Typology Methodology

A risk based approach

The Climate Risk Typology is formed around the risk framework applied within the latest report released by the Intergovernmental Panel on Climate Change (IPCC), the 5th Assessment Report. Here, risk is a function of climate hazards, and exposure and vulnerability (which encompasses sensitivity and adaptive capacity) to these hazards. The Typology is based around a set of indicators that reflect these risk elements.

Typology spatial scale

The Climate Risk Typology is formed at the scale of European NUTS3 regions, which are defined as ‘small regions’. NUTS3 regions are part of a system that subdivides the economic territory of Europe to support statistical data gathering, socio-economic analysis and the framing of European policies. There are 1379 NUTS3 regions in Europe. NUTS3 regions are a population-based classification system, and contain between 150,000 – 800,000 people. As a result, the density of NUTS3 regions across Europe varies. For example, there are 402 in Germany, 21 in Sweden and 5 in the city-region of Greater Manchester (located in North West England).

Typology methodology

Climate risk indicators sit at the heart of the typology. The typology methodology involved the selection, cleaning and processing of indicator data specific to the four IPCC risk domains (hazard, exposure, sensitivity and adaptive capacity). Originally the intention was to use existing indicator data to create the typology. However, due to issues of data coverage and quality, the majority of the indicators were in fact developed within the RESIN project in order to create the typology.

Having cleaned and processed the data and identified those indicators to develop the typology around, the next step was to determine the most appropriate method through which to cluster the indicators. The logic of cluster analysis is used to identify groups of objects (NUTS 3 regions in this case) based on their underlying characteristics. The cluster algorithm seeks to define homogenous groups of NUTS 3 regions that share similar climate risk characteristics, which were termed classes and sub-classes. The classes and sub-classes were then named and described.

The development of the typology followed a collaborative approach and was informed by inputs from individuals from within and beyond the RESIN project. The early stages of the typology work involved consultations that aimed to gather views about the purpose and function of the typology. Input and feedback was also gathered on the emerging typology outputs. A key point of engagement came with a consultation process on a prototype of the typology online portal. This sought to evaluate and improve the usability and application of the portal and the information contained within it.

A report on the typology includes further information on the methodology and output. This report contains a wide range of information on the typology methodology including details of:

Data transformation and standardisation approaches used to prepare the indicators for use within the cluster analysis, and tests performed on the indicators to identify those that are excessively correlated and which should therefore not be included in the cluster analysis.

The areal interpolation technique employed as a means of imputing missing indicator values in available data sets.

Climate risk indicator descriptions, including the source data and the method used to develop the indicators.

The cluster analysis approach used to define the typology classes and sub-classes.

Download the Report