"Seegrid will be due for a migration to confluence on the 1st of August. Any update on or after the 1st of August will NOT be migrated"

Project Scenarios


Contents

Related pages


I've added another page ProjectScenarioImplementations to deal with the detail of client requests for data and what SQL will need to cope with in terms of placeholders etc.

-- StuartGirvan - 11 Nov 2004

Note: The XML schemas used by this project are developed as part of the XMML. You will find cross-links where appropriate. Please try to add content with this in mind.

For discussion of reltive burden between client and server, number of transactions, etc, see MCAGeochemistryExamples.

Implementation will focus (initially at least) on the XMML measurement/specimen feature types.

PIRSA scenarios

Where do people start?
  • Spatial query or
  • analyte view

Scenario 1

Steps:
  1. Focus on small spatial area (eg. Exploration tenement)
  2. Select all geochemistry
  3. Subset to 1 analyte (eg. gold or silver) or (more likely) a combination of analytes
  4. Select analyte value for detail

  • Small area/all geochem "screenshot":
    GeochemScenario1Report.png

Questions

  1. How do we get to the "small" area- and how do we know its small enough? We can use a tenement, but that in turn implies we have a source of info about tenements. This will need to be a WMS that returns a tenement object including geometry as a GetFeatureInfo response I think - or a WFS and extra work creating style sheets etc.
  2. its not obvious to me which panel is created when - there seems to be a dependency - can we create a more step by step storyboard and specify excatly which queries against which feature type is required for each step. I assume that the two panels relate to different screens - or are they panels that interact on the same screen.
  3. What are you going to do with all geomchemistry - it seems that you really only want a summary of the data, per analyte and its probably best to do that with a backend query and new feature type
  4. Does selecting the site in the option list change the body of the screen on the left?
  5. is it going to be meaningful to list sites by name in a drop down list, or will we need a report the user can browse?

Issues to explore
  1. if we are using a tenement, we may need to have a WFS for the tenements so we can extract the polygon boundary and then send this to the WFS. We will need to test the WFS against filters other than BBOX - should be OK for oracle spatial - but will have to check how much of the plumbing is in place and working for geometryless (so far we extract the BBOX to do the first query and then it should refine the query in memory using the full geometry - but I havent yet tested this.
  2. we should quantify min/max quantities of info for each step

Scenario 2

looking for good analyte values

Steps:
  1. Start with broader geographic area (eg WA)
  2. Looking for 1 analyte with more than 50ppm - want to see scaled symbols (glyphs) (eg. "select gold values that are greater than 50 ppm?" - analyte unit handling issues)
  3. Select analyte value for detail

  • Broader area/single analyte "screenshot":
    GeochemScenario2Report.png

In general:
  1. Spatial Query
  2. Filter by geochemistry properties
  3. Select individual detail or report on collection (or XMML dump of entire collection)

Handling of multiple analytes portrayed on a single point location - over lap
  • much of the data is downhole data so is 3D but will be portrayed in 2D so a lot of overlap will occur


Generalised screenshot storyboard

All, my attempted visualised storyboard is attached. There are several things to note:

  • This is just the mock up please do not get hung up on style
  • Do highlight and place here any essential steps that you think might have been missed out
  • We need to keep the scenarios relatively simple
  • We probably need to highlight or distinguish between which data points are from which WFS onscreen because jurisdictions will be keen that their contirbutions are recognised.
  • We need to decide what backdrop WMS layers we'll need to pretty things up (I've started a list below).

GA will probably set up a feature type for distinct Analytes to be used by Web Map Composer for creating the filter query, but if any of the other participants are feeling keen please feel free to jump in.

GeochemScenarios.ppt

This has now been updated to better reflect scenario 1 described above. Scenario 2 described above needs further work by whoever wrote it.

-- StuartGirvan - 29 Oct 2004

Here's my first attempt at suggestd WMS backdrop layers:

  • Coastlines and State Borders
  • Geology (what kind by age and/or rocktype??)
  • Geological Structures

There are a bunch of others that GA can provide here http://www.ga.gov.au/map/broker/opengis.jsp

-- StuartGirvan - 29 Oct 2004

The download formats are controlled by which XSLT transformations we develop. Doing an Excel spreadsheet is probably a bit to much effort but doing say, CSV, which Excel can read, is relatively trivial (probably easier than formatted text). Once the first stylesheet is done the others should follow with almost the same structure. We can decide whether or not to pursue a few variations beyond the necessary ones at that point.

- RobertWoodcock - 29 Oct 2004

Strawman Storyboard (probably obselete)

For an effective demo to interested parties, we will need a fairly tight storyboard which illustrates the system capabilities in a way that the audience can relate to. This will define a more detailed set of requirements.

Here goes (this is fantasy - I have never been an explorer - please change this into something more plausible):

Explorer wants to assess an area of interest for Platinum anomalies
Operation implementation 1 implementation 2
Get results of measurements of Pt concentration in area of interest, visually inspect spatial variation Get point coverage of Pt from server; on client, if >200 samples, contour and display as coloured-map, if <200 plot individual colour-coded points Get normalised data from server; on client, process data to extract Pt results from Measurements, and associate these with the location of the associated Specimens; portray as in implementation 1
Identify an anomaly, determine subset of samples in proximity    
For subset, verify that the analytical procedure is consistent    
For subset, inspect spatial variation of other analytes    
For subset, plot Pt results against Cu results    
For a selected specimen, tabulate all analyte values    

to be continued ...
Topic revision: r19 - 15 Oct 2010, UnknownUser
 

Current license: All material on this collaboration platform is licensed under a Creative Commons Attribution 3.0 Australia Licence (CC BY 3.0).