On the potential of GPS tracking of fjord ice features for remote-sensing validation

Negrel, Jean; Gerland, Sebastian; Doulgeris, Anthony Paul; Lauknes, Tom Rune; Rouyet, Line


Event: Svalbard Science Conference (Oslo)

Date: 6. November 2017 –8. November 2017

Organiser: Research council of Norway

Climate change in the Arctic induces strong modification in precipitation and freezing regimes. In this context the monitoring of young thin sea ice is a key question. But the thinnest ice is also the most challenging to investigate, both in the field, because scientists cannot always step on it to sample it, and from satellites, as the low signal-to-noise ratio makes it difficult to distinguish from open water. Beyond that, thin sea ice is not a very stable feature: within a relatively short time it either grows and becomes thicker (first-year ice), or it gets broken up or melts. To monitor the changes of Arctic sea ice on various time scales, satellite remote sensing techniques remain preferable versus in situ work for reasons of cost, and spatial and temporal sampling. However, fieldwork remains crucial for the mandatory validation and calibration of these data. In April 2016 an Arctic fieldwork campaign was conducted at Kongsfjorden, Svalbard. This campaign was carried out in the framework of the Norwegian Polar Institute’s long-term monitoring of Svalbard fjord ice, in collaboration with the Center for Integrated Remote Sensing and Forecasting for Arctic operations (CIRFA).The field work has been an opportunity to combine various techniques to record properties of sea ice in an Arctic fjord ranging from local field measurements and sampling (ice coring and thickness drilling) to the broader ground radar (Ku-band radar sweeping the fjord every 2 min) and satellite radar remote sensing (acquisition of four Radarsat-2 high-resolution quad-pol scenes) of the fjord. The combination of the various techniques offers an opportunity to cross-validate all the data collected and investigate the capacities and limits of each. While visiting field sites in the fjords for sea ice studies, a hand-held GPS was set up for tracking the entire track of the small boat used. Beyond the support of the documentation of the fieldwork and operational support, the constant tracking offered a good opportunity to assess the limits of radar identification of small icebergs and thin ice. A total of 17 icebergs and the tracking of two thin ice edges were achieved over the week of fieldwork. The comparison of ice edge and transition mapping with the classification of the Radarsat-2 scenes resulted for different cases in both agreement and contrasting results. Here, we will discuss these results, along with a view on the next steps towards further development of such studies.