Introduction: Polygonal networks are a much more common surface feature on Mars than on Earth. Detailed images have shown large areas covered by widely diverse polygonal terrain [1-4], a fact that highlights the ubiquity of the presence of ice in the Martian ground. There is still much to be learned about the conditions that give rise to such a variety of morphologies and dimensions. Thus, a full understanding of their origin, age and evolution requires that their thorough characterization is attained. The number of polygons in a given network can be very large [5-7], depending on their average dimensions and the area occupied by the network. Collecting quantitative data on these networks is a daunting task that can be avoided through the use of an automated procedure [8]. This takes advantage of the characteristics presented by the networks on digital images to map the edges of the polygons. The application of this methodology to the vast northern plains of Mars around the Phoenix landing site [9-10] raised the possibility of using it to map and characterize the same type of feature on Earth. The advantages can be clearly seen: the automated results can be compared with a real groundtruth (and thus provide clues for refinement of the methodology when strictly applied to remotely sensed data, such as in the case of Mars), and some measure of comparison can be established between Martian polygons and terrestrial analogues, illuminating some issues and probably raising many interesting new questions. Three Portuguese research groups (CERENA/IST, CEG/IGOT and CGUC) together with one Norweagian research group (UNIS), organized around a research project, named ANAPOLIS, funded by the Portuguese Science Foundation (FCT), will try in the period 20102012 to give a major contribution to the understanding of polygonal networks on Mars through the combination of remote sensing analysis and in situ detailed characterization of terrestrial analogues. Field sites: Terrestrial polygonal networks will be studied in detail at test sites in Svalbard (Norway) (Figure 1), since field survey is crucial for gathering accurate data on the geometry of the polygons. The common and diversified occurrence of this type of patterned ground, previous basic process studies [1112] and the easy access to that archipelago make it a good choice for terrestrial analogue studies, as other teams testing probes for future planetary missions or working on similar studies on these and other geomorphological features are currently demonstrating [1314].