The aim of the project is analysing movements and mobility characters across Estonia. The analysis made during the project is based on open data. The primary source of the data is the Mobility Lab of the University of Tartu. I have downloaded the data from the Mobility Lab's website for 2016-17 years (January-March of 2018) by months. The data type is passive mobile positioning collected and processed by Telia, Estonia's mobile operator. The information contains unique observations of everyday commuters. The data aggregation level is neighbourhoods, also defined as territorial communities (mentioned on maps as territorial communities). The data also contains geographical information, in particular, coordinates of origin and destination. Thus, by converting given datasets into proper data formats, makes the possibility for spatial analysis. The second primary dataset used for the project is neighbourhoods downloaded from the Mobility Lab's website, though it is obtained from Statistics Estonia, updated by 2018. The data represents the geographical positioning of territorial communities. I have used this data to aggregate the number of commuters into neighbourhoods and introduce a "Choroplet" map.

In the project are involved different types of spatial analyses and geoprocessing methods. The whole process includes two main parts: data processing and analysing and, cartography. As mobility data contained big datasets stored into separated files, data processing required repeated actions. For avoiding reiteration, I have created the model by using the "Model Builder" tool. The model is valid for later usage for the same purpose. The vast part of the project includes statistical analysis. The main tools involved in the study are from data management, conversion, analysis sections of toolbox. Also, attributional and spatial joins and relations are used in several times.

The analysis is done in ArcMap and QGIS. Maps are made entirely in QGIS.

Flow maps

A flow map is one of the most used types for visualising mobility data as this type of visualisation shows the main spatial patterns.

This map shows the mobility flows between territorial communities for January 2016. The data classification method is used to distinguish different classes. High and low concentrations of the flows are noticeable on the map.

This map is made using the same method as the previous one; however, the data is aggregated by year and shows the number of commuters for the whole period of 2016.

Flow map with Time Series analysis

Time series analysis is convenient to see changes in time. The map represents the number of commuters for each month for 2016 and 2017 and shows seasonal patterns of flows.

Choropleth map

Displaying movements by flow map is an excellent way to encourage the visual reading of information. Though, for more detail analysis, I have created choropleth maps. This map is an example of one of them and represents the result of data aggregation. The data are aggregated by two different levels, by neighbourhoods and by municipalities. The trends differ from each other at smaller and more significant levels. Thus, this comparison refers to how manipulative can the maps be.

Dot density map

The dot density map represents another cartographical way to show the tendencies. On the map, we can see a high and low concentration of the commuters. Using the "Random Point Inside Polygon" tool, the appropriate number of points are created inside territorial communities. One point represents one commuter.