SPoRT/MODIS Real-time Vegetation Depicts Early Green-up

March 2012 was a month for the climatological record books, with anomalously warm temperatures prevalent across much of the Contiguous United States.  Numerous stations recorded their all-time warmest March mean temperature, including the greatest departure above normal for any month.

These unusually warm temperatures translated into an early, rapid green-up in the deciduous vegetation across much of the eastern half of the United States.  This rapid green-up is captured well by the real-time SPoRT/MODIS vegetation product, which is produced once per day at 0.01° resolution over the U.S. lower 48 states and adjacent portions of southern Canada and northern Mexico.  The figure below shows a one-year change in the SPoRT/MODIS Greenness Vegetation Fraction (GVF) product from 2011 to 2012, valid on 11 April.  Most of the eastern U.S. has larger GVF than last year by 10% or more.  In fact, a significant part of southeastern Canada has GVF increases over 40% compared to last year.  Parts of the Tennessee / Mississippi river valleys, and New England also have year-to-year increases of comparable magnitude.

Such dramatic inter-annual changes in the health and coverage of green vegetation can have significant impacts on the surface energy budget within numerical weather prediction models.  The default vegetation dataset in community models such as the WRF model consists of a climatology that depicts vegetation coverage the same from year to year.  In anomalous situations such as this Spring, the climatological dataset would likely mis-represent the coverage of vegetation in the model, thus negatively affecting the energy exchanges between the land surface and planetary boundary layer.  Therefore, the use of real-time vegetation, such as the SPoRT/MODIS product, has the potential to better represent vegetation coverage and the subsequent surface energy budget in real-time models.

One-year change in SPoRT/MODIS real-time Greenness Vegetation Fraction (GVF) from 11 April 2011 to 11 April 2012.

7 thoughts on “SPoRT/MODIS Real-time Vegetation Depicts Early Green-up

  1. Hi, this is pretty neat. Is there a way to access this data on the web in real time? Is there greeness climatology and not just a year to year compariso?

    Thanks, Jonathan Blaes WFO Raleigh,NC

    • Hi JonathanB:

      The only place we currently post real-time GVF images is on the 1-km real-time LIS output zoomed into Alabama: (look for the VEGFR links). We also have an internal page where we post the full-domain NDVI composites before converted into GVF for use in models.

      We began creating the real-time vegetation products on 1 June 2010. Because of the real-time methodology, we don’t have an easy way to produce a GVF climatology for the whole MODIS period of record. However, the climatology to compare against is the default monthly GVF dataset that comes with the community WRF model, which is based on AVHRR data and produced on a grid that is about 15 times coarser than our MODIS GVF grid.

    • I also wanted to mention that we make the real-time MODIS GVF available daily in the WRF model input format on our ftp site. If anyone is interested in using the MODIS GVF in your WRF Environmental Modeling System (EMS) runs, I can supply you with an instructions document on how to acquire the data and replace the default climatology with our real-time GVF data.

  2. Jon, do you know if the modeling group is working with anyone of our collaborating offices to run side-by-side comparisons using the MODIS GVF vs. the default climatological GVF data set…mainly during this early spring “greenup”? I wonder what kinds of differences would be seen in soil moisture values, surface air temperatures and even model precipitation output. There were several almost summer-like convective regimes that developed across the greater TN Valley and adjacent regions during the unusual warm spell of the past 4-6 weeks. Differences in model handling of these surface budget issues might have been magnified during this time period. This might be a nice project.

    BTW…one small correction…this past March was the 2nd most anomalous warm month on record for the U.S. The most anomalous warm month, at least according to the NCDC, was January 2006.

    • Hi Kris,

      We do have a collaboration with the Houston and Mobile WFOs, in which they are producing local WRF EMS runs with ALL SPoRT datasets (SSTs, LIS, and real-time MODIS GVF), while SPoRT generates the control runs without any SPoRT data. However, their southern locations probably won’t reveal as much impact from the anomalously high GVF this spring.

      A good hypothesis to test is that the use of real-time GVF would produce more accurate 2-m T/Td forecasts and subsequent boundary layer evolution, compared to using climatology during such an anomalous green-up. The forecast 2-m temperatures (dewpoints) should actually be lower (higher) in the model if the actual GVF is much higher than climatology, with all other factors being the same.

      Thanks for the clarification,

  3. Hi Jon,

    Thanks for the quick response. I thought it might be interesting to see the model runs with all similar data, physics, etc., except for the GVF data. One set of runs with the MODIS GVF and one set of runs with the GVF climatology. That way, the differences just the GVF would have on the model run could be demonstrated. I didn’t know if it was possible though, to run the models with just the GVF difference…or even would be a worthy cause.

  4. And yes, considering their location…the differences at Mobile or Houston might be small. It would be interesting to see differences in the mid-Mississippi Valley-Ohio-TN Valley regions where some of the differences were larger.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s