[udig-devel] Seeking more direction from the uDIG community

[Murray Richardson <murray.richardson@xxxxxxxxxxx>]

Hello Community,

Thanks for previous help you've offered.  I could use a bit more 
direction if possible to determine whether I should carry on with uDIG 
for a small project I am working on.  I'm finding the uDIG development 
learning curve to be a bit steep so far.

I've spent the last few days looking at uDIG (e.g. plugin tutorial and 
all the other online resources) and I am trying to figure out the 
simplest way to make use of powerful postgreSQL/postGIS capabilities. 
Previously I was concerned with constructing spatial queries on massive 
point data sets.  These processes are easy if I use a standard 
postgreSQL interface like psql or pgAdmin.  It seems like it should be a 
fairly simple task to construct plugins within uDIG that make use of 
these and other RDBMS capabilities through JDBC (if they don't already 
exist somewhere).  I read an entry on Cory Horner's community page about 
extending uDIG functionality to incorporate better RDBMS capabilities. I 
am most definitely interested in this because I am always faced with 
problems where this type of capability would be an asset.

To give you an idea of the type of functionality I am interested in, 
considering the following two examples:

(1) I am constantly using GIS and RDBMS capabilities separately (loosely 
coupled fashion)for information management in the field of hydrology. 
For example - I need to summarize the joint probability distribution of 
land-use and soil types to determine an average (lumped) runoff 
coefficient for nested watersheds (since runoff is a function of landuse 
and underlying soil properties, among other things).  I do this by 
exporting the polygon intersection of a soils layer, land-use layer and 
watershed layer to a RDBMS.  Then I use lookup tables to determine 
coefficients for individual landuse-soil combinations and use an 
aggregate query to determine the mean value for each watershed.  It is 
nice to then join the mean value back to the watershed layer and display 
this in a map.  I also use the RDBMS to do reporting of the individual 
intersected polygons by watershed.  This type of analysis and reporting 
is standard in flood risk assessment and could be very easily 
streamlined with a tool like uDIG and postgreSQL.

(2) Storing and accessing spatial-temporal hydrology data is an emerging 
area of interest (consider the ESRI ArcHydro framework if you're 
familiar with it) and an interesting direction for Open Source GIS. 
For example, I would like to be able to store decades worth of hourly 
streamflow data for numerous gauged weirs within a postgreSQL DB and 
have this linked to a uDIG application that allows me to extract the 
streamflow data by (e.g.) clicking on a point station and entering a 
date.  The query result could be just exported to a text file for now. 
This capability would open the door to a whole range of interesting 
spatial-temporal hydrology applications (e.g. recession analysis, 
climate and streamflow data interpolation, hydrologic model 
parameterization).

Based on these two particular examples, could someone perhaps direct me 
on how to proceed to develop a very simple proof of concept for enhanced 
RDBMS capabilities within a GIS environment?  Consider example 1 for 
instance - can I create a uDIG operation that is accessed by 
right-clicking on 3 highlighted layers (like the difference operation 
that is available when two layers are selected).  For my purposes I 
would like these to be postGIS layers.  The operation could then execute 
the postGIS intersection procedure, the attribute join and the aggregate 
function all within the postgreSQL service, and then map the results 
back to the watershed layer for display in uDIG?

I know this is a fairly specific question, but what it boils down to is, 
how do I access postGIS spatial operations as well as general postgreSQL 
RDBMS functionality and non-spatial tables within the uDIG environment?

Thanks in advance for any help.  This project is for a small Open 
Source/Open Access Student Experience program grant I am working on 
through University of Toronto that involves developing a 
proof-of-concept idea and some educational tutorials using 
postgreSQL/postGIS.  I would like to eventually be able to post the 
results on my own uDIG community page if things go well. I think this 
type of work could help open up uDIG capabilites to the slightly less 
technical users like myself (e.g. GIS and basic programming/RDBMS skills 
experience, but still need lots of examples and explicit documentation 
to follow!).

Look forward to hearing back from someone - hope it's not too much to 
ask over a mailing list!

--
Murray Richardson, M.Sc
PhD Candidate
University of Toronto
murray.richardson@xxxxxxxxxxx