Abstract
Several methods exist for extracting plant phenological information from time series of satellite data. However, there have been only a few successful attempts to temporarily match satellite observations (Land Surface Phenology or LSP) with ground based phenological observations (Ground Phenology or GP). The classical pixel to point matching problem along with the temporal and spatial resolution of remote sensing data are some of the many issues encountered. In this study, MODIS-sensor’s Normalised Differenced Vegetation Index (NDVI) time series data were smoothed using two filtering techniques for comparison. Several start of season (SOS) methods established in the literature, namely thresholds of amplitude, derivatives and delayed moving average, were tested for determination of LSP-SOS for broadleaf forests at a site in southwestern Germany using 2001–2013 time series of NDVI data. The different LSP-SOS estimates when compared with species-rich GP dataset revealed that different LSP-SOS extraction methods agree better with specific phases of GP, and the choice of data processing or smoothing strongly affects the LSP-SOS extracted. LSP methods mirroring late SOS dates, i.e., 75% amplitude and 1st derivative, indicated a better match in means and trends, and high, significant correlations of up to 0.7 with leaf unfolding and greening of late understory and broadleaf tree species. GP-SOS of early understory leaf unfolding partly were significantly correlated with earlier detecting LSP-SOS, i.e., 20% amplitude and 3rd derivative. Early understory SOS were, however, more difficult to detect from NDVI due to the lack of a high resolution land cover information.
Highlights
Phenology, the science of periodic events in plant and animal life cycle, has been widely studied and well documented for many decades (e.g., [1,2,3])
In addition the variance of start of season dates and their annual means obtained from different methods was better described using the Gaussian smoothed Normalised Difference Vegetation Index (NDVI) (Figure 3)
This paper aimed at studying the LSP of broadleaf forests and to assess its agreement with a rich set of Ground Phenological Data (GP) observations for specific species
Summary
The science of periodic events in plant and animal life cycle, has been widely studied and well documented for many decades (e.g., [1,2,3]). It has been a core parameter for demonstrating and studying the impact of climate change on terrestrial ecosystems. The advent of modern remote sensing techniques provides a promising alternative and new opportunities for phenological studies [4], a departure from the traditional ground based observations. In comparison to GP, remote sensing techniques provide a global coverage of data at various temporal and spatial scales, which can support the study of trends in phenology and its drivers.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
