In:
Geoscience Data Journal, Wiley, Vol. 7, No. 1 ( 2020-06), p. 13-26
Abstract:
The leaf phenology (i.e. the seasonality of leaf amount and leaf demography) of ecosystems can be characterized through the use of Earth observation data using a variety of different approaches. The most common approach is to derive time series of vegetation indices (VIs) which are related to the temporal evolution of FPAR, LAI and GPP or alternatively used to derive phenology metrics that quantify the growing season. The product presented here shows a map of average ‘amplitude’ (i.e. maximum minus minimum) of annual cycles observed in MODIS‐derived NDVI and EVI from 2000 to 2013 for Meso‐ and South America. It is a robust determination of the amplitude of annual cycles of vegetation greenness derived from a Lomb–Scargle spectral analysis of unevenly spaced data. VI time series pre‐processing was used to eliminate measurement outliers, and the outputs of the spectral analysis were screened for statistically significant annual signals. Amplitude maps provide an indication of net ecosystem phenology since the satellite observations integrate the greenness variations across the plant individuals within each pixel. The average amplitude values can be interpreted as indicating the degree to which the leaf life cycles of individual plants and species are synchronized. Areas without statistically significant annual variations in greenness may still consist of individuals that show a well‐defined annual leaf phenology. In such cases, the timing of the phenology events will vary strongly within the year between individuals. Alternatively, such areas may consist mainly of plants with leaf turnover strategies that maintain a constant canopy of leaves of different ages. Comparison with in situ observations confirms our interpretation of the average amplitude measure. VI amplitude interpreted as leaf life cycle synchrony can support model evaluation by informing on the likely leaf turn over rates and seasonal variation in ecosystem leaf age distribution.
Type of Medium:
Online Resource
ISSN:
2049-6060
,
2049-6060
Language:
English
Publisher:
Wiley
Publication Date:
2020
detail.hit.zdb_id:
2745699-7
