September 25, 2021, 01:00:49 PM

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.


Scenario scale-up: Max, mean, min

Started by rloiselle, March 24, 2010, 05:53:04 PM

previous topic - next topic
Go Down

rloiselle

I am a little unclear about how the SLR scenario "scale-up" works.  Can somebody clarify why the program produces output for an initial condition, then the next sequential year (I assume has to do with the scaling and setting DEM date to NWI date?), and then it starts the sequence of analysis for the specified time series??

For example, my NWI photo date is 2000, the DEM date is 1999.  The maps produced through the analysis are for the initial condition, condition at year 2000 (??), then 2025, 2050, 2075, and 2100.  There are very big differences between the initial land classes and year 2000 land classes, which seems questionable given the duration of the calculation (1 year).....

Another related question:  If I've already set my DEM elevations to NWI dates as indicated on page 9 and 14 of the Technical Manual, why does SLAMM need to upscale at all? 

I appreciate the clarification to my novice question!!   :D

Rebecca

Jonathan S. Clough

The initial condition year is the year of the NWI coverage.  That is the date of your land-cover and so by definition represents the "initial condition" for land cover (the date that forms the basis for your forecasting)

Therefore, what you are seeing is not a one year simulation but instead a "time-zero" run.  This is a run in which SLAMM tries to capture the current conditions within its conceptual model.  If you are seeing big differences in your "time zero" run then you need to figure out why and fix whatever problem is causing this to occur.  This may be considered a model calibration to site-specific data.  Until you get your "time-zero" run to reflect little or no change then you cannot have high confidence in your model forecasts.

Examine closely the units of your DEM, the MTL to NAVD88 correction, and the tide range parameters.  Then run an elevation analysis to examine the distributions of land elevations and how well they match the SLAMM conceptual model.  Ensure that there are not dikes or tide-muting considerations.  Note that tidal fresh marsh and tidal swamp elevations nearly always need to be adjusted from "SLAMM Defaults."   I think there is a note about that in the Technical Documentation.

Good luck with your model application...

rloiselle

So, if a user is working with a low resolution NED, and using the data specified by the user within the site parameters... The scale-up takes place within SLAMM prior to the time series calculations.  In other words, I don't have to calculate the new elevations based on the the date equation on pg13 and the MTL adjustment equation on pg 9 of the Technical Manual. SLAMM does this for me as long as I specify the NAVD88 correction factor within the site parameters or I use a VDATUM layer in addition to other required layer files....

Thanks for the clarification!

RL

Jonathan S. Clough

I think what you are saying is correct although I'm still not sure what you mean by "scale up."  Do you mean adjusting the date of the DEM so that it matches the land cover data?  If so, yes, SLAMM does the best that it can to do this using the assumptions as outlined in the Technical Documentation.

Note, if you are working with low resolution NED your model results will be considerably more uncertain.  Also, the "time-zero" run has considerably less utility with low-resolution elevation data as the pre-processor (assuming you're using that) will force the elevation data to conform to the SLAMM conceptual model prior to the running of "time-zero." 

-- J

rloiselle

Yes, sorry, I am referring to "scale-up" as the Date/MTL adjustment which sets the data to time-zero.

Thank you for the help!

Rebecca

Go Up