Study On Morphology Evolution And Surface Velocity Estimations Of Larsen Ice Shelf Using Multi-source Remote Sensing Data

The Antarctic ice sheet is closely related to some significant topics, such as global climate and eco-environment, as well as the future development of human society. Temperature warming displays strong regional variations, and it is also obvious in the polar region, where temperature warming is amplified. The Antarctic Peninsula has warmed substantially in all seasons, but most dramatically during the winter, at Faraday/Vernadsky station on Graham Land, the mean annual temperature has increased at a rate of 0.56℃ per decade, while temperature in winter have increased at 1.09 ℃ per decade.Ice shelves themselves are located at the intersection of the atmosphere, hydrosphere and the cryosphere-the air-ice-ocean boundary, and are sensitive to changes in any of these media. In addition to being particularly sensitive to changes in climate, ice shelves play an important role in controlling the flow of glaciers into the ocean, which has important implications for sea level changes. In a warming world, an increased understanding of how climate change is affecting Antarctic ice shelves is valuable for assessing vulnerable regions of the Antarctic that may be prone to further instability.In the present study, we focus on Larsen ice shelf, and a complete and long time series of monitoring extent, surface elevation, and estimating surface velocities was compiled. Based on the above results, we analyzed the relation between morphology evolution and global warming, discussed the mechanism for ice discharge, and predicted the stability of Larsen ice shelf in next century. The main contents and conclusions include:1. In the present work, we employed multiple historical data sources to monitor ice shelf extent changes, including the decryption of aerial photographs (Antarctic Single Frame Records) and other satellite images (Landsat images and MODIS images). A complete and long time series of monitoring the retreat of Larsen ice shelf was compiled. The results showed that the northern Larsen ice shelf had no significant changes until the late 1980s, while Larsen A began to major retreat in steps since 1986, and Larsen B followed a similar pattern from the early 1990s. The gradual retreat of the northern Larsen ice shelf has been punctuated by several catastrophic collapses. Larsen A and Larsen B have already diminished by about 14,000 km2 since 1968 under the background of globe warming. Larsen C ice shelf has been thinning, but otherwise did not exhibit signs with respect to obvious retreat. To date, Larsen C ice shelf has not shown evidence of climate driven retreat, as the variations in the ice front follow normal ice-shelf fluctuation patterns over past half century.2. We monitored the surface elevation changes of Larsen A and Larsen B in recent two decades by transforming the T/P and RA-2 data into a common ellipsoid system. Based on the limitations of calving events, the elevation monitoring was divided into four groups as follows:The remnant Larsen A (1992-2001), the front region of Larsen B (1992-2001), the remnant Larsen B (1992-2010) and Larsen C ice shelf (2002-2010). Overall, it exhibited a progressive thinned during 1992-2010 with the context of globe warming. It is worth noting that, there is a one-to-one correspondence between significantly trough of the elevation change curve and calving events. The remnant Larsen A showed progressive and drastically thinned during 1992-2001, it has lowered at a rate of 0.45 m/yr during 1992-2001. The front region of Larsen B has lowered at a rate of 0.19 m/yr during 1992-2001, and the elevation change curve of this region showed no marked fluctuations. The remnant Larsen B has lowered at a rate of 0.07 m/yr during 1992-2010. Beyond our expectation, the surface elevation of the remnant Larsen B tend to be more stabilized according to limited changes in extent. The lowering rate of Larsen C was significantly lower than that of its northern neighbors, which has lowered at a rate of 0.032 m/yr during 2002-2010.3. We estimated the velocities of the Larsen C ice shelf and Larsen B ice remains using co-registration of optically sensed images and correlation module (COSI-Corr) in the Environment for Visualizing Images (ENVI) based on TM images and MODIS images during 1986-2012, from which we concluded that the ice flow directions generally match the peninsulas pattern and the crevasse, ice flows mainly eastward into the Weddell Sea. In line with previous studies, and compared with the northern Larsen ice shelf, our study showed a continuous but moderate increase in speed on large portions of Larsen C. The overall velocity over the whole estimate area increased approximately by 13.7% froml986-1990 to 2006-2012. Larsen B ice shelf was influenced by calving event which happened in 2002, and the temporal changes in speed showed drastically fluctuates, the overall velocity over Larsen B was slightly lower than that of Larsen C during 1986-1990. However, the overall velocity of Larsen B increased around by 32% from 1986-1990 to 2006-2012, which was related to the catastrophic collapse in February-March 2002. At present, the mean velocity of Larsen B was slightly higher than that of Larsen C. In short, it showed a continuous increase in speed on large portions of Larsen ice shelf with the background of globe warming.4. We analyzed the temporal changes of surface air temperature in summer on the base of NCEP/NCAR data over Larsen ice shelf during 1968-2010. The results demonstrated that the Antarctic Peninsula exhibited a widespread warming over past 50 years. In contrast to the strong warming trends observed during the previous half-century, several stations over northern Antarctic Peninsula have shown much greater variability during 2000-2010, and it presented a slight cooling trend. There are several important mechanisms drive ice shelf surface elevation lowing, such as ice shelves surface and basal melt, ice discharge, ice calving and firn densification, etc. Air temperature increased over past 50 years, which led to longer melt seasons and an increase in melt pond extent on the northern Larsen ice shelf. The lowering circumstance of Larsen C ice shelf was dominated by the increased melt-water production/refreezingrather than increased basal melting. However, the key mechanism for surface lowering of Larsen A and Larsen B is ice fronts retreat and disintegration. The breakup of an ice shelf removed the longitudinal back stress acting on the grounded inland ice and led to flow acceleration, dynamic thinning and frontal retreat. Air temperature increases resulted in longer melt seasons and an increase in melt pond extent. Meanwhile, summer melt-water can descend via crevasses to the bed. Therefore, a key mechanism in the final collapse of these ice shelves has been the propagation of surface meltwater-filled crevasses through the entire ice thickness, rendering it into a series of narrow, elongated icebergs.In summary, global warming has important glaciological and rheology impacts, including widespread ice shelf collapse and fronts retreat, surface elevation lowering, and acceleration in speed, that being unprecedented during the Holocene. The-9℃ average surface temperature isotherm, suspected of being the northern limit of ice shelf viability, already crosses the northern edge of Larsen C. We found that Larsen B remain match the transverse crevasses pattern based on higher time-frequency optical images. In addition, Larsen B remain widely distribute melt ponds in summer melt season, thus we predict that Larsen B remain may disappear by surface melt water-driven crevassing in next 100 years. The surface morphology and extent of Larsen C do not suggest any dramatic changes at present. However, several interconnected mechanisms by which Larsen C stability could be compromised, and it emphasizes the uncertainty of Larsen C in terms of its stability and mass balance.