Spatiotemporal analysis of snow cover variations at Mt. Kilimanjaro using multi-temporal Landsat images during 27 years

Sung Hwan Park, Moung Jin Lee, Hyung Sup Jung

Research output: Contribution to journalArticlepeer-review

12 Scopus citations

Abstract

The Landsat TM and ETM+ images have been acquired for the long period from the 1980s until the present with the temporal resolution of a 16-day repeat cycle from the visible, near infrared (NIR), short wave infrared (SWIR) and thermal infrared (TIR) bands. The Landsat multi-temporal images have been successfully used to monitor variations of the Earth surface during 27 years. In this paper, we observe the variations of (1) the snow cover area, (2) the snowline height and (3) the land surface temperature (LST) lapse rate at Mt. Kilimanjaro using a total number of 15 Landsat-5 TM and Landsat-7 ETM+ images from June 1984-July 2011. Segmentation of normalized difference snow index (NDSI) images with a threshold of 0.6 is used to extract snow cover. Snowline altitude is then determined by combining the snow cover classification maps with a digital elevation model (DEM). And the LST lapse rate is also calculated from the TIR band in the forest area. The results from this study show that (1) the snow cover area largely decreases from 10.1 km2 to 2.3 km2 during about 27 years, which corresponds to a 77.2% reduction, (2) the snowline height rose from 4760 m to 5020 m by about 260 m, and (3) the LST lapse rate shifted from -5.2 °C/km to -2.7 °C/km. This study demonstrates that multi-temporal Landsat images can be successfully used for the spatiotemporal analysis of long-term snow cover changes.

Original languageEnglish
Pages (from-to)37-46
Number of pages10
JournalJournal of Atmospheric and Solar-Terrestrial Physics
Volume143-144
DOIs
StatePublished - 1 Jun 2016

Keywords

  • Landsat
  • Snow cover
  • Snowline
  • Spatiotemporal analysis

Fingerprint

Dive into the research topics of 'Spatiotemporal analysis of snow cover variations at Mt. Kilimanjaro using multi-temporal Landsat images during 27 years'. Together they form a unique fingerprint.

Cite this