Mis-reporting of the data in England biodiversity indicators Priority Habitats 2018

Summary

Way back in January 2019 I wrote to the Office for Statistics Regulation about the inaccuracies I thought were present within the Defra statistics for Lowland Meadow Priority Habitat; they agreed with my points and wrote to Defra asking them to make changes to their presentation of their statistics.

Defra wrote back to me at the end of September 2019 as they had promised (I’ve just not had time to write it up as a blog until now) to explain the source of their statistics more clearly and to say that they had added in extra information to the presentation of their statistics. (I’ve had this information before from them through previous email exchanges, but I wanted it stated in official documents because that is the official reference point for everyone quoting or using the statistics).

“The process of mapping inventories in England rounds areas up to parcel level, is based on old survey data so doesn’t reflect recent changes, and takes a broader definition of ‘grassland’ – including partially degraded and less species rich grassland than would be considered for designation. Overall, this leads to an overestimate of Priority Habitat cover in England.”

It was that “overestimate” that I was concerned about; the official government figure for Lowland Meadow Priority Habitat 36,129 ha, the reality is probably closer to 8000 ha and I say probably because no-one actually knows.

Why does this matter? If we are to conserve the genetic depth and breadth of our native grassland species, we must use seed from these sites and not from seed packets/commercial seed outlets in order to create new grasslands. Species-rich grassland acts like a glue that holds all the other habitats together to form a functioning ecosystem – destroy the glue and your ecosystem will function badly.

No doubt you will have noticed all the concern about the amount of ancient woodland the HS2 project will destroy. There is over 300,000 ha of ancient woodland in England but under 8000 ha of Lowland Meadow of which HS2 will destroy 1% or 80 ha (HS2 figures); of that Lowland Meadow Priority Habitat we don’t know how much is ancient grassland because there is no official definition of ancient grassland to assess it by.

The government has no plans in the future to map the extent of either known Lowland Meadow sites or to look for currently unknown sites other than a vague promise to improve the quantity and quality of the data, but I have been informed that they are developing drone technology. Currently Natural England have insufficient financial resources to check existing grassland SSSIs (I’ve asked and have local evidence) and the ministers at Defra have no plans to increase Natural England’s funding (I’ve asked that too). The government has no other public servants with sufficient expertise and experience to assess the most important parts of our shared natural heritage other than those employed within Natural England.

My thanks to Ed Humpherson and other team members at the Office for Statistics Regulation for their help with making government statistics more accurate.

 

Full details as below

This all grew out of my inability to understand government statistics for the extent of the Priority Habitat Lowland Meadows and I wrote about previous replies to my questions in this blog https://theintermingledpot.wordpress.com/2018/04/26/estimates/

The response from Ed Humpherson, Office for Statistics Regulation to my email regarding the Defra statistics can be found published here https://www.statisticsauthority.gov.uk/correspondence/response-to-england-biodiversity-indicators-statistics/

And there is a link to my original email questioning the quality of the data Defra were producing at the end of his response.

 

Email from Defra received 30/9/2019

“Thank you for getting in touch with us last year to highlight the differences that exist in the reported areas of lowland meadow grassland across England, and for giving us the opportunity to improve the quality of our priority habitats indicator.

I am aware that the Office for Statistics Regulation (OSR) have been in touch with you regarding your concerns and I can confirm that we have implemented the recommendations they made to us during the 2019 update of our Biodiversity Indicators. These indicators were published on 5 September and can be viewed here: https://www.gov.uk/government/statistics/england-biodiversity-indicators

I am sorry for the time it has taken for me to write to you about the differences in the reported areas of lowland meadow grassland having committed to do so earlier in the year. We have been working closely with colleagues in Natural England and the Joint Nature Conservation Committee to address the issues you raised, and I am now in a position to update you on the outcomes of this work.

As you know, there are several sources of published data for the extent of priority habitats in England, including: (i) those data based on the Biodiversity Action Reporting System (BARS); (ii) those in the 2013 priority habitats inventory, published by Natural England and used in the England Biodiversity Indicators (2a – Extent and condition of priority habitats); and (iii) those used for the selection of biological SSSIs. Given that BARS has now been archived, we have focused our efforts on providing additional clarity for the latter 2 sources of data.

In summary, the 2013 priority habitats’ inventory published by Natural England and used in the England Biodiversity Indicators covers 24 priority habitats. It addresses problems with the original Biodiversity Action Plan (BAP) priority habitat inventories by moving from multiple habitat layers to a single layer based on Rural Land Registry polygons (derived from OS MasterMap). In bringing the existing inventories into a single layer with no overlaps, a set of criteria were used to assign one main habitat (or no main habitat) to each polygon. Where other habitats may be present in the polygon, e.g. as part of a mosaic, these are attributed as ‘attribution habitats’. The process of mapping inventories in England is also based on old survey data so doesn’t reflect recent changes; it also takes a broader definition of ‘grassland’ that includes partially degraded and less species rich grassland than would be considered for designation. Overall, this leads to an overestimation of Priority Habitat cover in England, especially for MG5 communities (unimproved neutral grassland, including hay meadows) that are a subset of the Lowland Meadows Priority Habitat. Despite this overestimation, the 2013 priority habitats inventory map is still considered to be the best available national source of extent data for indicator 2a as it is the only one that allows for an assessment of how the condition of these priority habitats has changed over time.

The England Biodiversity Indicators (2a – Extent and condition of priority habitats)

Additional text based on the above has been added to the fiche for indicator 2a in order to explain why the areas of priority habitat in this indicator differ from those found in other sources.

The extent measure for indicator 2a shows that there are 1.87 million hectares of terrestrial and coastal priority habitats recorded in the 2013 priority habitats’ inventory for England; they represent around 14% of the total land area of the country. The 2013 inventory does not allow for assessing changes to the extent of these habitats, as changes to the inventory itself would represent improvements in current knowledge of the extent of priority habitats, rather than actual changes in the extent of those habitats. Furthermore, although there have been subsequent revisions to the priority habitats’ inventory, indicator 2a consistently assesses changes in the condition of habitats reported in the 2013 inventory.

This condition assessment is used to measure progress towards one of the higher-level outcomes from ‘Biodiversity 2020: A strategy for England’s wildlife and ecosystem services’ – to achieve “… 90% of priority habitats in favourable or recovering condition”. The most recent assessment shows that as of 31 March 2019, a little over 1.25 million hectares of priority habitats were in target condition. This equates to 66.9% of all priority habitats in a favourable or unfavourable recovering condition. Two of the 24 habitat types achieved or exceeded 90% of their area in favourable or unfavourable recovering condition, with a further 10 achieving or exceeding 80%, the target value for each individual habitat.

SSSI Guidelines

The following paragraph has been added to the guidelines for the selection of SSSIs. It has been linked to sections 4.10 and Annex 1 of chapter 3 of the report. The additional text explains why the areas of priority habitat in the SSSI guidelines are less than those reported in the England Biodiversity Indicators (2a).

“These areas are a subset of the national Priority Habitat areas used for England Biodiversity Indicator Report (Indicator 2A- Extent and condition of priority habitats). The process of mapping inventories in England rounds areas up to parcel level, is based on old survey data so doesn’t reflect recent changes, and takes a broader definition of ‘grassland’ – including partially degraded and less species rich grassland than would be considered for designation. Overall, this leads to an overestimate of Priority Habitat cover in England. This is particularly apparent for MG5 communities that are a subset of the Lowland Meadows Priority Habitat. While extent may differ, the England national inventory maps (used in England Biodiversity Indicator reporting) are a good indication of the location of known high quality sites.”

I hope this provides the additional clarity you were looking for.”

 

My reply on 8/10/19

“Thank you for taking the time to write to me on this issue.

I try very hard to understand and to consider what is written to me in reply to my questions and I fully appreciate the financial restrictions that you must work within, especially given the current political situation.

I have summarized what I think you have written to me:-

36,129 ha is the official government record of the extent of Lowland Meadow Priority Habitat in 2019; it is known to be an overestimate and to have been out of date since 2013 as the information it was based upon was out-of-date when the digital layer was created.

BARS has been archived and so data associated with BARS in regard to Lowland Meadows is no longer used in an official government capacity.

Defra uses the 2013 data set because it enables a condition assessment for Lowland Meadow over time and it is the best data set available for accurate location of the habitat.

The data that Natural England use regarding the extent of MG5 in England as evidence within a SSSI designation is a subset of Defra’s 2013 dataset [this bit is a puzzle as if Natural England have access to data at a higher resolution, then presumably they could share it with Defra, so I am assuming that it can’t be used because the condition assessment would not apply, except if a condition assessment applied to a larger area then it would surely apply to the smaller area within the larger area too? It may be that as I research further I will uncover the answer to that one] and I am inferring that the MG5 data is accurate because it is used as evidence in a legal situation.

I have noted that written within the 2a document it says that work is being carried out on improving the “quantity and quality of data on threatened habitats” and that this will be introduced at some unspecified date in the future.

I’m sorry that I have kept pressing you on this matter when I am sure you have little time available, but I do think it is important to have a clear government record on the state of our shared cultural heritage habitats, particularly given the Government’s continual reduction in funding for biodiversity and biodiversity expertise. I’m currently writing a book on England’s ancient grassland and in it feel I need to be able to explain clearly how the government officially records the Lowland Meadow Priority Habitat (within which some ancient grassland is a subset) and I don’t want to misunderstand or misrepresent the government’s official position, though obviously I will discuss how confusing it has been to understand it.

Many thanks for making the changes to the official record”

 

For further information on the SSSI problem see https://theintermingledpot.wordpress.com/2018/09/10/struggling/

This entry was posted in The I.Pot. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s