The Effects of Peripheral Canopy on DGPS Performance on Forest Roads
Authors
Audrey A. Martin
University College Dublin and Dublin Institute of Technology, Republic of Ireland
Nicholas M. Holden
University College Dublin, Republic of Ireland
Philip M. Owende
University College Dublin, Republic of Ireland
Shane M. Ward
University College Dublin, Republic of Ireland
Abstract
The purpose of this study was to evaluate differential global positioning system (DGPS) positional accuracy on Irish forest roads with typical peripheral canopies. The peripheral canopy obstruction at 20 forest road sites in Roundwood State Forest, was determined using a hand-held clinometer and magnetic compass. This simple field technique permitted quantification of the canopy obstruction using graphical means and resulted in a graphical skyplot of each site. The equipment, one Trimble ProXRS DGPS unit and two Trimble 4000SSi units permitted determination of the DGPS accuracy (average of 2.9 m) and precision (average of 2.1 m) with a range of peripheral canopies. DGPS performance was quantified in terms of the average absolute error in positional dilution of precision (PDOP) (DPDOP = 1.6). The relationship between DPDOP and percentage of open sky was found to be statistically significant (r = 0.706, r = 0.001). Statistical analysis also indicated a strong relationship between relative precision and DPDOP (r = 0.796, r = 0.000). Satellite constellation in the measurement period was not the sole factor affecting DGPS useability. Three distinct classes of peripheral obstruction at road sites were defined (Class I: 100-66 %; Class II: 65-33 %; Class III: 32-0 % obstruction) and it was found that both DGPS accuracy (3.70 m, 3.23 m, 1.91 m, respectively) and precision (4.10 m, 2.43 m, 0.83 m, respectively) improved with decreasing peripheral obstruction. These classes may be used as a means of predicting signal attenuation which might be expected under particular forest canopy conditions elsewhere.
