Estimating signal loss in pine forests using hemispherical sky oriented photos

Abstract

We depend on numerous technologies that use microwave signals. The reception of these signals is degraded by reflection, absorption, and scattering due to propagation through vegetation. An understanding of how these signals are influenced by vegetation structure allows for the determination of how specific technologies may be affected in certain forest environments. This study presents a model that predicts signal loss in forested areas using novel methods. We explore the relationships between forest parameters from traditional mensuration techniques and terrestrial-based hemispherical sky oriented photos (HSOPs), and GPS signal-to-noise ratios (SNRs). HSOPs can be used to rapidly estimate leaf area index (LAI) and canopy closure (CC) values at particular angles from zenith in forested areas. The relationships between changes in the observed SNR of received GPS L1-band signals under forest canopies and forest parameter estimates calculated using HSOPs and traditional forest measurements are used to model signal attenuation. Using ordinary least squares regression modeling, we present the Canopy Closure Predictive Model (CCPM). The CCPM outlines the key forest parameters used with an adjusted R2 of 0.71 and RMSE of 2.78 dB. The resulting CCPM predicts signal attenuation while using only the minimum number of statistically-significant parameters which, conveniently, are taken from sky oriented photos and GPS receivers allowing for simple and rapid replication.