I’ve been emailing different people at different government departments about the recording of ancient grassland (and also chatting to conservation charities) when I’ve had time. I just email a question and ask if it can be answered – please forward this to someone who might be able to help. I get back very detailed replies with links to different scientific papers or various writings – it’s been very interesting and people are extremely helpful where they can be.
One of the major sticking points has been the lack of an agreed definition of what ancient grassland is exactly, although people are happy to use the term in their replies.
The definition of Ancient Woodland relates more to the first mapped records that we have of woodland (“any wooded area that has been wooded continuously since at least 1600 AD (1750AD in Scotland)”) than to any particular woodland related ecological feature. The Inventory is provisional and has never been fully checked through survey and I doubt if there is evidence for every single wood; creating the Inventory was essentially a map and aerial photograph based exercise.
The Inventory also only relates to pieces of woodland of greater than 2 hectares, but recently West Sussex surveyed all their ancient woodland including pieces smaller than 2 ha, adding an extra 3500 ha to their previously mapped 16,874 ha from a 1989 report ( http://sxbrc.org.uk/projects/revised-ancient-woodland-inventory/ ). For context, according to 2016 government figures, there is only 7,282 ha of Lowland Meadow Priority Habitat in the whole of England http://jncc.defra.gov.uk/page-5848 although I noticed that in 2013 the Natural England Technical Information Note TIN147 states there is less than 6000 ha of unimproved neutral grassland remaining in England.
In two of the replies to my various emails I was directed to this paper http://onlinelibrary.wiley.com/doi/10.1111/avsc.12076/abstract * studying 150 years of plant community re-assembly on Salisbury Plain, UK and which used the term ‘ancient calcareous grassland’. If you use a term in a scientific paper then you must be able to define that term clearly and I had noticed within the paper there was only an implicit definition. After correspondence with the lead author (the same day!) and the involvement of another author of the paper, I had an explicit definition.
‘a semi-natural plant community maintained as grassland since 1840, on a site with no history of arable management or agricultural improvement since 1840 in any of the currently available land-use datasets.’
It may be that the part after the comma is not generally used, but at this stage is worth spelling out. 1840 is used as a starting date as that is usually around the time we first get good maps across the UK mentioning the land use (via the church Tithe maps); semi-natural plant community means that it has escaped all the various ravages of fertilizer and herbicide application (agricultural improvement); treated as a grassland since then means you avoid sites with resown wildflowers or other agricultural improvement i.e. it has been continuously grassland (c.f. the Ancient Woodland definition).
Yes, our records of grasslands are imperfect; we can never be sure it wasn’t ploughed once in 1862 (as a random example) but never ploughed again. However, I think it is a good start at a working definition…and more importantly for working towards an inventory, it has been intrinsic to the results of a published, peer-reviewed scientific paper and no-one has queried it yet.
*I got sent the full paper but there is a good article on the research here http://www.nerc.ac.uk/planetearth/stories/1722/
Ancient Woodland definition from https://www.gov.uk/guidance/ancient-woodland-and-veteran-trees-protection-surveys-licences