Regional Forest Monitoring Strategies Building on Multi-Sensor Integration Concepts
Hill, Joachim; Buddenbaum, Henning; Pueschel, Pyare; Stoffels, Johannes; Sachtleber, Thomas; Stern, Oksana; Nink, Sascha; Mader, Sebastian; Mandelkow, Alexander
University of Trier, Environmental Remote Sensing and Geoinformatics, GERMANY
Forest ecosystems cover large parts of the Earth's land surface and are among the most important providers of central ecosystem services. With the largest storages of biomass worldwide and acting as the most important terrestrial carbon dioxide sink, forests fulfill important ecological (e.g. biodiversity, feedback mechanisms), economic (e.g. timber production), and socioeconomic (e.g. recreation) functions. At the same time, forests and forest ecosystems are under accelerating pressure due to regional impacts of global warming and changing socio-economic conditions.
In most European countries operative and sustainable forest management institutions are facing the challenge of coping with the complexity of these issues, while additionally being under the pressure of increasing financial and personnel cuts. Against this background, there is a substantial need for developing forest monitoring strategies that make optimum use of the range of available and future remote sensing instruments.
Primary tasks to be considered include wall-to-wall mapping products of tree species and development stages (age groups), the disaggregation of state-wide timber volume and biomass estimates to the level of smaller forest management units, the assessment of climate-sensitive site conditions for key tree species, the design of integrated early warning systems for local outbreaks of insect infestations as well as the monitoring of changing habitat structures. The specific requirements emerging from these tasks can only be addressed by networking sensors with varying spatial, temporal and spectral coverage and resolution.
EO-derived maps of tree species distribution have already proven their competitiveness with field surveys (fig. 1). Due to the more frequent revisit capabilities and extended spatial coverage a combination of existing (e.g. RapidEye) and forthcoming earth observation systems (e.g. Landsat-8 OLI and Sentinel-2 MSI) will substantially increase the existing capacities to update state-wide mapping products; these are urgently required for closing the emerging gap with respect to terrestrial inventories. While these products are complemented by employing active remote sensing systems for supporting resource assessments, hot spot areas of structural and qualitative changes require the use of hyperspectral satellite systems (such as EnMAP) or airborne sensors (e.g. HySpex). With respect to triggering local interventions (e.g. processing of storm-damaged areas, or protective chopping within beetle attack areas) the process chain might integrate UAV deployments and specific target-oriented fieldwork.
This paper aims at demonstrating how systems operating at different spectral and spatial scales can be integrated to complement the existing but expensive ground-based surveys and thereby provide support for operational forest services. It builds on experiences from a long-term cooperation with the state forest administration of the Rhineland-Palatinate federal state in Germany.