reproject changing a dataset from one map projection (or measurement system) to another
Back in Chapter 2, we discussed the concept of having to transform data measured from different datums to a common datum so that everything would align correctly. It’s best to reproject your data (or transform all of the datasets to conform one to a common datum, coordinate system, and projection) before moving forward. Reprojecting data will translate the coordinates, map projections, and measurements from one system (for instance, SPCS) to another (such as UTM). Many geospatial software packages give you the capability to take your initial dataset (like your street map in State Plane NAD27) and reproject it to create a new dataset with new coordinates (so now you have a new street map measured in UTM NAD83). This can also be done to change data from one map projection into an other. In some cases, the software can “project on the fly”—calculate the necessary transformations without actually altering the base dataset. Whatever method you choose, you’ll save yourself a lot of headaches later on by using a common basis for the various geospatial datasets you use. This is especially important when you’re trying to get several different types of data to match up.
59
What Happens When Measurements Don’t Match Up?
It’s not uncommon to have multiple data sources as in the scenario described above, with different projections, coordinate systems, and datums that all need to be used together in a project. If everything doesn’t align properly, you can end up with points being plotted in the wrong location with respect to the base map they’re being located on, or street data that doesn’t line up with a satellite image of the same area. What kind of effects could this type of simple data mismatch have on real-world companies or government agencies? What if a road renovation crew is using sewer data and street data that don’t align properly because they use different datums or projections? What effects could this have? What could happen if the misalignment of data caused one dataset to be significantly off from its proper place on a map? Similarly, what effects could such misalignments have on projects like shoreline construction, zoning maps, or species habitat mapping?
60
spatial reference the use of a real-world coordinate system for identifying locations
Problems are going to arise if one (or more) of your pieces of data doesn’t have coordinates assigned to it, or if you don’t have any information as to what projection or datum was used to construct it. Suppose you scan an old map of your town into your computer, and you want to add other layers to it (like a road network or parcel layer) in a GIS to see how things have changed since that old map was made. You add the layers into the GIS, but the scanned historic map won’t match up with any of your data. This is because that image doesn’t contain any spatial reference (like a real-world coordinate system). Although it might show features like old roads or the dimensions of particular buildings, the only coordinate system the old map possesses is a grid that begins at a zero point at the lower left-hand corner and extends to the dimensions of the scanned image. For instance, coordinates would not be referenced by feet or degrees, but instead would be in whatever units you’re measuring with (such as 200 millimeters over from the left-hand corner and 40 millimeters up from the left-hand corner).
Items in the image can’t be referenced by latitude and longitude because those types of coordinates haven’t been assigned to anything in the image. Similarly, when you add this scanned image as a layer in a GIS, the software won’t know how to match it up with other data layers that do have referenced coordinate systems. Without knowing “where” your scanned image matches up with the rest of the data, you won’t be able to use this image as geospatial data or reproject it to match your other data. The same concept holds for other types of data used with geospatial technology: If all types of data automatically had the correct spatial reference when they were created, things would be a lot easier. So what you’ll have to do is match the data to the map, or assign a real-world set of coordinates to your data so it can synch up with other geospatial data. In other words, you’re going to have to georeference your data.