations 
on data quality for the entire project area. The horizontal 
accuracy analysis presented below helps to quantify the impact 
of scan FOV on expected data accuracy. 
3.4 GPS/INS System and Data Accuracy 
GPS/INS data quality is often considered a limiting factor in 
achieving the best accuracy of lidar data (Ussyshkin et al., 
2006b). To quantify the impact of GPS and INS data quality on 
the accuracy of a lidar system, we have launched a study on the 
attainable horizontal accuracy of an airborne lidar system. The 
results of this study have been presented recently (Ussyshkin et 
al., 2008a; Ussyshkin et al., 2008b). Figure 6represents some 
results of that study, which were based on theoretical modelling 
of the best achievable horizontal accuracy. While error due to 
laser footprint size was not taken into account in the study, the 
rangefinder and scanner error were limited to 5 cm and 0.001° 
respectively, and GPS/INS errors were modelled based on 
Applanix’s performance specifications for POS AV-510 and 
610 models (Mostafa et al., 2001). 
Horizontal accuracy modeling 
Figure 6. Theoretically achievable horizontal accuracy in an 
airborne lidar system equipped either with POS AV-510 or 
POS AV-610 (or equivalents). All solid lines represent 
modelling results with zero GPS error; dotted lines represent 
results with 5-cm GPS error. 
The results of the theoretical analysis of positional errors 
partially represented in Figure 5 and Figure 6 show that the 
combined impact of laser footprint size and GPS/INS system on 
lidar data accuracy may make data collected at very high 
altitudes and very wide scan angles not usable for most 
practical applications. Further details of this study, including 
comparison of theory versus practice, are found in our previous 
publications (Ussyshkin et al., 2008a; Ussyshkin et al., 2008b).