The phenological characteristics of the vegetation are key elements for understanding vegetation responses in different climate change scenarios, as well as indicators of ongoing processes of increasing aridity. Determination of phenological parameters for different types of vegetation in large areas help evaluate current and future impacts of climate change in ecosystems, specially in those more vulnerable. Moderate resolution remote sensing data, as provided by MODIS, has already been used to extract phenological characteristics from time series data of vegetation indices, most usually by data smoothing and fitting of polynomial models. In this work, we use hidden Markov models (HMMs) to define phenological parameters from MODIS derived NDVI time series data in a semiarid Mediterranean region. Different types of HMMs are applied in selected areas with well-defined vegetation communities, and their potentials for automatic phenological analysis at large scale are discussed.

Date Published: 12 August 2014
PDF: 8 pages
Proc. SPIE 9229, Second International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2014), 92291K (12 August 2014); doi: 10.1117/12.2066318

Published in SPIE Proceedings Vol. 9229:
Second International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2014)
Diofantos G. Hadjimitsis; Kyriacos Themistocleous; Silas Michaelides; Giorgos Papadavid, Editor(s)