Satellite time series analysis for land use/cover change detection

Maus, V. ORCID: https://orcid.org/0000-0002-7385-4723 (2014). Satellite time series analysis for land use/cover change detection. IIASA Interim Report. IIASA, Laxenburg, Austria: IR-14-017

Preview

Text IR-14-017.pdf Download (1MB) | Preview

Abstract

Currently, Brazilian land use data comes from the national agricultural census and land cover data comes from global data sets with sparse temporal coverage. This no longer meets the needs of the earth system modeling community. Long-term satellite image datasets with high temporal frequency yield a sequence of data points in a time series that can be used to detect and monitor land use and land cover changes. The vegetation phenological cycles are reflected in the satellite time series, allowing the classification of land cover types in time segments. This research aims at developing an automatic methodology to yield information about land use and land cover trajectories. To construct land use/cover trajectories maps, Dynamic Time Warping (DTW) is used to extract information from the MODIS 2-band Enhanced Vegetation Index (EVI2) time series. Validation tests were made in the areas of Mato Grosso state, Brazil. The preliminary results for the proposed methods are promising when compared with the official TerraClass land use maps in the Amazon Biome, finding 78.2% and 85.0% global accuracy for 2008 and 2010, respectively. Exploratory DTW results show significant potential to detect land use and cover changes.

Actions (login required)

View Item