27 October 2011 Estimating vegetation phenological trends using MODIS NDVI time series
Author Affiliations +
Proceedings Volume 8181, Earth Resources and Environmental Remote Sensing/GIS Applications II; 81810P (2011); doi: 10.1117/12.898062
Event: SPIE Remote Sensing, 2011, Prague, Czech Republic
Abstract
The method to extract phenological information for different land cover types is presented. Phenological features are two different start dates of growing season, date of maximum growth, end of growing season and two growing season lengths. Also, quality indicators are estimated for some phenological features. The method is based on NDVI-time series extracted from MODIS-images. The errors between extracted dates and in-situ measurements are reasonably small. For example, the residuals of the estimation of the start of Flux Growing Season are on only 2 days for broadleaf forest in one Southern Finland hydrological drainage basin. The method has been tested on Northern Boreal forest zone, where there are freezing temperatures and snow during winter.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Markus Törmä, Mikko Kervinen, Saku Anttila, "Estimating vegetation phenological trends using MODIS NDVI time series", Proc. SPIE 8181, Earth Resources and Environmental Remote Sensing/GIS Applications II, 81810P (27 October 2011); doi: 10.1117/12.898062; https://doi.org/10.1117/12.898062
PROCEEDINGS
14 PAGES


SHARE
KEYWORDS
MODIS

Vegetation

Feature extraction

Clouds

Satellite imaging

Satellites

Image processing

Back to Top