Spatial mapping of the extreme weather event datasets
Data were exported from R as ascii text files with grid cell centroid
locations provided as absolute integer coordinates in British National
Grid projection to facilitate import into ArcGIS 10.5 for visualisation
and further analyses. Null values (NA) representing offshore locations
were recoded to (-9999), ensuring compliance with numeric format prior
to import. The point locations were plotted and then spatially joined to
a pre-calculated vector 5 km grid, whereupon joined null values and
their corresponding grid squares were identified and removed. The
resulting datasets were then used to create thematic maps.
Geoprocessing (clipping) was used to extract underlying published land
cover data (Rowland et al, 2017). The resulting land cover data required
planimetric areas to be re-calculated, and these were subsequently
summarized by ecosystem type and aggregate area. The UK land cover
categories (Rowland et al. 2017) were grouped into four broad classes
each providing specific ecosystem services and levels of biodiversity:
(1) Agriculture, incorporating arable/horticultural and improved
grasslands (provisioning), (2) Woodlands, incorporating broadleaf and
coniferous woodlands (provisioning, regulating, and biodiversity), (3)
Conservation, incorporating National Parks and Sites of Special
Scientific Interest (SSSIs) (supporting, regulating and biodiversity),
(4) Carbon stores, incorporating heathland, heath grasslands and bogs
(regulating).
Where the analyses had revealed significant change, a field attribute
selection was used to identify the corresponding grid squares,
extracted, and then exported as separate geospatial datasets. To
facilitate further quantification of land cover types affected, the
boundaries between resulting significant grid squares were dissolved, so
that only the perimeters of aggregated squares remained. These two
datasets were combined to produce a map for each of the four land cover
categories overlain with areas of significant increase in frequency of
each extreme event metric.