AnisoVeg: Anisotropy and Nadir-normalized MODIS MAIAC datasets for satellite vegetation studies in South America

Dataset paper under review.

Contact Ricardo Dalagnol (ricds@hotmail.com) for more information.

Data: AnisoVeg: Anisotropy and Nadir-normalized MODIS MAIAC datasets for satellite vegetation studies in South America

Scale factor: 10000

obs: the no_samples layer does not have scale.

Coverage: South America land

Time period: 2000 to 2021 (starting in March 2000)

Spatial resolution: 0.009107388 degree equivalent to ~1 km

Temporal resolution: Monthly

Coordinate reference system: geographic projection, datum WGS-84

Processing details summary (more detailed explanation in the paper):

• The original MODIS (MAIAC) data were described by Lyasputin et al. 2011 (https://doi.org/10.1029/2010JD014986). The daily MODIS (MAIAC) surface reflectance data from collection 6, acquired from Terra and Aqua satellites, are available from the MCD19A1 product (https://ladsweb.modaps.eosdis.nasa.gov/archive/allData/6/MCD19A1). The Bidirectional Reflectance Distribution Function (BRDF) model parameters are available from MCD19A3 product (https://ladsweb.modaps.eosdis.nasa.gov/archive/allData/6/MCD19A3)
• The daily MCD19A1 data at 1 km spatial resolution were normalized using the BRDF parameters and Ross-Thick Li-Sparse (RTLS) model considering (1) a fixed nadir view and a 45 deg. solar zenith angle using the parameters from the MCD19A3 product; and (2) backward and forward scattering. For the nadir (NAD) product, the nadir normalization was taken. For the anisotropy (ANI) product, we calculated the backward minus forward surfaces for each layer, resulting in the ANI product.
• The daily data were aggregated into monthly composites by the pixel’s median.
• The tiles that cover the South America were mosaicked and re-projected from sinusoidal to geographic projection
• The Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) were calculated using standard formulas. The EVI parameters were: C1 = 6, C2 = 7.5, L = 1, G = 2.5

File(s) format:

• Zip files - one per year. NAD are the nadir-normalized products and ANI the anisotropy. Inside zip files there are raster files with ".tif" format, one per month window.
• The filename syntax is "maiac_southamerica_month_PRODUCT_YYYY_MM_LAYER_latlon.tif", where YYYY is the year (e.g. 2000), MM is the month with two digits (e.g. 03 for March), PRODUCT is either nadir or anisotropy, and LAYER can be bands 1-8, EVI and NDVI.
• There is also the number of samples (no_samples) layers for each date, which can be used to filter composites with a minimum desirable number of daily observations. 

Code: https://github.com/ricds/maiac_processing

Acknowledgements: R.D. was supported by Sao Paulo Research Foundation (FAPESP) grants 2015/22987-7 and 2019/21662-8. FHW was supported by FAPESP grant 2015/50484-0. Part of this work was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (NASA). The funders had no role in the study design, data collection and analysis, including the decision to publish or prepare the manuscript. We thank the MODIS MAIAC team from NASA for providing the freely available MODIS (MAIAC) daily dataset.

Auxiliary data of AnisoVeg:

• Backscattering data of AnisoVeg: https://doi.org/10.5281/zenodo.6040299 and https://doi.org/10.5281/zenodo.6040790
• Forward scattering data of AnisoVeg: https://doi.org/10.5281/zenodo.6048784 and https://doi.org/10.5281/zenodo.6048793

Layers available at Google Earth Engine (GEE):

• EVI Anisotropy: https://code.earthengine.google.com/?asset=projects/anisoveg/assets/evi_anisotropy
• EVI Nadir: https://code.earthengine.google.com/?asset=projects/anisoveg/assets/evi_nadir

Obs: these require a (free) google earth engine account.

Dataset usage: This dataset is a product of hundreds of hours of coding starting in 2016 with the first author PhD work and then into his Postdoc, and many more hundreds hours of data processing. Data is free to use, but if you use this dataset, please cite the dataset paper or this repository (while paper is under review). Invitation for collaboration are welcomed.

While the paper is under review, for use of this dataset please cite:

Dalagnol, Ricardo; Galvão, Lênio Soares; Wagner, Fabien Hubert; Moura, Yhasmin Mendes;  Gonçalves, Nathan; Wang, Yujie; Lyapustin, Alexei; Yang, Yan; Saatchi, Sassan; Aragão, Luiz Eduardo Oliveira e Cruz. (2022). "AnisoVeg: Anisotropy and Nadir-normalized MODIS MAIAC datasets for satellite vegetation studies in South America". (Version v1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.3878879