Cartographica's second calling, behind importing and displaying data, is analysis of that data. There are many options in the Tools menu for performing analysis and we will be adding more as time goes on.
There are two basic kinds of analysis functions, grouped by the resultant type of data. Analysis that result in a raster layer provide smooth visual output and selectable color maps, etc. Analysis that result in either new vector layers or modifications to existing vector layers provide high levels of scalability and are often more useful for further analysis. Some functions may be able to output either vector or raster data.
Kernel Density Maps are a way to look at the relative density of point data as spread over an area. The Cartographica Kernel Density Map option uses a highly-accurate, but somewhat inefficient relative distance Kernel for its analysis. By default, the kernel density analysis uses a set of "reasonable" parameters, however an advanced version of the feature is available by holding down the Option key while selecting the menu item which includes a number of tunable parameters.
To generate a Kernel Density Map, select a single point layer in your mapset and then select the Make Kernel Density Map item from the Tools menu. After a minute or two (depending on the size of your source data set and current underlying map), a new layer will be added to the mapset which shows the kernel density analysis. Depending on your particular data, you may find it helpful to either move this layer to the bottom of your layer stack, or use the Styles... item to change the Fill Color to be more transparent (on layers that have other coloring in them, the color is ignored, but the transparency is used
In the advanced version (which is used when you hold the Option key while selecting the menu item), you can change the various parameters that are used by Cartographica in performing the analysis. An explanation of these options follows.
Kernel Density Analysis Parameters
The analysis is performed using a grid of cells spread evenly across the field to be analysed. By setting the count of the grid cells, you can determine the courseness of the grid that is used. To change this, either choose the width and height in number of cells, or in map units (here in the example, it's in degrees, but if your map layer's CRS is set to feet or meters, the parameter will be in those units). If the Keep Proportionate check box is checked, it will tune the width to match the height you enter, or vice-versa. If you'd like a disproportionate grid, then uncheck the box.
The grid itself may be defined as having its extent over the layer, the map, or the visible map area. Map Bounds and Visible Map Area differ in whether the current margin is included in the visual area of the analysis.
The last 4 parameters change how the calculations are done. By selecting a column name for intensity and/or weight, you can change how Cartographica interprets a point. Without changing these parameters, the defaults are to use an intensity and weight of 1.0, which gives each point equal weight. However, if your data contains things like the number of incidents or sightings, you may want to set the intensity column to match that count column in your data. The bandwidth value determines the decay of the value from the central point. If the bandwidth is set to 8, then little or no effect will be generated by the 8th cell from the point. If you set this number larger, the effect will spread out to a wider radius. If it's set smaller, it will have less effect.
Merge Raster Layers is used when operating on raster data (such as satellite imagery from Landsat) that is split into separate layers for each color component. When using this data to show visible light, the layers (usually Red, Green, and Blue layers) must be merged into a single layer. Cartographica can do this automatically (using the Merge Selected Raster Layers item from the Tools menu) or manually, by selecting Merge Raster Layers... from the Tools menu (which appears when the Option key is held while pulling down the menu).When working automatically, Cartographica takes the selected layers in the order of the layer stack as Red, Green, Blue, and optionally alpha. When using the manual version, you have control over which layers to use, which can be helpful in re-arranging layers or in creating effects by using the same layer in more than one color band.
Merge Raster Layer... options
For each layer that you want to use, select a Raster Layer from the pop-up. Each layer may be selected zero or more times. The other control here is the Remove used layers control, which determines what to do with the selected layers when it is done merging them. If selected, the layers are deleted, if not, they are left in place.
Convex Hulls are geometric shapes that represent a completely convex "hull" or wrap around a set of data. Convex Hulls in Cartographica create vector polygon layers. If you select multiple objects, Cartographica will generate the Convex Hull around all of the objects as a set. If you instead select a layer with no selected items, the entirety of the occupied layer will be used as points to create the hull.
This function adds coordinate columns (X, Y, and optionally Z and M) from the selected point layer. These data are added to your data table.
This function adds a column for the area of the selected polygon layers. These data are added to your data table.
This function adds a column for the area of the selected polygon or line layers. These data are added to your data table. In the case of polygon layers, Cartographica uses the total length of all line segments in the polygons. For line layers, the data is the length of all line segments in each line.
This function adds columns for the coordinates of the midpoints of selected line layers. These data are added to your data table. Cartographica uses the midpoint based on the distance of all line segments in each line.
This function adds columns for the coordinates of the centroids of the selected polygon layers. These data are added to your data table. Cartographica calculates the centroid based on solid areas in the map, so if your polygons have holes, those will be removed when calculating the centroid.
This function adds a column to the selected polygon layer representing the count of points from another layer that are within the polygon. To use this function, select the polygon layer that will act as a boundary and then select the Count Points in Polygons... item from the Tools menu. Cartographica will prompt you for a layer to count the points from and a name for the column that will be created in your polygon layer.
Another form of analysis that Cartographica performs is Geocoding, or GEOgraphic enCODING of data that isn't native GIS data. Although similar in some ways to the process of importing other data, there are two kinds of data that we load in separately: Address information and photos.
Address geocoding is one part art, one part science. For more details on Geocoding addresses, please go to our knowlegebase and look up "Geocoding". We are always working on better algorithms and more ways to geocode, so watch this space in the future.
There are really two ways to do this: import files that have already been coded with geospatial data (usually Latitude and Longitude and using the EXIF extensions to the JPEG file format), and using the time code on digital photos to coordinate with GPS devices to render information.
Cartographica currently supports both the import form, and the timecoding form. To import data with existing geospatial information, either drag the files onto a map, or select the Plot GPS Encoded Files... item from the Tools menu. To find photos on a map based on GPS data, the process is a little more involved.
One interesting feature of Cartographica's photo geocoding is that all of the data remains in the Cartographica map files. Since we don't write data to your photos, they're safer than they would be if we opened them for write.