VIIRS NDVI/EVI (VNP13)¶

VIIRS Vegetation Indices (VNP13A1/A2/A3/C1/C2, VJ113, VJ213) Version 2 (NDVI, EVI and EVI2, 500 m to 0.05°, 16-day and monthly)

Available

At a glance¶

Note on period: VNP13 (Suomi NPP) 16-day products start on 2012-01-17 and monthly products on 2012-01-01; VJ113 (NOAA-20) starts on 2018-01-01; VJ213 (NOAA-21) 16-day products start on 2023-02-10 and monthly on 2023-02-01. A CMR check of the Korean tile (h28v05) on 2026-10-13 found the latest composites were 2026-09-14 to 29 for VJ213A1 and 2026-09-06 to 21 for VNP13A1. VNP13 is expected to stop when Suomi NPP data delivery ends in November 2026.

Overview¶

NDVI and EVI that continue the MODIS vegetation index (MOD13) system on the VIIRS sensor, with the addition of EVI2, which is computed from red and near-infrared only, without the blue band. Sixteen-day and monthly values are built with the same Constrained-View angle Maximum Value Composite (CV-MVC) as MODIS and provided on 500 m and 1 km tiles and a 0.05° global grid. Reflectances (red, NIR, blue, green and three SWIR bands), view geometry, composite day and quality information are included, so the VIIRS record from 2012 can carry phenology, drought and crop studies beyond MODIS.

What changes with the successor¶

VIIRS Vegetation Indices V1 (VNP13 V001) → VIIRS Vegetation Indices V2 (VNP13, VJ113, VJ213 V002)¶

Transition: 2023-07

Version 2 (Suomi NPP and NOAA-20) was released in July 2023; V1 production stopped on 17 June 2024 and its distribution ended in 2025 (the last V1 products on 2025-07-31). A NOAA-21 product (VJ213) was added to V2 in March 2026.

Sources: VIIRS Version 2 vegetation indices products released (Earthdata notice, 2023-07-19) · Earthdata catalog — VNP13A1 V002 (V2 changes) · VIIRS Version 1 data products decommissioned (Earthdata notice, 2025-07-08)

Available variables¶

From the layer tables of VNP13A1 and VNP13C1 V002 in the Earthdata catalog (VJ113 and VJ213 have the same structure). Layer names are prefixed with resolution and period (500 m 16 days … for 500 m 16-day, CMG 0.05 Deg 16 days … for 0.05°), and unlike MODIS (250m 16 days …) there is a space between the number and m. Index and reflectance int16 values are divided by 10000. Monthly layer names follow the LP DAAC VNP13A3 and VNP13C2 V002 product pages and are prefixed 1 km monthly … and CMG 0.05 Deg monthly ….

instantaneous mean accumulated maximum minimum static — instantaneous: value at the time stamp · mean/accumulated/maximum/minimum: over the indicated interval · static: does not change with time

16-day composites (VNP13A1 500 m; A2 1 km has the same structure)¶

16-day 0.05° global grid (VNP13C1)¶

Monthly (VNP13A3 1 km, VNP13C2 0.05°)¶

The full variable list is in the official documentation.

How to get the data¶

• NASA Earthdata Search — VJ213A1 (NOAA-21, recommended) — Web search and HTTPS download / earthaccess
  • Short names are VJ213A1, A2, A3, C1, C2 (NOAA-21), VJ113… (NOAA-20) and VNP13… (Suomi NPP), version 002. Files are HDF-EOS5 (.h5). Because Suomi NPP data delivery ends in November 2026, LP DAAC recommends VJ213 as primary and VJ113 as secondary.
• NASA Earthdata Search — VNP13A1 (Suomi NPP, 2012 onward) — Web search and HTTPS download / earthaccess
  • Only the Suomi NPP products cover 2012–2017. No distribution policy for existing files after delivery ends has been announced yet.
• AppEEARS (area and point extraction) — Web and API — area (GeoTIFF, NetCDF4) or point (CSV) extraction with optional reprojection
  • Delivers just your region without having to handle tiles or the HDF5 structure.

Example code¶

Downloads NOAA-21 VJ213A1 (500 m, 16-day) for the 2024 growing season with earthaccess and reads NDVI for the central Korean tile (h28v05). To avoid hard-coding paths inside the HDF-EOS5 file, h5py is used to find the layers by name. Stored values are divided by 10000 and the fill value (−15000) is removed.

# pip install earthaccess h5py numpy
import earthaccess
import h5py
import numpy as np

earthaccess.login()  # free NASA Earthdata account
results = earthaccess.search_data(
    short_name="VJ213A1", version="002",
    temporal=("2024-04-01", "2024-10-31"),
    bounding_box=(127.0, 36.0, 127.5, 36.5),   # (west, south, east, north)
)
files = earthaccess.download(results, "./vj213a1")

def find(h5, name):
    hits = []
    h5.visititems(lambda k, v: hits.append(k) if isinstance(v, h5py.Dataset) and k.endswith("/" + name) else None)
    return h5[hits[0]][()]

for f in sorted(map(str, files)):
    if ".h28v05." not in f:
        continue
    with h5py.File(f, "r") as h5:
        ndvi = find(h5, "500 m 16 days NDVI").astype("float32")
        doy = find(h5, "500 m 16 days composite day of the year")
    ndvi = np.where((ndvi >= -10000) & (ndvi <= 10000), ndvi / 10000.0, np.nan)
    print(f.split("/")[-1][:25], "mean NDVI:", round(float(np.nanmean(ndvi)), 3),
          "median observation day:", int(np.median(doy[doy > 0])))

Practical tips¶

• The scale is expressed the opposite way from MODIS. The LP DAAC table gives the scale as 10000, so divide stored values by 10000 (MODIS multiplies by 0.0001; the result is the same). The index fill is −15000 (MODIS −3000), the reflectance fill −1000, and angles are ÷ 100.
• Pixel reliability uses ranks 0–11 (fill −4), unlike the MODIS 0–3 classes. Do not reuse MODIS filters such as rel <= 1; check the rank table (Table 5) in the User Guide before choosing a threshold.
• Consecutive 16-day files overlap by 8 days. The series looks 8-day spaced, but neighbouring values are not independent, so do not inflate sample sizes in statistical tests.
• The blue band differs from MODIS (VIIRS M3 478–498 nm vs MODIS 459–479 nm), so EVI has weaker cross-sensor continuity. For long-term work joined to MODIS, consider NDVI or EVI2 first. Without the blue band, EVI2 is somewhat more sensitive to aerosol noise.
• VNP13 (Suomi NPP) ends in November 2026. VJ113 continues from 2018 and VJ213 from 2023, so choose a reference satellite and state where the satellite changes.
• Ocean tiles h33v07 and h18v14 are empty because critical inputs were missing (LP DAAC known issue).
• There is no 250 m product. To compare with MODIS 250 m (MOD13Q1), bring both to 500 m.
• The files are HDF5 (HDF-EOS5), so MODIS pyhdf code will not open them. Layer names contain spaces, so looking them up with h5py is the safer route.

Notices and change history¶

References for your paper¶

Dataset citation¶

VJ213A1 V002 (NOAA-21, 500 m, 16-day) — DOI 10.5067/VIIRS/VJ213A1.002

@misc{vj213a1_v002,
  author    = {Vermote, Eric},
  title     = {{VIIRS/JPSS2} Vegetation Indices 16-Day {L3} Global 500m {SIN} Grid {V002}},
  year      = {2026},
  publisher = {NASA Land Processes Distributed Active Archive Center},
  doi       = {10.5067/VIIRS/VJ213A1.002}
}

VJ113A1 V002 (NOAA-20, 500 m, 16-day) — DOI 10.5067/VIIRS/VJ113A1.002

@misc{vj113a1_v002,
  author    = {Vermote, Eric},
  title     = {{VIIRS/JPSS1} Vegetation Indices 16-Day {L3} Global 500m {SIN} Grid {V002}},
  year      = {2023},
  publisher = {NASA Land Processes Distributed Active Archive Center},
  doi       = {10.5067/VIIRS/VJ113A1.002}
}

VNP13A1 V002 (Suomi NPP, 500 m, 16-day) — DOI 10.5067/VIIRS/VNP13A1.002

@misc{vnp13a1_v002,
  author    = {Vermote, Eric},
  title     = {{VIIRS/NPP} Vegetation Indices 16-Day {L3} Global 500m {SIN} Grid {V002}},
  year      = {2023},
  publisher = {NASA Land Processes Distributed Active Archive Center},
  doi       = {10.5067/VIIRS/VNP13A1.002}
}

Reference papers¶

VIIRS Vegetation Index User Guide and abridged ATBD V2.1.2 — no DOI

@techreport{didan2018vnp13,
  author      = {Didan, Kamel and Barreto Munoz, Armando and Compton, Tucker J. and Pinzon, Jorge E.},
  title       = {Suomi National Polar-orbiting Partnership Visible Infrared Imaging Radiometer Suite Vegetation Index Product Suite User Guide \& Abridged Algorithm Theoretical Basis Document, Version 2.1.2},
  institution = {Vegetation Index and Phenology Lab, The University of Arizona},
  year        = {2018},
  url         = {https://lpdaac.usgs.gov/documents/1372/VNP13_User_Guide_ATBD_V2.1.2.pdf}
}

Continuity assessment of MODIS C6.1 and VIIRS C2 land products — DOI 10.1016/j.rse.2023.113963

@article{roman2024continuity,
  author  = {Rom{\'a}n, Miguel O. and Justice, Chris and Paynter, Ian and Boucher, Peter B. and Devadiga, Sadashiva and Endsley, Arthur and Erb, Angela and Friedl, Mark and Gao, Huilin and Giglio, Louis and Gray, Josh M. and Hall, Dorothy and Hulley, Glynn and Kimball, John and Knyazikhin, Yuri and Lyapustin, Alexei and Myneni, Ranga B. and Noojipady, Praveen and Pu, Jiabin and Riggs, George and Sarkar, Sudipta and Schaaf, Crystal and Shah, Deep and Tran, Khuong H. and Vermote, Eric and Wang, Dongdong and Wang, Zhuosen and Wu, Aisheng and Ye, Yongchang and Shen, Yu and Zhang, Shuai and Zhang, Xiaoyang and Zhao, Maosheng and Davidson, Carol and Wolfe, Robert},
  title   = {Continuity between {NASA} {MODIS} Collection 6.1 and {VIIRS} Collection 2 land products},
  journal = {Remote Sensing of Environment},
  volume  = {302},
  pages   = {113963},
  year    = {2024},
  doi     = {10.1016/j.rse.2023.113963}
}

Acknowledgements¶

Following the EOSDIS Data Use and Citation Guidance, LP DAAC asks you to cite the DOI of each product you used (by satellite, resolution and period). DOIs follow the pattern 10.5067/VIIRS/.002 (e.g. 10.5067/VIIRS/VNP13A3.002 for 1 km monthly). Cite the VNP13 User Guide and ATBD for the algorithm.

Information on the datasets used in this study was compiled with the help of the Korean Climate Data Navigator (KCDN) (Sim, 2026; https://doi.org/10.5281/zenodo.23300923).

Links¶

• VNP13 User Guide and ATBD V2.1.2 (PDF)
• Earthdata catalog — VNP13A1 V002
• Earthdata catalog — VNP13C1 V002
• VIIRS vegetation index validation page (VIIRS Land Team)
• NOAA NESDIS — cessation of Suomi NPP data delivery