Terrain Shape Index   Gary A. Beluzo
  May 23, 2005 09:01 PDT 
Dale, Tom, et al:

So.. what makes Zoar so special a growing environment? The TSI (Terrain Shape
Index), i.e. extreme concavity? Relative isolation from storm events? Have
ENTS folks found a good correlation between TSI (especially steep, narrow
ravines)nd maximum height for a given species? Henry McNabb suggests that
certainly Lirodendron tulipifera (being extremely site sensitive) shows one.

What say you ENTS?
RE: Zoar Valley... WOW!!!!!    Gary A. Beluzo
   May 23, 2005 19:36 PDT 
Hi Ed,

What follows is directly from Henry McNab's website...a great explanation as
to what both TSI (Terrain Shape Index) and LFI (Landform Index)a re. I want
to eventually characterize each of the sites I am working on with a TSI that
may be used to calculate derivative variables through GIS analysis (i.e.
geoprocessing and geoanalysis). Actually there are several ways to
approximate the TSI on a computer- for every pixel in a GIS image you apply
an algorithm that takes into account the neighboring pixel values.

Gary

----------------------------------------------------------------------------


LFI and TSI: Topographic Variables to Quantify Meso- and Micro-scale
Landforms

     
W. Henry McNab, Research Forester

USDA Forest Service, Southern Research Station

Bent Creek Experimental Forest, Asheville, NC 28806

   
WHAT ARE LFI AND TSI?--LFI (landform index) is a measurement that is
correlated with landscape-scale features of the earth's surface, such as
ridges and valleys. TSI (terrain shape index) is a variable that quantifies
surface shape of a sample plot, from convex to concave, and is the
microscale equivalent of LFI. Both indexes are continuous topographic
variables that allow quick, easy, and objective field measurement of
landforms that can be repeated among observers, much in the same way that a
compass is used to measure aspect. Because landform is one of the most
important topographic variables affecting environmental characteristics in
hilly and mountainous terrain, its measurement is important for activities
such as ecological classification and prediction of vegetation composition
and growth. Both variables are improvements of classes of landform (such as
ridge, slope, valley) because the indexes can be determined consistently and
used as numerical values in prediction equations. LFI for a site is
equivalent to the average slope gradient from the observer to the horizon.
TSI for a sample plot is the average slope gradient from the plot center to
the perimeter. Both landform measures can be determined on field plots using
a hand-held clinometer or estimated by geographic information systems (GIS)
from digital elevation databases...

http://www.srs.fs.usda.gov/bentcreek/LFIandTSI.htm  

SCIENTIFIC BASIS OF LFI AND TSI--The rationale of LFI and TSI as measures
of landform is presented in two publications:

McNab, W.H. 1989. Terrain shape index: quantifying effect of minor landforms
on tree height. Forest Science 35(1):91-104.

McNab, W.H. 1993. A topographic index to quantify the effect of mesoscale
landform on site productivity. Canadian Journal of Forest Research
23:1100-1107.

Examples of the relative importance of LFI and TSI for explaining variation
in composition of vegetation in relation to the topographic variables
commonly measured on sample plots (i.e. altitude, aspect, gradient) is
presented in:

McNab, W.H. 1991. Predicting forest type in Bent Creek Experimental Forest
from topographic variables. pp. 496-504. IN: S.Coleman and D.Neary (comps.)
Proceedings of the sixth biennial southern silvicultural research
conference; 1990 October 30-November 1; Memphis, TN. General Technical
Report SE-70. Asheville, NC: USDA Forest Service, Southeastern Forest
Experiment Station.

McNab, W.H.; Browning, S.A. 1993. Preliminary ecological classification of
arborescent communities on the Wine Spring Creek watershed, Nantahala
National Forest. pp. 213-221. IN: C.Brissette (ed) Proceedings of the
seventh biennial southern silvicultural research conference; 1992 November
17-19; Mobile, AL. General Technical Report SO-93. New Orleans, LA: USDA
Forest Service, Southern Forest Experiment Station.


Gary Beluzo
Terrain Shape Index    Gary A. Beluzo
   May 24, 2005 07:23 PDT 
Hi Ed,

Here is a link to an excellent example of terrain analysis in general and
some of the specific algorithms used in GIS to come up with derivative
layers. There are some very good images to look at to see the results of
each algorithm applied in GIS.   I will be working with these algorithms
over the next few weeks if others are interested.

http://www.lsc.usgs.gov/aeb/2077/envirogradients.asp
RE: Terrain Shape Index   Robert Leverett
  May 24, 2005 07:53 PDT 

Gary:

   I looked at the site. This is pretty potent stuff with applications
to what we do. I would be interested in seeing indices computed for the
tall tree hot spots in MTSF. Is this something you can do now?

Bob
RE: Terrain Shape Index    Gary A. Beluzo
   May 24, 2005 08:55 PDT 

Bob:

I am looking to see if there is an existing AML or VB script available for
ARCGIS 9 to specifically do the TSI (which looks at the relationship between
the elevation of each pixel and 8 or more of its neighbors depending on what
value of the plot radius or site resolution we want to use), if not I will
probably just write a short script myself in VB. Once I have the algorithm
to run in ARCGIS we can calculate the TSI for every pixel in the TIN
(triangular Irregular Network- a way to display surfaces in 3D)coverage. So
yes, within a few days we will be able to calculate slope, aspect, flow
accumulation, TSI, LFI, etc for any area in MA.

Gary