Articles | Volume 11, issue 12
https://doi.org/10.5194/gmd-11-5203-2018
https://doi.org/10.5194/gmd-11-5203-2018
Methods for assessment of models
 | 
21 Dec 2018
Methods for assessment of models |  | 21 Dec 2018

A generic pixel-to-point comparison for simulated large-scale ecosystem properties and ground-based observations: an example from the Amazon region

Anja Rammig, Jens Heinke, Florian Hofhansl, Hans Verbeeck, Timothy R. Baker, Bradley Christoffersen, Philippe Ciais, Hannes De Deurwaerder, Katrin Fleischer, David Galbraith, Matthieu Guimberteau, Andreas Huth, Michelle Johnson, Bart Krujit, Fanny Langerwisch, Patrick Meir, Phillip Papastefanou, Gilvan Sampaio, Kirsten Thonicke, Celso von Randow, Christian Zang, and Edna Rödig

Data sets

Observational data (10.5521/FORESTPLOTS.NET/2014_1) E. T. A. Mitchard et al. https://doi.org/10.1111/geb.12168

Inventory data (10.5521/ForestPlots.net/2014_4) R. J. W. Brienen https://doi.org/10.1038/nature14283

Download
Short summary
We propose a generic approach for a pixel-to-point comparison applicable for evaluation of models and remote-sensing products. We provide statistical measures accounting for the uncertainty in ecosystem variables. We demonstrate our approach by comparing simulated values of aboveground biomass, woody productivity and residence time of woody biomass from four dynamic global vegetation models (DGVMs) with measured inventory data from permanent plots in the Amazon rainforest.