News:

Due to SPAM attacks, new members must be approved before posting.  Please email jclough@warrenpinnacle.com when registering and your account will be approved.

Main Menu

NWI photo date

Started by cpapiez, September 21, 2010, 02:27:11 PM

Previous topic - Next topic

cpapiez

How does the NWI photo date effect the model?  Does the model set that date as the initial and then calculate the parameters from it for each timestep?  So if we put in 1991 but it was really a 1981 photo date would we see different results similar to those found when you change the timestep?

Thanks!

Jonathan S. Clough

The NWI photo date is the initial time condition for the model.  This corresponds to the date of the land cover imagery.  Hopefully the digital elevation map is of a similar date.  If not, the model adjusts the DEM to that photo date using available estimates of land uplift or subsidence for the region. 

If you put in 1991 for 1981 map two things would happen.  You would not be calculating effects from historical SLR measured during the 1981 to 1991 time-period.  The DEM would be adjusted to 1991 rather than 1981.  In sites where there is not significant land subsidence or uplift the effects of the latter are likely to be negligible.  The effects of the former may also be minor depending on the historical rate of SLR. 

In some areas where we have had a crazy patchwork of NWI photo dates with no obvious seams in the data, we have assigned the most prevalent photo date or an average of the photo dates for the region, the assumption being that the effects on model results will be negligible.

staylor

For scenarios of older NWI data and more recent Lidar... for example, 1983 NWI and 2002 Lidar -- one can assume that the Lidar reflects the most accurate conditions. Is there anyway to trick SLAMM into updating the vegetation class? Or would this be too arbitrary given that some vegetation class elevation ranges overlap.

Jonathan S. Clough

This is an interesting question.  The model is set up to  try to adjust the elevation data to the NWI date so that it has  data layers that are synoptic.  The model does this by adjusting the elevation for any uplift or subsidence that is estimated for the site in question.  Then the "time-zero" run does update the vegetation class taking into account the LiDAR elevations.  So we often see corrections in the beach/tidal-flat to open water interface, additions of channels to marsh surfaces, and other minor changes at "time-zero"  on the basis of the LiDAR elevations.

What you are suggesting is to have the "time-zero" (or simulation start date) reflective of the LiDAR date rather than the NWI date.  Probably the best way to do this would be to set the NWI date to the LiDAR date and watch to see what types of conversions occur at "time-zero" on the basis of the LiDAR data.  You can then use satellite imagery or site-specific knowledge to see whether the predicted changes have actually occurred in which case that would be a strong starting point for your simulation.  If your wanted to you could use the GIS output from "Time-Zero" and use that as your initial condition for future forecasts.

As you note, there is some uncertainty due to the overlap in vegetation classes.  At this time, SLAMM also will not model aggradation or additions of land on the basis of LiDAR elevations.

staylor

How does SLAMM handle elevations outside of the default range? For example, regularly flooded salt marsh less than 0 m and/or greater than 1.2 HTU?