Dr. Bryan Isacks
Department of Geological Sciences
Snee Hall
Cornell University
Ithaca, NY 14853

Co-Investigator:
Arthur L. Bloom, Cornell University




SIR-C/X-SAR Analysis of Topography and Climate in the Central Andes


OBJECTIVES

Understand large-scale interactions between tectonic and climate-controlled erosional processes that created the Andes.

Determine the modern and Pleistocene snow-line altitudes and gradients in a poorly known but critical latitude range of the central Andes, and interpret ice-age changes in atmospheric circulation.


PROGRESS

Since the launch of the April mission our project has focused primarily on the analysis of the enormous amount of new SIR-C/X-SAR data for the Patagonian Icefields. The North and South Patagonian Icefields represent 62% of the glacial area in the southern hemisphere outside of Antarctica, with the South Patagonian Icefield (SPI) ranking as the third largest ice mass on Earth after Antarctica and Greenland. Although little studied, these glaciers are some of the most dynamic in the world, with annual precipitation exceeding 7 m in the accumulation areas and ablation rates as high as 6 cm/day at sea level, allowing them to respond quickly to climatic variations. Monitoring the response of these glaciers is important for the study of the regional climate as well as the relationships between climate changes in the northern and southern hemispheres. The new radar data will have a major impact on the understanding of these important icefields and glaciers.

Radar interferometry analysis of the San Rafael glacier, one of the fastest moving glaciers in the world, has produced spectacular results revealing new information regarding its flow characteristics. The extremely fast velocities at the terminus are not characteristic throughout its length and slow down by an order of magnitude up glacier. Within the interior of the icefield the glacier acts as a well defined ice stream with slower moving ice on its flanks.

Analysis of the Patagonian SIR-C/X-SAR data has lead us to new discoveries regarding the spatial and temporal characterization of snow and ice conditions on the Patagonian Icefields, and has demonstrated the utility of combining multitemporal SAR with meteorological and hydrological data to detect the effects of synoptic weather systems and seasonal changes on snow and ice conditions.

The Patagonian results have spawned collaborations with the Chilean and Japanese glaciologists who have been responsible for much of the previous research on the icefields. We are planning to conduct a joint field trip with them to measure ice depth profiles and snow and ice conditions on the San Rafael glacier. The ice depth measurements will be made with an ice penetrating radar and are needed in order to use the ice velocity measurements to calculate ice flux and test models of glacier flow.

In April 1995 we received our first four TOPSAR scenes processed through the new AIRSAR integrated processor. These four scenes cover two glacier sites in the Cordillera Real, Bolivia and the Quelccaya Ice Cap, Peru. The focus of our recent research has been to assess the usefulness of this data for glaciological studies, especially as a high resolution data source for detection of glacier change. We are collaborating with researchers at the Unidad de Glaciology Recursos Hedricos de Electroperu to use TOPSAR data to update the glacial inventory of the Cordillera Blanca which was originally done using aerial photographs taken in the 1950s and 60s.


FUTURE PLANS

Spaceborne Imaging Radar (SIR-C/X-SAR) Glaciological Observations of the Patagonian Icefields

A mosaic of the North Patagonian Icefield and the northern half of the South Patagonian Icefield (SPI) is being constructed from the 14 SIR-C scenes acquired during SRL-1. Because of the persistent cloud cover this is the first complete mosaic of the icefields since a series of TM acquisitions in 1986. Terminus position changes of 15 glaciers are determined and the spatial distribution of these changes correlated with glacier characteristics including aspect, area, length, accumulation area/ablation area ratio, equilibrium line altitude (ELA) and gradient. Snowlines are mapped and compared to previous estimates of ELAs. Several ice divides on the SPI are defined for the first time. Other previously undetected glacial features include: buried crevasses, small icefalls, and ogives. The complete set of SRL-2 data provides a comprehensive view of seasonal change for the northern SPI.

Snow and Ice Conditions inferred from Scattering Mechanism Decomposition of SIR-C/X-SAR images of the Patagonian Icefields

The scattering mechanism decomposition technique of Cloude, modified for SIR-C data by van Zyl and Rignot, is used to interpret the snow and ice conditions of the Patagonian Icefields. Several large areas on the icefields are dominated by a double bounce scattering mechanism which has never been reported for a snow and ice target. We presently interpret this to be the result of subsurface refrozen ice on a 5 cm scale. Inverse backscatter modeling together with the limited published field data will be used to constrain the physical parameters of the snow and ice such as: grain size distribution, snow wetness, density, surface roughness and the orientation and structure of subsurface refrozen ice. We hope to further constrain the snow and ice parameters with our own measurements from snow pits on the Patagonian Icefields.

Ice Velocity and Surface Topographic Mapping of the Moreno Glacier area of the South Patagonian Icefield from Spaceborne SAR interferometry

Interferometric analysis similar to that done on the San Rafael Glacier will be applied to the fast-moving Moreno Glacier and several slower glaciers in the area. The glaciers in the southern portion of the SPI act as a coalescence of individual glaciers rather than outlet glaciers draining a common icefield as they do on the NPI, therefore, comparisons of the velocity and strain maps of this area with those of the San Rafael area will illustrate the differences in flow characteristics of these two glacial regimes.

Results from San Rafael Glacier Field Experiments

If we are successful in obtaining ice depth profiles and characterizing the snow and ice conditions on San Rafael Glacier during our planned trip to Chile then we will be in the position to provide important ground validation results for both the San Rafael interferometry results (Rignot, et al., 1995) and the Patagonian radar glacier zones (Forster and Isacks, 1995).

Mapping of Glacial Geomorphology in the Cordillera Real, Bolivia with SIR-C/X-SAR images

Various geomorphic features are easily identifiable on the SIR-C/X-SAR images that are not as evident on VNIR images. The extent of Pleistocene glaciation can be mapped from moraine positions prominently displayed on the SAR images because of the enhancement of subtle topography to the side-looking geometry. Rock glaciers, talus and bedrock outcrops can be distinguished through the variations in surface roughness.

A comparison of DEMs generated from SAR interferometry and stereo pairs in the Cordillera Real, Bolivia

We have a unique opportunity to compare DEMs generated from SIR-C and TOPSAR interferometry, SPOT stereo pairs, and from aerial photograph stereo pairs for an area with extreme relief for which we also have good ground control from 1:50,000 maps. This will serve as a test for the accuracy and coverage of a spaceborne interferometer in the presence of large topographic variations.

Himalayan Radar Glacier Zones observed with SIR-C/X-SAR

We will apply our interpretation of the Patagonian radar glacier zones to the SIR-C/X-SAR data acquired over the Himalayan Glaciers. The climatic setting in the Himalayans is dominated by monsoonal and winter precipitation as opposed to the year-round precipitation in Patagonia. We anticipate this to be reflected in the spatial and temporal distribution of the radar glacier zones. The seasonal changes induced by the 1994 monsoon season (mid-summer) should be recorded by the differences in the April and October radar glacier zones. Analysis of the radar zones will not be limited to relatively flat areas (as they were in Patagonia) because we can correct the SAR data for radiometric and geometric topographic influences with the high resolution DEM available at Cornell.

A multitemporal analysis of the Radar Glacier Zones in Patagonia and British Columbia with ERS-1 and SIR-C/X-SAR

Based on the Patagonian radar glacier zones defined from our SIR-C/X-SAR analysis we are observing those zones as shown by C-VV from a continuing series of ERS-1 and 2 SAR acquisitions of the Patagonian Icefields and British Columbia. These images are correlated with elevation of the snow and ice surfaces and with weather conditions to determine the dynamics of climate/glacier interactions over time scales of synoptic weather to seasonal change.

TOPSAR Data for Bolivia and Southern Peru

Our research has focused primarily on assessing the geometric accuracy of the data to determine if it can be used as part of a glaciological time series. Glacier boundaries are clearly discernible on the SAR imagery and can be easily mapped. Rectification of TOPSAR data to other datasets is more problematic. The glacier recession that has occurred in the region over the past 30-40 years is measured in 10's of meters, so accurate coregistration is required. The small footprint of a TOPSAR scene has hindered coregistration simply because there is often a limited number of good ground control points that can be used to accurately define the transform. In an attempt to improve coregistration we have begun to focus on better visualization of the radar imagery. We have also recently begun to integrate TOPSAR digital elevation data with SAR imagery to create stereo radar imagery. This combination has resulted in dramatically improved visualization of radar data. It offers several advantages over even geocorrected data in mountainous regions. First and foremost it allows radar backscatter signatures to be seen in their true landscape position. This is crucial for proper interpretation of geomorphic features such as moraines or debris left by retreating glaciers as well as accurate mapping of the position of glacier termini. Secondly, it allows much better registration with other datasets as ground control points become much easier to identify. Lastly, It also allows radar imagery to be used by a much wider community. Nonradar scientists can find using SAR data, especially in mountainous regions, a frustrating task as layover and foreshortening render the landscape nearly unrecognizable. However, through the creation of stereopairs, radar data is presented in a form identical to widely used aerial photographs that researchers, even in underdeveloped countries, are familiar working with. We intend to use this technique with the DEM that will be created for the Cordillera, Real region using SLR-2 data as well.

Anticipated new data requests:
SIR-C/X-SAR
Patagonia sites (10)
Bolivia, Southern Peru and other South American sites (15)
Himalayan glaciers (10)

Integrated TOPSAR processing:
We require the entire coverage of the Peruvian Cordillera Blanca and the Bolivian Cordillera Real.


PUBLICATIONS

Forster, Richard R. and Bryan L. Isacks, Spaceborne Imaging Radar Reveals South Patagonian Icefield Responses to Seasonal and Synoptic Weather Changes, submitted to JGR , Planetary Science, 1 October 1995.

Rignot, E. J. M., R. R. Forster and B. L. Isacks, Interferometric radar observations of Glaciar San Rafael, Chile, J. Glaciology , (in press), 1996.

Rignot, E. J. M., R. R. Forster and B.L. Isacks, Mapping of glacial motion and surface topography of Hielo Patagonical Norte, Chile, using satellite SAR L-band interferometry data. Annals of Glaciology , vol. 213, 1996, (in press).


Presentations at Scientific Meetings

Forster, R. R., and B. L. Isacks, The Patagonian icefields revealed by space shuttle synthetic aperture radar (SIR-X/XSAR), EOS , Transactions, American Geophysical Union (Abstracts), v. 75 (44), p. 226, 1994.

Forster, R. R., B. L. Isacks, and A. G. Klein, Modern and LGM glaciation of the Bolivian and Patagonian Andes, Geological Society of America annual meeting (Abstracts) , v. 26 (7), p. 129, 1994.

Forster, R. R., B. L. Isacks, and A. G. Klein, Spaceborne Imaging Radar (SIR-C/X-SAR) Glaciological Observations of the Patagonian Icefields, submitted to AGU Fall 1995.

Table of Contents


Converted to HTML by Alvin Wong, al.wong@jpl.nasa.gov

The Jet Propulsion Laboratory
4800 Oak Grove Drive
Pasadena, Cailfornia 91109