[kepler-dev] Re: [Fwd: Biodiversity indices update]

Mon Jul 12 10:11:32 PDT 2004

Deanna,

I did look over the Biodivesity Index document you sent last week and I 
set down this weekend and created a Composite actor that calculates many 
of the indices using only the 'expressions' actor that is part of PTII. 
That sample workflow has been checked into CVS (and is attached to this 
email). You might want to try it out by downloading PTII4.0 from the 
Ptolemy site and then opening the file.

The actor calculates Richness-Margalef, Richness-Menhinick, Shannon's 
Index, Simpson's lamda, Hill's 3 'Nx' indices, and the 5 'Evenness' 
parameters from the document you provided.  The input is an array of 
number densities for a set of species. It is ridiculously simple in the 
example, but can be tied to any array of arbitrary size with some 
additional data prep actors. Some screen shots are attached.

This is an example of how someone can create an actor for a specific 
task without having to write any Java code. In particular, the 
'expressions' actor can be used to evaluate a wide variey of equations. 
Also, Composite actors are easier to work with in PTII 4.0 than before, 
so that one can easily create conceptual workflow (i.e. generic 
workflows without implelemtations by simply creating a set of 
unimplemented Composite actors. Adding ports to these Composite actors 
is much easier that it used to be,  as can be seen in the Port dialog 
shown below. Note that port names can be displayed, and some unused 
ports can be hidden.

Comments and corrections always appreciated!

Dan

Part of Main screen

Inside the Composite actor

Composite actor Port dialog

Deana Pennington wrote:

> Bill's list of biodiversity indices...
>
> Let me know which (if any) you have time to work on.
>
> Thanks,
> Deana
>
>
> -------- Original Message --------
> Subject:     Biodiversity indices update
> Date:     Tue, 25 May 2004 16:11:26 -0600
> From:     Bill Michener <wmichene at lternet.edu>
> To:     'Shawn Bowers' <bowers at sdsc.edu>, 'Bertram Ludaescher' 
> <ludaesch at sdsc.edu>, 'Deana Pennington' <dpennington at lternet.edu>, 
> 'Mark Schildhauer' <schild at nceas.ucsb.edu>, 'Kristin Vanderbilt' 
> <vanderbi at sevilleta.unm.edu>, 'Rich Williams' <rwilliams at nceas.ucsb.edu>
>
>
>
> Hi all,
>
> Thanks for the update Shawn!  I have made a first pass at compiling a 
> set of
> biodiversity indices.  Most of the common ones are included in the 
> document
> that I have attached.  I am still looking for the definitive list of
> indices, but have not found it yet.  Suffice it to say that several other
> indices will likely be added, but that this is a good start. 
> If you know of other sources for this info let me know.
>
> More as I come across them.
>
> Cheers,
>
> Bill
>
> -----Original Message-----
> From: Shawn Bowers [mailto:bowers at sdsc.edu] Sent: Monday, May 24, 2004 
> 3:50 PM
> To: Bertram Ludaescher; Deana Pennington; Bill Michener; Mark 
> Schildhauer;
> Kristin Vanderbilt; Rich Williams; Shawn Bowers
> Subject: Notes from the biodiversity breakout in Edinburgh
>
>
> Hi everyone,
>
> Here are the notes I took (converted from PPT to ASCII) from the 
> breakout session in Edinburgh concerning biodiversity.  There are 
> deliverables and dates from the meeting listed in the notes.
>
> Thanks,
> Shawn
>
>
> ---------------------------------
>
> Breakout: Dataset finding
>
> * Join on Space, Time, and Taxon
> * What property measured, joining on taxon
> * Infer co-occurrence (through data); to "fill-in" information (e.g.,
>   required parasitism)
> * Simple treatment of density vs. count div area (Matt/Mark)
>   - Taxon column; species name embedded as label
> * Current Garp/Beam Niche Modeling
>   - Issues are: occurrence data, resolve taxa;
>   - Deana: no integration problems currently in BEAM use case (Mark got
>     this in writing)
> * Mixing Biodiversity Data with Modeled Data
>   - Biodiversity indices
>   - Scaling problems
>   - Structure could vary widely across datasets
>   - This is our choice ...
>
>
> Semantic Registration of Biodiversity Data
>
> * Goals:
>   - Biodiversity ontology that knows about biodiversity indices
>   - Semantically annotate heterogeneous biodiversity data to the
>     ontology
>   - Do "deductive closure" (what follows from the ontology, and applies
>     to the dataset) of datasets based on registrations
>     * Property computed with other properties (expanded properties)
>     * Can I compute a biodiversity index for some of these datasets?
>     * Determine which indices are computable for one or more datasets
>     * Determine which datasets permit the computation of a particular
>       index
>     * Determine which datasets absolutely don't permit the computation
>       of a particular index - flag to say why
>     * Determine appropriateness for a particular purpose
>   - Data integration
>     * Taxonomic, distinct sites, scaling, ?; applying rules to compute
>       when integration is possible
>     * Explore additional applications (queries, e.g.), for annotated
>       datasets (what are the hard problems in Biodiversity data .)
>   - Keplerize the execution and reasoning components
>
>
> Semantic Registration of Biodiversity Data (final deliverables)
>
> * Deliverables
>   - Biodiversity indices (Bill; June 1st)
>     * 20 to 30, e.g., Shannon-wiener index
>   - Datasets: Biodiversity (Kristen, Mark; June 15th)
>     * Kristin survey LTER sites useful to compute indices; various
>       species (Kristen)
>   - NCEAS / PISCO (Mark)
>   - Ontology for indices (Rich; July 1st)
>     * Shawn will start annotations in parallel
>   - Semantic Annotations complete (Shawn; July 15th)
>   - Rules for when integration can occur (Mark et al.)
>     * Depends on the analytical goal/purpose . (the index)
>   - Develop inference/integration procedures (Shawn, Rich, Bertram; .)
>   - Figure out best way to put examples into Kepler
>     * not sure how to do this
>
>
>
>
>
>

-- 
---------------------------------
Dan Higgins
Software Developer
National Center for Ecological Analysis and Synthesis (NCEAS) 
735 State St. Rm 205 
Santa Barbara, CA 93101 
805-892-2531 
higgins at nceas.ucsb.edu 
---------------------------------

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mercury.nceas.ucsb.edu/kepler/pipermail/kepler-dev/attachments/20040712/60949b31/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: biodiv1.gif
Type: image/gif
Size: 30927 bytes
Desc: not available
URL: <http://mercury.nceas.ucsb.edu/kepler/pipermail/kepler-dev/attachments/20040712/60949b31/attachment.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: biodiv3.gif
Type: image/gif
Size: 31918 bytes
Desc: not available
URL: <http://mercury.nceas.ucsb.edu/kepler/pipermail/kepler-dev/attachments/20040712/60949b31/attachment-0001.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ports.gif
Type: image/gif
Size: 15852 bytes
Desc: not available
URL: <http://mercury.nceas.ucsb.edu/kepler/pipermail/kepler-dev/attachments/20040712/60949b31/attachment-0002.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: compositeBiodiversityExample.xml
Type: text/xml
Size: 41250 bytes
Desc: not available
URL: <http://mercury.nceas.ucsb.edu/kepler/pipermail/kepler-dev/attachments/20040712/60949b31/attachment.xml>