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 to SLAMM land cover classes

Started by Dean, August 18, 2009, 04:05:01 PM

Previous topic - Next topic

Dean

I am using the VLOOKUP function in Excel to reclassify the NWI land cover to the SLAMM5 categories as shown in the Tech. Doc. from October 2008.  I am having trouble understanding Table 5 in this Tech Doc and discerning exactly which NWI classes are being changed.  It appears as though some NWI classes are double counted, ie appearing twice in the new SLAMM5 categories.

For example, SLAMM5 class 6 (TidalFreshMarsh) shows that it is derived from NWI classes R1EM and PEM ["K"-"U"]. However, SLAMM5 class 23 appears to overlap, showing NWI classes PSS, PFO"K"-"V"/EM1 "K"-"V".  To me, it looks as though PEM1K - PEM1U are doubly represented.  Am I missing something here.  An explanation of shorthand of the NWI classes may help.

On another note, is there an explanation somewhere why the modifier "K" was chosen to be included with tidal, in most if not all cases of reclassification?

Is there a way to perform this process in ArcMap, or perhaps using AML code?

Thanks!
Dean

Jonathan S. Clough

#1
Thanks for this question.  The tech. doc table is now a bit out of date and is not the best way to map these categories.  There is an Excel spreadsheet for that.  The latest spreadsheet can be found here:

http://warrenpinnacle.com/prof/SLAMM6/SLAMM6_nwi_codes_2009.xls

Bill Wilen, the head of the National Wetlands Inventory has carefully vetted and approved this version of the spreadsheet.  If you come across any codes that are not in the spreadsheet feel free to ask for guidance here.

On your second note, the modifier "K" was chosen to be included with tidal because of an error in some of the (now obsolete) NWI documentation which included that under the tidal classification.  So that has been fixed in this spreadsheet.

Finally, Excel's VLOOKUP is what we usually use, if you come up with a script in ArcMap I hope that you are willing to share it.

Thanks again -- Jonathan

Dean

Thanks, Jonathan!  That file is going to be very helpful.  I appreciate your quick response too, which is also helpful.

Thanks for starting this forum, I think it's a great idea. I don't think everyone modeling SLR should use the same model, for diversity's sake, but I think SLAMM is one of the most comprehensive (definitely the most I've seen).  So having a forum for it is a brilliant idea!

-Dean

Dean

Jonathan,

For filling in the gaps between NWI polygons (ie developed dry land and undeveloped dry land), are you using NLCD data? Just curious.

Thanks,
Dean

Jonathan S. Clough

Yes, generally we use the 2001 NLCD percent impervious as the best available dataset unless a better one is available.

In fact, you can input a percent impervious integer raster into SLAMM and it will differentiate dry land and developed dry land on that basis (I think using 25% developed as the cutoff).

Thanks for your kind words about the Forum.  I think it will grow as the new version of SLAMM becomes available this fall and more communication is warranted.

-- Jonathan

chayek

The link for the excel file provided earlier in this thread is not working.  Can you please direct me to the latest lookup table?

Thanks

Jonathan S. Clough

The newest Excel file is installed along with SLAMM6.EXE in the program directory.  It is called SLAMM6_nwi_codes_2009.xls

If you're having trouble finding it, another copy may be downloaded here: http://warrenpinnacle.com/prof/SLAMM6/SLAMM6_nwi_codes_2009.xls.

Also see Table 5 in the SLAMM 6 technical documentation.

-- Jonathan

chayek

So I am running into several situations where the NWI classification is a mix of two different wetland types (denoted by a slash), each of which falls into a different SLAMM category.  How do I know which SLAMM category to use?  Is the first of the two wetland types dominant?

Jonathan S. Clough

Yes, based on communication with Bill Wilen (head of NWI) the first class is dominant, at least in the newer NWI coverages.  That is apparently the new protocol but I'm not sure it applies to all historical datasets.  Here's an excerpt from an email I wrote in Sept 2009:

QuoteI spoke to Bill Wilen on the phone today.  When looking at "split" NWI codes, the historical datasets did not specify the dominant class first, but rather classes were just listed in order.

yanan

hi Jonathan,

When i use the VLOOKUP function, i have come across many codes that are not in the spreadsheet(SLAMM6_nwi_codes_2009.xls), then i look up these NWI codes in the technical documentation (p43-p44) and the open source code, a lot of them are determined, but still there are some codes that i cannot find, such as "E2UB3H", "E2UB3L", "E2UBM", and etc. In the open source code i have found a code named 'E2UBN' in the comment line, which is called "Tidal Creek" in SLAMM's category. Are there any relationship between them,or is there something wrong with the comment line?

thank you.

Jonathan S. Clough

The definitive sources are the Tech. Doc. tables as derived and refined by Bill Wilen and the Excel spreadsheet which is very thorough but does not include all NWI classes.  Through these two resources you should be able to assign a given NWI class to a SLAMM category.

The comments in the code may provide useful hints but they are dated and have not been closely maintained.

-- Jonathan

yanan

I have chosen North Carolina as my study area,and  it's NWI codes contain  "E2UB3H", "E2UB3L", "E2UBM" and .etc,but l just could not find any information about E2UB in the Tech. Doc. I don't know where did i go wrong.

Jonathan S. Clough

You haven't done anything wrong, it is just that there are nearly infinite NWI classes so the spreadsheet is not perfect...

"E2UB3H", "E2UB3L", "E2UBM"

Here are some useful resources: 

http://www.fws.gov/wetlands/Data/wetlandcodes.html  and
http://137.227.242.85/Data/interpreters/wetlands.aspx

These codes are one sort of open water or another due to the "UB" designation and the model is not particularly sensitive to differences in classes of open water.

These codes also seem to be outdated or invalid when using the wetlands interpreter

I would put them as class 17 EstuarineWater.

-- J