Influence of the juxtaposition of trees on consumer-grade GPS position quality

Until now, limited observational data have suggested that the juxtaposition of trees with respect to place where a GPS position fix is collected may affect static horizontal position accuracy of that determined position. Our goal was to assess GPS accuracy with respect to the spatial arrangement of nearby trees, and determine whether correlations existed or whether trends were evident. Therefore, static horizontal position accuracy of a consumer-grade GPS receiver was estimated in a young loblolly pine (Pinus taeda) plantation in Georgia (USA) to determine whether the arrangement of trees had any influence on position quality. Thirty visits to twenty-nine test points, randomly ordered, were made to collect the necessary data regarding positional accuracy. No significant relationship was observed between static horizontal positional accuracy and environmental variables (air temperature, relative humidity, and atmospheric pressure) or the planned positional dilution of precision (PDOP) of the NAVSTAR satellite configuration. However, we found moderate correlation between average positional error and a few forest structure measures. For example, we observed that as hardwood (deciduous species) basal area and hardwood tree count within 4 or 5 m of a test point increased, the average positional error tended to increase. No significant correlation was observed using forest structure values obtained within 3 m of each test point. Using rose diagrams (circular histograms), we observed that in some cases there seemed to be a negative attraction between the location of live trees and the position determined by the GPS receiver. Using vectors to represent magnitude and direction of both GPS error and forest conditions, we found evidence to conclude that the average distance and direction to live deciduous (hardwood) trees within this young pine forest may have some influence on position quality.