Terrestrial remote sensing-based estimation of mean trace length, trace intensity and block size/shape

The primary objective of this paper is to investigate the potential of terrestrial remote sensing techniques for the estimation of mean trace length, trace intensity and block size/shape. Sampling window mapping is applied, specifically adapted for terrestrial remote sensing data, and compared with field-based scanline measurements. The authors introduce what is referred to as a “topographic” circular sampling window, which, when used in combination with trace count estimators, minimizes the bias potentially introduced by the application of planar sampling windows on remote sensing 3D models. In particular, circular window mapping avoids underestimation of trace intensity and overestimation of mean trace length. An investigation into the quantification of block size/shape distribution through the generation of discrete fracture network (DFN) models highlights some current limitations in remote sensing estimation of both trace intensity and mean trace length. These limitations include both sampling bias and modeling parameters whose combined effects contribute to the overall uncertainty associated with the quantification of fracture network parameters. A number of recommendations, based on a preliminary set of remote sensing derived DFN models, are suggested to optimize remote sensing based quantification of mean trace length, trace intensity and block size/shape. ► Remote sensing-based quantification of trace intensity parameters is evaluated. ► A “topographic” circular sampling window method is introduced. ► Block size/shape distributions are derived from discrete fracture network models. ► The effect of sampling bias and modeling parameters are highlighted. ► Recommendations are provided for the quantification of trace intensity parameters.