Different Methods of Detection of Glacier Surface in the Central Andes, Argentina

MALDONADO, GABRIELA I.; ESPIZUA, LIDIA; DELGADO, SILVIA

Buenos Aires, Argentina

Conferencia; International Conference “Environmental change and rational water use”.; 2005

Comisión for Water Sustainability of the International Geographical Union

This work deals with the results from comparison of different methods for glacier mapping with remote sensing data, applied to two glaciers in several years in Mendoza Province, Argentina. Detection of glacier changes is an important issue not only for discussions about global climate change, but also because their disappearance might have large economic and social impacts as water is the principal component in the regional economy. Satellite remote sensing is an excellent tool for monitoring glacier fluctuations, because it gives information about areas that are very difficult to access. There are frequent images since 1975 and they offer a less expensive alternative. In this study we used four Landsat 5 satellite images  for the study of two glaciers: De las Vacas Glacier, located north of Mt. Aconcagua,  the highest peak in the Western Hemisphere, (from 1987 and 2004); and El Peñón Glacier in the south of Mendoza Province (from 1990 and 2004). All satellite images were geometrically and atmospherically corrected, because the Landsat scenes were acquired on different dates and under varied atmospheric conditions. Digital methods were used for the glacier fluctuations study and were compared with manually-derived outlines from the same satellite image. Four different ice-snow surface detections were applied: Dozier method, ratio TM4/TM5, ratio TM3/TM5 and Normalized Difference Snow Index (NDSI). The accuracy of each method is presented here. The best result was obtained with the Dozier method, with a 99 percent of accuracy.