68th Canadian Geotechnical Conference and 7th Canadian Permafrost Conference

QUALITY ASSESSMENT OF PERMAFROST THERMAL STATE AND ACTIVE LAYER THICKNESS DATA IN GTN-P

Session: Characterization of Permafrost State and Variability II / Caractérisation et variabilité du pergélisol II

Boris Biskaborn, Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (Germany)
Jean-Pierre Lanckman, Arctic Portal (Iceland)
Hugues Lantuit, Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (Germany)
Vladimir Romanovsky, University of Alaska Fairbanks (United States)
Dmitry Sergeev, Institute of Environmental Geoscience RAS (Russia)
Gonc¸alo Vieira, CEG/Instituto de Geografia e Ordenamento do Territo´rio, Universidade de Lisboa (Portugal)
William Cable, University of Alaska Fairbanks (United States)
Paolo Pogliotti, Environmental Protection Agency of Valle d'Aosta (Italy)
Jeannette Nötzli, epartment of Geography, University of Zurich (Switzerland)
Hanne Christiansen, Arctic Geology Department, University Centre in Svalbard (Norway)
Halldo´r Jo´hannsson, Arctic Portal (Iceland)

The Global Terrestrial Network for Permafrost (GTN-P, gtnp.org) established the new ‘dynamic’ GTN-P Database (gtnpdatabase.org), which targets the essential climate variable (ECV) permafrost, described by the thermal state of permafrost (TSP) and active layer thickness (ALT). In this paper we outline the requirements for assessing the GTN-P data quality. Our aim is to conceive and discuss useful data quality indices as a basis for the 2nd official GTN-P National Correspondents Meeting in Quebec, September 2015. We describe the TSP and ALT data structures and the importance of precise metadata for the reliability of sound statements on the state and changes of permafrost. We define the most critical parameters related to quality assessment of TSP (borehole depth, number of sensors per depth, recording interval, sensor calibration) and ALT (grid structure, null values and exceeded maximum values, time consistency). We conceive and discuss a set of potential (to be reviewed at the GTN-P meeting) data quality indices by distinguishing between different borehole depths and spatial and temporal data dimensions of TSP and ALT datasets.