This road map was a request by the Ghana Institute of Horticulture to members of the Canadian Agricultural Rural Extension Society (C.A.R.E.S.). Commercial maps in rural Africa tend not to include smaller villages, and so road maps for rural areas are typically hand drawn and localized. This kind of map is insufficient for the practicalities of monitoring multiple project sites over more than one district.
We had previously extracted the geographical co-ordinates for small villages in Lake Zone, Tanzania by bringing a hand-held GPS into the field, but at the time of the above request there was no funding available for this kind of venture. So we improvised. Many of the places to be mapped had co-ordinates available in digital format online, and those that were not available in this way could be identified on Google Earth in terms of proximity: “X miles south-east of Y village along public road number Z”. A “thumbtack” could be placed on the Google Earth map, converted to a digital format and then brought into the map. Of course, feedback by people living in the area was necessary to ensure that the placements were accurate.
The road data retrieved from an online source was an almost exact match with the satellite data on Google Earth – only a small amount of updating was necessary. More up-to-date road information on a local scale can be acquired later by means of a hand-held GPS.
As with the previous map, this one was also created using QuantumGIS, a free, open-source mapping application. Open source software is always a work in progress, but much more financially accessible to locals in Ghana than the commercial GIS software in widespread use across North America.
The Shuttle Radar Topography Mission (SRTM) created a database of elevation data for most of the globe and was part of an 11-day mission of the Space Shuttle Endeavor that went into orbit in February, 2000. Thanks to this data made freely available online, topographic data is available for developing countries where on-the-ground surveying is often not done. This is a topographical map done for the Ghana Institute of Horticulture in partnership with the Canadian Agricultural Rural Extension Society for project sites in Upper West Ghana.
I am working with an open source program called QuantumGIS, which has many great features but is still a work in progress. Work-arounds needed to be done in order to have the legend and scale bar visible in the map layout, but I expect these problems will be resolved in future versions. Training in open source GIS would allow locals in developing countries to create their own maps without the oppressive price tag of commercial software.
(Click on the map for a full-sized version, or on the link above for a more detailed pdf.)
The Canadian Fossil Discovery Center is currently located in Morden, Manitoba. This part of Manitoba used to be under water many thousands of years ago; this museum actually holds the largest collection of marine reptile fossils in Canada, including “Bruce”, a massive 43 foot Mosasaur. School children are invited to go on fossil digs in the area. It’s a great travel destination for anyone going exploring Manitoba (http://www.discoverfossils.com/).
The museum is looking at a location change, which becomes an educational opportunity for those of us interested in surveying. The five of us in the GIS program at Assiniboine Community College produced the elevation map shown above on a proposed site just outside Morden. We used a survey-grade GPS system (a base station and rover) and an ATV.
The process was as follows: we first set up the base station in an open area so it would have uninterrupted access to the sky (and the GPS satellites). The rover was another receiver that was mounted on the ATV, which traveled up and down the field. Picking up a correction signal in real time from the base station, the rover took and stored accurate elevation points for the survey area. These points were then downloaded into ESRI’s ArcGIS software, which computed an estimated shape of the entire field.
(Click on the map for a full-sized version, or on the link above for a more detailed pdf.)
This is again a fictional map, showing part of the water main network in Brandon, along with the most current sites of breakage. Not all water main materials are created equal; some types are more susceptible to breakage than others. This map displays the data for further analysis.
(Click on the map for a full-sized version, or on the link below for a more detailed pdf.)
This map is pretty self-explanatory; it is a drawing of two city blocks in Brandon, Manitoba along with their taxation classes. The individual land parcels were drawn, guided by legal descriptions alone. Also of note is the area of “No Data” – you can only map the information you have been given.
(Click on the map for a full-sized version, or on the link above for a more detailed pdf.)
This is a drawing of the registered plan for Butler, Manitoba. If you were to go to the site of this drawing, either physically or on Google Earth, you would find nothing there now, except for the roads and the rail line, which appear to be still in use.
This drawing was created using ESRI’s ArcEditor, with orthophoto from the Manitoba Land Initiatives website as background.
(Click on the map for a full sized version, or on the link above for a more detailed pdf.)
This map is a fictional study of statistics as it relates to duck nests. Each nest location in the study area was marked and the number of eggs in each nest noted. With that data in hand, the question is, is the number of eggs in each nest really an indicator of duck suitability for a particular place, or will ducks lay a certain number of eggs that is appropriate for ducks and nest where it’s most hospitable? I concluded that the type of land cover created a hospitable environment for ducks to nest and that the number of eggs was a relative constant.
Then, using a density analysis in ArcGIS, a probability map was made based on the frequency of duck nests in the study area. The majority of the duck nests were found in two quarter sections: SE-19-14-17W1 and SW 20-14-17W1, most likely because of the combination of grassland, forest, marsh and agricultural land that provides an ideal habitat for them (see blog for May 25).
(Click on the map for a full-sized version, or on the above link for a more detailed pdf.).
Information about the way land is used in an area is an invaluable resource for those who work in environmental management. The land use in these two quarter-sections is being evaluated in terms of its ability to support duck populations. Note that some of the land is farmed, but much of it is left to its native landscape of grassland, forest and marsh, great habitat for all kinds of wildlife, ducks included.
(Click on the maps for a full-sized version, or on the above links for a more detailed pdf.)
The data for this example comes from the course “Introduction to ArcGIS 9 Geostatistical Analyst” by ESRI and Jay Ver Hoef, Ph.D., available on http://training.esri.com.
This study (most likely fictitious) of ozone levels in California comes to us in the form of a set of points linked to a coordinate. That point’s ozone level in parts per million is the recorded data for that particular point. When this information is loaded into GIS software, in this case, ESRI’s ArcGIS, one can see how the points are related in space, and estimates can be made of how the entire state of California measures up for ozone levels.
To estimate the ozone levels for all of California based on the ozone levels at particular points throughout the state is to use a technique called “interpolation”. There are many different kinds of interpolation methods, all of them derived from statistics. The definitions of all the methods employed here are shown on the maps, but you might ask, ‘Which one is the most accurate?’ That will depend on the available data and which method fits the data best.
(Click on the map for a full-sized version, or on the link above for a more detailed pdf.)
A Regional Health Authority in Manitoba is a division of the province, related to population, that is responsible for administering health care to its residents. For the purposes of this map, I have included only the RHA’s that are involved in the care of those residents that were subjected to above-average West Nile Virus exposure in 2007 (based on fictitious data, as we’ve talked about). Mosquito trap counts and confirmed cases of West Nile Virus are documented by RHA; you can look at the most current reports online at http://www.manitoba.ca/health/wnv/
Alanna's GIS Blog 