TH4.TO4.1.ppt

Editor's Notes

• #2: This presentation is entitled: ..... My name is ... From German aerospace center My coauthors are ... From and ...
• #3: Ice surface velocity is crucial for estimating the magnitude of ice flux and the mass balance of glaciers and ice sheets. These are fundamental factors to understand the dynamics of the East Antarctica fast glaciers that discharge into the Ross Ice Shelf and to predict climate change relevant parameters, such as sea level variations. In the lower left image you can see the ALOS PALSAR Mosaic of Antarctica: despite L band is good for preserving coherence a large portion of the ice sheet is not covered. C-band: 4-8 GHz (3.75 – 7.5 cm) X-band: 8-12 GHz (2.5 – 3.75 cm)
• #4: Imaging modes: Scansar, Stripmap, Spotlight, High Resolution Spotlight. These characteristics are very well suited to the estimation of ice surface velocity by means of amplitude tracking and DInSAR !!!!
• #5: This image shows the acquisition plan over the Ross Sea Sector The superimposed colors show the acquisition tracks on different glacier basins. Data are acquired moving the satellite to left looking position, in STRIPMAP mode which has a swath width of 30 km and spatial resolution lower than 3 m. The outlet glaciers flow from the East Antarctic Ice Sheet (EAIS) into the Ross Ice Shelf through rocky channels crossing the Transantarctic Mountains (TAM) . In this study we focus on the Nimrod and Byrd glaciers, in the lower part of the Ross Ice Shelf. ------------------- Q: ACQUISITION STATUS? A: missing only MacAyel, Bindschadler, Kamb
• #6: The investigated glaciers drain ice from the East Antarctic Ice Sheet (EAIS) through the Transantarctic Mountains into the Ross Embayment. Nimrod Glacier consist of two main tributaries which merge downwards the Kon-Tiki Nunatak (shown).
• #7: {Photo: view from Mt. Quackenbush towards Lonewolf Nunataks.} Catchment basin ~size of France and Spain together Changes in its flow dynamics would have an impact on the mass balance of East Antarctica and on the stability of the Ross Ice Shelf !!! The behavior of the TAM outlet glaciers, and especially of Byrd, is influenced by : The upstream catchment area The shearing on the rocky walls and on the bed (where lubrication by subglacial water is possible) of the glacier channel The reaction of the floating ice shelf to the ice input (buttressing) The dynamical behavior of these glaciers has been relatively steady over the past 1000 years. For Byrd some changes in the upstream flow have been recently linked to changes in the subglacial water flow. (discharge of subglacial lakes 200 km upstream of Byrd’s Grounding Line [Stearns, L., 2008])
• #8: Ice velocity is measured by means of Speckle Tracking: this consists in computing the ccf for a regular grid of patches on the 2 images and measuring the displacement of the peak of the ccf. We used a patch size of 256² with an overlap of 50%. Input images: Stripmap (SM) geocoded amplitude images (detected) (EEC) Pixel spacing = 1.25 m Spatial resolution ~ 3 m Looks ~ 1.3 Typical processed SM image size: 32000 x 50000 pixel Comment the images instead of last sentence. Left: CCF of slow and smooth area, and the RGB composition ot the master and the slave images. Coherence stems from the speckle pattern. The ccf shows a well defined peak with a small displacement, leading to high coherence and accuracy. Right: CCF of fast and crevassed area, coherence is given by the texture pattern. The ccf shape depends on the texture pattern with a larger and more displaced peak. This leads to lower coherence and accuracy.
• #9: TerraSAR-X Velocity map detail, the arrow size and color indicates the absolute velocity. The red squares show the patch extent and the patch overlap. Speckle tracking (vs. DInSAR): Absolute surface displacement is obtained Unambiguous measurements (no phase unwrapping) Robust to lower of coherence Lower accuracy.
• #11: Longitudinal profile was selected along the central line of the glacier and parallel to the flow direction. STRAIN RATE: dertivative of the STRAIN E=(l-l0)/l0. Gives an idea of the degree of compression/extension the ice is subject to. A positive value means the ice is expanding, while a negative value means that the ice is compressing. TENSILE STRENGTH: maximum stress that a material can withstand while being stretched before necking (-> crevasses onset). For glaciers: in an area of increasing expansion the tensile strength is reached when the ice „breaks“, i.e. starts to be crevassed (here is where we compute the TS), afther breaking the strain rate normally decreases since the material expands abruptly. FOR QUESTIONS: These are standard parameters to asses the state of a glacier (eg find GL) They help the interpretation and the combination of different measurements.
• #12: Constant velocity in the upper part (0.3 m/d) of the Nimrod and Marsh glaciers Approaching the Kon-Tiki Nunatak the flow becomes abruptly constrained because of the narrowing of the glacier width and sustains a rapid velocity increase , with the appearance of strong crevassing!! Maximum speeds of 2.4 m/d are reached downstream of the Nunatak where the 2 flows merge Ice velocity decreases abruptly as the fjord gets wider and possibly deeper It then stabilizes to a constant value of around 0.6 m/d crossing the grounding line and entering the Ross ice Shelf, which offers resistance (butressing) to the flow, avoiding further acceleration (this is in contrast to David Glacier which velocity increase after the grounding line into the floating Drygalski Ice Tongue) The transverse profile, extracted just above the GL, shows a plug like behavior, with abrupt velocity variation on the side bands and an almost constant value in the central part: this suggests that flow is mostly constrained by side drag. Technical info: BASELINE INTERVAL (Height of Ambiguity): -533.9 (-24.4) – 456.9 (17.78) m
• #13: The strain rate graph shows an equilibrium status in the upper slower part of Nimrod glacier, then increases (0.02 – 0.05 a -1 ) as the flow gets faster. In point (2) major crevassing appears, here we computed a value of 0.03 a -1 with a tensile strength of 487 – 562 kPa. In the fastest sector of Nimrod, the tensile strength follows an irregular behavior with a peak of 0.1 a -1 followed by a strong compression peak (-0.15) as the glacier slows down abruptly. It then maintains an equilibrated behavior as the speed remains constant. PROCESSING DETAILS: Velocity image (45m pixel spacing) was filtered with: Median filter, window size = 7x7 Low pass filter, window size = 3x3 Velocity profile (distance between values ~63m) was FILTERED with: Median filter, window size: 15 Smoothing filter, window size: 21
• #14: The velocity field and the longitudinal profile show a rather different pattern from Nimrod Gl. Ice velocity increases steadly in response to increased surface slope: the increase is initially moderate and then more consistent approaching the fjord. The maximum speed of 2.5 m/d is reached just after the GL. Ice velocity decreases slowly as it flows through the lower part of the fjord and into the Ross Ice Shelf, here flow from Byrd remains well distinguishable for its higher velocity. There are no abrupt variations of speed as in Nimrod. Two transverse profiles were extracted: The upper T1 shows a parabolic shape indicating strong basal drag In proximity of the GL the T2 profile shows a plug-like behavior indicating a prevalence of side drag over basal drag due to a well-lubricated bed. Technical info: BASELINE INTERVAL (Height of Ambiguity): -393.5 (-23.0) – +154.0 (52.9) m
• #15: Byrd glacier shows a much smoother longitududinal velocity profile than Nimrod, the Strain Rate graph is also much more regular. It shows an expansive behaviour till the velocity peak, with values oscillating between 0.002 and 0.15 After the GL the strain rate graph moves into the compression zone as the decelerates smoothly. Tensile strenghts have been computed for 3 points of crecasses onset, shown on the map. [read values 2,3]