This data set comprises the Normalized Differential Vegetation Index (NDVI) together with some information about data quality, derived from reflectances observed by channels 1 and 2 of the Advanced Very High Resolution Radiometer (AVHRR) as part of the Global Inventory Modeling and Mapping Studies (GIMMS) project.
The NDVI is based on the difference: reflectance observed in the near-infrared (NIR) portion of the electromagnetic spectrum minus reflectance observed in the red portion of the electromagnetic spectrum. Areas with photosynthetically active vegetation absorb much more radiation in the red portion of the electromagnetic spectrum than areas without any or with photosynthetically inactive vegetation. This is due to the pigments in the leaves, needles, etc. In contrast, absorption of NIR radiation is low for both surface types. However, the amount of NIR radiation transmitted and/or scattered is a function of the surface type / roughness. Therefore, the difference of the reflectances is small for areas without any or with photosynthetically inactive vegetation and high for areas with photosynthetically active vegetation. Note that in areas with a lot of vegetation the dependence of the NIR-radiation scattering and/or transmission on the surface type / roughness helps to overcome saturation effects in the red portion of the electromagnetic spectrum and therefore allows a better separation of high NDVI values.
The impact of clouds, aerosols, sensor noise (and other error sources not mentioned here explicitly) is mitigated as are effects by changes in sensors, sensor degradation or changes in local sensor overpass time. For more details we refer to Tucker et al. and Pinzon and Tucker.
We offer the GIMMS NDVI3g version 1 data set with bi-weekly temporal resolution.
The data set was included into the ICDC data holding: November 7, 2019