For Windows, there are two installation options, which are as follows:

Follow these steps to install QGIS on a Mac:

There are two options when installing QGIS on Ubuntu: installing QGIS only, or installing QGIS as well as other Free and Open Source Software for Geographical Information Systems (FOSSGIS) packages. Either of these methods requires the use of the command line, sudo rights, and the apt-get package manager.

deb        http://qgis.org/debian trusty main
deb-src    http://qgis.org.debian trusty main

sudo apt-get update
sudo apt-get install qgis python-qgis

sudo apt-get install qgis-plugin-grass

sudo apt-get install python-software-properties
sudo add-apt-repository ppa:ubuntugis/ubuntugis-unstable
sudo apt-get update
sudo apt-get install qgis python-qgis qgis-plugin-grass

The QGIS interface is divided into four interface types: Menu Bar, Toolbars, Panel, and Map Display. The following screenshot shows QGIS Desktop with all four interface types displayed:

The map display shows the styled data added to the QGIS project and, by default, takes up the majority of the space in QGIS Desktop. The menu bar, displayed across the top, provides access to most of QGIS Desktop's functionality. The toolbars provide quick access to QGIS Desktop functionality. The toolbars can be arranged to either float independently or dock at the top, bottom, left, or right sides of the application. The panels, such as Browser and Layers, provide a variety of functionality and can be arranged to either float independently or dock above, below, right, or left of the map display.

There are four toolbars that are particularly useful, and it is recommended that you enable them:

QGIS Desktop offers a number of customization options. You can toggle the visibility of toolbars by navigating to View | Toolbars, or by right-clicking on the Menu Bar or the enabled Toolbars button, which will open a context menu allowing you to toggle toolbar and panel visibility. You can assign shortcut keys to operations by navigating to Settings | Configure shortcuts. You can also change application options such as interface language and rendering options by navigating to Settings | Options.

A new feature of QGIS Desktop is user interface themes. These are similar to skins for the application. These can be found by clicking Settings on the Menu bar and choosing Options. On the General tab, there is an option for UI Theme. If you are doing some late-night mapping, you might experience eye fatigue from the normal bright interface. Click the dropdown arrow and choose Night Mapping. The QGIS interface will switch to a dark skin.

You can also create your own themes. Start by reviewing the QSS style file that creates the Night Mapping theme from GitHub: https://github.com/qgis/QGIS/tree/master/resources/themes. To create a new theme, modify this file and save it to your .qgis2\themes folder.

The QGIS Browser interface (shown in the following screenshot) is composed of three parts: Toolbar, Data Tree view, and Information Panel:

The data tree view is an expandable tree listing of all geospatial data files on your computer and through connections. The information display, which takes most of the space on the application, contains four tabs that provide different views of the selected data in the data tree listing, and they are as follows:

The toolbar provides access to four functions. The Refresh function reloads the data tree view, while the Manage WMS function opens the WMS management screen allowing you to manage the WMS connections. The New Shapefile function opens the new vector layer dialog, allowing new shapefiles to be created. Finally, the Set layer CRS function allows you to define the coordinate reference system of the geospatial data file that is selected in the data tree view.

To load vector files, click on Add Vector Layer by navigating to Layer | Add Layer. This will open the Add Vector Layer dialog that will allow us to choose the source type and the source of the dataset that we wish to load.

The source type contains four options: File, Directory, Database, and Protocol. When you choose a source type, the source interface will change to display the appropriate options. Let's take a moment to discuss what types of data these four source types can load:

To load raster data into QGIS, click on Add Raster Layer by navigating to Layer | Add Layer. This will open a file browser window and allow you to choose a GDAL-supported raster file. The commonly used raster types supported by GDAL are as follows:

To add an Oracle GeoRaster, click on Add Oracle GeoRaster Layer by navigating to Layer | Add Layer, then connect to an Oracle database to load the raster. More information about loading database layers is given in the following section.

QGIS supports PostGIS, SpatiaLite, Microsoft SQL Server, and Oracle databases. Regardless of the type of database you wish to load, the loading sequence is very similar. Therefore, instead of covering specific examples, the general sequence will be covered.

First, click on Add Layer under Layer and then choose the database type you wish to load. This will open a window with options for adding the data stored in a database. As an example, the following screenshot shows the window that opens when you navigate to Layer | Add Layer | Add SpatiaLite Layer:

To load data from a database, we must first create a connection to the database. To create a new connection, click on the New button to open a connection information window. Depending on the database type you are connecting to, different connection options will be shown. Once you have created a database connection, select it from the drop-down list and click on Connect; you will see a list of all layers contained within the database display. If there are a large number of tables, you can select Search options and perform a search on the database. To load a layer, select it in the list and click on Add. If you only wish to load a portion of the layer, select the layer and then click on Set Filter to open the query builder. If you set a query and then add the layer, only the filtered features will be added.

QGIS supports the loading of OGC-compliant web services such as WMS/WMTS, WCS, and WFS. Loading a web service is similar to loading a database service, in that you must first set up the connection to the service, then connect to the service to choose which layers to add to the map canvas.

As an example, to add a WMS service, click Layer | Add Layer | Add WMS/WMTS Layer. On the window that appears, click New to open the Create a new WMS connection window (shown in the following figure) where you can define the parameters of the new service. Enter a Name for the service, the URL of the service, set any of the remaining options that may apply, and then click OK to add the service as an available connection.

With the new service added, make sure it is selected, then click Connect to connect to the service, and list the available layers. To add a layer to the map canvas, select it, choose the Image encoding, Coordinate Reference System, and Layer name, and then click Add.

Let's say that you need to make a map of the total population by county. However, the counties' GIS layers do not have population as an attribute. Instead, this data is contained in an Excel spreadsheet. It is possible to join additional tabular data to an existing attribute table.

There are two requirements, which are as follows:

To create a join, load both the GIS layer and the standalone table into QGIS Desktop. QGIS will accept a variety of standalone table file formats including Excel spreadsheets (.xls and .xlsx), dBase (.dbf) files, and comma separated value (.csv) files. You can load this tabular data using the Add Vector Layer menu by navigating to Layer | Add Layer and setting the file type filter to All files (*) (*.*) as shown in the following screenshot:

Once the data is loaded, a join can be completed by following these steps:

Snapping is an important editing consideration. It is a specified distance (tolerance) within which vertices of one feature will automatically align with vertices of another feature. The specific snapping tolerance can be set for the whole project or per layer. The method for setting the snapping tolerance for a project varies according to the operating system, which is as follows:

In addition to setting the snapping tolerance, here the snapping mode can also be set to vertex, segment, or vertex and segment. Snapping can be set for individual layers by navigating to Settings | Snapping options. Individual layer snapping settings will override those of the project. The following screenshot shows examples of multiple snapping option choices:

Note that there are also checkboxes for Enable topological editing and Enable snapping on intersection, which are covered in more detail in Chapter 6, Advanced Data Creation and Editing.

When you load spatial data layers into QGIS Desktop, they are styled with a random single symbol rendering. To change this, navigate to Layer | Properties | Style.

There are several rendering choices available from the menu in the top-left corner, which are as follows:

The following screenshot shows the Style properties available for a vector data layer:

In the preceding screenshot, the renderer is the layer symbol. For a given symbol, you can work with the first level, which gives you the ability to change the transparency and color. You can also click on the second level, which gives you control over parameters such as fill, border, fill style, border style, join style, border width, and X/Y offsets. These parameters change depending on the geometry of your layer. You can also use this hierarchy to build symbol layers, which are styles built from several symbols that are combined vertically.

You also have many choices when styling raster data in QGIS Desktop. There is a different choice of renderers for raster datasets, which are as follows:

The following is a screenshot of the Style tab of a raster file's Layer Properties, showing where the aforementioned style choices are located:

Another important consideration with raster styling is the settings that are used for contrast enhancement when rendering the data. Let's start by loading the Jemez_dem.img image and opening the Style menu under Layer Properties (shown in the following figure). This is an elevation layer and the data is being stretched on a black-to-white color ramp from the Min and Max values listed under Band rendering. By default, these values only include those that are from 2 percent to 98 percent of the estimation of the full range of values in the dataset, and cut out the outlying values. This makes rendering faster, but it is not necessarily the most accurate.

Next, we will change these settings to get a full stretch across all the data values in the raster. To do this, perform the following steps:

As shown in the following figure, by enhancing the contrast, the raster is displayed with much more detail, and is more visually attractive, too:

Default singleband contrast enhancement (left) and more accurate contrast enhancement (right)

You can specify the default settings for rendering rasters by navigating to Settings | Options | Rendering. Here, the defaults for the Contrast enhancement, Load min/max values, and Cumulative count cut thresholds, and the standard deviation multiplier, can be set.

The blending modes allow for more sophisticated rendering between GIS layers. Historically, these tools have only been available in graphics programs and they are a fairly new addition to QGIS. Previously, only layer transparency could be controlled. There are now 13 different blending modes that are available: Normal, Lighten, Screen, Dodge, Addition, Darken, Multiply, Burn, Overlay, Soft light, Hard light, Difference, and Subtract. These are much more powerful than simple layer transparency, which can be effective but typically results in the underneath layer being washed out or dulled. With blending modes, you can create effects where the full intensity of the underlying layer is still visible. Blending mode settings can be found at the bottom of the Style menu under Layer Properties in the Layer Rendering section (for vectors) and Color rendering section (for rasters).

As an example of using blending modes, we will show vegetation data (Jemez_vegetation.tif) in combination with a hillshade image (Jemez_hillshade.img). Both datasets are loaded and the vegetation data is dragged to the top of the layer list. Vegetation is then styled with a singleband pseudocolor renderer; you can do this by performing the following steps:

The following screenshot shows what the Style properties should look like after following the preceding steps:

At the bottom of the Style menu, under Layer Properties, set the Blending mode to Multiply and the Contrast to 45, and click on Apply. The blending mode allows all the details of both the datasets to be seen. Experiment with different blending modes to see how they change the appearance of the image. The following screenshot shows an example of how blending and contrast settings can work together to make a raster pop off the screen:

A new feature of QGIS since version 2.12 is custom variables. Variables can be set at several different scopes or levels:

Global variables are available at the application level in any QGIS Desktop instance. These can be managed from Settings | Options | Variables. Project variables are available throughout a given project. These can be set from the Project Properties window on the Variables tab. Layer-level variables are available from Layer Properties on the Variables tab and are available only for that layer. The other important aspect of these variables is that they cascade from most specific scope to least specific scope. In other words, a @my_var set at the Global level will be overwritten by a @my_var set at the Project level.

To begin, we will explore the variables for a project that was just saved. We will click on Project | Project Properties | Variables (see the following figure). Here you can browse both the default Global and Project variables. When a project is saved, three Project-level variables are populated by default: @project_filename, @project_folder, and @project_path. @project_title will also be populated if you have set that on the General tab. Note that when referencing variables, they are preceded by an ampersand to differentiate them from fields or functions. These variables can be used anywhere and expressions can be written such as the Field calculator, Label expressions, or Map Composer labels. With that introduction, we will go through several example use cases:

Using Project level variables in the Print Composer

One use case for variables is in the Print Composer. Say you want to have variables for the cartographer's name and the project path to be used on each map in a project. You can add a new Project-level variable by clicking on the green plus button and typing in a variable name and the value. In this example, a new variable has been added named @Cartographer and populated with the name. (Note: if you want to add a Global variable, you can do that from Settings | Options | Variables.)

Once the Print Composer is open, a label can be added to a map composition using these variables. To reference a variable in a Print Composer Label, use the following syntax: [%@variable%]. If you have multiple map compositions in a project, these variable can be used in each. Better yet, the composition can be saved as a template for future use, along with the variables. Using custom variables can be a real time saver and can help avoid mistakes. In this example, the path to the project is important to have on a map. If the project is moved, you may forget to update that item. Using a variable causes that information to be updated automatically.

Any text that needs to be used repeatedly can be stored in a custom variable. Remember that Global variables are also available in this context. If you want information to be accessible in any QGIS project, make it a Global variable. Additionally, creating a Project level variable with the same name as a Global variable will overwrite the Global variable. The Global variable can be used as a default setting. If that needs to change in a particular map project, simply create a Project level variable with the same name. In the following example, the Global variable @department has been overwritten by a Project level variable:

Using variables in data defined overrides

Data defined overrides are available for most layer labeling parameters. For example, these can be used to control label placement by linking to attribute columns containing the X/Y coordinates for the label position. Now custom variables can also be used as inputs to Data defined override. If we want use the font Calibri for all the labels on a map, we can set that as a Project variable called @label_font with a value of Calibri. Then when we open Layer Properties | Labels | Text, we can set Data defined override for the font to that variable by choosing the Variable option from the context menu (see the following figure).

We can repeat this step for each labeled layer. If we then decide Calibri is not working, and we are not using a variable for font, we have to open Layer Properties | Labels | Text for each layer and make the change. Using a variable, we can simply change the value of the @label_font variable to Franklin Gothic. All the labels will be updated automatically! This makes seeing the effects of such changes quick and easy. Variables can be used for any Data defined override.

Using Layer level variables

Layer level variables can be used in expressions that are evaluated within a layer. This includes rule-based styling or labeling expressions, diagrams, and the Field calculator. In this example, we frequently need to update the acreage for a set of polygons in a State Plane CRS. We can store the conversion factor from square feet to acres as a Layer level variable called @feet_2 acres. To manage Layer level variables, open Layer Properties | Variables. You will be able to see Global, Project, and Layer level variables:

When we open the Field calculator to update the acreage field, we can use the @ft_2_acres variable to calculate the acreage with the following expression: $area * @ft_2_acres. All the available variables are listed in the Variables section:

Variables are still a very new feature, but have a lot of potential for creating more efficient workflows. You will learn additional use cases for variables in the final section of Chapter 8, The Processing Toolbox, where the Atlas feature is described. Once you begin using them, you will undoubtedly discover many other use cases.