In this article I will present some links to the shapefiles, describe their internals, suggest areas of use, and finally describe how the files were made. This article is a rewrite/upgrade of an earlier article.

I should perhaps mention that the main reason why I do this work with QDGC-files is not that I am obsessive about leaving my computer to crunch data in the afternoons. The motivation for producing QDGC shapefiles and facilitating for downloads, conversions, use of and general knowledge about the QDGC concept, is my firm belief in the use of this standard could serve as a building stone for ecological monitoring.

CDs with the shapefiles are available at the Conservation Resource Center or by request to Ragnvald Larsen.

Table of contents



QDGC Shapefiles

The file gallery with the QDGC files is available in the link below. New files will be added to this gallery. Suggestions to make qdgc-files for other countries will be made upon request - given that I have spare time to do the work needed.

I should perhaps mention that the compression system used for some of the files gives a compression rate of up to 1:45. So the qdgc_04_america_s.7z file with the modest size of 6.79 Mb ends up being 289 Mb. The 7-Zip compression/decompression program is a free tool and available for download at this website:

Shapefiles relevant for Africa

The shapefiles are available on the TZGISUG website for download. The direct links are here:

Shapefiles relevant for others

Using QDGC is best for areas from around 60 degrees north and south of equator. Above or below the squares tend to be elongated. The areas calculated are still correct. But for use aimed at visualization one might be better served with squares. One possibility is to divide the squares into two above or below a certain latitude. To do this we will have to do some changes to the QDGC nomenclature. But the whole thing is doable. Just ask.

Technical details

The shapefiles have the following attributes:
  • LON
  • LAT
  • AREA
  • QDGC

The longitude/latitude values are shape center point values in decimal degrees. All area values has been calculated in square kilometers based on Albers projection and Ecker VI (for the QDGC world data set). QDGC is calculated based on the level and according to the QDGC standard as described her:


Areas of use

Lets be honest. The QDGC standard is not THE columbi egg of ecological monitoring. Most available data tends to have it's origins in point observations. The most valuable data around are point based data! BUT data tends to be collected by researchers with funding where the continuing support relies on them proving that they are able and effective. If you give away your raw data - you are not very smart.

Using QDGC as a basis for exchanging data might therefore be useful for all of us. The researcher gets to keep his/her raw data, and the research community get's access to important building stones of higher understanding. At least that's the theory.

Let me give you an example. Three researchers (and/or even offices) own different data sets. The data sets are detailed and there might be good legal reasons not to share the datasets. So what do they do? They export generalizations of their data sets using the QDGC references. Given access to the other entitys data each of them can then build models based on other datasets.



Maybe, if the QDGC level is high enough will this lead to good conclusions. If the level is low (the data resolution low) this could at least give the researchers an idea to work on.

In the above example the combined plus-zones indicate habitable areas for a species. This is something we can infer from the observations. Based on these three pieces of information we can build a habitat model. This is of course horribly simplified, but the concept should be evident. The recently published Bird Atlas of Uganda (Carswell, M., Pomeroy, D., et al. 2005. The Bird Atlas of Uganda) uses a method similar to the above, but faar more detailed.

Other uses are:
  • Make maps for presenting your data
  • Calculate density values
  • Make survey intensity maps
  • Interface data with other projects through QDGC.


Description of the process


To make QDGC fishnet we need to use a tool. ArcGIS does not handle this without some external help. XTools provides this tool and a lot of others. You can download a test-version of XTools from their website:

We start the XTools program and select Create Fishnet from the menu:



This opens a fairly complex dialog window. The extent is set either by using a current layer as source, or we could set the extent manually. The extent for all QDGC in the Tanzania series is left: 29, right: 41, top: -1 and bottom -12. 12 full degrees east-west and 11 full degrees north-south. The rows and columns are where we decide how many squares we will end up with. For QDGC we need to keep to the rule for such squares. The following table describes the number of squares (and rows/columns) per degree square:

LevelColumns/RowsSquaresExample
0 11 S01E010
12 4 S01E010A
24 16 S01E010AD
38 64 S01E010ADC
416 256 S01E010ADCB
532 1024 S01E010ADCBD
664 4096 S01E010ADCDBA


Remove the index columns alternatives. They are contextual based on the extent of the fishnet, and as such not relevant in most cases where we will use the QDGC shapefiles. In the bottom of the dialogue we fill inn the output feature class:



Next thing we need to do is to make center points LON LAT columns for all the squares in the shapefile. This is also done using XTools. Then we calculate the area, just in case someone would like to use the shapefiles to measure density. Again we are using XTools. The area of the shapefiles varies somewhat. Remember that the further north or south the more elongated will the squares become. The area is calculated by using a projection which is also used for QDGC-files covering all of Africa. The reason for not usinga local projection (which would give a fraction more accurate area) is that we would like the squares used in the different parts of Africa to be compatible. If you need to use a more precise, or for some reason a different projection, it is easy to change recalculate the area value. Which brings me to the procedures.

Find center point and assign it to columns LON and LAT:



Calculate area in square kilometers:



Before leaving ArcGIS we add the field QDGC in the attribute table. Field type is text, and length is set to 100. It is left empty:



ArcGIS leaves the .dbf-file with the data type F for LON, LAT and AREA. The Python library I am using for reading .dbf-files does not accept data type F. The .dbf-file is thus opened in a hex editor and data type is manually changed from F to N. Seems like this doesn't harm the data as such.



Then lets move on to the core of the matter - the QDGC string. We need to calculate and insert QDGC for all rows in the shapefile based on the LON/LAT values. As earlier stated this used to be a process involving a web-server, a PHP script, MS Access and a lot of cutting and pasting. The method was practically useless for shapefiles with more than 50.000 objects (squares).

The new method is to use a python script which accepts the .dbf-file as input. A value for the appropriate level is set in the script. Then it is run, and after some minutes the QDGC values are in place. I plan to publish the script as an ArcGIS toolbox thus making it possible for anyone to to the procedure themselves. For now you may download the tool and run it as a command line tool:

The script is currently run through a developers tool called Wings:



You may read more about using the tool on this web page:


Tool help text:
NAME
Countless Calculate DBF Conversion Utility - Version 0.5

FILE
lonlat2qdgc.exe

DESCRIPTION
The program accepts a .dbf file as input and modifies the
contents of the file. LON and LAT column values are used
to generate a QDGC string on a level determined by the
users input.

Read more here:
http://www.mindland.com/countless/

No warranties. Use at own risk. Do not use on your original
files without having a backup.

SYNTAX
lonlat2qdgc drive:pathfilename QDGC level

drive:pathfilename input .dbf-file to be calculated

QDGC level Quarter Degree Grid Cell level to
be calculated. Read more on QDGC
levels at Wikipedia:
http://en.wikipedia.org/wiki/QDGC

The input .dbf-file should have the following columns:

LON - FLOAT
LAT - FLOAT
QDGC - STRING

Any values in the QDGC column will be overwritten with the
new QDGC value.



Conclusions

I have now made QDGC shapefiles available for several important areas. To make the system more available on a national scale subsets of the continental scale files needs has to be made. This involves a lot of work with ArcGIS. It should be fairly easy for the users on a national scale to do this, but requests for help or consultancy will be handled accordingly.

Other specialized QDGC shapefile sets are available upon request. Please inquire by dropping me an email.