University of Hamburg - to research, to teach, to educate and form, to homepage
ICDC
CENIntegrated Climate Data Center ICDC

Photo: getty images/L. Calvetti

Leaf Area Index LAI and FAPAR from MODIS

MODIS LAI, Aug 2016
MODIS leaf area index (LAI), Aug 2016. Copyright: UHH/CEN/S. Kern

Topics

Access

UNRESTRICTED:

RESTRICTED: This link to the data set is only available for a restricted user group.  The data set is only accessible in CEN/MPI net or accessible from external nets with a customer account.  Please contact ICDC if you would like to access this data from outside the network.

to top

Description

Atmospheric influence corrected surface reflectances of up to 7 spectral bands of reflectance measurements of the MODIS sensors aboard both EOS platforms: TERRA and AQUA, are used to derive the Leaf Area Index (LAI) and the fraction of the absorbed (by green vegetation) photosynthetically active, i.e. 400-700 nm, radiation (FAPAR). For high enough solar elevation angels a radiative transfer modeling (RTM) approach is used. For small solar elevation angles and/or if the main algorithm based on the RTM fails due to other reasons an empirical back-up algorithm is used which - like the RTM based one - includes MODIS and other land cover products and the respective distribution of vegetation types (biomes) but which is less physical in the setup.

The key features of the RTM based approach as well as of the empirical algorithm are described in the Algorithm Theoretical Basis Document (ATBD, see references). This approach has been developed further since the first MODIS LAI data set release, which was collection 3); the current version belongs to the Collection 6.1 suite of MODIS data products. An extensive literature list about the different retrieval stages can be found at th website of the Climate and Vegetation Research Group, see references.

One of the key improvements towards collections 5 and 6 (see LAI/FPAR change report in references) was the inclusion of more biomes; different vegetation causes different canopy reflectance and a more detailed and more reliable classification into different biomes was found to increase success for LAI and FPAR retrieval.

We have the entire data set MOD15A2H of Collection 6.1 available - organized into tiles with 500 m grid cell resolution and also gridded onto a global 0.5 degree x 0.5 degree resolution climate modeling grid (CMG).

Last update of data set at ICDC: March 28, 2024.

to top

Parameters

Parameters of the LAI Tiles data set
NameUnitComments /Valid values
Fraction of photosynthetically active (400-700nm) i.e. absorbed by green vegetation radiation (FAPAR) % 0 ... 1
Leaf area index = Lead area per unit ground area (LAI) m² / m 0 ... 10
FAPAR retrieval standard deviation -  
LAI retrieval standard deviation m² / m²  
Detailed quality flag -

0, 1, 5, 6, 9;

added to 0, 10, 20, 50, 100, 110, 120, and 150, respectively
Quality flag for land - 0, 10, 20, 40, 60, 100, 120, 140, 200, 220, 240
Quality flag for cloud and aerosol -

0, 1, 10, 11, 20, 21, 30, 31,

100, 101, 110, 111, 120, 121, 200, 201
Quality flag method -
  • 0: no retrieval;
  • 1: very good, main method;
  • 2: good, empirical method
Parameters of the LAI Climate modeling grid data set
NameUnitComments /Valid values
Mean fraction of photosynthetically active (400-700nm) i.e. absorbed by green vegetation radiation (FAPAR) % 0 ... 1
Mean leaf area index = Leaf area per unit ground area (LAI) m² / m² 0 ... 10
Variance of mean FAPAR % 0 ... 1
Variance of mean LAI m4 / m4 0 ... 10
Mean FAPAR retrieval standard deviation -  
Mean LAI retrieval standard deviation m² / m²  
Number of valid FAPAR or LAI values per grid cell - 0 ... 14196
Number of valid retrieval standard deviation values per grid cell - 0 ... 14196
Primary (most common) quality flag in grid cell - 0 ... 700
Secondary (2nd most common) quality flag in grid cell - 0 ... 700
Primary (most common) land surface type in grid cell - 0 ... 900
Secondary (2nd most common) land surface type in grid cell - 0 ... 900
Fraction of pixels with cloud cover % 0 ... 100
Fraction of pixels with considerable aerosol load % 0 ... 100

to top

Coverage, spatial and temporal resolution

modis-sinusoidal-grid-tiles-colored

Period and temporal resolution:

  • 2000/02 - 2023/12
  • 8-daily

Coverage and spatial resolution:

  • Global or split into tiles of ~10° latitude and 10+° longitude (see image)
  • Spatial resolution: Climate Modeling Grid (CMG): 0.5° x 0.5°; tiles: 500m x 500m, MODIS sinusoidal grid
  • Geographic Latitude: CMG: -89.75 to 89.75; tiles: depends on tile
  • Geographic Longitude: CMG: -179.75 to 179.75; tiles: depends on tile
  • Dimension: CMG: 360 rows x 720 columns; tiles: 2400 rows x 2400 columns per tile
  • Altitude: following terrain

Format:

  • NetCDF

to top

Data quality

The data set contains retrieval uncertainties for both parameters LAI and FAPAR. In addition the data set contains a number of quality flags.

The most simple one distinguishes between the methods used and allows us to discriminate between areas with very good retrievals, where the main method based on radiative transfer modeling was used and areas with "just" good retrievals, where the main method needed to be replaced by an empirical method because, e.g., a too low solar elevation. This latter issue is the reason why in some maps - depending on geographic latitude and season - LAI and FAPAR retrieval standard deviations are missing.

The other quality flags (3 layers) give information about the land cover other than the biomes used, e.g. whether snow has been detected or whether the area is mainly a built-up one, about the influence of clouds and aerosol (with or without aerosol or clouds, cirrus or no cirrus, cloud shadows, etc.) and for those who prefer really detailed information there is a detailed quality flag - where one can find for instance where a pixel was just assumed to be cloud free.

The references list selected literature dealing with MODIS LAI and FAPAR evaluation - mainly for the collection 5 MODIS LAI and FAPAR data. The paper by Disney et al., 2016 (see references) contains a good overview about differences in methodology between different approaches and points to the potential limitations of this data set.

We recommend to also take a look at the comments about data quality for the combined AVHRR-LAI data set.

to top

Contact

Ranga Myneni
Climate and Vegetation Research Group
Department of Earth and Environment, Boston University, MA, U.S.A.
email: myneni (at) bu.edu

Stefan Kern
University of Hamburg
E-Mail: stefan.kern (at) uni-hamburg.de

to top

References

Literature:

Websites:

to top

Data citation and License

Please cite the data as follows:

Tiles:

MODIS LAI and FPAR data from Myneni, R., Knyazikhin, Y., Park, T. (2024). MOD15A2H MODIS/Terra Leaf Area Index/FAPAR 8-Day L4 Global 500m SIN Grid V061. NASA EOSDIS Land Processes DAAC. http://doi.org/10.5067/MODIS/MOD15A2H.061 [last access: March 13, 2024], distributed by the Land Processes Distributed Active Archive Center (LP DAAC), located at the U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center (lpdaac.usgs.gov), provided with de-coded quality flags and per grid cell latitude / longitude information in netCDF format by the Integrated Climate Data Center (ICDC), CEN, University of Hamburg, Germany: https://doi.org/10.25592/uhhfdm.14184 [last access: MM-DD-YYYY].

Global:

Globally gridded MODIS Terra Collection 6.1 8-daily LAI and FAPAR at 0.5 degree grid resolution derived at ICDC, https://doi.org/10.25592/uhhfdm.14185 [last access: MM-DD-YYYY], from Myneni, R., Knyazikhin, Y., Park, T. (2024). MOD15A2H MODIS/Terra Leaf Area Index/FPAR 8-Day L4 Global 500m SIN Grid V061. NASA EOSDIS Land Processes DAAC. http://doi.org/10.5067/MODIS/MOD15A2H.061 [last access: March 13, 2024].

Licence: The data was published under the Creative Commons Attribution 4.0 International License (CC-BY-4.0).

to top