Following a review of approaches used in Ireland and other countries, it was clear that the approach should consider the number of occupied hectads (10 km × 10 km squares). Following discussion with other experts, we set the following thresholds:

• Rare species occur in 10 hectads or fewer. 10 hectads equates to about 1% of the land area of Ireland.
• Scarce species occur in between 11 and 50 hectads. 50 hectads equates to about 6% of the land area of Ireland.

We tested these on existing datasets for macro-moths and water beetles. We developed a set of PostgreSQL/PostGIS views that consistently processed and analysed these raw biodiversity datasets. We also generated standardised distribution maps for all 861 insect species through a novel use of atlas functionality in QGIS.

We also pushed the process further to provide additional insights into the distribution of each species. This included:

• Distinguishing coastal species, defined as species where 75% of occupied hectads were within 1 km of the coastline. To aid interpretation, inland and coastal hectads were shaded differently on the maps.
• Identifying geographic skew in species distributions, calculated using the spatial distribution of the underlying records. This skew was presented on each map as standard box-and-whisker plots, and using cardinal and intercardinal compass directions (e.g. north-eastern).
• Recognising species that had a patchy distribution using a DBSCAN cluster analysis algorithm. This helped to identify species that had no clear geographic skew but were nonetheless geographically restricted.

We also considered whether it was possible to assess the abundance of a species in the hectads from which it was recorded. This needed to overcome the effect of variable recording effort, which means that abundances in different datasets are not comparable. We therefore developed a Normalized Hectad Abundance Score, which scored the abundance of each species from 0 (lowest abundance, which is usually an average of 1 record per hectad) to 1 (the species with the highest average number of records per hectad) within each dataset.

All this information was clearly included on the maps, to facilitate the expert review of the results.

Further information from Crispin Flower.

Deer are large animals that are a potential hazard to vehicles on the road. Each year in the UK alone, it is thought that over 700 people suffer injuries or fatalities and over £17 million is spent on vehicle repairs because of Deer Vehicle Collisions (DVCs). In Scotland, a combination of increasing deer populations, especially in the central belt, and a growth in traffic has led to an increase in DVCs.

Exegesis was awarded a contract to collect and analyse data on Deer Vehicle Collisions (DVCs) in Scotland between 2019 and 2021. This work involved obtaining DVC records from a variety of sources in the public, private and third sectors, collating it into a consistent database, and analysing the data to identify trends and highlight areas of concern. This continued work undertaken since 2008 that had established data collation, processing and analysis protocols.

We added 5,479 records of DVC incidents in Scotland, bringing the total in the database for 2008 to 2021 up to 22,753 records. These were analysed in detail and a full report produced. Key findings included:

• Whilst there has been no recent significant increase in DVCs in Scotland overall, there has been a general increase in the central belt that has been offset by a decrease in northern Scotland (see map).
• Dusk is the worst time for DVCs, especially in May and June when overall numbers of DVCs are higher. This information can be used to target mitigation activities.
• There was no evidence of a decrease in DVC incidents resulting from travel restrictions introduced due to the Covid-19 pandemic (shown below). This contradicts claims of an anthropause where human impacts on wildlife were reduced. Nevertheless, the number of DVCs reported by members of the public, who would have been affected by the restrictions, were clearly suppressed. This suggests that freight traffic, which was largely unaffected by travel restrictions, was the cause of many DVCs during this time.

We also developed a new approach to analysing DVC risk on sections of the trunk road network, by using a moving window algorithm to identify the parts of roads with the most DVCs. We identified the blackspots with the highest DVC risk from the results. Most of these proved to be junctions surrounded by woodland in the more urban central belt. We recommended actions to mitigate the risk at the 10 worst blackspots.

The report has been published by NatureScot. We also produced a short non-technical summary of the key results.

Further information from Richard Griffiths.

Zero Waste Scotland contracted Exegesis to support Scottish local authorities in meeting the requirement for land zoning data. This involved:

• Providing practical technical advice to local authority staff on the use of GIS to zone land.
• Creating first cut spatial datasets of relevant land to be zoned.
• Processing data on behalf of local authorities in various ways.
• Supporting local authority staff as they determine and apply zones to their data.
• Quality assuring data and preparing it for the Litter Monitoring System.

Data processing was generally undertaken in the PostgreSQL extension PostGIS, using QGIS to view the data in PostGIS. Some processing was also undertaken in GRASS where this was beneficial, such as cleaning and removing small areas from the spatial data (using v.clean, especially the rmarea tool). Advice to local authority staff was tailored to their corporate GIS, mainly ESRI ArcGIS and Pitney Bowes MapInfo.

One of the requirements was that zoned land was split into 1,000 m² sample locations. Each sample location defined an area of land that could be surveyed as part of litter monitoring. Part of the support provided to local authorities was to run the pre-existing QGIS Polygon Divider plugin on supplied data to create the sample locations. This highlighted the need to enhance the Polygon Divider, which we undertook following discussion with Zero Waste Scotland. This included:

• The addition of an approach to split complex polygons into more manageable chunks to the Polygon Divider interface.
• Improved data handling to use memory more efficiently and reduce crashes and failures.
• Implementation of PostGIS compatibility.
• Improved message logging.
• Various minor bug fixes.

The changes to the Polygon Divider were submitted for inclusion in the released tool.

Further information from Richard Griffiths.

Prior to the National HLC (NHLC) project, virtually all of England had been characterised through individual HLC projects, but these were at county or sub-regional level, with each of the studies using slightly different methodologies and categorisation methods. In early 2016, Natural England appointed Exegesis to compile a national HLC dataset, drawing together the existing sub-regional HLC datasets and applying a common framework, structure and terminology to a new unified dataset. The aim of the project was to improve the awareness, understanding, appreciation and ability to manage and monitor the historic dimension of England’s landscape at a national scale, for both professional and non-professional end-users.

The database containing the terminology applied to all the different sub-regional datasets was cleaned and then the original terms used were mapped to a bespoke thesaurus, created to ensure that all could be mapped to a suitable, equivalent new term. The associated spatial data was also processed so that the irregular polygons were firstly imported to a merged dataset and then generalised into a gridded dataset. The gridded approach eliminated discrepancies in polygon size or accuracy, and overcame issues where there were small amounts of missing data in the original datasets.

The final gridded dataset was produced at 500m and 1000m scale, with each grid cell having a single associated record with information about the historic landscape character types (both broad types and narrower, character types) and period information. The datasets can be viewed and interrogated in GIS software packages, and when viewed alongside other datasets, the NHLC data can act as an indicator of historic landscape character and can therefore feed in to decisions about landscape management, planning, heritage asset management or research priorities.

View the outputs.

Further information from Abby Hunt.

The surprising answer is 'probably yes'.

We identified the Saproxylic Quality Index (SQI) as the best available measure of quality, as it was felt that a site's saproxylic (wood decay) beetle fauna would be a good surrogate of wider wood decay habitat quality, which is one of the most important elements of wood pasture and parkland. We combined available SQI scores with the wood-pasture and parkland inventory we had previously created for Natural England to create a sample dataset of 98 sites. We then analysed the sample data against other available data, including tree records from the Ancient Tree Hunt, geography, climate, designations and habitat networks using PostGIS and multiple regression analysis in the statistical package R. The models produced were refined to identify the model best able to predict wood-pasture and parkland quality.

Eleven site attributes were included in the final model for predicting SQI score. Sites in the south east and lowland sites tended to have higher scores, suggesting that warmer and less exposed sites were better for saproxylic beetles. Unsurprisingly sites with higher numbers and densities of veteran or ancient trees also tended to have higher scores, owing to the increased volume of wood decay habitat present.Sites that were in a landscape with other wood-pasture and parkland were also found to have higher SQI scores.

The best model proved to be remarkably good at predicting the SQI score of the sample. It suggested that sites with a predicted score of 600 or more were likely to have an actual score of at least 400 - itself quite a high score. This suggests that if the model were applied to the wider wood-pasture and parkland inventory it could be used to predict SQI scores and identify sites that could be targeted for survey.

However, this was a proof of concept study, so the model needs more work before we can conclusively state that wood-pasture and parkland quality can be predicted. This includes further refinements to identify attributes that help predict quality. It then needs to be tested to determine how well it predicts SQI scores for sites that were not part of the sample. We hope to be able to continue this work, as we believe the improved model could be very valuable for targeting entomological survey effort.

Further information from Crispin Flower.

To support local authorities in meeting their duty to provide route maps of existing safe footpaths and cycle-ways, Exegesis were contracted to specify the structure of datasets and design a web based data management system. At the same time Sustrans carried out an initial survey of existing paths and infrastructure to populate the system.

Exegesis created a structured national road and public rights of way network ready for the survey, then presented the survey data via a web mapping interface. The Sustrans survey data was added and thematically displayed. The LAs then defined their routes by clicking on the existing paths and roads and entering route details. The routes were inspected to assess the condition, hazards and suitability for an Active Travel route.

The resulting data could be visualised and edited on the web site by all of the Welsh LAs, minimising costs and helping to ensure that a consistent data set was created. The public consultation maps could then be printed directly from the website, providing consistently styled and annotated maps.

The Active Travel data set became the first national system for managing Active Travel paths and cycle-ways in the UK and will be used to improve the network and plan for the future.

Further information from Crispin Flower.

The work involved consultation with staff at JNCC and the British Oceanographic Data Centre (BODC) to ensure a faithful import and production of meaningful layers that are useful to JNCC marine staff when assessing the current extent of UK marine monitoring according to each, broad marine discipline (e.g. biodiversity, physical oceanography and fisheries). In addition, users can create custom layers, selecting only the monitoring parameters that may be relevant to a particular analysis.

Further information from Andy Brewer

Natural England had collated a series of spatial datasets relating to surveys of subtidal and intertidal marine habitats. Though all were in digital Geographical Information System format, they were from a variety of sources with differing formats. These were standardised to MESH Translated Habitat DEF, including an assessment of MESH confidence, data cleansing and validation, quality assurance and production of MEDIN metadata. We used a range of tools within ArcGIS, QGIS, GRASS and MapInfo in order to quickly and efficiently produce accurate and reliable outputs. We also utilised four standard habitat translation tables to manually assure the EUNIS habitat identified. A generic process diagram was also created, that can be used to guide all future MESH translation work.

The data to be imported into Marine Recorder were reviewed so that issues could be identified and the approach to import could be agreed. Biotopes were assigned through expert assessment of the data, based upon faunal groupings from Bray-Curtis analysis, particle size distribution, geographic location and field survey notes for each sample. These were then compared with the Marine Habitat Classification for Britain and Ireland (Connor et al., 2004) for the final biotope assignment. The Marine Recorder spreadsheet import function was used to import the resulting data, with particle size and biotope entered manually. The imported data in the snapshot were then independently checked to ensure they were correctly attributed.

Further information from Claire Lush.

Exegesis were contracted to investigate the distribution of non-native species on protected sites in England, in order to help develop a programme of work to tackle INNS. This involved:

• the development of a master list of 3,687 non-native species (NNS)
• the collation of nearly five million records from the National Biodiversity Network (NBN) Gateway and a range of Natural England and third party datasets
• spatial analysis of NSS distribution against Special Area of Conservation (SAC), Special Protection Area (SPA) and Site of Special Scientific Interest (SSSI) boundaries in SQL Server

We presented the results in Microsoft Excel workbooks showing the NNS that had been recorded from each site. Each spreadsheet could be filtered, allowing only the data within a specific site or region to be displayed. This showed that 98% of SACs, 99% of SPAs and 87% of SSSIs had records of NNS. Potential INNS intersected with 90% of SACs, 96% of SPAs and 75% of SSSIs.

The results for seven sites were reviewed against on the ground knowledge by Natural England site staff to determine any differences. We also made a comparison against data in Natural England’s ENSIS database. These comparisons demonstrated the case for increasing data flow by increasing awareness.

Based on the results, we undertook a review of the recording, systems and data flow processes within Natural England. We recommended and costed the use of the NBN Gateway as a data repository and the use of existing Natural England systems for recording and interrogating INNS data. Additional recommendations were made to improve data flow more generally.

Read the report.

Further information from Claire Lush.

PDNPA has maintained SHINE (Selected Heritage Inventory for Natural England) records for the Derbyshire part of the Park since the inception of the scheme. Once it was decided that PDNPA would maintain SHINE data for the whole of the Park, the need for a Park-wide HBSMR dataset became apparent.

In Spring 2014 exeGesIS undertook the merging of the five constituent local authority HBSMR datasets. Although the five datasets from Cheshire, Derbyshire, Greater Manchester, South Yorkshire and Staffordshire all share the same structure, the technical process of merging the data was nonetheless challenging. Specific issues that had to be overcome were different back-end database engines, different GIS software, spatial selection of records for inclusion in the merged dataset, differences in look-up table data, differences in thesaurus candidate terms, different period/date interpretations, different use of user defined-fields etc. Most importantly, the merged dataset had to provide an accurate image of the contributing datasets and, although new Peak District UIDs were generated, the original references needed to be maintained.

The merged dataset was installed together with HBSMR v4 in the Bakewell offices of the PDNPA in early May 2014. The next phase of the project will be to define effective data exchange mechanisms between the Park and the other authorities that use HBSMR, so that all records benefit from the emergence of new heritage information.

Further information from Crispin Flower.

The review of other habitat mapping activities highlighted the range of approaches in use, including the incorporation of existing data, remote sensing and field survey. These approaches were summarised to form a list of possible options for inventory creation, from which a preferred approach was selected based upon the options available and some innovation where solutions more appropriate to England were required.

The prototype inventories were created based upon data already available to Natural England. These covered the whole of England for all 64 terrestrial, freshwater and coastal Annex I habitats. The presence of Annex I habitat was identified based upon existing habitat data, automatically searching for habitat codes that correspond to each Annex I habitat. Coverage, gaps and confidence was assessed for each source dataset and resulting inventory.

Further information from Crispin Flower.

The survey was undertaken by our mini-plane less than 24 hours after the initial request was made. In order to attain the highest level of spatial accuracy, ground control points were taken with a high precision GPS unit. The images were orthorectified and mosaicked to create a single image using Pix4UAV advanced processing software.

We also used Pix4UAV to generate a digital surface model (DSM) to enable 3D analysis to be undertaken. This DSM was compared against lidar data, which provided a baseline height for the site before the addition of material. The lidar data was subtracted from the DSM to give the change in height, which was then analysed to determine the volume of additional material within discrete patches identified from the aerial photography. The volumetric analysis was undertaken using GRASS and verified with MapInfo Engage 3D.

The aerial images, DSM and outputs of the analysis were provided to the client, along with a report describing the methodology used and the results.

Further information from Crispin Flower.

The objective of this project was to process licence data from their legacy database to improve the accuracy of the spatial information. This was completed using a series of data validation rules.

For sites with no spatial object, the location description was used to geocode each license so that they could be analysed spatially.

In addition, many licences require survey data to be collated and supplied to the MMO. A list of keywords were identified and used to tag the survey metadata to improve searches for similar surveys within a specified radius. The keywords significantly improve the results returned and allow data to be compared over time.

Further information from Claire Lush.

In addition to the survey, the actual path line for the Wales Coast Path was digitised and categorised based on BS7666 standards.   

The survey was completed with the goal of providing a summary of the condition of the Wales Coast Path. It was also intended to act as a resource for local authorities responsible for maintaining the path as it passes along the coast.

After the survey all data was made available to NRW via a dedicated data management system. The Wales Coast Path Quality Management System is hosted and supported by Exegesis and available to NRW staff for the purposes of managing the Wales Coast Path.

For more information, please contact Simon Allen. 

The task required analysing approximately 1500km of path network around the coastline of Wales. A route was supplied by NRW, this was then split and labelled with attributes according to the surface of the path, the requirements of NRW and BS7666 guidelines.

The path was redrafted several times to improve accuracy and reflect changes agreed between landowners and the NRW that occurred during the project. The end result was a path digitised to OS Mastermap data to an accuracy of <5m. 

For more information, please contact Simon Allen. 

Over successive years these data were assessed to refine the boundary and attribute information associated with each polygon in the inventory and additional data were sourced regionally for inclusion. A data capture rule base was created and maintained throughout these projects, to act as a guide to wood-pasture and parkland assessment. Data were assessed against historic maps and aerial photography, as well as modern data sources. Ground truthing was undertaken in the west midlands and south west, through which the inventory data capture rule base was refined to ensure the best assessment of available datasets.

The resulting provisional inventory mapped 156,838 ha of wood-pasture and parkland, though this was known to overestimate the actual extent, and it was estimated to include over 5,000 sites made up of over 7,500 polygons.

Further information from Claire Lush.

Data were collated to create a dataset of candidate Greenspaces, an access network dataset and an access point dataset, all of which were checked by local authority staff. In addition the data was validated by undertaking a survey of greenspace type, condition and access points for a sample of 250 non-Local Authority sites. 

Population for each settlement was calculated from national census data. Analysis was undertaken to determine the area within the distance criteria for each Greenspace access point.

These data were then used to calculate the percentage of the population in each settlement which passed the criteria. The results were reviewed by local authority staff via a website, which allowed comments to be left and read by others. The data was then reprocessed to produce the final outputs.

Further information from Claire Lush.

Data were collated to create a dataset of candidate Greenspaces, an access network dataset and an access point dataset, all of which were checked by local authority staff. A dataset showing population distribution was created from national census data. Analysis was undertaken to determine the distance between each postcode and the nearest access point on the nearest Greenspace. With 14,000 postcodes, 12,000 candidate Greenspaces and 12,000 kilometres of access network this represented a significant data processing task.

Additional analyses were carried out that assessed the slope of each section of the access network and reanalysed the data excluding sections of the access network regarded as being too steep.

Further information from Kathryn Steemson.

In order to create the indicative map we combined all relevant datasets into a single ArcGIS personal geodatabase containing over 1.8 million polygons. This geodatabase was linked to an Access database application that assessed the likely habitat based upon the reliability of each source dataset and suggested the most likely habitat for each input polygon. The assessment was made based upon an agreed rule base. Using this assessment the application generated a new personal geodatabase that incorporated the classification of each polygon by likely habitat and an indication of the reliability of the assessment. In this way we were able to incorporate all available data in the assessment of each polygon, rather than prioritising the data contained in certain datasets over others.

Click on the map to the right for a larger image.

Further information from Crispin Flower.