Period: 1981 – 2014
Sensor: MODIS & AVHRR
Frequency: Yearly
Resolutions: 5.6km
Extent: Global
Data can be accessed here.
Algorithm description: Following information is available in this product: Start, End and Length of the growing season, Day of peak, Magnitude of peak, Rate of greening, Rate of senescence, Cumulative green.
The algorithm is similar to that for NASA MODIS/AVHRR VIPPHEN EVI2, however it is implemented on NDVI time series.
Accuracy: Overall errors due to reflectance input uncertainty are small (~0.05) except sparsely vegetated areas due to very small signal. Errors are largest in winter.
It has also been noted in the ATBD of the product that errors associated with the indices and phenology parameters can be large and complex; however, it is difficult to characterise them due nature of the index itself.
Marshal et al (2016) compared VIP collection (both EVI and NDVI) with GIMMS collection version 3 on global scale. GIMM NDVI data are purely based on AVHRR sensor reflectance unlike VIP. They found VIP NDVI and EVI2 (version 3) performed worse that GIMMS NDVI. They were higher in magnitude, noisier and less inconsistent in terms of long term trend compared to GIMMS. Majority of these discrepancies were attributed to the fact that VIP relies on blending of AVHRR and MODIS that have different sensitivities
Limitations:
- Product is generally suitable for studying long term phenological changes.
- Inconsistencies in MODIS and AVHRR derived products.
Reference:
Didan, K. et al. Multi-Sensor Vegetation Index and Phenology Earth Science Data Records Algorithm Theoretical Basis Document and User Guide Version 4.0. 1–84 (Vegetation Index and Phenology Lab., University of Arizona, 2015).
Marshall, M., Okuto, E., Kang, Y., Opiyo, E. & Ahmed, M. Global assessment of Vegetation Index and Phenology Lab (VIP) and Global Inventory Modeling and Mapping Studies (GIMMS) version 3 products. Biogeosciences 13, 625–639 (2016).