Question: Is there a landcover class definition somewhere? i.e what's the difference between High producing exotic grassland and low producing grassland? If there is area that was high producing exotic grassland and is now covered with 50% manuka but is still grazed, does that make it low producing grassland?

Answer: We've put links to LCDB2's Illustrated Guide to Target Classes on the Collaborator's page. LCDB3 is still adhering to the LCDB2 class definitions where applicable so hopefully this will address the first part of your question. Obviously, pasture species are not evident without field inspection so you have to go on other visual cues like vigour, intensity of fencing, presence/absence of weeds and scrub etc. 50% ground cover of manuka would not be characteristic of High producing exotic grassland so its probably justifiable to revert such areas to Low producing or even Manuka/kanuka shrubland. Note the land cover group, to which Manuka/kanuka belongs, is described as Scrub or Shrubland with the latter term implying an open canopy association of shrub and herbaceous covers.

Question: Do you have a land cover class change table or other summary statistics for <region>? 

Answer: We don’t have figures for individual regions but have provided a spread sheet table showing class areas for the whole country at 2001 and 2008 and the changes between them. You can download this from our portal at http://lris.scinfo.org.nz/#/file/330-lcdb-cover-class-change-summary/.

 

However it is relatively easy to produce this table or similar summary statistics for a region, district or other sub-area of New Zealand. The instructions below show how the table referred to above was generated. No doubt you could also generate a simpler table, for instance just listing the total areas for the three years, by manipulating the data within Excel.

 

The easiest approach to this exercise is as follows:

 

1)    download the cropped LCDBv3 data as a shapefile from the LRIS Portal (http://lris.scinfo.org.nz/#/layer/304-lcdb-v30-land-cover-database-version-3/). Note you crop to a regional, district or other boundary before downloading by using the crop button at the bottom of the page.

2)    In ArcGIS (or equivalent) add a “hectares” column to the downloaded shapefile’s attribute table

3)    In ArcGIS right click on the column name and “calculate geometry” to update the areas in the table (you have to do this because the clip may have cut polygons along your regional boundary – and in fact this is why we don’t have an area column in our tables – to force users to update areas).

4)    Exit ArcGIS

5)    With shapefiles the attribute table is stored in .DBF format which you can now open with Excel (look for <shapefilename>.dbf).

6)    Immediately “save as” to make a working copy in Excel format (this is to avoid corrupting the .DBF file for use with the GIS)

7)    Click on the insert tab right of the home tab (top left of Excel window)

8)    Left-most icon is “PivotTable” – click on this and just confirm that the Table/Range covers all data in your file and the pivot table is going into a new worksheet and click OK

9)    You should now see and empty worksheet but with a “pivot table field list” menu box.

10)  Drag LCDB2 field into the row labels box (all unique codes should appear down the left column of the worksheet)

11)  Drag LCDB3 field into the column labels box (all unique codes should appear across the top row of the worksheet)

12)  Drag the hectares field into the values box (lower right – and all of the sums of hectares should appear in the body of your pivot table)

 

The diagonal of cells from top left to bottom right should contain the areas that are unchanged – that is they have the same code in LCDB2 and LCDB3.  Anything off that diagonal has changed – and you can see from-what-to-what by tracking back left and up from a cell to see the codes that it matches.  Some things like plantation forest will show quite a bit of change – other things should show very little or none.

 

Note that you can drag things out of the boxes and change the summary as well – so this is quite a powerful tool.  

 

We have done this exercise for the whole country – so the size of the table for a single region certainly shouldn’t crash Excel.  

 

Note also you do need to be able to recalculate areas for your district LCDB clip – so if you don’t personally have access to ArcGIS or some other GIS, you may need help from your GIS staff for steps 2–4. It is trivial – a 2 minute job – provided you have access to ArcGIS or equivalent.