Inventory of standing dead trees in the surroundings of communication routes - The contribution of remote sensing to potential risk assessments

We present a comparison of three approaches to support risk assessment of dead standing trees near communication routes. The three methods consider different possible data available, workflow complexity, results accuracy and practical implementation. The following inventory methods were implemented and evaluated: (1) passive remote sensing (RS), (2) passive remote sensing + GIS (RS + GIS) and (3) active remote sensing (RS + ALS). The study was carried out in the Polish part of Bialowie2a Forest. The roadside analysis performed throughout Biatowieia Forest showed that 1671 sections of routes (72.1% of all those analysed) have standing dead trees in their vicinity. The comparison of the LIDAR based method with the passive remote sensing based methods demonstrated the significant advantage of the former in this kind of research, where precision, accuracy and level of detail of information enable the reliable assessment of risk (even at the single tree). However, this does not preclude the use of the RS methods in this type of analysis; the categorisation of at-risk routes obtained using the latter methods also proved to be satisfactory (with 81% compliance in threat categorisation when compared with the use of the LIDAR based method). Nevertheless, the importance and potentially fatal consequences of RS-based underestimation of potential risk in 16% of cases (episodes) cannot be ignored. At a practical level, our results describe time-and cost-effective methods that bring sufficient information for land managers to handle traffic management in areas with large scale tree mortality. (C) 2017 Elsevier B.V. All rights reserved.