Spatial Data Analysis
Datasets are often complex with multiple tables and values. Spatial data analysis is used to analyse raw data or processed data so that the patterns within the data can be easily interpreted.
The type of spatial analysis required will depend upon the question that you wish to answer and the raw or processed data to be analysed. Some examples of questions which could be answered using spatial analysis include:
- Where is the best location for a new development?
- How does accessibility of services vary? How many people do not have access to services?
- How has the landscape changed over time?
- Which factors affect the quality of habitats and number of species?
- Does the quality of priority habitats vary across the country?
Some types of spatial analysis which may be used include:
Least Cost Analysis
Least Cost Analysis (LCA) uses cost layers to calculate the best path to follow between locations. Cost layers are created by assigning a cost to each square in a grid. Depending on the application, the cost layer could be based on factors such as slope, habitat type, wetness. Maximum cost can also be applied to the analysis to limit possible routes.
It can be used to model the preferred least cost paths for a variety of scenarios, including:
- Animals migrating between habitats, as in the Econet project
- Walkers visiting attractions in a National Park
Network analysis analyses the links within a network e.g. roads and footpaths. It is generally used to analyse walking or driving distance/time between points of interest. This analysis can be applied in a variety of ways:
- Determining whether pupils live within walking distance of school
- Determining the catchment area for a doctor's surgery
- Performing accessibility analysis to determine availability of Rights of Way for different users e.g. pedestrians, cyclists, horse riders and wheelchair users
- Performing greenspace analysis to calculate access to greenspaces of a particular size
Census data can be combined with the results of network analysis to determine how many people have access to services and identify under or over-provision.
Statistical analysis is useful in analysing large volumes of data to determine correlations between variables and datasets. These analyses can be performed on sample data and used to generate a model to predict values at another location.
Regression is one of the statistical methods that may be used to determine patterns within datasets which may be spatial. It is used to identify, from a list of candidates, those variables that affect a dependent variable. As a simple example, temperature may be effected by latitude, altitude, exposure and a range of other spatial and non-spatial variables; regression analysis can identify these variables and their importance.
Terrain analysis analyses height data, particularly contour data and digital terrain models (DTM). The height data may be processed prior to the analysis to calculate slope, aspect and relief.
Terrain data can be used to calculate the impact or feasibility of a new development e.g. wind turbines, radio masts using viewshed analysis or intervisibility analysis:
- Viewshed analysis uses the height of the new object to model the area over which the development will be visible. Additional datasets such as woodlands and housing can be used to improve the quality of this model.
- Intervisibility analysis uses the height of each object to determine whether there is a clear line of sight between the objects. This analysis could be used to calculate whether signals would pass between radio masts effectively.
DTMs acquired on different dates can also be used to calculate changes in the landscape over time, e.g. dune or saltmarsh accretion, erosion, quarrying. Differences in both height between DTMs can be calculated and used to calculate the volume.
The DTMs used may be derived from a variety of sources including: synthetic aperture radar (SAR), lidar or high resolution aerial photos.
exeGesIS offer a cost-effective gradient analysis service for Rights of Way. This service uses Rights of Way data in conjunction with contour data to determine the gradient for each link within the network.