Have you ever imported data that doesn’t quite line up how you’d expect? It may be that you’ve fallen victim to a common workflow error when importing GIS data. Some file types such as CSV can be used for GIS data but don’t contain coordinate system information. When you are importing data from this format, you first have to define the correct coordinate system.
In this example, we’re going to look at the common mistakes people make and how to avoid them. We’ll start with a world map in the Robinson projection.
We have a CSV file containing points for large cities that we’d like to add to the map. We know from our data source that the CSV uses the WGS84 coordinate system. After selecting the file for import, the MAPublisher Import dialog box helpfully notifies us that some required settings are missing. We’ll click the blue ‘Required settings are missing’ link to continue.
Setting up the import, the coordinate column settings are easy since we have an X_COLUMN and a Y_COLUMN, but we can’t forget to check that the format is correct. The default is Projected units, but we know the file uses WGS84, and can tell by the numbers in the column that the coordinates are in decimal degrees, so we’ll change the format to reflect this information and click OK.
Back at the import window, we see the message ‘Data loaded successfully’. Great! Let’s click OK and add the large cities to the map.
The data has been imported but the result isn’t what we expected. The new layer has been added to a new MAP View, so let’s try dragging it into the Robinson MAP View with the world map.
We get a prompt saying that there isn’t any coordinate system information. We want it to be in Robinson like the rest of the map so we’re going to leave the default setting of Same as: Robinson.
The data has moved, but it still doesn’t look like we were expecting. Where did we go wrong here?
There are actually two places in the workflow where we could have avoided this common mistake. When we dragged the point layer into the Robinson MAP View, the pop-up dialog box prompted that a coordinate system wasn’t specified. We specified Same as: Robinson, thinking this was the correct choice, but we had already determined during import that the CSV was in WGS84. What we should have done here was to specify the coordinate system as WGS84.
The other place where we could have avoided this error was right after setting up the CSV file for import. In MAPublisher 9.4, there’s a new button on the Import dialog box that allows you to see more detailed information about files being imported. By clicking the Advanced button in the Import dialog box, we would have noticed that there was no coordinate system specified.
Even here, it might have been tempting to choose Same as: Robinson to add it to the Robinson MAP View, but this would import the points exactly the same as before – all in one location in the middle of the map. Instead, what we want to do is click the blue ‘No Coordinate System Specified’ link and choose WGS84. After this is set up, we’ll click OK to add the data to the map.
The data still isn’t quite right – it looks the same as when we first imported it. But again we notice that it has been imported into a new MAP View, so we’re going to drag the layer into the Robinson MAP View and see what happens.
Perfect! By assigning the correct coordinate system to the data during import, the points have been imported correctly!
Mistakes during data import are common amongst GIS users, especially those who are just starting out. In the first scenario, when we imported the CSV and added the data directly to the Robinson MAP View, we thought we were telling MAPublisher that we’d like it to match up with the map. What we really did was tell MAPublisher that the data was already in the Robinson projection, even though we knew it was in WGS84. What we should have done first was to define the projection by telling MAPublisher what coordinate system the data is already using. Once MAPublisher knows what system the data is starting in, we can then ask it to project or transform the data into the coordinate system that we’d like to use.
When working with data that doesn’t have coordinate systems already defined, it is very important to follow the workflow in the correct order to avoid frustration when the data doesn’t line up as expected. Always check your sources when using data that isn’t defined, and make sure you’re assigning the correct coordinate system before performing any transformations or projections.